The Dialogical Self of Migrants

Home , Dialogical self

THE DIALOGICAL SELF OF MIGRANTS:

EXPLORING THE MIGRATION EXPERIENCE THROUGH SELF-NARRATIVES

Mariel Sanchez-Rockliffe

Faculty of Health, Arts and Design

Swinburne University of Technology

Australia

A Report of an investigation submitted

as partial requirement for the degree of

Doctor of Psychology (Counselling Psychology)

December 2015

Declaration

I declare that that this thesis does not incorporate without acknowledgment any material previously submitted for a degree in any university, or other educational institution; and that to the best of my knowledge and belief it does not contain any material previously published or written by another person except where due reference is made in the text.

I further declare that the ethical principles and procedures specified by the

Psychology Department’s document on human research and experimentation have been adhered to in the preparation of this report.

Mariel Sanchez-Rockliffe

Abstract

Hermans’ theory of the dialogical self and the self-confrontation method (SCM) are used to study the effect of migration on the self. A radical change in the social environment, in this case migration, is shown to be reflected in the structure of the dialogical self. Relationships between migrants’ I-positions relating to life in their countries of origin and in Australia are explored.

The study is based on a sample of 38 migrants to Australia who perceived themselves as having two identities, one for their country of origin, the other for their new country. Participants were interviewed under the procedures of the SCM to elicit a total of 1500 valuations (narratives) relating to their I-positions. These were scored along the standard four SCM dimensions: self-enhancement (S), union-with-the-other

(O), positive affect (P) and negative affect (N). Participants also provided demographic data, and self-reports of psychological well-being and distress.

The first study employed cluster analysis to group participants into four types.

“Progressives” steadily moved towards a valued goal; “Low-stables” experienced little affective change, remaining at a lower level than “High-stables”, who likewise changed little but were at a higher level of well-being. “Regressives” were the opposite of

“Progressive”. The types were then characterised in terms of self-confrontation indices, demographics and psychological self-reports, and their differences identified. In the second study participants were classified according to the nature of their valuations and a transition matrix constructed linking the major themes of their valuations for country of origin and for Australia. The transition matrix was statistically tested for the null hypothesis that migration had no effect on I-position. The dialogical self was treated as a hypothetical probability distribution of themes of valuations whereby the SCM is a technique for generating a sample of valuations. The properties of the underlying

iii distribution could then be inferred, as in standard sampling theory. The third study was idiographic, analysing case studies representing each of the participant types identified by the first study.

The main finding is the overall stability of the thematic content of valuations.

Where differences occur, they are intelligible. For example, many respondents seemed to be responding positively to the peaceful, liberal social conditions they find in

Australia. The research corroborates previous research that identifies two modes of biculturalism: compatible and oppositional. Low-stable and High-stable migrants viewed their heritage and host cultures as complementary, whereas Progressive and

Regressive migrants viewed them as discrepant and conflicting. Many migrants had I- positions with unambiguously different thematic content. This is evidence for Hermans’ dialogical view of the self and against the hypothesis of an underlying integrated self.

Most migrants showed self-enhancement (S) in their Australian than in their country of origin I-positions, whereas union-with-the-other (O) did not differ significantly, suggesting that migration accommodates self-enhancement (S) proportionately more than union-with-the-other (O). Migrants varied in their psychological states in a plausible and consistent way. In general, Progressive and High-stable migrants were lowest on distress and highest on well-being; Regressive migrants were the opposite; and the Low-stable migrants were intermediate. Lastly, good matches were found between the case studies and the participant types they represented.

Acknowledgments

It is pleasure and deep gratitude that I acknowledge the help, support and advice extended to me in completing this thesis.

I am indebted to Swinburne University for the opportunity to undertake it, and for their financial support and encouragement in attending international conferences pertinent to my research.

To my supervisor, Glen Bates, I extend my deepest thanks for his guidance and support. Without his patience and understanding this thesis would surely never have been completed.

To my friend and mentor, James Symons, I extend my thanks for his untiring efforts and advice in developing and testing hypotheses relating to the thesis. To my husband, I extend my gratitude for his assistance in data management and editorial matters. And to both I offer my sincere appreciation for their keen critical insights, which were all the more valuable as they were drawn from disciplines outside my own.

Lastly, I extend my thanks to all those—friends, family and colleagues too numerous to mention—who have stood by me.

Table of Contents

Declaration i

Abstract ii

Acknowledgments iv

Table of Contents v

List of Tables vii

List of Figures ix

CHAPTER 1: Introduction and Overview 1

CHAPTER 2: Self and Identity 3 2.1 James’ Theory of the Self 3 2.2 Symbolic Interactionism 6 2.3 Identity 11 2.4 The Dialogical Self 24 2.5 Summary 34

CHAPTER 3: Migration and Acculturation 36 3.1 Australian Migration 36 3.2 Acculturation 44 3.3 Acculturation Outcomes 58 3.4 Individualism versus Collectivism 67 3.5 Summary 73

CHAPTER 4: The Dialogical Self and Acculturation 75 4.1 Globalisation and the Illusion of Stability 75 4.2 A Dialogical Theory of Acculturation 82 4.3 Cultural Continuity 87 4.4 Research Aims and Questions 94

CHAPTER 5: Research Methodology 99 5.1 Participants 99 5.2 Self-report Measures of Psychological Distress and Well-being 104 5.3 Self-confrontation Method (SCM) 114 5.4 Procedure 118 5.5 Treatment of the Interview Data 121 5.6 Comparison with Previous Studies 127 5.7 Personal Reflections on the Experience of Migration 130

CHAPTER 6: Study 1 – Identification and Comparison of Migrant Groups 135 6.1 Preliminary Analysis 135 6.2 Distinguishing I-positions 143 6.3 Analyses of Participant Type 157 6.4 Comparison of Participant Types 170

6.5 Summary 177

CHAPTER 7: Study 2 – The Dialogical Self as a Probability Distribution 180 7.1 The Dialogical Self as a Probability Distribution of Themes 180 7.2 Hypotheses 184 7.3 Analysis Based on Major Themes 185 7.4 Analysis Based on Self-confrontation Indices 195 7.5 Summary 199

CHAPTER 8: Study 3 – Case Studies 200 8.1 Euclidean distance 200 8.2 Case study: The ‘Progressive’ Migrant (Type 1) 204 8.3 Case study: The “Low-stable” Migrant (Type 2) 211 8.4 Case study: The “High-stable” Migrant (Type 3) 218 8.5 Case study: The “Regressive” Migrant (Type 4) 227 8.6 Summary 233

CHAPTER 9: General Discussion 236 9.1 Summary of the Empirical Work 236 9.2 Main Findings 239 9.3 Directions for Future Research 265 9.4 Conclusions 271

Glossary 273

References 275

vii

List of Tables

Table 3.1 Migrant stock: selected countries of large-scale immigration 39 Table 5.1 Characteristics of the sample, by gender 101 Table 5.2 Means and standard deviations for Years in Australia, Age and Individualism, by gender 102 Table 5.3 Number of participants, by Hofstede’s individualism index 103 Table 5.4 Dimensions of Ryff’s Psychological Well-being Scale (PWB) 110 Table 5.5 Questions of the Self-confrontation method 116 Table 5.6 Composition of indices in the SCM 122 Table 5.7 Composition of major themes 125 Table 5.8 Empirical studies using the Self Confrontation Method 128 Table 5.9 Demographic difference between minority and majority cultures for selected studies 128 Table 6.1 Means and standard deviations for the self-report measures 136 Table 6.2 Correlations among self-report measures 137 Table 6.3 Correlations among self-confrontation indices 139 Table 6.4 Correlations among self-reports and self-confrontation indices 141 Table 6.5 Means and standard deviations of self-confrontation indices 144 Table 6.6 Number of participants by cluster: country of origin 149 Table 6.7 Mean self-confrontation indices for country of origin I-position, by cluster in country of origin 151 Table 6.8 Number of participants by cluster: Australia 154 Table 6. 9 Mean self-confrontation indices for Australian I-position, by cluster in Australia 155 Table 6.10 Number of participants, by cluster in country of origin and in Australia 157 Table 6.11 Mean self-confrontation indices for the I-position, by location and participant type 159 Table 6.12 Tabulation of demographic characteristics, by participant type 171 Table 6.13 Mean and standard deviation of demographic characteristics, by participant type 172 Table 6.14 Mean psychological distress, by participant type 173 Table 6.15 Mean psychological well-being, by participant type 174 Table 6.16 Mean self-confrontation indices for country of origin I-position, by participant type 175 Table 6.17 Mean self-confrontation indices for the Australian I-position, by

viii

participant type 176 Table 7.1 Rules for assigning valuations to major themes 186 Table 7.2 Number of valuations, by participant type and major theme 187 Table 7.3 Thematic profiles showing “important” major themes, by participant and location 190 Table 7.4 Average differences in proportions of major themes 194 Table 7.5 Thematic transitions 197 Table 7.6 Average differences in thematic scores (Australia minus country of origin) 198 Table 8.1 Valuations with major themes, by location and period: Yoana 210 Table 8.2 Valuations with major themes, by location and period: Juan 217 Table 8.3 Valuations with major themes, by location and period: Luisa 225 Table 8.4 Valuations with major themes, by location and period: Carla 234

List of Figures

Figure 6.1 Agglomeration schedule: cluster in country of origin 147 Figure 6.2 Dendrogram of 2-cluster solution: cluster in country of origin 148 Figure 6.3 Sum of squares from centroid: cluster in country of origin 150 Figure 6.4 Mean self-confrontation indices for country of origin I-position, by cluster in country of origin 151 Figure 6.5 Agglomeration schedule: cluster in Australia 152 Figure 6.6 Dendrogram of 2-cluster solution: cluster in Australia 153 Figure 6.7 Sum of squares from centroid: cluster in Australia 154 Figure 6.8 Mean self-confrontation indices for Australian I-position, by cluster in Australia 156 Figure 6.9 Mean self-enhancement for country of origin and Australian I- position, by participant type 161 Figure 6.10 Mean union-with-the-other for country of origin and Australian I-position, by participant type 162 Figure 6.11 Mean positive affect for country of origin and Australian I- position, by participant type 163 Figure 6.12 Mean negative affect for country of origin and Australian I- position, by participant type 164 Figure 6.13 Differences in self-confrontation indices for country of origin and Australian I-positions, by participant type 165 Figure 6.14 Mean self-confrontation indices, by participant type 167 Figure 7.1 Hypothetical probability distribution of valuation themes: oblique view and plan view (inset) 182 Figure 8.1 Euclidean distance between profiles for country of origin and Australia based on mean self-confrontation indices of valuations 202 Figure 8.2 Euclidean distance between profiles for country of origin and Australia based on proportions of major themes of valuations 203 Figure 8.3 Proportion of valuations, by major theme and location: Progressive migrants and Yoana 206 Figure 8.4 Mean indices of valuations, by location: Yoana 207 Figure 8.5 Mean indices of valuations, by location and period: Yoana 208 Figure 8.6 Proportion of valuations, by major theme and location: Type 2 participants and Juan 213 Figure 8.7 Mean indices of valuations, by location: Juan 214 Figure 8.8 Mean indices of valuations, by location and period: Juan 215 Figure 8.9 Proportion of valuations, by major theme and location: Type 3 participants and Luisa 220

Figure 8.10 Mean indices of valuations, by location: Luisa 221 Figure 8.11 Mean indices of valuations, by location and period: Luisa 222 Figure 8.12 Proportion of valuations, by major theme and location: Type 4 participants and Carla 229 Figure 8.13 Mean self-confrontation indices, by location: Carla 230 Figure 8.14 Proportion of valuations, by location and period: Carla 231

CHAPTER 1: Introduction and Overview

The overarching aim of the thesis is to strengthen the empirical evidence base of the theory of the dialogical self. To this end, it explores the effect of migration on the self within the theoretical context of Hubert Hermans’ theory of the dialogical self. It does so by means of Hermans’ self-confrontation method (SCM), an investigative tool widely used to access and explore the dialogical self. It shows how a radical change in the social environment—in this case migration to a new country—is reflected in the repertoire of I-positions of migrants. In so doing, it develops the nomothetic aspect of the SCM by identifying significant associations between migrants’ I-positions relating to life in their countries of origin, and their I-positions relating to life in Australia.

The thesis is structured as follows. Chapters 2, 3 and 4 review the relevant literature. Chapter 2 (Self and Identity) explores the psychological concept of the self.

This is central to the thesis as it is what is accessed and measured by the SCM. The thesis discusses current theories, including the theory of the dialogical self, and explores the proposition that people possess multiple I-positions. Particular attention is paid to previous studies of migrants.

Chapter 3 (Migration and acculturation) describes psychological theories of how and why people migrate, and what happens to them at a psychological level when they do. It explores the proposition that some migrants preserve their former cultural identity while taking on a new one for the their country of settlement; it is “integrated” migrants such as these who were the subject of the current research.

Chapter 4 (The Dialogical Self and Migration) brings together the previous two chapters — one on theories of the self, the other on theories of migration and acculturation. It shows how the theory of the dialogical self applies to migration and the acculturative stress that it entails. From this synthesis, research questions are developed.

Chapter 5 (Method) explains how the data were gathered from guided interviews of 38 migrants. The data collection process is described and the analytical procedures set out. The data were analysed in two ways, termed Studies 1. 2 and 3, that differed in approach.

Chapter 6, 7 and 8 present the results of three separate studies using the same data set. Study 1 uses cluster analysis as a data reduction technique to group participants into four types, and analyses their characteristics. Study 2 classifies participants into groups on the basis of their “valuations” (to use Hermans’ terminology, which is defined later).

By treating the dialogical self as a hypothetical probability distribution of themes of valuations, it permits formal hypothesis testing of transition matrices linking the participants’ group in country of origin, with their group in Australia. Study 3 presents four case-studies identified in the course of the preceding studies, one to exemplify each of the four migrant types identified under Study 1.

Lastly Chapter 9 discusses the results of all three studies. In terms of external validity, the thesis had the advantage of being based on a non-clinical respondent sample; but its conclusions are limited to migrants of a particular kind (“integrated”), and particular countries of origin and settlement. More work would be needed to extend the validity of its conclusions. In terms of method, in Study 2 the thesis introduced a novel way of characterising the dialogical self: as a frequency distribution of valuations.

A frequency distribution of this kind does not capture the full richness of the dialogical self nor the I-positions that compose it, but it does offer powerful insights for exploring the changes in the dialogical self that may be of value to future researchers, particularly in the area of formal hypothesis testing.

CHAPTER 2: Self and Identity

This chapter identifies the antecedents of the theory of the dialogical self and traces its evolution. It begins with William James’ theory of the self, since it was James who arguably set the agenda for future social psychological inquiry about self and identity (Suls, Tesser & Felson, 2000). Next, it discusses contemporary issues of self and identity: the collective self, multiple selves versus a unified self, and identity— personal, social, and bicultural. The concept of the dialogical self is then introduced.

Dialogical self theory, it is argued, is a powerful conceptual model for understanding the process of acculturation in that it addresses weaknesses of previous acculturation models, principally that of Berry (1990).

2.1 James’ Theory of the Self

James (1890) was not the first thinker to explore the nature of personal identity; the Ancient Greek philosophers, notably Heraclitus, had pondered the notion. But James was one of the first to articulate a distinction between the I (or self-as-subject) and the

Me (or the self-as-object)—a notion that came to be known as the duality of the self. For

James, the I was the knower, and the Me was the known, being the empirical aggregate of everything objectively known about the self. James viewed the I as the subjective self because it organised and interpreted one’s experiences.

According to Westen (1995), it is in James’ I that the sense of self resides; and it is in James’ I that individuals experience themselves as the thinkers of their thoughts, feelers of their feelings, and agents of their actions. Harter (1996) notes that, for James, the “I-self” also demonstrated more specific types of awareness, such as a concern with one’s personal continuity through time, and an awareness of the uniqueness of one’s life experiences leading to the distinctiveness of oneself as a person. Although it was

4 essential for James to posit a knowing “I-self” as the active agent responsible for constructing the “Me-self”, it is the Me that came to be labelled the self-concept and that has received most attention in the field of self psychology (Harter, 1988; 1996).

James (1890) used the term “empirical self” to refer to the ways in which people think about themselves:

The Empirical Self of each of us is all that he is tempted to call by the name of me. But it is clear that between what a man calls me and what he simply calls mine the line is difficult to draw. We feel and act about certain things that are ours very much as we feel and act about ourselves. Our fame, our children, the work of our hands, may be as dear to us as our bodies are, and arouse the same feelings and the same acts of reprisal if attacked. And our bodies themselves, are they simply ours, or are they us? (p. 291)

James’ (1890) definition of the “empirical self”—or “Me”—is broad:

In its widest possible sense, … a man’s Self is the sum total of all that he CAN call his, not only his body and his psychic powers, but his clothes and his house, his wife and children, his ancestors and friends, his reputation and works, his lands and horses, and yacht and bank-account. All these things give him the same emotions. If they wax and prosper, he feels triumphant; if they dwindle and die away, he feels cast down—not necessarily in the same degree for each thing, but in much the same way for all. (pp. 291–292, emphasis original)

This definition is broad enough to encompass most aspects of personhood

(Brown, 1998).

James subdivided his Me into three: the spiritual Me (the inner self or the psychological self) at the top, the material Me (tangible objects, people, or places designated as “my” or “mine”) at the bottom, and the social Me in the middle. It is

James’ notion of the social Me that appears to be the forerunner of most psychological research and theories concerning personal identity. In James’ conception, the social Me

5 consists of the recognition a person receives from others, and hence there exists a different social Me for each group whose opinion one is concerned about:

… a man has as many social selves as there are individuals who recognise him and carry an image of him in their mind… But as the individuals who carry the images fall naturally into classes, we may practically say that he has as many different social selves as there are distinct groups of persons about whose opinion he cares. He generally shows a different side of himself to each of these different groups. (p. 294)

It is through the recognition of others that what he calls “the different sides of oneself” begin to develop. Thus, by having different people or groups respond to one in different ways, one begins to acquire various social roles—that is, distinctive patterns of behaviour that are displayed in, and because of, different social situations.

For James, each of these social selves (for instance, “mother” versus “student”) has its own set of expectations and behaviours; which are sometimes at variance and occasionally in conflict due to competing demands from others.

Many a youth who is demure enough before his parents and teachers, swears and swaggers like a pirate among his ‘tough’ young friends. We do not show ourselves to our children as to our club-companions, to our customers as to the laborers we employ, to our own masters and employers as to our intimate friends. From this there results what practically is a division of the man into several selves; and this may be a discordant splitting, as where one is afraid to let one set of acquaintances know him as he is elsewhere; or it may be a perfectly harmonious division of labor, as where one tender to his children is stern to the soldiers or prisoners under his command. (p. 294)

For James’, how people think of themselves largely depends on their social roles

(Roberts & Donahue, 1994). Difficulties arise, however, where roles clash (Brown,

1998). People furthermore seek recognition; they affiliate, he argued, not just for the company but to seek recognition and status, otherwise known as social climbing.

A man’s Social Self is the recognition which he gets from his mates. We are not only gregarious animals, liking to be in sight of our fellows, but we have an innate propensity to get ourselves noticed, and noticed favorably, by our kind. No more fiendish punishment could be devised, were such a thing physically possible, than that one should be turned loose in society and remain absolutely unnoticed by all the members thereof. (p. 293)

These social selves have come to be referred to as social identities. Deaux, Reid,

Mizrahi, and Ethier (1995) distinguished five types: personal relationships (husband, wife etc.), ethnic or religious affiliation, political affiliation, stigmatised group, and vocation. Some, one is born with, others are attained. So for example, a person might in this way see herself as “mother”, “Hispanic”, “Catholic”, “left-wing” and a “teacher”.

2.2 Symbolic Interactionism

Symbolic interactionism is a conceptual model of the self-concept first conceived by Cooley (1902) and Mead (1934). It is concerned with the process of socialisation, that is, how culture is acquired and perpetuated, and how people come to adopt the values, standards, and norms of the society into which they are born. It posits that the self-concept, being a product of social interaction with others, is in constant flux. For both Cooley and Mead the self is fundamentally social in origin. Mead held that social interaction creates an individual’s sense of self. People learn to take the role of others and in this way learn who they are and how they should behave. Cooley, with his notion of the “looking glass self”, held that people obtain a sense of who they are by how others perceive them. The implication of their theories for the present thesis is that since society itself is not constant (as is the case for migrants), the change of environment creates a new self, which arises as a result of the new social commerce in the new society; migrants experience a social rebirth in fact.

Cooley (1902), who gave a sociological perspective to the self, acknowledged that he was strongly influenced by William James. He developed a conceptual framework for dealing with the relationship between self and society, which he considered to be aspects of the same whole. In explaining the origin of the self, he said, “self and society are twin born … and the notion of a separate and independent ego is an illusion” (p. 8, cited in Breakwell, 1983). The mind is a social construct, and society a mental construct, and the relationship between them is best understood by the metaphor of a

“looking glass self”, by which he meant that what one internalises as one’s own is based on information about oneself that one receives from others (Weinreich, 2003). In other words, people see themselves as others see them because they learn about themselves from others, who act as mirrors in which a person sees himself or herself.

James’ idea of the formative effect of the responses of others on the development of a social self was pursued by social psychologist George Mead, who in a development of Cooley’s theory elaborated on James’ social self and produced a more extensive theory of its development (Mead, 1934). Mead differed markedly from James in his conception of the social self: for James it was one constituent of the self; for Mead it was the entire self, for the self is wholly a social construct, and is created particularly through the use of language (Breakwell, 1983). According to Mead (1934, p 135):

the self is something which has a development; it is not initially there, at birth, but arises in the process of social experience and activity, that is, develops in the given individual as a result of his relations to that process as a whole and to other individuals within that process.

Mead (1934) conceptualised the self “as that which can be an object to itself, is essentially a social structure, and it arises in social experience” (p. 140). The person gets to know himself only through others; that is, he cannot experience himself directly as such, “but only indirectly, from the particular standpoints of other individual members

8 of the same social group, or from the generalised standpoint of the social group as a whole to which he belongs” (p. 138). In this way Mead affirmed the primacy of the social context—particularly the family—in which an individual is born and socialised that defines a station in life in terms of kinship relationships, race and class. For Mead the self-concept arises in social interaction. It is in the course of this socialisation that people learn to take the “perspective of the other”, to place oneself in the other’s shoes.

According to Mead, the self is defined through linguistic exchanges (that is, symbolic interactions) within communities. Initially people take on the roles of the significant others in their immediate world: parents, playmates, and so on. Over time, they extend their framework to take on the role of the generalised other—their culture. This abstract and generalised other represents the broader society and culture into which they are born. Socialisation is conceived as role learning; it forms a person’s identity as a consequence of social interaction. The acquisition of roles results in the development of self-conception. In time, the person is able to view the self, as it were, from outside the self—as mirrored reflections of others. People learn how their culture interprets gestures, and in this way they develop a self-concept: they see themselves through a cultural lens. Through the learning of a culture (that is, a sort of symbolic “language”) one can both predict the behaviour of others and anticipate the predictions that others make of one’s own behaviour (Mead (1934); Breakwell, 1983). The opposite is also true: migrants and others, who are immersed in a foreign culture, are unable to gauge others’ reactions, and hence may become unsure of who they really are.

Mead (1934) conceptualised his “looking glass self” as a consensus about the self’s typical characteristics as reflected by the “generalised other” that represents an amalgamation of others who view the self’s enduring ways. The Meadian “self” is firmly situated within a nexus of others, without which there would be no notion of

“self”. Identities are given by the roles people occupy as defined by society—one’s identity as “mother”, “student”, “migrant”, and so on. In Mead’s symbolic interactionist view of the self, identities are not self-contained but are the products of social interactions (Weinreich, 2003). Thus James’ (1890) argument that the individual has “as many social selves as there are individuals who recognise him” (p. 178) is further elaborated in Mead’s view that multiple roles are a normal consequence of life in a complex social world.

Mead’s focus was primarily on the Me, or self as object. With his “generalised other”, Mead claimed to have solved the problem of how one can be an object to oneself

(Burns, 1979). He differentiated between the self as known (the Me) and the self as knower (the I) by the context in which each operates. The Me operates in the context of group membership, status, roles, and interactions with others; the I operates in the context of asserting individuality:

The simplest way of handling the problem would be in terms of memory. I talk to myself, and I remember what I said and perhaps the emotional content that went with it. The “I” of this moment is present in the “Me” of the next moment. There again I cannot turn around quick enough to catch myself. (Strauss, 1956, p.242).

So as Mead’s I moves along, it leaves Me in its wake and provides for the self a way of thinking about the self—self-knowledge. Burns (1979) noted that Mead’s “I-

Me” dichotomy differs in a major way from James’ initial formulation: Mead’s I is the impulsive tendency, the undisciplined, disorganised potential of the individual (almost parallel to the Freudian id). Every behaviour commences as an I but evolves into a Me as it comes under the influence of society’s constraints. In Mead’s analysis, the Me is essentially a social construction, eminently compatible with James’ social self.

While James, Cooley, and Mead often refer to the self in the singular, its lived presence in the world of everyday life needs to be plural (Holstein & Gubrium, 2000).

James called attention to the socially manifold self when he stated: “Properly speaking, a man has as many social selves as there are individuals who recognise him and carry an image of him in their mind” (emphasis in the original, p.189-190).

Mead (1934) saw the self as developing out of social interactions and, as in

James’ theory of multiple social selves, allows for the existence of several selves within one person: “… we divide ourselves up in all sorts of different selves with reference to our acquaintances. We discuss politics with one and religion with another. There are all sorts of different selves answering to all sorts of different social situations” (p. 140).

As Holstein and Gubrium (2000) asserted, James’, Cooley’s, and Mead’s selves were plural because their attention was focused on the varied “we’s”, “us’s”, and

“them’s” that people are to one another as they go about their daily lives. The social self was always the “self-at-hand, the socially operative sense of who we are to ourselves and to others” (p. 20). Both Cooley and Mead stress the role that significant others play in shaping the self-concept through social interactions. Here the individual considers not only feedback from others, but the process through which more generalised attitudes towards the self are adopted. Cooley moreover hints at a developmental process of internalisation whereby the reflected appraisals of others are incorporated in the form of enduring attitudes about the self—a process with implications for the stability of the self-concept. Given that the self emerges in the context of a complex, organised, differentiated society, it too must be complex, organised, and differentiated, reflecting

Stryker’s (1980) view that “self reflects society”, and James’ (1890) notion that an individual has as many selves as social roles . As Stryker (1980) noted, a person has “an internalised positional designation” (p.60), for each role the person holds in society.

To summarise, this section has reviewed some important antecedents of many current theories of the self that hold that individuals’ self-concepts derive meaning

11 through relationships with other people and groups. The next section examines contemporary issues that have emerged from these historical roots and are relevant to the present thesis’ emphasis on understanding the philosophical and theoretical underpinnings of dialogical self theory.

2.3 Identity

Personal versus social identity: one self or many? The core tenet of the theory of the dialogical self is that individuals have within them not one self but many

(Hermans & Kempen, 1993). It is therefore appropriate to investigate where this notion arose, and how well-founded it is. This section outlines prominent psychological and sociological theories that embody this idea.

It was James (1890) who first drew the distinction between the spiritual self

(being one’s inner or subjective being and dispositions) and the social self (being the recognition one receives from others). Since then, psychologists and sociologists have long debated whether the personal (internal) or social (external) aspects of identity are more important in understanding human behaviour. Tappan (1998) pointed out that much of the theoretical and empirical work on the self has been conducted from two perspectives: psychological and sociological. The multiplicity of identity, or senses of self, was proposed early in the twentieth century. As noted earlier, James (1890) observed that there may be incompatibility between certain role-related selves leading to the “conflict of the different Me’s”.

By the mid-twentieth century, however, the consensus of opinion had reverted to a unitary conception of the self. In the explicitly psychological perspective (Damon &

Hart, 1988), theorists such as Jung (1957) and Maslow (1961) asserted that a person is most truly his or herself and closest to his or her authentic identity when experiencing a deep sense of personal uniqueness. Allport, Rogers, and Maslow, for example, spoke of

12 the self in the singular; they accepted that a person might exhibit subsidiary social selves or roles, but saw them as less real, central, or authentic (Martindale, 1980). And although Jung (1956) and Assagioli (1976) accepted that people were indeed possessed of multiple subselves or subpersonalities, they argued that these are (or could be) integrated in a deeper-level unitary self. In such a paradigm, identity was simply a personal characteristic, and its development a result of internal cognitive processes.

Debate over the unity of the self is not new. As Martindale (1980) observed, it has concerned scholars for centuries. Does a person have a single dominant and enduring self, or a multiplicity of subselves that compete for control? In tracing the dominant viewpoints about the self through time, he noted that in the early nineteenth century the theoretical consensus, exemplified by theorists such as Kant and Bain, favoured a unitary self or transcendental ego. But by the end of the century opinion had swung the other way: psychiatrists and psychoanalysts explicitly denied the existence of a superordinate self and instead postulated the existence of multiple selves. The phenomenon of multiple personality was adduced as evidence; supposedly, in the normal personality these different selves were elicited appropriately as circumstances required, but in the abnormal personality one or more attained an unwonted autonomy

(Martindale, 1980). Other social psychologists and sociologists defined identity almost exclusively in social terms: “Identity is socially bestowed, socially sustained, and socially transformed” (Berger, 1963, p.98). They focused mainly on the social component of identity because “the shaping of one’s identity is dependent on the valuations (sanctions, reinforcements) placed on one’s public conduct by relevant others” (Sarbin & Allen, 1968, p.550).

Sociological role theorists such as Goffman (1959) took the distinction between components of identity to its logical extreme. They maintained that people adopt short-

13 term selves, donning them as roles and shedding them as easily. People are actors in a drama where, for a time, the role dictates the appropriate actions. Goffman did not explain who the actors are behind the roles they play, nor who they are when no role is available, for according to Goffman they always have roles to perform. As Breakwell

(1983) noted, one could conclude from Goffman that the individual is nothing but an agglomeration of roles, and that the self is always socially constructed. That individuals present differently in different social settings casts doubt on the stability of a self that transcends social roles. Gergen (1968) argued as much: “the popular notion of the self- concept as unified, consistent, or perceptually whole psychological structure is possibly ill-conceived” (p. 306). In his experiments he showed that individuals do indeed modify their self-presentation, and also the way they see and experience themselves, with different people in different situations; and he concluded: “The individual has many potential selves. He carries with him the capacity to define himself as warm or cold, dominant or submissive, sexy or plain. The social conditions around him help determine which of these options are evoked.” (Gergen,1972, p.66). By maintaining that the self consists solely of social roles, theorists such as Gergen (1982, also Shotter & Gergen,

1989), deny the existence of a separate self; they see the self and identity as exclusively social and lacking personal agency or authenticity. Other theorists find this position extreme, however, and adhere to James’ position that a person’s social identities share an enduring sense of self (Brown, 1998).

Further evidence for the multiplicity of the self was provided by the Twenty

Statements Test (TST) developed by Kuhn and McPartland (1954). Linking identity to the question “Who am I?”, they found that responses are typically couched in terms of social positions or social roles that respondents perform (or aspire to), that is, they name their “masks” or personas (defined as that aspect of one’s personality that one displays

14 to others). Since a person occupies multiple social positions and performs multiple social roles, it follows that he or she would have multiple social identities. Indeed, Kuhn and McPartland further point out that when asked to give twenty separate answers to the question “Who am I?”, nearly all participants named masks or personas, and many named up to twenty. Gecas (1973, 1982; Gecas & Schwalbe, 1983) found that research using the TST has focused on socialisation into the identity (e.g., gender-roles), evaluations of the identity (e.g., racial and ethnic), and role transitions (e.g., adjustment to widowhood, identity loss in the family).

Social roles also feature in research by Rosenberg and Gara (1985), who argued that personal identity consists of a person’s various social and “personal” roles—kinship roles, occupational roles, religious affiliation, group membership, intellectual concerns, and so on. Thus, a person could, for example, have an identity consisting of such roles as mother, stockbroker, Catholic, university student, avid reader, and so on. Each role, or “identity” as Rosenberg and Gara call them, is “an amalgam of features—personal characteristics, feelings, values, intentions, and images—experienced by the individual”

(p. 90) that can be enacted in different situations. They conceptualised relations between identities in various ways, including that of subset-superset and disjunction. The subset- superset relation occurs when one identity includes all the features of another identity; thus someone’s identity as a mother might include all the features of her identity as a caregiver. The disjunction relation occurs when two or more identities have nothing in common, as where the identity of a devout church-goer might share no features with the identity of an aggressive stockbroker. Although their main concern is with the social roles played by the individual, these roles are defined in terms of feelings, values, and so on, experienced by the individual rather than in terms of social interaction. This resembles James’ (1890) view that the social self is in the mind, the major difference

15 being that for Rosenberg and Gara the social self is in the mind of the person playing the role, while James’ social self is in the minds of those who recognise the person.

In another insight, Markus and Kunda (1986) noted that “the self is not a unitary structure or even a generalised average of images and cognitions. Instead, the self- concept encompasses … a wide variety of self-conceptions—the good selves, the bad selves, the hoped-for selves, the feared selves, the not-me selves, the ideal selves, the possible selves, the ought selves” (p. 858). They reason that at any given moment people function according to a working self-concept: the most currently available view of ourselves.

In his more recent discussion of the postmodern era, Gergen (1991) observed that in the modern world, people confront “a swirling sea of multiple social relationships”.

They adapt by constructing multiple selves, but sometimes at a cost to themselves, for their different selves may not necessarily harmonise. Their enduring, core self is compromised by playing the “social chameleon”. The postmodern identity, dubbed

“multiphrenia” by Gergen (1991), is “the splitting of the individual into a multiplicity of self-investments” (p. 73–74). He claimed that “In place of the enduring and identifiable self, we find fragmentation and incoherence… the postmodern sensibility questions the concept of ‘true’ or ‘basic’ self, and the concomitant need for personal coherence or consistency” (p. 172–178).

Lifton’s (1993) postmodern “protean self” (named after Proteus, the shape- shifting Greek god), develops the same theme. The protean self emerges out of

“confusion, from the widespread feeling that we are losing our psychological moorings”

(p. 1); “the older version of personal identity, at least insofar as it suggests inner stability and sameness, was derived from a version of tradition culture in which

16 relationships to symbols and institutions are still relatively intact—hardly the case in the last years of the twentieth century” (p. 4).

In summary, this section outlines prominent theories that hold that within every individual there is not a single self but many. This notion is a central to the concept of the dialogical self, which holds that not only are individuals composed of multiple selves, but those selves are in constant dialogue (Hermans & Kempen, 1993). The attraction and strength of dialogical self theory is that it integrates these multiple strands of thought into a single and unifying theoretical framework that is amenable to empirical testing, conceptual refinement, and widespread application. It is furthermore for this reason that dialogical self theory underpins the exploratory tools used in this thesis to explore the acculturative aspects of migration.

Identity as memory. The previous section discussed the multiplicity of the self; this section broaches another key strand—identity as continuity of memory. Both are core aspects of the theory of the dialogical self, since the dialogical self is accessed through narratives, which themselves are founded in memory. As stated before, the singularity and unity of personal identity has long held a central position in Western philosophical and psychological thought (Tafarodi, Lo, Yamaguchi, Lee, & Katsura,

2004). James (1890) challenged this orthodoxy by arguing for a plurality of selves:

“Each of us, when he awakens says, here’s the same old self again, just as he says, here’s the same old bed, the same old room, the same old world” (p. 334).

While Tafarodi et al. (2004) agree that James’ “same old self” is the familiarity of some “formless, elusive I”, they recognise that it is also much more. It lies in the perceived continuity of the traits, intentions, beliefs, attitudes and emotions that people use to define themselves in society (Tafarodi et al., 2004). So, although people conduct

17 themselves differently according to the social demands of the circumstances, they do not lose their essential sense of self.

People are not static. Just as James’ material Me grows old, so his social Me evolves: roles are added and moulded, and old ones discarded (James, 1890). One can assume that James’ spiritual Me changes as well. Over a lifetime, hardly a single aspect of the person remains the same. Nevertheless, most people have an unmistakable sense of continuity. James (1890) was one of the first to consider the sense of personal continuity and the sense of distinctness from others as constructs of identity. For James

(1890), the sense of personal identity was most closely associated with the I or the “self- as-knower”. A stable self-identity derives from a sense of continuity of the self-as- knower. According to James, “each of us spontaneously considers that by the “I” he means something always the same” (1961, p.63). James believes that “the worst alterations of the self” (1948, p.207) are associated with disruptions of the sense of personal continuity. For James, this feeling of individuality derives from the subjective nature of the self-as-knower: “Other men’s experiences, no matter how much I may know about them, never bear this vivid, this peculiar brand” (1961, p.71). It is this special feeling associated with one’s own experience which is the basis for one’s sense of personal uniqueness. But according to James the identity of self-consciousness across time is “only a loosely construed thing”: James gives the example of adults failing to recognise photos of themselves as children. In this case, the disjunction is not only due to a lack of memory for “what it was like to be” (Nagel, cited in Tafarodi et al., 2004) but also to dissimilarity in physical appearance. People furthermore may change in other ways and so revise aspects of their inner selves, and accordingly revise specific beliefs about who and what they are (Bem, 1972).

Giles (1997) commented that the way in which James (1890) developed his argument about personal identity as memory was by noting that everyone is familiar with an aspect of existence—a central nucleus or core—that seems peculiarly one’s own. The I is the stream of consciousness that creates one’s knowing of the Me, which is the aggregate of things known about the self. James’ I is fundamentally a stream of thoughts in which each thought can “remember” the one before, and it is this that lends a sense of continuity across time. James’ solution to understanding personal identity is that personal identity is to be found in treating the spiritual self in a concrete way—that is, treating it as a seamless flow of uninterrupted perceptions. Each perception flows by and they are joined because they are part of the same stream of consciousness. It is this consciousness that provides a feeling of continuity of the self. Even if there are gaps in this continuity (due to sleep or otherwise) there is a sense in which consciousness before the gap “goes together with” consciousness after the gap. For James, consciousness coheres over time, and mainly through the effects of memory (Apter, 1983).

Moreover, according to James (1890), it is the feeling associated with each perception that provides the tie that binds them. Each perception carries a distinctive feeling that people recognise as theirs and theirs alone:

Each thought, out of a multitude of other thoughts of which it may think, is able to distinguish those which belong to its own [self] from those which do not. The former have a warmth and intimacy about them of which the later are completely devoid … (p. 330).

So, for James, personal identity is one’s uninterrupted memory for prior perceptions and the affect that accompanied them. Giles (1997):

A uniform feeling of “warmth”…pervades [our various selves] and this is what gives them a generic unity, and makes them the same kind…where the

resemblance and continuity are no longer felt, the sense of personal identity goes too. (p. 335).

For James (1890), it is not only the continuity of consciousness but also the continuity of warmth and intimacy of a person’s memories that underlies personal identity. In this way, personal identity requires more than memory; it requires an ability to recapture the feelings associated with the experience:

If a man wakes up some fine day unable to recall any of his past experiences, so that he has to learn his biography afresh, or if he only recalls the facts of it in a cold abstract way … he feels, and he says that he is a changed person. (p. 336)

According to James, then, one’s sense of personal identity is underlain by one’s autobiographical memories and what it felt like to experience them. The notion of identity as a felt sense of similarity and continuity became better known through the work of Erikson (1950, 1956, 1959), who sought to establish a bridge between personal and social phenomena:

The term identity expresses a mutual relation in that it connotes both a persistent sameness within oneself (self-sameness) and a persistent sharing of some kind of essential character with others. At one time it [identity] will appear to refer to a conscious sense of individual identity; at another, to an unconscious striving for a continuity of personal character; at a third, as a criterion of the silent doings of ego-synthesis; and finally, as a maintenance of inner solidarity with a group’s ideals and identity. (p. 102)

In this way he characterised identity by two essential experiences: an enduring and unique personal existence over time, and the sharing of an essential unity at some level with others. In his definition, Erikson follows the general definition—of continuously being one and the same person—but adds a social component, namely the recognition of this sameness and continuity by others and the wider sociocultural context the person lives in (Erikson, 1950). His conception of identity invokes the

20 interplay of individual and context: a person can feel embedded in his or her context. In contrast, Erikson (1964) has discussed the feelings of “uprootedness” of migrants. They are missing that sense of embeddedness.

In conclusion, the theory of the dialogical self, which is elaborated later, inherits a long philosophical and psychological tradition that holds that identity is founded in memory—a crucial aspect of dialogical self theory and the interview technique that is employed to access it.

Identity as narratives. The previous section argued that identity is founded in memory. This section explores how memories are accessed and marshalled to create identity, and how both memory and identity are in a sense social; for memory comes alive in the act of narration, and identity is realised when the self is presented to another.

Human identities are emergent constructs based on social interactions over a lifetime. Narratives on the other hand are specific to time and place, use a particular language, and refer to a particular stock of historical conventions, beliefs and values

(Crites, 1971). Narrative theories have in the past tended to focus on textual analysis

(such as rhetorical devices and form). More recently, theories have focused on the meanings, relationships and socio-cultural context (such as motives and plots) (Gergen

& Gergen, 1986; Hermans & Hermans-Jansen, 1995; McAdams, 1985, 2014;

Polkinghorne, 1988, Hoshmand, 2005). For example, Mancuso and Sarbin (1983) focus explicitly on the integration and coherence of self. The individual is the narrator of the story, constructing plots that give meaning to important events. They distinguish between the “I” as narrator, agent, organiser, and the “me” as actor, player, performer; and “emplotment” is the process by which meaning is assigned so as to develop a unified story:

The underlying assumption in employing the narrative as a root metaphor is: Human beings think, perceive, imagine, and dream according to a narrative structure … When one probes into the ordered or patterned perceptual response, it becomes immediately apparent that a plot is imposed on the disparate inputs … Only when the raw data are placed in the context of a plot structure does the possibility exist for establishing a coherent plot. (p. 234-235).

The self-as-a-story-or-theory approach, also known as the narratology approach

(Shotter & Gergen, 1989, McAdams, 2014), has two central tenets. One is the appeal to an informal story that is a theory of self; it is the story that the self uses to explain to the self and to others why one is as one is, and why one acts as one does. For example, a person’s much rehearsed story of disadvantage and abuse that explains a proneness to being victimised. The second is the notion that certain people express a dominant view of themselves and the world in terms of their story, to which all else is subordinated; for example a story in which one is a victim of circumstances that are continually recalled as being the reason for all subsequent experiences—a “poor me” dominant story.

In the “identity is a life story” approach (McAdams, 1998), “the problem of identity is the problem of arriving at a life story that makes sense—provides unity and purpose—within a socio-historical matrix that embodies a much larger story” (p. 18.).

Thus, the individual recurrently constructs and reconstructs the story that is ultimately the answer to the existential question, “Who am I?” Furthermore, “the story provides a coherent narrative framework within which the disparate events and the various roles of a person’s life can be embedded and given meaning” (p. 19). McAdams’s narrative approach to identity focuses explicitly on the integration and coherence of self, which are key aspects of the sense of identity in Erikson’s terms. McAdams (1996) argued that people organise their life stories in the context of a personal narrative. The individual is the narrator of the story, constructing plots that give meaning to important events.

Adopting a symbolic interactionist perspective, Gergen and Gergen (1988) hold that self-identity grows out of the narratives people tell to others and to themselves.

Stories, they contend, “serve as a critical means by which we make ourselves intelligible within the social world … We use the story form to identify ourselves to others and ourselves’ (p. 35). People’s experiences of self and others, the Gergens say, is not based on “a series of discrete, endlessly juxtaposed moments, “but is instead derived from “goal-directed, coherent’ narratives in which they attempt to “understand life events as systematically related”. And moreover, they also emphasise that

“narratives of the self are not fundamentally possessions of the individual; rather they are products of social interchange: “Narratives are essentially social constructions, undergoing continuous alteration as interaction progresses” (pp. 20-21).

Gergen and Gergen (1988) proposed some of the rules of narrative form and the types of narratives that are commonly employed, maintaining that cultural conventions determine the validity of the story rather than the “absolute match between word and thing”. Narrative structure often revolves around temporal forms or recurring sequences.

These rules of narrative structure lend coherence and directionality to a life. Plots are made chronologically linear with respect of evaluative shifts over time; and within them, sub-plots and overarching themes may involve stability, progressive change, or regressive change (Gergen & Gergen, 1983). A paradox of human relationships, however, is that “functioning viably in a relationship often depends on one’s ability to show that one has always been the same, and will continue to be so, and yet, contrapuntally to show how one is continuing to improve” (Gergen & Gergen, 1983, p.266). Focusing on the protagonist’s developmental movement, Gergen and Gergen

(1988) differentiated between “progressive” narratives, where there is a positive change in self-evaluation; “regressive” narratives, where there is a negative change in self-

23 evaluation: and “stable” narratives (elaborated later), where there is no change in self- evaluation. In their schema, a life story establishes the main plot and, within it, which incidents are construed as advancing (“progressing”), retreating (“regressing”) or digressing. They proposed that progressive and regressive narratives be viewed as moving through a two-dimensional space. One dimension refers to its evaluation, positive or negative, the other its direction: “progressive”, “regressive” or “stable”. In a progressive narrative, events are linked “in such a way that one steadily progresses toward a goal” (p. 27): and in a regressive narrative one is “continuously moving away from a valued state” (p. 27). Thus Gergen and Gergen’s (1986) framework for the analysis of progressive and regressive narrative structures accounts for direction, where direction has to do with moving towards or away from the goal.

Gergen (1991) noted that in different situations people view themselves differently: that is, they make up a different story about themselves. The multiplicity of the self here is understood to follow from the proposition that individuals harbour the

“capacity for a multiplicity of narratives forms” (Gergen & Gergen, 1988, p.35). They emphasise the changes in self-identity as the person engages in different interactions in different situations. The point is that in the postmodern world, our main task is just to be different people in different situations.

Because narratives are communicative acts with a particular listener in mind, the same story requires major adaptation to appeal to different audiences, who value different things and judge narratives by different criteria. It is commonplace to tell conflicting stories in different contexts, as the different voices in Hermans’ dialogical self cannot all be satisfied with the same narrative (Hermans, 1996a). Thus, the demand for narrative consistency, highlighted by McAdams et al. (1997), clashes with the need to please different audiences with different accounts of the same actions. As has been

24 shown, some contemporary theorists challenge the view that the mature individual necessarily strives towards a consistent identity. Instead, they argue that personal sameness and continuity are a social construct, and since modern societies are constantly in flux, personal sameness is no longer a sign of maturity (Hermans,

Kempen, & Van Loon, 1992; Gergen, 1991; Lifton, 1993; Sampson, 1995).

To summarise, the literature contrasts traditional and post-modern conceptions of personal identity: the former holds that good mental health is founded on a largely coherent and harmonious life story; the latter holds that lives—even healthy lives—are messy and disorganised in narrative terms.

2.4 The Dialogical Self

Since the 1980s there has been a resurgence of interest in the self (Harter, 1988).

New models have proliferated, though researchers have sought guidance from past scholars, typically paying intellectual homage to James (1890) and symbolic interactionists such as Cooley (1903) and Mead (1934). Hermans’ (1992) theory of the dialogical self is one of the more influential models that have appeared since that time.

As previously discussed, it draws on James’ (1890) distinction between I (the “self-as- knower”) and Me (the “self-as-known”) as the two main components of the self. This section describes the building blocks of dialogical self theory, and explains how they fit together.

I-positions and voices. According to Hermans (1996a), dialogical self theory

… conceptualized the self in terms of a dynamic multiplicity of relatively autonomous I positions in an imaginal landscape. In this conception, the I has the possibility to move, as in a space, from one position to the other in accordance with changes in situation and time. The I fluctuates among different and even opposed positions and has the capacity to imaginatively endow each position with a voice so that analogical relations between positions can be established. The

voices function like interacting characters in a story, involved in a process of question and answer, agreement and disagreement. Each of them has a story to tell about his or her own experiences from his or her own stance. As different voices, these characters exchange information about their respective Mes, resulting in a complex, narratively structured self (p.33).

This quote captures the two main ways in which Hermans’ view differs from that of James. Where James saw the I “as a unifying principle that is responsible for organizing the different aspects of the Me as parts of a continuous stream of consciousness” (Hermans, 2001, p. 246), Hermans sees it as possessing “the possibility to move from one spatial position to another in accordance with changes in situation and time” (p. 248). The distinction is one of stasis versus flux: James’ I is continuous and corporeally bound—both essentially static notions; Hermans’ I possesses discontinuity as well as continuity, and combines temporal characteristics with the spatial—both essentially dynamic.

Dialogical self theory furthermore posits that multiple I-positions can be occupied by the same person in space and time (Hermans, 2001), giving rise to identities that are fluid and multiple. At the same time it accepts a self that remains continuous through space and time (Richardson, Rogers, & McCarroll, 1998). In this way it belongs to the

“multiple selves” school of thought in the sense that it echoes ideas current at the turn of the century, such as those of James, to which Hermans and colleagues (Hermans,

Kempen & Van Loon, 1992; Hermans & Kempen, 1993) pay explicit homage; but it also transcends them.

Derivatives of I-positions. The concept of the I-position has generated a number of derivative concepts, of which three—the “meta-position”, the “coalition” of I- positions, and the “third position”—are central to a dialogical understanding of acculturation.

I-positions form a “society of mind” (Hermans (2004_, p. 2), and like members of any society they can form coalitions and hierarchies. Primary I-positions can therefore be grouped in the mind of the observer into “coalitions” of I-positions. Likewise, “meta- positions” stand in a hierarchically superordinate relation to coalitions of primary I- positions in that they allow individuals to observe their own I-positions “from the outside, as an act of self-reflection” (Hermans and Hermans-Konopka, 2010; p. 9).

According to Hermans (2001)

… a meta-position has several specific features: it creates a certain distance toward the other positions; it provides an overarching view; it enables the participants to interrelate the positions as part of their personal history; it provides an opportunity for evaluating the several positions and their organization; and, finally, the direction of change and the importance of one or more positions in view of future possibilities become apparent. In summary, the meta-position contributes, more than most other positions, to the integration and unity of the repertoire. (p. 354)

In this manner, acculturating individuals are able to group their I-positions into two coalitions (among others), one comprising I-positions that relate to their country of origin, the other to Australia; and to survey each coalition independently but as a collectivity by means of a meta-position unique to each coalition. It is this integrative function of meta-positions and coalitions of I-positions that render the concepts crucial in understanding acculturation.

Lastly, “third positions” (Hermans and Hermans-Konopka, 2010) are formed from the dialogical creative interaction of I-positions. Far from being a simple amalgamation of their formative primary I-positions, third positions may be completely or partly novel, whereby ‘relationships of tension are transformed into conciliation and creation’

(Surgan & Abbey, 2011; p. 165). Surgan & Abbey and Valsiner (2002) propose various

27 mechanisms by which this takes place in the case of transcultural migrants, many of which fall under the broad category of cultural “hybridisation” (see Chapter 4).

Emotional voices. “Dialogical voices can be reasoned or emotional” (Hermans,

2010; p. 41). It is natural to suppose that the “reasoned” content of dialogical voices should matter, as reason facilitates rational dialogue. But not all dialogue is rational in the narrow sense. By colouring the dialogical relationship, emotion constrains the range of possible dialogical outcomes in terms of ethnic identification: majority culture, minority, or hybrid. Dialogue that is coloured by antagonism is unlikely to lead to resolution; conversely, shared emotional ground provides a basis for negotiating an outcome that benefits both parties; and both are reflected in mental wellbeing. In the cultural context, this may make all the difference between adaptive acculturation and maladaptive.

What actually determines this emotional content is, however, another matter.

Hermans and Hermans-Konopka (2010) and Kinnvall (2004), point to power disparities between minority and majority groups. When confronted by racism, whether outright or in its more subtle manifestations, migrants may seek to counter its impact though strategies that ascribe comforting emotional content to acts and objects—strategies such as localisation and homesteading (see Chapter 4). Mahmoud (2009), Malhi (2009) and

Verkuyten (2002) see emotion (or “feeling”, as they put it) as a mediating factor in determining ethnic identity. The relationships they describe are complex, but in essence ethnic identity is partly a personal choice, but one that is governed by social context and constrained by social norms. In dialogical terms all these models require that dialogical voices be freighted with emotion.

Dialogue. Hermans (1996a) describes dialogue as implying “an interchange between mutually influencing voices” (p. 31). The concept draws on the Russian

28 literary critic Bakhtin’s (1973) concept of the “polyphonic novel”. Hermans, Rijks and

Kempen (1993) conceive of the polyphonic novel as one “where different voices, often of a markedly different character and representing a multiplicity of relatively independent worlds, interact to create a self narrative” (p. 208). In the polyphonic novel the author adopts a number of viewpoints corresponding to different imaginary characters, who are not necessarily consistent in their attitudes, knowledge or beliefs

(Hermans, Kempen & Van Loon, 1992). According to Hermans and his colleagues, the self is best regarded as a composite of characters that exhibits the variety one might find in such a novel. Some—the I-positions—correspond to different authorial viewpoints; others—the Me-positions—are the characters. Like the characters in such a novel, they engage in dialogue, hence the term “dialogical self”. This “multivoiced” self contrasts sharply with the conventional view of the self as organised around a single centre or core. As author of its own life story, the I moves from one I-position to another, adopting different perspectives. The self emerges as the totality of such I-positions, together with their inter-relationships, that is, dialogues.

Self-narratives. Hermans, Rijks and Kempen (1993) hold that the metaphor of the polyphonic novel extends the original narrative conception of the I as author and the

Me as actor (Sarbin, 1986). But unlike Sarbin, who assumes that a single narrator tells a story about himself or herself as actor, Hermans, Rijks and Kempen consider that narrators adopt numerous vantage points. According to Hermans and colleagues (1993;

Hermans, 1996a; 2001; 2002) the conception of the self as a polyphonic novel goes even further in that it assumes that:

The one and the same individual lives in a multiplicity of worlds with each world having its own author who may tell a story relatively independent of the authors of the other worlds. It is thus assumed that the individual consists of multiple

authors entering into dialogical relationships with each other and creating a more complex organization of the self (p. 213).

In their view each author relates a story that is entirely independent of that told by another. Authors relate to one another, but they do so dialogically, with none having ultimate power, though each may dominate at different times and in different places. It is in the spatial dimension that Hermans’ dialogical model departs from those of other advocates of a narrative approach, such as Sarbin (1986), Bruner (1986), Gergen and

Gergen (1988), and McAdams (1993). Following Bakhtin, time and space are seen as equally important. This is suggested by the words “position” and “positioning”, which

Hermans employs in place of the traditional term “role” (Hermans, 2001); and also by

Bakhtin’s (1929/1973) usage of the term “juxtaposition”, which presupposes a plurality of voices that are heterogeneous and even opposed, much less identical or unified.

Valuations. In the theory of the dialogical self, a “valuation” is “anything that a person finds to be of importance when thinking about his or her life situation”

(Hermans, 1988, p.792). Valuations include loved ones, those one dislikes, disturbing dreams, problems, cherished opportunities, memories of important events, plans or goals, and so on. Each valuation is a “unit of meaning”, and each has either a positive, negative or mixed (ambivalent) emotional quality. Through self-reflection, people organise their valuations into self-narratives that situate them in time and space—a valuation system. This means that the “person lives in the present and is, from a specific point in space and time, oriented toward the past as well as the future” (Hermans &

Kempen, 1993, p.81). By reflecting on past and future experiences, the individual composes a personal story consisting of significant events by. Moreover valuations may differ through the life-cycle because the individual’s reference point changes (Hermans,

2006; 1991).

The latent–manifest distinction. Hermans (1993) introduced a latent-manifest distinction into valuation theory. It assumes

that a small set of basic motives are latent in the affective component of the valuation system. These motives, moreover, are… assumed to be similar across individuals and to be continuously active within each individual. But, at the manifest level, valuations vary phenomenologically not only across individuals but also within a single individual across time and space. (p. 442)

Hermans (1989) notes that the reflective individual, as an observer of his or her own life, “can look back with nostalgia on the past, be worried about the present, and have hope for the future” (p. 14), but is always limited in his or her perspective by the circumstances of the present. Since Hermans conceives personality as a process in time and space, the perspective of the “I” is not fixed but changes from moment to moment:

The question is now how this concept of change can be related to the manifest and latent levels of life meaning representation. It is expected that when the perspective of the “I” changes, there may be a concomitant change in the manifest experience of an event. Whether or not this change also implies basic reorganization at the latent level remains to be seen. (p. 14 )

According to Hermans, valuations can be interpreted in terms of two primary motives: the self-enhancement motive and the union-with-the-other motive. Following

Bakan’s (1966) concepts of agency and communion, the self-enhancement motive is self-oriented—a striving for superiority, expansion, power and control. The union-with- the-other motive is other-oriented—a longing for contact, union and intimacy. The two motives give direction and organisation to the valuations in a person’s life story.

Sampson (1988) echoed the same idea when referring to the work of Bakan (1966). In every individual there coexist two mutually opposing motives. “Agency”, manifests as the urge for self-assertion, self-protection, and self-aggrandizement; “communion”, manifests as the urge to seek union with something larger than the individual. These two

31 motives are paralleled in other research. McAdams (1985) distinguished power and intimacy as basic motives. Many researchers in cross-cultural psychology (e.g.,

Triandis, Bontempo, Villareal, Asai, & Luca, 1988; Schwartz, 1990) make use of a notion that is variously labelled “individualism–collectivism” and “idiocentrism– allocentrism”, which in all its guises closely resembles Herman’s “self-enhancement” and “union-with-the-other”.

Another way of describing valuations is by their affectivity, positive and negative.

Each valuation reflects positive and negative feelings, which indicate the extent to which basic motives referred to above are fulfilled. Different I-positions may be compared and contrasted, then, with respect to the extent to which their valuation systems exhibit different degrees of self-enhancement motive, union-with-the-other motive, positive affect, and negative affect. The following example clarifies the relationship between the motivational and affective components of a valuation. The valuation, “I gained my doctorate by hard work”, evinces pride and strength; the self- enhancement motive is manifest in these affects as part of the valuation. Similarly, the evaluation, “I enjoyed the time when I was together with my family,” evinces tenderness and intimacy.

Identity. That personal identity is essentially dialogical is a natural extension of the notion of a narrative self, and follows from the proposition that self-narratives are collaborative, involving one or more listeners (Sarbin, 1993). In the contextualist view of human conduct (Sarbin, 1976, 1986), human identities are evolving constructions; they emerge out of continual social interactions in the course of life. This is a theme that permeates the novels of Amin Maalouf, a Lebanese migrant to France: “Identity isn’t given once and for all: it is built up and changes throughout a person’s lifetime.”

A person’s social identity may designate the collectivity with which a person identifies (e.g., an ideology or religion) or with a location on the social map (Berger,

1966). Greenwald (1988) used the term collective self to refer to the individual’s identification with a reference group, which leads to adoption of the group’s values and goals. He pointed out that a range of groups can provide the basis of the collective self

(e.g., religious, racial, ethnic, national, political, corporate, and professional groups).

Although dialogical self theory posits that the dialogical self is “social” (Hermans,

2001a), this sociality is “not in the sense that a self-contained individual enters into social interactions with other outside people, but in the sense that other people occupy positions in the multivoiced self” (p. 250). Commenting on the relationship between self and collectivity (e.g., Miller & Prentice, 1994; Turner, Hogg, Oakes, Reicher, &

Wetherell, 1987), Hermans (2001a) notes that the

idea of the Cartesian individualist and centralized self, which is separated from group and culture, can be contrasted with his dialogical approach that opens the realm of collective voices. (p. 262)

For Hermans, Baktin’s multivoicedness entails not only “the simultaneous existence of different individual voices in the self, but the simultaneous existence of an individual voice and voice of a group” (p. 262). In answer to his question regarding how are these individual and collective voices related to one another, Hermans comments:

Empirical evidence suggests that the two voices function as relatively autonomous parts of the self …The relative autonomy of the personal and the collective parts of the self requires us to study their dialogical relations. As individuals people may agree or disagree with the collectivities to which they belong. (p. 262)

Individuals engage in dialogue at a collective level as well as an individual one

(Hermans, 2002a, 2002b, 2003). The individual and collective voices within the dialogical self parallel the personal and social positions in which people may find

33 themselves. Social positions are defined by the expectations and prescriptions of society, personal positions by how individuals choose to organize their lives. Hermans

(2001a) illustrates this point with the following example:

when a person in a particular culture is defined as a woman, this social position carries specific expectations regarding the person’s dress, movements, behaviour and emotion regulation. However, from the perspective of her personal point of view, she may feel feminine in some situations (e.g. in dress), but masculine in other situations (e.g. in sexual behaviour). In this case, the meaning system of this person is constructed in a field of tension between her social position and one or more of her personal positions. (p. 263)

Thus, in Hermans, dialogical voices are not always harmonious but are engaged in

“negotiation, disagreement, power, negation, conflict, domination, privileging and hierarchy” (Bhatia & Ram, 2001, p.301). For Hermans (2001a) these individual and collective voices coexist and are in constant tension between “(symmetrical) interchange and (asymmetrical) social domination” (p. 266). In this way the theory accommodates the ways in which power asymmetries emerge between cultures and social groups.

In this sense, culture operates as a collective voice that organises the meaning of the dialogical relations (Hermans, 2001a). Because the self and culture are conceived in the theory in terms of a “multiplicity of positions among which dialogical relationships can develop” (p. 243), it then follows that cultural changes bring about changes in the dialogical self by creating a movement of I positions as well as a change in how these positions relate mutually. The culturally constituted dialogical self can be observed and understood from personal narrative accounts (Hermans, 2004; McAdams, 1993;

McAdams & Pals, 2006). Since a personal narrative embodies memories, feelings, and experiences, narrators generally tell their stories in highly selected ways that depend on

34 changing situations and social expectations (Hermans, 2000). This means that the created world of the self-narrative, where the recounted actions and the dramatised voices take place, is always context-dependent. Different narrative identities are constructed according to linguistic, social and physical context. Speakers make narrative choices in order to display a particular portrait of themselves, constructed for a particular context and a particular audience. They may also tell different and sometimes contrasting stories from different positions at different times (Hermans 2001a).

In short, dialogical self theory provides a conceptual toolkit with which to explore the psychological aspects of culture and acculturation. Applying it, however, demands elaborations and extensions of the theory, which are described in Chapter 4.

2.5 Summary

This chapter discusses the concept of the self, showing how and why it relates to the thesis. A pervading theme is the notion of multiple selves. This idea is not new, but has now become mainstream, beginning with James’ theories, and progressing through

Cooley, Mead, and Hermans. Their theories differ, sometimes substantially, in detail and approach, but all have in common the notion that people possess multiple selves, and that different selves are manifest in different circumstances.

Since Hermans developed his theory of the dialogical self, researchers have refined and extended his theory into practical areas of application, important among them being migration and acculturation. Their work is also described, but because it relates closely to theories of migration and acculturation, it is presented after the next chapter on migration and acculturation theories.

The idea that people have multiple selves that coexist in time and space is central to this thesis. It is furthermore necessary, for the kind of Cartesian individualist and centralized self, separated from group and culture, is incapable of explaining the

35 changes that often (though not always) occur when migrants migrate. For that one needs a multivoiced approach, and furthermore one that accepts that the self is moulded by the external environment, in this case culture. The empirical component of the thesis therefore focuses on participants’ multiple selves, namely those corresponding to the participant as a member of the culture of their country of origin, and those corresponding to the participant as a member of the culture of their country of settlement, in this case Australia.

CHAPTER 3: Migration and Acculturation

Ce qui fait que je suis moi-même et pas un autre, c’est que je suis ainsi à la lisière de deux pays … C’est précisément cela qui définit mon identité. Serais-je plus authentique si je m’amputais d’une partie de moi-même? [What makes me myself rather than anyone else is the very fact that I am poised between two countries … It is precisely this that defines my identity. Would I exist more authentically if I cut off a part of myself?]

Amin Maalouf (1949– )

In my case, the effort for these years to live in the dress of Arabs, and to imitate their mental foundation, quitted me of my English self, and let me look at the West and its conventions with new eyes: they destroyed it all for me. At the same time I could not sincerely take on the Arab skin: it was an affectation only …. Sometimes these selves would converse in the void; and then madness was very near, as I believe it would be near the man who could see things through the veils at once of two customs, two educations, two environments.

T. E. Lawrence, Seven Pillars of Wisdom (1922)

This chapter describes the migration experience from a psychological and sociological perspective, with particular focus on the Australian experience. It explores the process of migration and acculturation in order to understand the various outcomes of intercultural contact, and concludes with a synopsis of acculturation theories and their implications for the thesis.

3.1 Australian Migration

The global context. It is commonplace to say we now live in a ‘globalised’ world. One measure of this is language. Language is not the same thing as culture, but being relatively easy to observe, it is a serviceable indicator of it. For instance, the

European Commission (2012) reported that just over half of Europeans were ‘able to

37 hold a conversation in at least one additional language…’ (p. 5). By that criterion, most people in Europe at least have to engage at some level with a culture not their own.

Global figures are harder to come by, but the British Council (2014) estimated that ‘one out of four of the world's population speak English to some level of competence; demand from the other three-quarters is increasing.’

If second language acquisition is a sign of potential acculturation, migration is even more so. The OECD (2013) recently reported:

Some 232 million international migrants are living in the world today. Since 1990, the number of international migrants in the global North increased by around 53 million (65%), while the migrant population in the global South grew by around 24 million (34%). Today, about six out of every ten international migrants reside in the developed regions.

In short, migration is large and growing, particularly from the ‘South’ (which the

OECD broadly defines as developing economies), to the ‘North’ (developed economies). Given the differences between their cultures, it is clear that acculturation issues are greater than they have ever been, and are moreover set to increase.

The historical perspective. Australian immigration has been characterised by several phases. Since European settlement began over 200 years ago until the end of

World War 2, Australia was essentially a “British” country—British in population in that its people hailed overwhelmingly from the British Isles; and British in customs, more so than the British in some ways, since Australians saw themselves as “peoples of the three kingdoms [England, Scotland and Ireland]… equal sharers in this new community” Hirst (2014; p. 146). Even the Irish, whose country at that time was part of the United Kingdom, were happy enough to regard themselves as “British” in Australia, when they would have resolutely rejected the notion back in Ireland.

All that changed in the aftermath of World War 2, by which time Australia’s proportion of immigrants had fallen from a high of about 30% in 1893, to a low of near

10% in the late 1940s (Australian Bureau of Statistics 2013, December). After the near invasion of their country by Japan, Australians rallied to the slogan: ‘Populate or perish!’ True, institutionalised racism (present since Federation) remained a formal part of the new immigration policy, but given the shortfall of British migrants, Australians were persuaded to accept—reluctantly at first—other northern Europeans, then successively more southerners, and ultimately non-Europeans. And accept them they did (Hirst, 2010, 2014; Markus et al., 2009). By the 1970s the proportion of foreign- born neared 25%.

The White Australia Policy was finally abolished in the mid-1970s, leading to a surge in non-European immigration, mostly from Asia and the Middle-East. By the mid-1990s nearly 50% of new entrants came from Asia, although Britain and New

Zealand were still major source countries (Castles, Iredale & Vasta, 1994).

Current situation. Australia is exceptional in its migration history, which is reflected in its current migrant composition. Other countries may have a higher proportion of foreign-born residents, but those countries are not traditionally countries of settlement in the way that Australia is: some are burdened by refugees, such as

Jordan; some have large guest worker communities, such as Saudi Arabia and

Switzerland; and many are small island countries that have little impact globally, such as Bermuda (OECD & UNDESA, 2013). Only Australia has so many foreign-born, increasing so rapidly. Even Israel, with which Australia is often compared in matters of migration, has these days a lower proportion of foreign-born (and that against a backdrop of Israel’s Law of Return, which accords any Jew the right to migrate to

Israel).

Table 3.1 compares Australia with comparable countries from a migration perspective. Australia’s migrant stock, though proportionately lower than Israel’s in

1990, now exceeds it; and its rate of growth exceeds that of all countries but New

Zealand. Canada, with which Australia is also commonly compared, is culturally more diverse (Hawkins, 1989) but has proportionately fewer migrants. Because of its size the

United States receives many more immigrants than Australia in absolute terms, but proportionately its migrant stock is half that of Australia’s and in recent years has grown half as fast.

Table 3.1 Migrant stock: selected countries of large-scale immigration

Migrant stock

Growth rate 1990 2013 2010-2013

Australia 22.7% 27.7% 2.5% pa

Israel 36.3% 26.5% 0.5% pa

Canada 16.3% 20.7% 1.3% pa

New Zealand 15.4% 25.1% 3.4% pa

United States 9.1% 14.3% 1.2% pa

Source: OECD & UNDESA (2013).

Demographic analysis based on the 2006 national census showed that out of a population reaching 21 million, nearly a quarter of the population was born overseas, and a third had at least one overseas-born parent. The UK and New Zealand still remain the most common birth countries. The dominant culture remains European, with British,

Italian and German still the main non-Australian ancestries, but Australians are becoming less European and more Asian. Chinese are now the fourth largest overseas- born group, with Italian third. Overall, Australia has 250 ancestries speaking 400

40 languages at home, but 79 per cent only speak English at home. Italian, Greek and

Cantonese are the most common non-English tongues.

Policy settings. When mass immigration began after the World War II, the policy objective was officially one of cultural assimilation: migrants would become Australian in the quickest possible time, and to that end were termed ‘New Australians’ by

Government and agencies (Hirst, 2010, 2014; Markus et al., 2009). By the mid-1960s, this had changed: cultural “integration” was officially the goal, by which was meant that the authorities “still sought social cohesion but more gently” (Hirst, 2014; p. 160).And by the mid-seventies the term “multiculturalism” was officially adopted in recognition of a social environment became much more tolerant and accepting of difference, despite the racist ideas of the past (Castles & Vasta, 2000; Castles, Iredale, & Vasta, 1994;

Ward, Bochner and Furnham, 2001). Although few countries are completely monocultural, some are more culturally diverse than others, and Australia is often cited as an example of a culturally diverse society, as are the United States and Canada

(Berry, 1997; Bochner, 1986; Bochner & Hesketh, 1994).

There is a pattern to this policy progression: the foundation for what Hirst (2014) terms Australia’s migration “success” rested on Australians’ fears being allayed. At first, non-British, European migrants were accepted, grudgingly, and then only because they were needed: this was the era of ‘Populate or perish’. Even so, ‘old’ Australians demanded that they adopt Australian ways. Later, when Australians realised the newcomers caused no trouble (and indeed, brought interesting cuisines with them), they were allowed to keep their cultural identity (Hirst, 2014; Markus et al., 2009). Finally, those customs were accepted as part of the Australian social fabric. It took 30 years, but the transformation was dramatic, and still continues.

Push and pull factors. In a review of post-war literature on Australia’s immigration program, Betts (1996) observed that immigration was largely explained in terms of pull factors, that is, ones intrinsic to the host country that draw migrants to it.

Such theories have flourished in Australia and overseas since the mid-1970s, despite the fact that since 1974 changes in immigration were already beginning to make these theories incomplete. Instead, she suggested push factors are now more important, driving migration in the absence of the former official policy of active recruitment, and often in the face of attempts at exclusion. She saw a need to further define the terms

“push” and “pull” factors more clearly—that is, not from the point of view of the host society but of the potential immigrant. For example, in the “new economics” theory, potential migrants weigh up the net costs and benefits of attempting to emigrate

(Massey et al., 1993). This means that they consider the pull of the attractions in the new country together with the push of negative factors in their home country and evaluate these against the expense and risk of moving.

Betts (1996) also noted that the distinction between push and pull is confusing when perceived from the migrant’s point of view. If a person is contemplating a move he or she must see some attraction (pull) in the target country as well as some reason to depart (push) in the home country. It follows that a potential migrant’s actions are neither governed by push or pull: if benefits are to be perceived, both push and pull are required. The decision to move involves weighing up the benefits of both against the cost of breaking old ties, making a journey, and so on.

Motivations. Young and Madden (1992) examined how people arrive at the decision to migrate to Australia. In a study of over 500 migrants from the United

Kingdom, Ireland, Lebanon, Malaysia/Brunei, and Vietnam, they found that women generally migrated for “family reasons”, while men more often migrated as skilled

42 workers. Men were more likely to be ‘primary’ migrants in all categories and women were ‘secondary’ migrants.

Maydell-Stevens, Masgoret & Ward (2007) found that motives were a key influence on migrants’ adaptation to a new culture, to the extent even of determining their acculturation strategy. Boneva and Frieze (2001) found that achievement motivation, power motivation, and affiliation motivation all influence the desire to migrate: intending migrants tend to be above average in achievement and power motivation and lower in affiliation motivation. Migration is an expression of an individualistic orientation: those high in achievement motivation may migrate in order to seek opportunities, and those high in power motivation may be more willing to take risks in reaching their goals. The authors interpret migration as a life-changing event that offers opportunity but also carries risk: by moving into a new and usually larger society, migration offers career and economic enhancement but impedes social ties with home.

Migrants also differ from other cross-cultural travellers in motivation and commitment to country of resettlement. Unlike refugees, who are “pushed” into an alien environment, migrants are “pulled” towards a new country in pursuit of personal, familial, social, financial and political goals (Winter-Ebmer, 1994). Economic factors exert the strongest influence, though family reunification and political pressures play a part. In a psychological analysis of migration intention, Winchie and Carmen (1988) found that migrants are more likely to be dissatisfied with their employment in their home country and come from a lower socio-economic stratum than non-migrants.

A typology of migrants. Taylor (1969) identified three types of migrant.

“Resultant” migrants are pressured by circumstances to move; they move essential because they have to. “Dislocated” migrants move to avoid “dislocation” from their

43 primary group, often to join family or friends who have already migrated. But the most are “aspirers”, who migrate to better themselves and their children. The “dislocated” and “aspirers” are responding mainly to “pull” factors; the “resultant”, to “push” factors.

Modern economic theories of migration place the cause for moving within the immigrants’ perceptions of their situations rather than within the host countries’ economic and political institutions, which may explain the selectivity of migration

(Massey et al., 1993). These factors point to economic, demographic and network factors in order to explain why people leave their countries of origin (Massey et al.,

1993). They do not, however, explain why some people become migrants and others in similar socio-economic conditions do not.

It may be that some people are psychologically predisposed to migration, that is, they may possess characteristics that predispose them to migrate and help them endure the challenges of migrant life. Jennings (1970) coined the term “mobicentric man” for individuals who value action and are constantly on the move. Morrison and Wheeler

(1976) coined the term “pioneering personality” and suggested that the desire for novelty might be as important a motivator as the desire for material advancement.

Other studies support the notion of a “migrant personality” and specify its characteristics. Boneva and Frieze (2001) note that not everyone in poor countries wants to migrate to richer ones, and would-be migrants are not all from the poorest class. They accept that economic prospects, migration policies, and network support in the receiving country create the preconditions for the desire to migrate, but argue that at the level of the individual, migration is largely driven by personality. They found that primary migrants (that is, those making the initial decision) to be more work-oriented, and motivated more by achievement and power than by affiliation or family centrality, than

44 secondary migrants (that is, those who follow in order to be reunited with family)

(Boneva & Frieze, 2001; Frieze et al., 2004).

3.2 Acculturation

Terms and concepts. Acculturation is “the process by which individuals change, by being influenced by contact with another culture and by being participants in the general acculturation changes under way in their own culture” (Berry, 1990, p.204). It is important to differentiate acculturation of the group from acculturation of the individual.

Sociologists and anthropologists were the first to study acculturation, and were mainly interested in group-level changes; Redfield, Linton, and Herskovits (1936) were among the first to provide a formal definition:

Acculturation comprehends those phenomena which result when groups of individuals having different cultures come into continuous first-hand contact, with subsequent changes in the original culture patterns of either or both groups. (p. 149)

Later, Graves (1967) introduced the concept of “psychological acculturation” to describe changes as a result of personal contact with the other culture, and participation in acculturation at the group level (Berry, 1990), so recognising migrants’ emotional attachment to the host culture and country of origin (Betancourt & Lopez, 1993).

According to Berry, the phenomena are different at the two levels: at the group level, changes occur mainly in the group’s social structure, economic base, and political organisation; at the individual level, changes occur in identity, values, attitudes, habits and motives (Berry, Kim, Power, Young, & Bujaki, 1989; Berry, Poortinga, Segall, &

Dasen, 2002).

Acculturative changes are normally adaptive in that they help individual to transact two different cultures. This is illustrated in Figure 3.1.

Societal-level variables Society of origin Society of settlement . Social factors . Social factors . Political factors . Political factors . Economic factors . Economic factors . Cultural factors . Cultural factors

Responses: Outcomes: Cross-cultural Stress affective, psychological transition and skill behavioural, sociocultural experience deficits cognitive adaptation

Individual-level variables

Person Situation . Personality . Length of cultural contact . Language fluency . Amount of intra- and inter-group contact . Training and experience . Quality of intra- and inter-group contact . Cultural identity . Cultural distance . Acculturation strategies . Amount of life changes . Values . Social support . Reasons for migration

Figure 3.1 The acculturation process (adapted from Berry, 2005)

Group-level factors (at top) precede acculturation in the society of origin, and pre- exist in the society of settlement. Individual-level factors (at bottom) precede acculturation in the migrant. In the middle are factors that arise in individuals during the course of acculturation. Both societies act jointly to initiate and shape psychological acculturation, which is described in terms of stressors, responses, and outcomes (for a review, see Berry & Kim, 1988; Berry, Kim, Minde & Mok, 1987; Berry, 2005).

Individuals differ in how they experience acculturation. Some perceive acculturative changes as stressors that expose deficits in coping skills; others perceive them as benign opportunities for growth. Acculturative stress is manifest in affective, behavioural or

46 cognitive responses. Variations in adaptation (psychological or sociocultural) range from the well-adapted to maladapted.

Intercultural contact. The social psychology of cross-cultural relations and the psychological effects of culture contact are important areas of research in cross-cultural psychology, (e.g., Ward, Bochner & Furnham, 2001; Bochner, 1982; Brislin, 1981).

Bochner (1982) found intercultural contact to be of two kinds: that between members of a culturally diverse society, and that which takes place when a person from one culture travels to another with a particular objective, for example, to work, play, study, exploit, convert, or assist. Over the last 30 years empirical research on intercultural contact has flourished, particularly as concerns sojourners, refugees, migrants and expatriate workers. Migrant adaptation has been found to be influenced by pre-migration conditions, by the transitional experience in moving from one country to another, by the characteristics of the migrants themselves, and by the conditions of the receiving country, including government policies and economic factors. Migrants generally relocate with the intention of long-term, if not permanent, resettlement (Ward, Bochner

& Furnham, 2001).

Berry (1990) observed that, technically, overseas-born, first generation settlers should be described as migrants, and their descendants as members of ethnocultural groups. Ward and colleagues (2001) pointed out that it is the initial contact between the first generation migrants and the members of the receiving culture that most closely resembles a between-culture contact. Over time, this interaction merges into a within- culture contact, with successive generations as migrants groups evolve into established ethnocultural communities within the larger society.

Migration and identity. As common as migration is, it is not the experience of most people. Migration is a complex psychosocial process with lasting effects on

47 individual identity. For most migrants, their identity impinges on almost everything they do. So how do migrants, who move between cultures, reconcile their social identities?

Oddly, James (1890, p.213), who “regarded [personal identity] as the most puzzling puzzle with which psychology has to deal,” had little to say about migration; but as Brown (1998) points out, James wrote at a time when psychology was mostly the province of white males operating in their own cultural milieu. Not so Freud, possibly because as a Jew in ethnically diverse Vienna he felt his ethnicity more personally. In a

1926 address to the B’nai B’rith (a Jewish community organisation) he referred to his

“irresistible” attraction to being a member of an ethnic group, so echoing his own writings, which claimed the search for identity to be one of the mind’s basic motives because it guaranteed safety and wish fulfilment (Graafsma, 1994).

Whenever I felt an inclination to national enthusiasm, I strove to suppress it as being harmful and wrong, alarmed by the warning examples of the peoples among whom we Jews live. But plenty of other things remained over to make the attraction of Jewry and Jews irresistible—many obscure emotional forces … as well as a clear consciousness of inner identity, the safe privacy of a common mental construction. And beyond this there was a perception that it was to my Jewish nature alone that I owed two characteristics that had become indispensable to me in the difficult course of my life. Because I was a Jew, I was free from many prejudices which restricted others in the use of their intellect; and, as a Jew, I was prepared to join the Opposition and to do without agreement with the “compact majority” (Freud, cited in Zavalloni, 1983, p.205)

According to Brown (1998), two aspects of collective identity have received attention by researchers. One is how people evaluate identity: some minority members resent or even disown their ethnic identity to avoid the stigma they feel (Lewin, 1948).

The other aspect explored how people maintain their ethnic identity as a minority in a

48 dominant culture. Amin Maalouf (2000), a Lebanese-born writer who has lived and worked in France for decades, wrote:

How many times, since I left Lebanon in 1976 to live in France, have people asked me, with the best intentions in the world, whether I felt ‘more French’ or ‘more Lebanese’? And I always give the same answer: ‘Both!’ I say that not in the interests of fairness or balance, but because any other answer would be a lie. What makes me myself rather than anyone else is the very fact that I am poised between two countries, two or three languages and several cultural traditions. It is precisely this that defines my identity. Would I exist more authentically if I cut off a part of myself? So am I half French and half Lebanese? Of course not. Identity can’t be compartmentalized. You can’t divide it up into halves or thirds or any other separate segments. I haven’t got several identities: I’ve got just one, made up of many components in a mixture that is unique to every individual. Sometimes, after I’ve been giving a detailed account of exactly why I lay claim to all my affiliations, someone comes and pats me on the shoulder and says, “of course, of course – but what do you really feel, deep down inside?” (pp. 3-4)

Maalouf has a life story that according to McAdams’s (2003) theory of identity captures what he is in all his complexity. Unity is important for Maalouf, but he wants a unity that is comprehensive, not one that cuts out portions of his story to arrive at a false coherence. His audiences, however, demand a simplification that is fundamentally in error. Biculturalism can foster pride in one’s origins, solidarity with compatriots, and a rich appreciation of history; equally it can bring confusion, disappointment and cultural conflict (Haritatos & Benet-Martınez, 2003). Migrants find themselves simultaneously immersed in multiple cultures, and so typically have multiple group allegiances.

Liebkind (2008) brought another perspective when he observed that ethnicity is both ascribed (in that one cannot choose one’s ethnic origin) and achieved (in that it is a matter of choice). Identification with one’s ethnic roots therefore involves both a

49 cognitive self-categorisation and a sense of attachment. Liebkind (1996) notes that ethnic/cultural identity has been empirically treated not only as “mere self- identification” but also as “feelings of belongingness and commitment” towards one’s group (p. 164).

On its own, ethnic self-identification says little about a migrant’s attitudes towards the heritage culture or strength of identification with the self-applied category. For example, Ethier and Deaux (1990) asked Hispanic American college students to name their ethnic identities and list the characteristics they associated with each. Most (but not all) claimed a Hispanic identity but associated with it differed meanings: some were positive, others ambivalent. Likewise, in a study of students from ethnically diverse high schools in the United States, Phinney and Devich-Navarro (1997) found no statistical link between self-ascribed ethnicity and degree of involvement in their origin ethnic group and/or American society. Likewise, in a study of first- and second- generation Chinese Australian and Chinese American adolescents, Rosenthal and

Feldman (1992) found that ethnic identification (that is, labelling oneself as Chinese) and knowledge and practice of Chinese culture declined over time, but subjective evaluation of their Chinese identity and the importance attached to Chinese cultural practices did not. They concluded that although the behavioural aspects of ethnic identity diminished, its internal components did not.

In a review of relevant studies, Phinney (1990) found contradictory evidence to link ethnic identification with degree of acculturation. Some studies showed the two dimensions to be unrelated: Hutnik (1986) for instance found no significant correlation between ethnic identity and cultural attitudes among young Indians in Britain. But other studies gave a different picture: in separate studies of Israeli students in the United

States, Elias and Blanton (1987) found some correlations of behavioural and affective

50 aspects of ethnic identity. Phinney concludes from such studies that “ethnic identity is not necessarily a linear construct” (p. 509). At a conceptual level, therefore, there exist two main acculturation models that differ according to whether acculturation is seen as unidimensional or bidimensional. Gordon (1964) proposed a unidimensional model. At one end of the spectrum, migrants cling to their own culture; at the other, they adopt the host culture’s values, attitudes, and behaviours. In a refinement of this simplistic view, the so-called ‘balance’ model permitted a both cultures to coexist in the form of biculturalism. However, it still viewed heritage and host cultures as interdependent rather than orthogonal, and failed to accommodate those who identified strongly (or weakly) with both cultures. Biculturalism was considered a temporary state in the transition to full adoption of the host culture (Goldlust & Richmond, 1974;

LaFromboise, Coleman, & Gerton, 1993; Ryder, Alden, & Paulhus (2000).

To address these difficulties, acculturation models were devised that considered heritage and host cultures as independent influences. These so-called “bidimensional” models accommodate biculturalism and intensity of cultural involvement as orthogonal dimensions (e.g., Berry, 1980; La Fromboise et al., 1993; Szapockznik, Kurtiness, &

Fernandez, 1980). They are based on two core assumptions: that individuals differ in the degree to which their identity includes values, attitudes and behaviours from their heritage culture; and that individuals can possess multiple cultural identities of varying independence and strength (Ryder, Alden, & Paulhus, 2000). In models based on the first assumption, immigrants can indeed relate to both heritage and host cultures differently, but in different domains of life; they may for instance learn the host language but still cling to the values they were brought up with. In models based on the second assumption, the influences of both cultures are independent of each other. The

51 leading model of this kind is the one proposed by Berry (1980, 1984, 1997), described next.

Berry’s acculturation model. Berry (1980, 1984, 1990, 1997, 2005) developed a widely accepted framework for acculturation that embodies factors pertaining to both the group and the individual, and to both the society of origin and that of settlement.

Berry (1980, 1984; Berry & Sam, 1997) was the first to propose that cultural identities were separate dimensions rather than extremes of a single dimension; or to put it another way, identifying more with one culture did not mean identifying less with the other. There were instead four possible psychological responses to cultural change, which Berry calls “acculturation strategies”: identification with the societies of both origin and settlement indicates “integration” (real biculturalism); identification with the culture of settlement alone indicates “assimilation”; identification with the culture of origin alone indicates “separation”; and identification with neither culture indicates

“marginalisation” (e.g., Berry, 1980, 1985, 1990, 1997; 2005; Berry & Sam, 1997; Sam

& Berry, 2006). Most long term migrants adopt a strategy of “integration” or

“assimilation”, depending on whether or not they respectively retain or discard their culture of origin. But not all migrants take this path: some either choose or are forced into a strategy of “separation” or “marginalisation”, again depending on whether or not they respectively retain or discard their culture of origin.

According to Berry’s (1980, 1984) psychological acculturation model, migrants settled in the host society face two fundamental choices: whether or not to retain their migrant culture; and whether or not to seek relations with the host society. There are four possible psychological response combinations, which Berry calls “acculturation strategies”: integration, assimilation, separation, and marginalisation. Integration means maintaining some features of the migrant culture while adopting other features of the

52 host culture, and integrating the two. Assimilation means relinquishing the migrant culture in its entirety and replacing it with that of the host culture. Separation is its opposite: maintaining the migrant culture in its entirety and rejecting the host culture.

Finally, marginalisation means rejecting both cultures—their own and that of the majority. These four strategies can be adopted at the individual or level (Berry, 1980,

1984). Acculturation may, however, vary across the domains of behaviour and social life: “For example, one may seek economic assimilation (in work), linguistic integration

(by way of bilingualism), and marital separation (by endogamy)” (Berry, 1990, p.217).

Numerous empirical studies have shown the Berry model to have strong descriptive and explanatory power (Berry, 1997; Redmond, 2000; Sam, 1995). Berry and his associates (Berry, 1980, Berry, Kim, Power, Young, & Bujaki, 1989) investigated the acculturation strategies of the following first-generation immigrants to

North America among others: Portuguese, Hungarians, and Koreans in Canada (Berry et al., 1989); Lebanese in Canada (Sayegh & Lasry, 1993), and Indians in the United

States (Berry & Krishnan, 1992). They found integration to be the preferred acculturation strategy, followed by assimilation and separation; marginalisation was least preferred. Other studies confirm migrants’ preference for integration, and furthermore see flow-on mental health benefits (Niles, 1999). For example, migrants to

Belgium and the Netherlands seek to combine their own culture with that of the mainstream (e.g., Phalet & Hagendoorn, 1996; Van de Vijver, Helms-Lorenz, &

Feltzer, 1999). In a Canadian study, migrants who opted for integration were minimally affected by acculturative stress, whereas the marginalized and separated tended to be highly stressed (Berry, Kim, Minde & Mok, 1987); a tendency confirmed by Berry

(1997), who found that the assimilated experienced intermediate levels of acculturative stress. However, none of this negates the distinction migrants draw between public and

53 private domains; for example, Turkish migrants in the Netherlands were found to favour separation in the private domain but integration in the public (Ourasse & van de Vijver

(2005).

Beyond Berry. For all its elegance, Berry’s acculturation model has come under increasing criticism. In particular, Berry’s “integration” says little about how people maintain two cultures or why biculturalism is hard to sustain. He assumes that acculturation to the new culture and to the culture of origin are independent (Berry,

1980; 1990); and although he offers a taxonomy of acculturation, he does not explain how acculturation type is manifest in migrant identity, nor how migrants negotiate two cultures. Ryder, Alden, and Paulhus (2000) remedied this by showing that migrants who prefer integration (in Berry’s terminology) are bicultural in the sense of reporting more than one cultural identity; and that cultural identifications are independent of measures of personality, self-identity, and adjustment.

Some researchers consider Berry’s model to be too simplistic to account for the complexities of acculturation (e.g., Boski, 1998; Weinreich, 1998). Bourhis, Moise,

Perreault, and Senecal (1997) replaced Berry’s social contact and participation dimension with one pertaining to host culture adoption: “Is it considered to be of value to adopt the culture of the host community?” (p. 378). Ward (1999) likewise identified different underlying dimensions of cultural identity. Just as acculturation strategies are classified on two dimensions (own cultural maintenance and involvement with host society), migrants’ self-perceptions are also constructed along two dimensions: identification with their country of origin’s heritage (their cultural or ethnic identity), and identification with the host society (their created national identity) (Phinney, 1990;

1999; 2000). Moreover, the two dimensions of acculturation strategy are orthogonal, as are the two dimensions of cultural self-perception. The strategies that emerge, mirror

Berry’s four acculturation strategies. Phinney (1990) described four possible outcomes based on strength of identification with the majority and minority groups. Those who identify with the dominant culture, while still identifying strongly with their cultural background, are said to be acculturated, integrated, or bicultural. Those who abandon their ethnic identity for that of the dominant culture, are said to be assimilated. Those who decline to identify with the dominant culture are said to be separated. And those who lose their ties to both cultural groups are said to be marginalised. Research suggests that a bicultural identity is the most adaptive kind for migrants (Phinney, Horenczyk,

Liebkind, & Vedder, 2001), and bicultural competence, or the ability to interact in two cultures without tension, is essential (LaFromboise et al, 1993). In short, it is possible to possess two cultural identities at the same time.

Birman (1994) suggested two models of biculturalism: the alternation model and the fusion model. In a review of biculturalism similar to that of LaFromboise et al.

(1993), she draws a distinction between the behavioural aspects of biculturalism

(LaFromboise et al.’s “bicultural competence”) and the psychological (“bicultural identity”), recommending that both be considered. By “alternation” she means the type of biculturalism where individuals feel acculturated to either one culture or the other depending on context, and feel a sense of belonging to both. By contrast, according to the fusion model, migrants exhibit a blended pattern of multicultural behaviour; rather than having two distinct cultural repertoires that are invoked in different cultural contexts, they belong to neither culture exclusively but instead to both (Birman, 1994;

Hermans & Kempen, 1998; LaFromboise et al., 1993). In Hermans and Kempen’s

(1998) view an acculturating individual fuses both cultures in a “hybrid” identity that combines the best of each.

In a study of Mexican Americans and African Americans in California, participants provided quantitative and qualitative data (Phinney, 1992). The study generated a typology of bicultural identification. “Blended” biculturalism describes those who identified equally with both cultures and dealt with them by combining them; they are migrants who, when asked their ethnicity, might answer, say: “I am both

Mexican and American” (see for instance Phinney & Devich-Navarro, 1997).

“Alternating” biculturalism describes those who identified more with their ethnic culture but were comfortable in both; they are migrants who, when asked to describe their ethnicity, might answer, say: “I am a Mexican in America,” but not “I am Mexican

American” (Phinney & Devich-Navarro, 1997). Alternating biculturals identify with both cultures (as do blended biculturals) but are also conscious of cultural discrepancies that could be a source of internal conflict, so they keep their cultural identities separate

(Phinney & Devich-Navarro, 1997; Vivero & Jenkins, 1999).

There are thus two apparent modes of biculturalism: compatible and oppositional.

The one views their heritage and host cultures as complementary, the other as discrepant or conflicting. Neither depends on how long an individual has lived in the host culture

(Phinney & Devich-Navarro, 1997) nor on attitudes toward biculturalism (Vivero &

Jenkins, 1999). However, as Vivero & Jenkins note, oppositional biculturalism can generate internal conflict, which reduces well-being.

Benet-Martinez, Leu, Lee, and Morris (2002) elaborated the compatible-versus- oppositional dichotomy by proposing that it arose from “bicultural identity integration”

(BII), an individual difference that moderates cultural “frame-switching”, or the act of moving between cultures in response to situational cues. Bicultural individuals, whether low or high on BII, exhibit Berry’s acculturative strategy of integration and hence show

56 a bicultural/bilingual pattern of cultural identification and language proficiency (Benet-

Martinez et al., 2002; Benet-Martinez et al., 2006).

In a later study, Benet-Martinez and Haritatos (2005) found that migrants internalize potentially conflicting cultural identifications. Using their purpose-designed

Bicultural Identity Integration Scale, they scored a sample of first-generation Chinese-

American migrants on BII and correlated the results with measures of personality traits and bicultural stressors. They inferred the existence of two modes of bicultural identity integration: cultural conflict, which perceives that cultures clash, and was correlated with neuroticism, discrimination, and strain in intercultural relations; and cultural distance, which perceives that cultures are separate, and was correlated with low openness and living in a culturally isolated environment. They then identified two groups: the high BII, who feel their cultures can coexist, and who identify more with the mainstream culture, even though competence in their ethnic culture of origin is often similar for both groups; and the low BII, who feel conflicted about their two cultures and see them as oppositional. These findings accord with the “alternation” model for second culture acquisition (e.g., LaFromboise, Coleman, & Gerton, 1993), as experimental research showed that often those high in BII also take part in cultural frame-switching.

Benet-Martinez and Haritatos (2005) also found low-BII persons to be culturally more complex than high-BII persons, which they attributed to the greater cultural conflict that low-BII migrants experience. They suggested that low-BII people, particularly migrants, feel “culturally homeless”, quoting from Vivero & Jenkins’s

(1999) study, in which one bicultural individual is further recorded as stating:

You start building a home in one place within one culture … but do not complete it. Then you continue to build your home within another culture …. At the end,

you have different pieces of home in different places. You can never put them together, because they may contradict or conflict with one another (p. 7).

Expanding on Benet-Martinez et. al’s (2002) concept of bicultural identity integration, Boski (2008, p.148) distinguished “harmonious” and “conflictual” biculturality. The harmoniously integrated experience a consistency between the two identities: when either is activated, both are parts of the same cultural context and correspond to each other. But the conflictually integrated experience dissonance between cultures. This implies two modes of integration, irrespective of competence in both cultural contexts. According to Boski (2009, p.148), to be bicultural means to experience a “psychological parallelism” where a person switches between cultures and functions effectively and coherently in both. But it can also mean that a person experiences both antagonistically, to the detriment of psychological well-being. Both depart from the premise that the person is equipped with competence in both cultures: this is the paradox. What comes out of this is that integration in Berry’s occurs only when the two identities are harmonious, as it is a contradictory to say that a person is biculturally integrated when the two identities are in conflict. And yet, Cheng, Lee and

Benet-Martinez (2006) report that both harmoniously and conflictually bicultural

Chinese-Americans endorse Berry’s integration style.

Psychological theories of acculturation focus on changes in attitudes and mental health (Berry, 1990). According to Phinney and Flores (2002), such changes must be explained in terms of a two-dimensional model of acculturation. Like Berry and Sam

(1997), they propose that acculturation is not one-dimensional in that an increase in involvement in one culture implies a decrease in involvement in the other; but two dimensional in that involvement in one culture is independent of involvement in the other. To test this proposition, they examined measures of acculturation as predictors of

58 traditional gender role attitudes in a sample of first-, second-, and third-generation

Hispanic migrants. Acculturation was measured in terms of ethnic retention (Spanish language, Hispanic friendships), mainstream involvement (English language, non-

Hispanic friendships), and traditional gender role attitudes. English language and non-

Hispanic friendships correlated negatively with traditional gender role attitudes; Spanish language and Hispanic friendships were uncorrelated. Thus the bicultural Hispanic can be highly involved in, and share the values of, mainstream America, and yet retain

Hispanic social networks and Spanish language. Furthermore, “migrant generation”

(that is, first, second, third etc.) was unrelated to social networks, Hispanic and non-

Hispanic, or to English language usage. Thus, knowing a person’s migrant generation reveals little about the effects of cultural contact on the individual. Such findings powerfully support Berry and Sam’s (1997) two-dimensional model of acculturation in that mainstream adaptation and ethnic retention are revealed to be independent of each other and to have different outcomes.

Researchers (e.g., LaFromboise, Coleman & Gerton, 1998) agree that

“successful” migrants acquire social skills, cultural knowledge, and other aspects of the society of settlement. Those who do not achieve bicultural competence may experience acculturative stress in the form of hostility, anxiety, identity confusion, depression, and feelings of marginality and alienation. The implication is clear: acculturation strategy matters, because it affects well-being; likewise cultural identity, since an acculturation strategy is defined by the migrant’s cultural identity (or identities).

3.3 Acculturation Outcomes

The literature views acculturation outcome from three conceptual angles: psychopathology, behaviour, and acculturative stress. Psychopathology sees it as a stressor, hence potentially problematic. Migrants, it implies, require assistance in

59 coping, But according to Bonanno (2004) and Ward (1996) this cannot be so, as most migrants relocate successfully, with health, psychological and social outcomes like those of the mainstream. When viewed in terms of observable behaviour, acculturation is usually seen as unproblematic. It comprises three sub-processes: culture shedding, culture learning, and culture conflict (Berry, 1992). In a process of “adjustment” (Ward

& Kennedy, 1993b), culture shedding and culture learning entail replacement of inappropriate behaviours with ones that sit better with the host culture. They are perceived as accomplished with little difficulty, but if conflict occurs, it is resolved by the migrant’s yielding to the norms of the mainstream. Assimilation is the likely outcome. Lastly, when viewed in terms of acculturative stress, acculturation is judged to be problematic but surmountable (Berry, Kim, Mindle & Mok, 1987). Migrants experience changes that challenge their cultural norms in ways that cannot be dealt with by adjusting or assimilating.

The psychopathology of culture shock. Migration always causes a certain

“culture shock” (Garza-Guerrero, 1974). The anxiety that ensues challenges the stability of the migrant’s “psychic organisation” and

a growing sensation of discontinuity of identity emerges. It is as if, out of his usual habitat, the newcomer no longer has the necessary corroborative environmental feedback for his ego identity …. The severity of the threats to the individual identity runs parallel to the severity of concomitant mourning. Thus, the more serious the break with the newcomer’s continuity of his identity, the greater this yearning for those lost love objects (abandoned culture) which in the past provided a comfortable sense of continuity. On the other hand, the greater the longing for those love objects, the more afflictive the threats to his identity. (Garza-Guerrero 1974, pp. 418–419)

For Volkan (2004), issues of personal identity are paramount for migrants.

Migrants go through a mourning process, which may be complicated by feelings of guilt

60 over loss of parts of the self (e.g., the migrant’s previous identity, his or her investment in the language, culture, land, and people left behind). Guilt may trigger other psychological processes, such as depression, identity confusion, shame, helplessness, and culture shock. It is this culture shock that embraces adaptive processes that may include reactions of anxiety, depression, fear, uncertainty, and anger.

Among the best known early writers on international relocation is Oberg (1960) who discussed “culture shock”, the disorientation and confusion that results when a person enters a second culture and familiar cues are absent. This results in feelings of loss, confusion over roles and role expectations, and feelings of loss of control. Oberg’s detailed four phases of emotional reactions: (a) the “honeymoon”, characterised by euphoria, enchantment, fascination, and enthusiasm; (b) the crisis, characterised by feelings of inadequacy, frustration, anxiety and anger; (c) the recovery, including crisis resolution and culture learning; and finally (d) adjustment, characterised by enjoyment of, and functional competence in, the new environment. Oberg had in mind people who were more or less suddenly exposed to a completely unfamiliar setting and largely overwhelmed by it. Some of the affective responses consistently mentioned in the literature included confusion, anxiety, disorientation, suspicion, bewilderment, perplexity and an intense desire to be elsewhere.

This “honeymoon period” of culture shock may be contrasted with hypotheses arising from the stress-and-coping literature which predict that, in contrast to “entry euphoria”, immigrants suffer the most severe adjustment problems at the initial stages of transition when the number of life changes is the highest and coping resources are lowest (Ward, Bochner & Furnham, 2001). Early research on culture shock compared it to a disease, to be cured with appropriate treatment. This approach highlights the factors that reduce the distress of culture contact, including personal resources such as self-

61 efficacy and emotional resilience, and interpersonal assets such as social support. The affective outcome of a well-adapted migrant in the new culture is a psychological adjustment.

Acculturation stress. The migrant must accommodate different beliefs, routines, and social roles (Hovey, 2000). The term “acculturative stress” captures the stress and loss this entails. For Lazarus & Folkman (1984), acculturative stress is an outcome of

the losses that occur when adopting a new system of beliefs, routines, and social roles (Hovey, 2000). It draws on the broader paradigms of stress, adaptation, and coping. Berry, Kim, Minde & Mok (1987) concur, seeing acculturative stress as a reaction to life events rooted in the experience of acculturation, with depression, anxiety and psychosomatic problems as some of its adverse consequences (Berry, 1997; Berry

& Sam, 1997).

Under this approach, the long-held prevailing view was that acculturation inevitably brought about a decline in mental health. However, this view is no longer supported (Berry & Kim, 1988). Many factors mitigate acculturative experiences such that mental health is unaffected or only slightly so. Through adaptation the consequences of acculturative stress may be positive as migrants develop skills and behaviours that enhance mental and physical health, and social functioning.

In this regard, Okazaki (1997) explored the role of ethnicity in emotional distress in a sample of Asian-born migrants to the United States. Drawing on Markus and

Kitayama’s (1991) model of self-construals, she found that Asian Americans are socialised to hold highly interdependent self-construals; in consequence they are more attuned to social cues and so more susceptible to social anxiety than others. After controlling for depression and social anxiety, ethnicity and self-construal predicted for social anxiety, and self-construal alone predicted fear of negative evaluation.

Building on work by Okazaki (1997), Norasakkunkit and Kalick (2002) measured self-reported depression, social avoidance and distress, and fear of negative evaluation in a sample of European Americans and Asians. They found that for the Asians, but not the others, the measures were positively related to interdependent self-construal, and negatively related to independent self-construal and self-enhancement. This was explained in terms of the Asians’ greater sensitivity to relationships, given their greater interdependent self-construal.

On the assumption that acculturative stress diminishes with time, researchers in the 1980s and 1990s began to use acculturation level as a proxy for acculturative stress.

Their findings are contradictory. For example, studies of migrants’ attitudes in Canada showed that an integration strategy produces the lowest acculturative stress (Berry,

1990; Berry & Kim, 1988; Berry et al. 1987). Conversely, in a study of acculturative stress and depression, Hovey and King (1996), found that acculturation level and stress to be unrelated. It may therefore be wrong to assume that the less acculturated (e.g., separated) experience more acculturative stress than the more acculturated.

Many factors contribute to successful acculturation at the individual level: personal experience of each individual, often associated with age and gender; social status such as education, prior contact with diverse cultures, familiarity with the language and culture of the host community; and personality. Moreover, the greater the cultural distance between the origin and host society, the more difficulties migrants encounter (Ward & Kennedy, 1992; Ward & Searle, 1991; Hirst, 2014).

Acculturation adaptation. Except perhaps for social identity theory (Ward,

2001), acculturation studies usually conceptualise individual changes either as a response to stress (Berry, Kim, Minde & Monk, 1987), or as a response to a need, in this case to learn cultural skills so as to thrive and survive (Furnham & Bochner, 1986).

Acculturation results in the migrant’s learning skills that are adaptive and functional

(Berry, Poortinga, Segall & Dasen, 1992). Adaptation takes various forms, though all are largely stable changes at an individual or group level in response to external demands (Berry, 1997; Ward, 2001). It does not require that migrants become more like their environment (a strategy of assimilation); it may imply resistance, efforts to change the mainstream, or distancing (a strategy of separation). And the outcome varies accordingly, from material, social and mental success, to the inability to function effectively in the new society (Berry, 1997).

As illustrated in Figure 3.2, the literature distinguishes three theoretical approaches to acculturation, one that invokes psychological models of stress-and-coping as applied to cross-cultural adaptation; a culture learning approach; and a social identification approach (Ward, Bochner & Furnham, 2001).

The stress-and-coping approach views cross-cultural transitions as a series of stress-provoking life changes that require coping responses from the migrant. The culture learning approach focuses on the skills and behaviours required in order to succeed in a new culture; adaptation is a result of learning new and appropriate skills

(Bochner, 1972, 1986). Lastly, the social identification approach is associated with social identification theories, and is concerned with values, attitudes, beliefs, expectations, self-identity, cultural identity (Taft, 1977).

Another distinction found in the literature is that between psychological and sociocultural adaptation (Searle & Ward, 1990). Psychological adaptation covers depression, anxiety and psychosomatic problems, which are the most common mental health consequences of acculturating individuals (Berry, 1997; Berry & Sam, 1997).

Sociocultural adaptation covers the relationship between individuals and their cultural context, and concerns the success or otherwise of participation in the host society

(Ward, Bochner & Furnham, 2001; Searle & Ward, 1990; Ward et al., 2001; Ward &

Kennedy, 1994). In this framework, the stress of acculturation results in the acquisition of adaptive and functional coping skills (Berry, Poortinga, Segall & Dasen, 1992)..

Behaviour Processes Culture learning theories

Coping Acquiring with Outcomes specific cultural change skills

Behav- Affects iours

Affect Stress and coping theories Percep- tions

Cognition Social identification Developing theories changing, maintaining identity

Figure 3.2 The “ABC model” of acculturation, adapted from Ward, Bochner & Furnham (2001)

Though separate, psychological and sociocultural adaptation are linked, for success in the host culture are likely to boost feelings of well-being, leading to a virtuous circle. Two studies provide empirical support for this: Berry (2003) found correlations in the range of .4 to .5 between measures of psychological and sociocultural

65 adaptation; Ward and Kennedy (1999) also found them to be positively correlated.

Moreover, the strength of the association was positively related to with cultural proximity to, and integration in, the host society (Ward, 1999; Ward & Kennedy, 1996;

Ward & Rana-Deuba, 1999).

Several studies on cross-cultural adjustment have focused on differentiating psychological and sociocultural forms of adjustment. Berry and colleagues (2002) argued that, while empirically related to some extent, psychological and sociocultural adaptation are distinct in time-scale and causation. Psychological problems often increase soon after contact, declining over time, whereas sociocultural adaptation typically proceeds linearly. Likewise, psychological adaptation is mainly predicted by personality and social support, whereas sociocultural adaptation is predicted by other factors: cultural knowledge, degree of contact, and intergroup attitudes (Berry et al.,

2002).

In order to elaborate and test the distinction between psychological and sociocultural adjustment Searle and Ward (1990) investigated a sample of Malaysian and Singaporean students in New Zealand. Depression (a psychological measure) and social difficulty (a sociocultural one) were tested for their relationship to expected difficulty, cultural distance, quantity and quality of extraversion, life changes, and other variables. They found that social difficulty, satisfaction with host contacts, extraversion, and life changes accounted for 34% of the variance in psychological adjustment and cultural distance, expected difficulty, and depression accounted for 36% of the variance in sociocultural adjustment. No significant differences were found for either psychological or sociocultural adjustment on length of residence. They concluded that satisfying interactions with members of the host culture predicted psychological adjustment. More specifically, they found that while the quantity of host national

66 contact predicted socio-cultural adjustment, it was the quality of the contact that predicted psychological adjustment. For relationship satisfaction implied adequate social support, and relationship dissatisfaction implied increased psychological distress—findings that are congruent with the clinical literature on social support and mental health (Ward & Kennedy, 1992). In contrast, Ward and Kennedy (1993) found that it was relations with co-nationals which form the important social support network, and that it was satisfaction with these relations that underpins psychological adjustment.

Acculturation strategies and adaptation. Any acculturation strategy that migrants prefer can be experienced as stressful if they have to strive hard to realise their aims regarding successful adaptation to a new culture (Berry, 1992; Berry & Sam,

1996). In her review of research on acculturation, Ward (1996; Ward & Kennedy, 1993) found that both psychological and sociocultural adaptation are helped by adopting an acculturation strategy of integration and by minimising cultural distance. Research relating adaptation to acculturation strategies allows for some generalisations (Berry,

1997; Ward, 1996). In most cases, the integrated appear to be best adapted, the marginalised are worst adapted, and the assimilated and the marginalised lie in between.

Other research into acculturation strategies and psychological adaptation as an outcome variable has been contradictory. Sing (cited in Ourasse & van de Vijver, 2005) found that identification with the host culture was positively related to stress; Padilla

(1986) found the opposite. Ghaffarian (1987) found that identification with the host culture was negatively related to depression; Kaplan and Marks (1990 found the opposite. Buriel, Calzada, and Vasquez (1982), moreover, found that a balanced identification with both cultures was positively related to psychological adjustment;

Ward and Kennedy (1994) found differently: identification with the host culture was

67 positively associated with sociocultural adjustment, but identification with the migrant culture was negatively related to psychological problems.

For many researchers, integration is the preferred acculturation strategy, mainly because it is seen to promote mental health (Niles, 1999). They find, for example, that most migrants to Belgium and the Netherlands want to combine their native mainstream cultures (e.g., Phalet & Hagendoorn, 1996; Van de Vijver, Helms-Lorenz, & Feltzer,

1999). Likewise, the psychological adjustment of migrants to Canada could be ranked from highest to lowest as follows: integrated, assimilated, separated and marginalised

(Berry, Kim, Minde & Mok, 1987; Berry & Annis, 1974)

3.4 Individualism versus Collectivism

This section examines the individualism–collectivism dimension that researchers such as Hofstede (1980, 1984) have used to characterise cultures. Individualism– collectivism is not the only dimension that has been used in this way (Hofstede initially defined four) but it holds special relevance for this thesis because it parallels the dimensions by which the dialogical self is characterised, and hence is discussed in depth here and referred to in later chapters.

Using the previously mentioned self-confrontation method (Hermans & Hermans-

Jansen, 1995), the dialogical self can be characterised in terms of two motives, self- enhancement and union-with-the-other, whereas cultures can be characterised as

“individualistic” or “collectivist”. Although the parallel is not straightforward, it suggests that there is a link between self and identity on the one hand, and migration and acculturation on the other.

Anthropologists and cultural psychologists have long argued that personhood is constructed differently in different cultures (for instance Shweder & Bourne, 1984).

Geertz (1975) noted the distinctive features of the Western conception of personhood:

… the Western conception of the person as a bounded, unique, more or less integrated motivational and cognitive universe, a dynamic centre of awareness, emotion, judgement and action organised into a distinctive whole and set contrastively both against other such wholes and against its social and natural background is, however incorrigible it may seem to us, a rather peculiar idea within the context of the world’s cultures. (p. 48)

In the West, the self is generally conceived of as autonomous and separate from other people or social contexts (Geertz, 1975). Non-Western countries tend to stress group solidarity and connectedness, viewing the self as embedded in social roles, duties, and responsibilities. Shweder and Bourne (1984) suggested that the concept of the person varies across cultures. They distinguish two main types of personal organisation:

“egocentric-contractual” (individualistic, subjected to and limited by sociocultural rules), and “sociocentric-organic” (outward-directed, closely integrated with the surrounding environment). Various other distinctions and perspectives exist: achievement orientation and affiliation orientation, self-directed and other-directed or field dependent.

For Sampson (1988), individualism is more a philosophical and practical principle than a universally espoused standard of social life. Referencing Heelas and Lock (1981), he observes that all cultures distinguish self and the nonself, but do so very differently.

American society promotes a self-contained individualism that fosters individual freedom and achievement, whereas “elements of Japanese, Chinese, and Islamic indigenous psychologies offer ensembled type [sic] of individualism” (Sampson, 1988, p.17). Other researchers have largely concurred. For example, societies can be characterised by their location on the individualism-collectivism dimension (Triandis,

Bontempo, Villareal, Asai, & Lucca, 1988). According to Hofstede (1980, 1984), in individualistic societies, individuals are expected only to look after themselves and

69 immediate family; in collectivistic societies individuals are integrated into cohesive in- groups that protect them in return for group loyalty. In such a culture the individual’s

“selves” largely derive from, and are sustained by, the groups to which they belong.

Research by Hofstede (1980), Triandis (1990), Triandis, Bontempo, et al. (1986) and Triandis, Bontempo, Villareal, Asai, & Lucca (1988) found collectivism to be highest in traditional societies, and in East Asia and most of Africa, Asia, Latin

America, and the Pacific Islands. Hofstede (1980) found that the most individualistic countries were the United States, Australia, and the UK,. Moderate levels of individualism are found in the rest of Europe. In a refinement of Hofstede’s (1980) conceptualisation of individualism and collectivism as bipolar opposites, Triandis et al.

(1986) found that at the cultural level, each construct of individualism and collectivism consists of two factors. Individualism he defines as separation from in-groups, plus self- reliance combined with hedonism; collectivism he defines as family integrity, plus interdependence combined with sociability. They suggested that individualism and collectivism are not opposite poles of a single bipolar construct, but are independent constructs.

Triandis, Leung, Villareal, and Clark (1985) distinguished between cultural individualism and collectivism, and psychological individualism and collectivism, which they term allocentrism and idiocentrism. Allocentric individuals subordinate their personal goals to the interests of their collectivity and may derive satisfaction in doing so. Idiocentric individuals, on the other hand set their personal goals above the interest of any collectivity. While allocentrics in individualist cultures join communes and other collectives, idiocentrics in collectivist cultures try to escape the dominance of their in- groups (Triandis et al, 1985).

Along similar lines, many researchers contrast Western conceptions of the self, which emphasise the separateness of the individual, with non-Western conceptions that emphasise relatedness. For example, Markus and Kitayama (1991) found that the

Western cultures seek to maintain the independence of the individual as a separate, self- contained entity; individuals are socialised to be unique, to express themselves, and to realise and actualise their inner self and promote personal goals. The emphasis is on personal attributes, demonstrated in public and confirmed in private through social comparison (Shweder & Bourne, 1984), and to “be true to one’s own internal structures of preferences, rights, convictions, and goals” (Markus & Kitayama, 1994, p.569).

Non-Western cultures, by comparison, do not value overt separateness but tend to emphasise the “fundamental connectedness of human beings” (Matsumoto, 2000).

Under the interdependent self-construal (Markus & Kitayama, 1991), individuals focus on their interdependent status with other people: the self is unbounded, flexible, and contingent on context. Furthermore, the most salient aspects of the self are defined in relationships—that is, those features of the self related to and inseparable from social contexts. The overall aim is to maintain social harmony.

Pervin (1999) observed that although the distinction between the independent and interdependent self is helpful, it is far from the whole story. Selves are not dichotomously independent or interdependent; they may be both (Singelis, 1994). That is, collectivism and individualism are two dimensions, not opposite ends of a single dimension (Rhee, Uleman & Lee (1996). Moreover, there is considerable variation in self-construal within a single culture. Matsumoto (2000) points out that the concept of independent versus interdependent selves is not unlike other dualities of the self and human nature, including Freud’s (1930/1961) “union with others” versus egoistic happiness, Balint’s (1959) ocnophilic and philobatic tendencies, Bakan’s (1966)

71 communion and agency, Boen’s (1966) togetherness and individuality, Bowlby’s (1969) attachment and separation, Franz and White’s (1985) individuation and attachment,

Stewart and Malley’s (1987) interpersonal relatedness and self-definition, and Slavin and Kriegman’s (1992) mutualistic and individualistic urges (all cited in Guisinger &

Blatt, 1994).

In elaboration of this notion, Guisinger and Blatt (1994) acknowledged that although mainstream psychology has historically placed more importance on self- development for autonomy than for interpersonal relatedness, the two processes are not mutually exclusive. Subsequently, Niedenthal and Beike (1997) proposed the existence of both interrelated and isolated self-concepts. Like Guisinger and Blatt (1994), they argue that these concepts exist not as dichotomies but as interrelated dualities, and that an individual can possess both self-concepts in the same domain simultaneously.

According to Matsumoto (2000) the implication of the existence of these dualities means that cultures may emphasise both types of self-construal rather than only one.

Finally, Hermans and Kempen (1998) argued that

Globalization involves social processes that are complex and laden with tension. These processes fall squarely outside the scope of cultural dichotomies, which by their nature are oversimplifying and insensitive to the apparent tensions that are so typical of the relationships between cultural groups. (p. 1112).

They concluded that theories involving cultural dichotomies are flawed since they take cultures to be homogeneous, distinct, and geographically delineated. According to

Hermans and Kempen, globalisation shows that cultural identities are neither fixed nor static, but fluid, dynamic, negotiable, and in flux; they cite the example of individuals who grow up in bicultural environments and experience cultural dialogues within themselves that represent a dynamic and interactive multiplicity of different and contrasting voices and roles.

Hermans is not without his critics. Roland (2001), a psychoanalyst who has worked in India and Japan and with patients from diverse cultures and ethnic backgrounds in New York, has expressed his reservations in these words:

[Hermans’] dismissal of the work of Markus and Kitayama (1991), Marsella (1983), Shweder and Bourne (1984), Triandis (1980) and others as too dichotomous, as being too governed by a Cartesian dualist of the ‘West against the rest’, is really a continuation of his agenda to flatten out radically different cultural/psychological worlds (p. 317).

He notes that his own research in inter-Asian analyses indicated that “radically different cultural psychological worlds do exist” (Roland, 1988).

Tweed, Conway and Ryder (1999) have also responded to Hermans and

Kempen’s (1998) criticism that mainstream cross-cultural psychology sees cultural differences in terms of strict, stable dichotomies defining homogeneous groups. They note that Triandis’s (1989) research “has already hinted at the complexity of self and identity with his assertion that each individual has multiple selves from which to sample” (p. 838). In terms of cultural conceptions of the self, Triandis (1989) argued that there are three kinds of self-conceptions: private, public, and collective selves.

Private selves are the aspect of self-conceptions that are shown and known to oneself; public selves are those aspects that are presented to others; and collective selves represent the self as members of various groups. Markus and Kitayama’s conceptualisation of interdependent self-construal “conflates Triandis’s public and collective selves into one type of conception of the self that emphasises its social and contextual embeddedness” (Kashima, 2001, p.213). Kashima further notes that the argument has been that Triandis’s public and collective selves, and Markus and

Kitayama’s interdependent self-construal, are more prevalent and emphasised in collectivist cultures such as East Asia and Latin America, while public selves and

73 independent self-construal are more prevalent and emphasised in individualist cultures such as Western European-based cultures including North America and Australia.

Triandis’s (1989) differentiation of the self along three dimensions has been further developed by other researchers. For example, Brewer and Gardner (1996) proposed a distinction between three sources of self-representations: the personal, the interpersonal, and the collective. They propose that

At the individual level, the personal self is the differentiated, individuated self- concept most characteristic of studies of the self in Western psychology. At the interpersonal level, the relational self is the self concept derived from connections and role relationships with significant others. This corresponds most closely to the interdependent self as defined by Markus and Kitayama (1991). Finally, at the group level, is the collective self, which corresponds to the concept of social identity as represented in social identity theory and self-categorisation theory. (Turner, Hogg, Oakes, Reicher, & Wetherell, 1987, p.84)

Along these same lines, Kashima (2001) postulated a concept of self in relation to

“at least three major classes of other entities: goals, individuals, and groups” (p. 214):

A self in relation to a goal is an agentic self; a self in relation to another individual is a relational self; and a self in relation to a group is a collective self. These three aspects constitute three components of the meaning of a self. (p. 214)

Recently something of a consensus has developed. Most researchers now agree that all cultures accommodate multiple self-construals of the self to varying degrees, and that different selves manifest in different contexts (Church, 2000).

3.5 Summary

This chapter first described Australia’s migration program: its history, the policies that shaped it, and the social forces that sustained it—and still do. The chapter then explored the psychological aspects of the acculturation that commonly accompanies

74 migration. James and the early psychologists paid scant attention to acculturation, probably because they lived in an era of limited international migration by the bourgeoisie, who were then the only ones to attract the serious attention of the psychologists. Later, anthropologists and sociologists took an interest, but did so from the perspective of their own disciplines. Arguably, Berry was the first to develop a psychological model of acculturation, which subsequently attracted great attention from psychologists and others; and for that reason it is explained at length. Also explained are the many models that succeeded Berry but which, like Berry, are essentially static in nature and not dynamic. The next chapter demonstrates how the theory of the dialogical self—as a dynamic, process-oriented model—has the potential to address many criticisms of Berry’s model and those of his successors.

CHAPTER 4: The Dialogical Self and Acculturation

The two preceding chapters described theories of the self and acculturation independently. This chapter brings them together by showing how one particular theory of the self, the dialogical self, offers a framework to analyse and explain the psychological dimensions of acculturation in a way that Berry’s acculturation model

(and other models constructed along similar lines) do not.

The chapter opens with a discussion of the phenomenon of globalisation, for it is this that has caused researchers from Hermans (1998) onwards to question previous models of acculturation. Next follows a section showing how dialogical self theory accounts for aspects of acculturation in an era of globalisation—in particular cultural continuity—that were troubling for previous models. The chapter concludes with the research questions addressed in this thesis, having first placed them in the context of the theories of migration and acculturation.

4.1 Globalisation and the Illusion of Stability

Globalization is “a process by which cultures influence one another and become more alike through trade, immigration, and the exchange of information and ideas”

(Arnett, 2002; p. 774). Although globalisation began centuries ago with the rise of the great trade-driven European maritime empires (Friedman, 2005), it has come to the fore in the public consciousness only in recent decades (Kinnvall, 2004; Hermans &

Dimaggio, 2007). But for all its boons, globalisation has a “shadow side”, to which

Hermans and Dimaggio attach the descriptor of “uncertainty”. It is against the backdrop of globalisation, and the psychological influence that it exerts, that traditional models of acculturation have been found wanting, and which dialogical self theory addresses.

The challenge of globalisation. By revealing weaknesses in previous acculturation models, globalisation has thrown up challenges for cultural psychologists.

According to Kadianaki (2009; p. 477), whereas migration is “complex and dynamic”, past theories focus “on states rather than processes”; and what is needed instead is a theory that captures “life histories of individuals as they develop in space and time”.

Dialogical self theory addresses these issues: as a theory it is inherently process- oriented (Andreouli (2013), and as a metatheoretical substrate for a range of psychological “complex and dynamic” phenomena such as acculturation, it lends itself to ready elaboration (Hermans & Dimaggio, 2007). It is, thus, a “language” in which dynamic models and theories can be couched.

The remainder of this section makes the case that previous models, being static and dichotomous in nature, do not adequately account for cultural identity on a world increasingly populated by cultural hybrids.

Blurred boundaries: the hybridised identity. The impact of globalisation on culture is nowhere better captured than in the concept of the “hybridised identity”. The term seems to be of recent origin, along with globalisation itself. Over a century ago,

James (1890) recognized that the self is shaped by cultural and community, but wrote little on the subject. A century later, Bruner (1986, 1990) provided a cultural perspective, as did Hermans, Kempen, and van Loon (1992), who argued that ”the concept of the dialogical self is based on the notion that people have told each other stories and listened to stories across cultures and across times” (p. 108; emphasis added).

In this vein, Hermans (2004) observed that cultural identities and boundaries are far from fixed. With the metaphor of the dialogical self as a “society of mind”, he highlighted its similarity to society itself, replete with shifting coalitions, tensions and

77 factional conflict (Hermans, 2002a). This multivoiced aspect of the dialogical self therefore renders it particularly suitable for analysing cultural identity (Hermans &

Dimaggio, 2007) because the pace of social change has led to new forms of cultural identity and “hybridized” identities (Hermans & Kempen, 1998):

...[such] as Mexican school girls dressed in Greek togas dancing in the style of Isadora Duncan, a London boy of Asian origin playing for a local Bengali cricket team and at the same time supporting the Arsenal football club, Thai boxing by Moroccan girls in Amsterdam, and native Americans celebrating Mardi Gras in the United States. (p. 1113).

Hybridised identities are not a developmental end-point but rather they shape a person’s dialogical negotiations in the context of racism, gender, imperialism, and power (Bhatia & Ram, 2001; Bhatia, 2002; Da Cunha, 1995). Hermans (2001a) explained how dialogical self theory accommodates cultural misunderstanding:

When people are raised in one culture and then migrate to another, they arrive in a situation in which two or more heterogeneous internal positions (e.g., I as Egyptian and I as Dutch) interact with a multiplicity of very heterogeneous external positions (e.g., the family of one’s culture of origin and individuals and groups representing the host culture). Such positions (e.g., Egyptian versus Dutch) may be felt as conflicting or they may coexist in relatively independent ways or even fuse so that hybrid combinations emerge in the form of multiple identities (p. 258).

It is the multi-faceted nature of cultural identities (Hermans & Kempen, 1998) that prompts the question of “how people, involved in a process of acculturation, organize and reorganize their self-system in such a way that they are able to share with other people cultural elements that may be highly divergent, partly unknown and laden with power differences” (Hermans, 2001a, p.258). For Hermans, acculturation is an

“intercultural phenomenon par excellence because it takes place on the contact zone between someone’s home culture and host culture” (p. 269).

Hermans (2001a) noted that Bhatia and Ram (2001) called into question Berry’s

(1980) acculturation model and strategies:

One of the main assumptions is that, although there are considerable variations in the life circumstances of the cultural groups involved, the psychological processes that operate during acculturation are essentially the same for all groups. (pp. 269- 270)

Berry, as has been stated, claimed that his “integration” strategy was optimal for the acculturating migrant since it provided the best social support, allowing involvement in the host culture while maintaining connections to home (Berry, 1980). But Hermans

(2001a), again echoing Bhatia and Ram (2001), observed that Berry’s model mistakenly implied that the psychological processes were the same for migrants from Western

Europe as for those from former colonies. Berry’s model also considered integration “to be a linear trajectory that leads to an end-goal” (Hermans, 2001a; p.271), but did not show how that goal could be achieved given the power and status differences between the majority and minority cultures.

In his review of Hermans (2001a), Roland (2001) agreed with Hermans in rejecting cross-cultural psychology’s “assumption that [acculturation] is the same for everyone and that everyone comes to it with the same psychological make up” (p. 318).

He further agreed that it is the process of acculturation that is more important than its outcome or endpoint. Roland only questions whether an endpoint is ever achieved, as

“there is always the issue of how much of the indigenous self is kept or disavowed in relationship to a newer self in the eventual makeup of a bicultural self” (p. 318), especially when a person migrates from a different “civilisation”.

Nevertheless, Adams and Markus (2001) acknowledged that the dialogical self model, with its emphasis on the multiplicity of identity and the agency of the self as a constructor of identity, makes an important contribution as a model of identity negotiation. They saw the model as illustrating how the self functions as a “zone of proximal development”, where both personal-identity and cultural-identity categories

“make each other up” (p. 293). The dialogical self supersedes the one-dimensionality of other paradigms, such as the acculturation studies paradigm (Berry & Sam, 1997), which view cultures as mutually distinct and internally homogeneous, and presupposes a linear conception of cultural change. In the dialogical self paradigm, identities can be hyphenated and hybrid constructions made of heterogeneous and contrasting elements.

Lastly, Falmagne (2004) welcomed the dialogical self model for its analytical sophistication in respect of the self, although she criticised it for its lack of a societal- level frame of analysis. The model sees the self as “continuous with culture” (Hermans,

2001a), but according to Falmagne it has little to say about the politics of culture

(Squire, 2000), and the societal system of which culture is a constituent.

False dichotomies and reification. Thus far, traditional acculturation models have been criticised on empirical grounds. Criticism has been levelled on methodological grounds as well. Traditional acculturation models, it is said, force acculturation into a false dichotomy. This leads to the logical fallacy of reification, whereby a category that is implied by theory is treated as if it existed in reality when in fact (given that the theory is incorrect) it does not (Schwartz, 2010).

Foremost among these critics is Rudmin (2003), who observed that a common and serious failing of theorists was to ignore the obvious fact that in any acculturative setting there is not one culture but two; so each of Berry’s four acculturative strategies

80 applies not only to the minority culture but to the majority culture as well, making 16 possible combinations, not the mere four that Berry and his fellow theorists discuss.

To suggest that minorities are psychologically reactive to intercultural contact and that dominant groups are not almost implies that minority people are a different species of psychological being, one distinct from the majority…. To the contrary, as a result of the speed and ease of world travel, global communications, and international marketing, all humans, everywhere, are subject to acculturation processes, whether they know it or not and whether they like it or not (p. 6).

He went on to criticise researchers’ lack of logical rigour, which explains why their studies have variously found measures of incompatible acculturative attitudes to be positively correlated, and why bicultural integration and marginalisation are confounded constructs, and why there is no robust evidence that biculturalism is most adaptive. This latter presumption he found so irksome that he was moved to complain that “studies showing minorities to have superior health are called ‘counter-intuitive’ and a

‘paradox’…. In fact many studies using large samples show immigrants to have superior mental health…” (Rudmin (2010b, p. 302). He also drew attention to other errors and claims that acculturation research has produced little and has been guilty of confused concepts, circular reasoning, wrong research questions, and “mistaken citations of mistaken interpretations of studies using mistaken measurements” (Rudmin

(2010a; 2010b, p. 299).

Gillespie, Howarth and Cornish (2012) have also criticised traditional acculturation models on methodological grounds. Social categories, they claim, are

“perspectival” in that categorization stems from social position, history and particular interests, none of which is objective; social categories are “historical” in that they and their members are in constant flux; social categories are blurred by the movement of people between categories; social categories are “re-constitutive of the phenomena they

81 seek to describe” (p. 1) because people react to categorisation; and social categories oversimplify by blurring intra-group differences and accentuating inter-group differences.

Even dialogical self theory itself has not escaped methodological criticism, though unlike the criticism directed at older models, it is not deemed to be fatal. Adams and

Markus (2001) voiced misgivings as to “whether the dialogical self meets the objective of a non-reifying account of culture and self” (p. 284). They observed that Hermans’

“treatment of culture as positions” as in itself entailing the “reification of dynamic flowing [cultural] patterns into static, fixed entities” (p. 284). Further, they see

Hermans’ notion of positioning as based on the premise that “selves and ‘cultures’ can be represented or ‘fixed’ as locations in space” (p. 284); and that the Hermans’ metaphor of dialogue “restricts cultural influence to knowable, objectified ‘me’ aspects of the Jamesian self and appears to neglect or understate the possibility of cultural influence on knowing, subjective ‘I’ aspects”, with the implication of “an essential, core self—whether unitary and continuous or multiple and discontinuous—that is insensitive to cultural variation” (p. 284). Notwithstanding, they conclude that

As a model of identity negotiation, Hermans’ dialogical self makes important contributions: it emphasizes the multiplicity of identity, highlights the agency of the self as a constructor of identity, and suggests the importance of psychology— and the study of self, in particular— for the study of culture (p. 283).

In summary, researchers have criticised traditional acculturation models on two main grounds, empirical and methodological. At the empirical level, researchers have argued that traditional models do not adequately fit the facts since they cannot accommodate the blurred boundaries of cultural hybrids; and at the methodological level, researchers have argued that traditional models and the methods by which they are applied are logically flawed.

4.2 A Dialogical Theory of Acculturation

The previous section made the case that traditional acculturation models do not adequately explain acculturation. This section makes the converse case by showing how dialogical self theory explains the epiphenomena of acculturation, such as localisation

(as a counterforce to globalisation), asymmetric power relations, and cultural continuity.

It is argued that all such epiphenomena have as their driving force a single impetus: the desire by migrants and others to mitigate the debilitating and disturbing impact of uncertainty.

Uncertainty and its discontents. Uncertainty has long been recognised by all branches of the social sciences as undesirable and undesired. For instance, Hofstede

(1980) recognised uncertainty avoidance as one of his cultural dimensions, and in the field of cross-cultural psychology, Hermans (1999) linked the discontents of uncertainty to globalisation:

Globalization is easily understood as contrary to living one's “authentic life” in peace, partly because authenticity and pureness, however conceptualized, is better suited to a homogeneous, stable, localized, and predictable society than an increasingly heterogeneous, changing, translocal, and unpredictable global world. (p. 1118).

He went on to argue that uncertainty and the strategies mitigating it must be understood in a dialogical framework; and that “uncertainty is not primarily in a culture's core but in its contact zones” (pp. 1119), where migrants meet their hosts. In his later studies, Hermans (2001a; Hermans & Dimaggio, 2007) saw the progression towards hybrid identity in the following terms: globalization leads to identity hybridisation, which leads to uncertainty, which leads to anxiety. The logic of this progression is found in dialogical self theory. In dialogical terms, the hybrid cultural

83 identity is characterised by an unprecedented density and heterogeneity of I-positions, and the self is subjected to larger “position leaps” than ever before.

To recap, uncertainty is an unavoidable outcome of globalisation, particularly for migrants, who themselves are a product of globalisation. But uncertainty alone does not create anxiety; for that to happen there must also exist the possibility of undesired outcomes among life’s many uncertainties. This is where power relationships enter the argument, for it is the asymmetry of power as between migrant and host country that diminishes the control that migrants have over their lives, and hence the likelihood of undesired outcomes. Therein lies anxiety.

Power relationships. Disparities of power colour the acculturation process

(Hermans (2001). Migrants find themselves in a strange and unfamiliar environment where even the simplest exchanges can conceal pitfalls as words are misunderstood or unintended offence given. Hermans (2001) depicted the problem in dialogical terms by characterising cultures

...as collective voices that function as social positions in the self. Such voices are expressions of embodied and historically situated selves that are constantly involved in dialogical relationships with other voices. At the same time these voices are constantly subjected to differences in power. (p. 272, italics added)

Identity formation takes place by means of an internal dialogue between personal and collective voices, in which external power relations play a role (Hermans &

Hermans-Konopka, 2010). Psychological and cultural processes are moreover intimately connected, such that cultural diversity is reflected in the self.

Subsequent researchers have expanded on this theme. Kinnvall (2004) observed that migration “is often characterized by a sense of powerlessness and dependence as insecurity is increasing among many migrants. This is frequently mixed with an acute anxiety about their new circumstances and strong feelings of homelessness” (p. 747).

Bhatia (2002) and Chaudhary & Sriram (2001) drew attention to the role of race and the history of colonialism that colours much North–South migration. Raggatt (2007) identified power relationships as central to his taxonomy of positioning in the dialogical self. Lastly, much of the work of O'Sullivan-Lago (2011a; 2011b; O’Sullivan-Lago & de Abreu, 2008; 2010; O’Sullivan-Lago, de Abreu & Burgess, 2008), which concerns immigrants to Ireland, is conducted against a backdrop of power asymmetry, exacerbated by the inclusion of asylum-seekers in her sample (see also later section on cultural continuity).

Power relationships are of importance because of their function—or rather, lack of one—in traditional acculturation models. Models such as that of Berry (1980) are limited largely because they do not recognise and formally incorporate the reality of power relationships. According to Bhatia and Ram (2001), such models implicitly assume that psychological processes are the same for all migrants, weak and strong.

Migrants, it is supposed, are free to choose a preferred acculturation strategy, integration being touted as the one with the best outcome in terms of mental well-being.

This, argue Bhatia and Ram, simply misrepresents the reality: asymmetries of power and status can often rule out integration, and sometimes even assimilation, leading some migrants to choose separation and leaving others to be marginalised.

In dialogical terms, migration is often less a “blissful marriage of … cultures” and more a series of “contested, negotiated and painful, rupturing experiences associated with living between cultures.” (Hermans, 2001; p. 271). When people migrate, they naturally take with them an existing heterogeneous set of internal I-positions, developed in their home country in dialogue with its culturally specific external I-positions; which then confront an equally heterogeneous, but thoroughly unfamiliar, multiplicity of external positions I-positions from the adoptive culture. In Hermans’ (2003) words:

Voices from the culture of origin do not simply disappear when people are involved in an acculturation process. Instead, the older or deeper voices are often established parts of the self, and they are challenged, evoked, repressed, or simply ignored (p. 95).

Strategies for uncertainty mitigation. According to Hermans and Dimaggio

(2007), in an effort to calm the anxiety arising from the turbulence and unpredictability of globalisation, people may resort to “localising” and “homesteading” (best seen as a type of localisation) as a counterforce to the globalising forces they see all around. The authors couch their explanation in terms of the dialogical self, which, they argue, is the appropriate theoretical framework for the task. They grant that dialogical self theory is not the only theory that claims to explain localisation: the others are social identity theory, in which membership creates self-categorization to favour the in-group; optimal distinctiveness theory, in which people seek a balance between desire to ﬁt in or stand out; and intersectionality theory, in which race, class, and gender are seen as intersecting and interlocking. But in their view, none of these alternative theories adequately explains the epiphenomenon of localisation.

Hermans and Hermans-Konopka (2010) identify “Five strategies to cope with the heightened level of uncertainty” (pp. 44–47). The first is to reduce the number of a person’s I-positions and their heterogeneity—to simplify life and so restore tranquillity.

The second strategy is to let a single, powerful I-position take the lead; leaders may rely on this in their followers either for good (when they termed “inspirational”), or ill (when they are termed “populists”). The third strategy is to delineate more firmly the boundary between in-group and out-group (“We are not like them”), whereby multiple internal I- positions effectively become external and so can be subordinated and subdued. The fourth strategy is to augment the number of I-positions by adding others that provide solace—a strategy not without risk as the same turbulent I-positions remain in place.

Lastly, the fifth strategy is what Hermans terms “going into uncertainty”, that is, to confront and wrestle with uncertainty dialogically, as one would in constructive debate.

Significantly, all five strategies draw on dialogical self theory, but in different ways. Strategies one through four are dialogical only in the sense that they intentionally manipulate the number and type of I-positions. The fifth strategy is exceptional in that the intention lies in the dialogue, not the I-positions that may or may not result. Any change in I-positions is therefore unintentional in the sense of being unpredicted; what is intentional is the dialogue that causes them. The I-position changes are necessarily unpredictable, emerging perhaps serendipitously, for if not they would have been managed already by one or other of the other strategies. Therein lies the real power of dialogical thinking.

Kinnvall (2004; 2010), like Hermans, argues that globalisation diminishes people’s sense of security. Majority and minority communities in Europe have reacted by searching for one stable and secure identity, which she terms “the securitization of subjectivity” (Kinnvall, 2004; p. 763). One strategy for achieving this aim—in essence, localisation—is what Kinnvall calls “homesteading”, a term coined by Kronsell (2002) for the different physical circumstances of an academic establishment, but co-opted by

Kinnvall to describe “making and shaping a political space for oneself in order to go beyond and surpass the life of contradictions and anxieties that characterizes homelessness” (Kinnvall, 2010, p. 601). Another, topical, example (not Kinnvall’s) is provided by the independence movements of such places as Catalonia and Scotland, which can be interpreted as localising resistance to the globalising impact of the Spanish and British states. And though it seems paradoxical that both regions should so strongly support membership of the European Union, the paradox is resolved when if Brussels is perceived as a more distant, and therefore acceptable, alternative to the stifling embrace

87 of the mother country, providing enough political certainty to mitigate the uncertainty and insecurity of smallness.

4.3 Cultural Continuity

Thus far, the discussion has centred on the mechanism by which globalisation generates anxiety, and on the means of mitigating that anxiety. This section demonstrates how this process commonly manifests itself in practice: the search for cultural continuity, that is, the efforts that migrants (and more broadly anyone experiencing acculturative stress) make to preserve their mother culture, sometimes with success, sometimes not.

Fear of fragmentation. Culture metaphorically cradles the self in the sense that it creates the “others” to which the self must relate dialogically, and so generates many of the external I-positions that constitute the dialogical self (Salgado and Hermans, 2005).

And in so doing, it shapes and moulds the self in its image—or a reflection of it. But attractive as this metaphor is, it does not of itself explain what happens to the self when culture changes. For whatever else takes place, migrants (except in exceptional and possibly pathological cases) do normally preserve their sense of personal continuity. It seems, then, that individual and cultural continuity are intimately entwined in the sense that the one depends very much on the other. It is for example impossible to imagine cultural continuity without individual continuity (or indeed any kind of continuity for that matter). Conversely, it may not be impossible, but it is certainly problematic, to imagine individual continuity without cultural continuity.

This theme was ably explored by Chandler and Lalonde (1998; 2004; 2008) in their research into First Nations of Canada. Like many indigenous communities, the

First Nations are migrants of a kind—not geographical but temporal: they stayed put while the world changed around them. But the acculturation problems they faced were

88 nonetheless real, and the worse for being involuntary and infected with the power dynamics of racism. In an extensive survey of First Nations communities, Chandler and

Lalonde (2008) found that those “that preserve heritage culture, and control destinies, are dramatically more successful in insulating their youth against the risks of suicide”

(p. 6). The reason is that the adolescents of communities, already at an age of precarious selfhood, see no future for themselves, literally, unless sustained by the visible and emotional signifiers of cultural continuity. Chandler and Lalonde (2004) state:

If, owing to some train of personal or collective mishaps, single individuals or whole communities lose track of themselves in time, and so suffer some disconnect with their past or future, then old responsibilities and new promissory notes fly out the same window, and life, we argue, becomes cheap. (p. 3)

It remains to explain this phenomenon in dialogical terms.

Salgado and Hermans (2005) argued that the self exists only by virtue of its context. Rejecting both Thomist monism and Cartesian dualism, they see the self as an emergent property of dialogism: “the I emerges by reference with an Other” (p.10). In this way they reconcile the notion that the self must change as its dialogical others change. The implication is that if one’s culture ceases to exist, then the external I- positions that it sustains likewise are extinguished, which goes to the very core of selfhood.

An alternative, but still consistent, explanation is afforded by the phenomenon of fragmentation. Fragmentation is one form of dissociation, the others being omission, subduing and distortion (Hermans & Hermans-Jansen, 1995). Unlike the postmodern self, the dialogical self does not normally give rise to fragmentation; instead, it bridges the gap between the modern ideas of discontinuity and continuity (Hermans (2001a).

Voices from different I-positions may disagree, but they are normally in dialogue, that is, the oppositional I-positions acknowledge the other’s existence and may compromise.

Where the strains are too great, however, fragmentation may occur, such as that experienced by Chaudhary (Chaudhary & Sriram, 2001), but it falls far short of the pathological:

Whatever I do or say… either exotic or weird… I resist, sometimes consciously, sometimes unknowingly… but the foreign culture is affecting my psyche… nothing I do is irrelevant, it is either a reaction or an act of conformity… sometimes I go out of my way not to resist… I want to be open… beyond culture… (p. 383)

In summary, cultural and personal identity are inseparable but it is nevertheless possible for the self to persist throughout the acculturation process and still retain its sense of continuity. It does so by means of dialogue with multiple cultures,. Strategies for doing so are described next.

Strategies for engaging alterity. In a program of research spanning two decades,

Bhatia and colleagues explored the acculturation of mainly South Asian migrants to the

United States. What they found resembled less a linear transition from monoculture to one of Berry’s four acculturated states, than a dialogically mediated struggle.

Bhatia & Ram (2001; 2007) found that race, gender, imperialism and power relations coloured and shaped cultural choices. Migrant women in particular, including second-generation daughters, were caught in a dynamic of push-and-pull dialogical negotiation, alternately tugged this way and that by their separate cultural identities.

According to Bhatia & Ram (2001):

These negotiations are multi-layered and complex because the voices of parents, peers, language, siblings, homeland and American society are all represented in the dialogical self. (p. 305)

In an investigation of how diasporic women negotiate multiple and conflicting cultural identities, Bhatia and Ram (2004; 2007) found that they adopted three

90 dialogical strategies: polyphonization, expropriation, and ventriloquation. By

‘polyphonization’ is meant a strategy whereby selves oppose each other’s voices in a complex dialogue that despite its contradictions does not fragment; by ‘expropriation’ is meant a monologue whereby social institutions that demands total obedience to social norms and institutions attempt to suppress dissenting voices, usually without success; and by ‘ventriloquation’ is meant the phenomenon of one voice infusing the other voice to express its own voice or I position. These are all dialogical strategies that attempt to mitigate (whether in a healthy way or not) acculturative stress; and consequently they demand a dialogical theory of the self in order to be understood.

Bhatia and Ram (2009) describe the experience of South Asian migrants to the

United States during the terrorist attack on the Twin Towers. Having taken for granted that they were “integrated”, they found to their shock that they were not; overnight they came to be treated as the “Other” for no other reason than they looked Arab, at least in some Americans’ eyes. Bhatia (2002; 2003; 2014) echoes this theme by describing

Edward Said’s struggles with forming a coherent identity, and describing the cultural constraints places on migrants in an occasionally racist host culture: “Will we ever belong?” The implication is that “an immigrant’s journey through acculturation is not straightforward, direct, self-evident or ever complete” (p. 147).

Although integration and bicultural competency may be worthy goals to achieve… for most people living in contemporary diasporas, their negotiation with multiple cultural sites is fluid, dynamic, interminable and often unstable. (p. 148)

Homesteading is another dialogical strategy to assuage the existential insecurity engendered by encountering alterity. According to Hermans and Hermans-Konopka

(2010), “A primary way of responding to these experiences is to seek reaffirmation by drawing closer to any localized group that is seen as capable of reducing uncertainty and

91 insecurity” (pp. 41–42). Bhatia (2007) explains how South Asians migrants to the

United States have done so: by re-creating India in the home, through media consumption (mainly movies and music), membership of social clubs etc.

Homesteading, like localisation, is an antidote to the homogenising impact of globalisation in that it sustains external I-positions reflecting the home culture.

Yet another strategy is to develop “adaptive” I-positions, by which is meant ones that are conducive to functioning in the host culture. Aveling and Gillespie (2008) identify three distinct I-positions among adolescents of Turkish origin in the UK, and show that they are adaptive to the fractured and asymmetrical social contexts in which they are embedded. As an example, the researchers show that one such I-position can

undermine these constraining racial chauvinisms, allowing them to build cultural bridges, as, for example, in the diversity of their friendship groups. Thus while reflecting persistent and unresolved dialogical struggles resulting from the asymmetries of power which structure their socio-cultural context, this movement between positions is nonetheless adaptive. (p. 21)

Lastly, Gillespie, Kadianaki, and O’Sullivan-Lago (2012) and O’Sullivan-Lago

(2008; 2010a) draw on the theory of semiotics for ameliorative strategies. Semiotic regulation refers to the way in which meanings and actions regulate the thoughts and actions of self and other:

Encountering alterity cannot be reduced to contact in the sense of co-presence in the same geographical space. For a genuine encounter with alterity to occur, the perspective of the other has to permeate the self and to some extent, transform the self.… In the same way that living tissue expunges alien objects, semantic barriers operate within the semantic realm to expunge discomforting ideas; just as the body regulates what enters it, the self regulates what enters the semiotic realm. (p. 706)

In summary, what these strategies have in common is the attempt to mould the migrants’ mental and physical environment to accommodate the demands placed upon them by the imposition of otherness, and this way to create a familiar and hence restorative mental and physical space.

Strategies for identity maintenance. O’Sullivan-Lago and her collaborators

(O’Sullivan-Lago, 2011a; O’Sullivan-Lago & de Abreu, 2008; 2010a; 2010b) have for many years researched the acculturative experiences of immigrants to Ireland. Ireland has traditionally been a land of emigrants, so to start receiving immigrants was therefore a novel, and to many Irish people, disturbing experience. O’Sullivan-Lago and collaborators saw this therefore as fertile ground for employing dialogical self theory to explore how both immigrants and Irish adapted to the situation.

O'Sullivan-Lago de Abreu and Burgess (2008) describe a strategy whereby both immigrants and Irish adapted an I-position that permitted both groups to accept the dignity of the other. As a basis for their research they drew on Chandler and Lalonde’s

(1998) research into youth suicide in Canadian First Nations communities, which found enormously elevated rates in certain communities, and linked these to a lack of cultural continuity. According to O'Sullivan-Lago et al. (2008), the relevance to Ireland is two- staged: first, “an individual’s decision to end their own life might be made far easier if they were suffering from broken personal continuity because they would lose their personal meaningful connection to who they will become in the future” (p. 350); and second, “we can infer from Chandler’s theory that… cultural continuity has a profound impact at the individual level” (p. 355). Or to trace the causal link from its genesis, cultural continuity affects individual continuity, which in turn affects suicidal behaviour. Completing the syllogism, immigration has caused cultural discontinuity, which has caused, not suicide, but mental ill-health—or would do so but for the

93 ameliorative I-position adopted by both sides: “I as a human being”, the function of which is to maintain cultural, and hence individual, continuity in the face of demonstrated stresses on cultural continuity:

The strategy of evaluating all as equal due to the possession of the ‘I as a human being’ identity allowed each group to attempt to restore their broken continuity. All are, and always will be, human beings despite divisions such as culture and ethnicity. (p. 361)

In an elaboration of this theme, O’Sullivan-Lago (2011a) explored “how solidarity might be formed between the groups now resident in the multicultural space and examine[d] what identity spaces are available that will bridge the gap between self and other” (p. 3.18). Besides the shared-humanity perspective identified by O'Sullivan-

Lago et al. (2008), several other representations of intercultural solidarity were found, concerning religion, collective history, and EU membership. In an analysis of Roma

(otherwise termed Romanies or Gypsies), O’Sullivan-Lago (2011b) explored the same theme from a negative perspective, that is, one where the “I as a human being” position is not established. By constructing the Romany other as less than human… the relationship is no longer dialogical but monological: the possibility of a relationship, of dialogue, is removed” (p. 68). Though not achieved in this case, the shared-humanity I- position is seen as supremely important for dialogue and thereby the maintenance of culture.

O'Sullivan-Lago and de Abreu (2010b) approached cultural continuity from a different angle, by asking not about coping strategies as such (as O'Sullivan-Lago de

Abreu and Burgess, 2008) but about their proposed identity strategies given the cultural futures they foresee for themselves. Accepting the proposition that the dialogical self is a “self-organizing system”, they asked “what happens in the Dialogical Self when individuals experience cultural discontinuity” (p. 284). The answer depended on the

94 group in question. The native-born Irish represented Irish cultural ways as “normal”, which carried the suggestion that migrants should assimilate. The voluntary immigrants stressed their integration into Irish culture but at the same time emphasized the maintenance of their own cultural identity positions—an “integration” strategy in

Berry’s terms. Lastly, the asylum seekers espoused a more or less assimilationist position because they needed to stay in Ireland. What is interesting about these strategies is not that they fit Berry’s taxonomy (which anyway covers the range of what is possible) but that they are negotiated, malleable (the voluntary migrants might want to repatriate, so keep their cultural options open), and above all constrained. All seek to restore continuity, but find different ways to do so given their differing circumstances.

4.4 Research Aims and Questions

Aims. The preceding review of the research to date strongly suggests that dialogical self theory has the potential to explain the nature of acculturation better than previous theories. Supporting evidence is important, given that dialogical self theory is a relatively new model of personality. Quantitative research is particularly needed, to build on existing qualitative research. The present thesis seeks to investigate the migration experience by utilising a dialogical-self perspective to add to this endeavour.

There are two broad aims. One is to test the theory of the dialogical self in a natural context, namely the phenomenon of migration. The logic of the investigation is as follows: the migration experience constitutes a transition in the lives of migrants, which places them under acculturative stress; this cultural rupture, which broadly takes the form of uncertainty, is then manifest in the repertoire of I-positions of migrants when probed using the self-confrontation method. Verifying this analysis would buttress dialogical self theory as both an explanatory and a diagnostic tool in a range of applications.

The second aim of the thesis follows from the first. To recap, the first aim is to use the theory of the dialogical self to establish whether or not migration is reflected in the migrant’s repertoire of I-positions; the second is to determine the magnitude of any such change.

Research questions. Five research questions (some being multi-part) are investigated.

Migrant identity. The present thesis has presented theories of acculturation that have implications for I-positions. In essence, a number of theories claim ethnic identification as a causal factor in acculturation outcome. Research questions 1a and 1b explore this proposition.

Research question 1a: Biculturalism. How is biculturalism reflected in I- positions?

Research question 1b: Ethnic identification. How is ethnic identification reflected in I-positions?

Stability and change in the dialogical self. The present thesis has shown that people have multiple selves that coexist in time and space, and that this proposition is widely, though not universally, accepted by contemporary theorists in the field. It is, furthermore, a central and defining plank of the theory of the dialogical self. For these reasons it is central to this thesis. The thesis moreover tests the existence of multiple selves, for migration qualifies as exactly the type of life change that might propagate a corresponding I-position. At the same time, however, it is not expected that all migrants will exhibit changes in I-position, for even theories of multiple identity acknowledge that change must take place against a backdrop of stability, without which it is impossible to talk about the individual as an enduring construct. Research questions 2a

96 and 2b pertain to migrants who claim to have two cultural identities, and were explicitly selected for the study sample.

Research question 2a: Stability of the dialogical self. Are country-of-origin and

Australian I-positions similar, and if so, how?

Research question 2b: Changes in the dialogical self. Do country-of-origin and

Australian I-positions differ, and if so, how?

Composition of I-positions. The present thesis has presented theories of identity some of which suggest possible parallels with aspects of the dialogical self. Research questions 3a, 3b and 3c explore aspects of this and similar propositions.

Research question 3a: Evidence of multivalence of the dialogical self. Can individuals simultaneously possess I-positions that differ significantly in their major themes.

Research question 3b: Evidence of adjustment to migration. If so, do individuals’

I-positions correspond to those for country of origin and Australia? In what way?

Research question 3c: Motives for migration. What motivates successful migrants: the desire for self-enhancement or union-with-the-other?

Psychological state of migrants. The surveyed literature touched on migrants’ psychological response to acculturation in terms of culture shock and acculturation stress. Although marked distress was not expected (the study sample had lived in

Australia long enough to have passed through any period of culture shock, or cultural

“honeymoon”), it was thought that psychological responses to migration might reveal the degree of adaptation to life in Australia. Research question 4 is grounded in theories that invoke a cultural adaptation as a response to the stresses imposed by trying to survive and prosper in an alien culture, particularly the “ABC model” of acculturation.

It reveals migrants’ attempts to adapt to circumstances by exploring the relationship between their I-positions and their psychological distress and psychological well-being.

Research question 4: Psychological distress and well-being. How do migrants, when classified by their I-positions pertaining to country of origin and Australia, vary in psychological distress and well-being?

Case studies. Research question 5 seeks to corroborate the answers to research questions 1 and 2 by taking an idiographic approach.

Research question 5: Idiographic confirmation. Are differences between individuals classified by their I-positions reflected in the life stories of specific case studies?

Plan of the empirical work. The present research took the form of three parallel studies. Study 1 was exploratory in nature. It investigated whether or not there might exist naturally forming groups among migrants. Its specific aim was to investigate if migration brings into being separate I-positions relating variously to the migrant’s country of origin and to Australia; and if so, to determine if the nature of those I- positions were related to the demographic characteristics, psychological self-reports and self-confrontation indices of the migrants possessing them. Study 1 therefore addressed the first stated aim of the research, namely to establish whether or not migration were reflected in the migrant’s repertoire of I-positions.

The aim of Study 2 was to elaborate and thereby corroborate Study 1. Unlike

Study 1, which used cluster analysis to group individuals, Study 2 classified them by comparing their valuations in their country of origin and Australia. Despite their differences in analytical approach, both studies investigated similar patterns of stability and change in I-positions. However, it was in the context of Study 2 that hypotheses concerning stability of the self were developed and formally tested. Study 2 therefore

98 addressed the second stated aim of the research, namely to determine the magnitude of any change observed under Study 1.

Study 3 adopted an idiographic approach to complement the nomothetic approach of Study 1. Four case studies will be discussed, one of each migrant type: Progressive,

Low-stable High-stable, and Regressive. The aim is to exemplify and corroborate the nomothetic principles identified in Study 1, and thereby to furnish a more complete understanding of the individual.

All three studies employ data gathered from the same sample of 38 adult migrants to Australia. To maximise comparability they were selected according to five criteria: they perceived themselves as having two separate cultural identities, one for their country of origin and the other for Australia; they were all born overseas and migrated to Australia after the age of 12; they were voluntary migrants (as opposed to refugees); they had lived in Australia for at least five years; and they had a good knowledge of

English. Chapter 5 documents the rationale for these criteria and details the characteristics of the sample.

CHAPTER 5: Research Methodology

This chapter describes the research method adopted in the thesis. First, the characteristics of the sample are described. Next, details are given for the data gathered under the study; these were of two main kinds: self-report measures and self- confrontation indices. After that, the self-confrontation method (SCM) is described, this being the method by which the self-confrontation indices were derived. The chapter concludes with a description of the procedure for gathering data, and the procedure for computing the self-confrontation indices from the participants’ responses.

5.1 Participants

Sample selection. Data for the present thesis came from a sample of 38 migrants

(14 men and 24 women), who were recruited using a variety of methods. Initially it was thought that the sample could be recruited on campus, and to this end advertisements were placed on notice boards posted around the university campus (see Appendix A). In the event only two student participants were obtained in this way. A “snowball” sampling method (Valentine, 1997) was adopted instead, which had the advantage that the respondents were drawn from the broader community. This was deemed appropriate as the major research goal was to study a range of migrants’ valuations to see whether patterns existed across them. First, a few participants were recruited from interested applicants who contacted the researcher after she was interviewed about the study by a local community radio station. These then suggested others, who in some cases suggested still others.

To maximise comparability of participants, they were selected according to five criteria:

100

Criterion 1: On questioning, the participants reported that they perceived themselves as having two separate cultural identities—that is, they thought of themselves as having achieved bi-cultural identification, one identity belonging to their country of origin and the other identity belonging to Australia.

Criterion 2: The participants were all born overseas and had experienced the same life changing event: migration to Australia after the age of 12. As such, they are categorised as first-generation migrants (Rumbaut, 2004) since those who arrive in their adolescent years have similar experiences and adaptive outcome as adult migrants.

Criterion 3: The participants were voluntary migrants (as opposed to refugees).

Reasons for migrating have been studied using the concepts of “pull/push motivations”

(Berry, 1990), where push motives characterise the involuntary or forced migration of refugees. This is generally negative and mostly reactive, whereas pull motives characterise voluntary migration of migrants, which is generally positive and mostly proactive, seeking a better future (Richmond, 1993).

Criterion 4: The participants had lived in Australia for at least five years so as to allow for sufficient time to fully experience migration and also to allow for enough time for the adaptation process to take place (Taft, 1986).

Criterion 5: The participants had a good knowledge of English to avoid translation difficulties. Also, following La Fromboise et al.’s (1994) focus on the concept of bicultural competence, English language proficiency underlies the ability to function effectively in mainstream Australia. A good language competence is known to facilitate the adaptation process and be an important indicator of acculturation level (Berry,

1990).

Sample characteristics. Table 5.1 sets out the salient characteristics of the sample.

101

Table 5.1 Characteristics of the sample, by gender

Gender Total

Male Female N %

Marital status Single 4 6 10 26.3%

Married 7 15 22 57.9%

Divorced 3 3 6 15.8%

Education Before Year 10 2 2 5.3%

Year 10 1 1 2.6%

Year 12 1 3 4 10.5%

Undergraduate 5 4 9 23.7%

Postgraduate 8 14 22 57.9%

Language at home English only 7 8 15 39.5%

Bilingual 6 6 15.8%

Mother tongue only 7 10 17 44.7%

Ethnic identification Belongs to ethnic group 1 7 8 21.1%

No ethnic belonging 13 17 30 78.9%

Fourteen participants were male and 24 female. Most were married and the rest either single or in a few cases divorced. Most were highly educated—possibly as a consequence of using a snowball sampling method. Just under 40% spoke only English at home, slightly more spoke only their mother tongue at home, and a few were bilingual. About one in five identified with their ethnic group, and the rest claimed no particular ethnic affiliation. At first sight this is unexpected, as a criterion for selecting participants was that they perceived themselves as having two cultural identities, one for

102 country of origin, the other for Australia. One would not, therefore, expect them to express “no ethnic belonging”. A plausible explanation is that they perceived their ethnic identity as private rather than public. Thus they have two private cultural identities while at the same time not identifying publicly with their ethnic group.

Table 5.2 presents means of continuous variables by gender. No significant differences were found for years in Australia, t(36) = –.46, p = .65 (two-tailed), or for age, t(36) = –.33, p = .74.

Table 5.2 Means and standard deviations for Years in Australia, Age and Individualism, by gender

Gender

Male Female All

Years in Australia Mean 18.7 20.4 19.8

Standard deviation 10.7 10.6 10.5

Age Mean 41.1 42.2 41.8

Standard deviation 10.4 9.6 9.8

Individualism Mean 38.4 43.1 41.4

Standard deviation 12.5 24.5 20.8

Hofstede’s (1980) concept of individualism was discussed in Chapter 2. Table 5.3 ranks the sample by individualism score of their country of origin. About three-quarters of the sample was drawn from countries such China and Colombia that are low on individualism (arbitrarily defined as 50 on Hofstede’s scale) and consequently high on collectivism. These are cultures that stress collective identity, in-group loyalties, emotional interdependence, sharing, duties and obligations, a need for stable and predetermined friendships, and subordination of individual interests.

103

Table 5.3 Number of participants, by Hofstede’s individualism index

Country Individualism index n United States 91 1 Australia (listed for comparison only) 90 0 United Kingdom 89 2 Italy 76 2 South Africa 65 1 Poland 60 1 Spain 51 2 India 48 4 Argentina 46 1 Japan 46 1 Iran 41 1 Egypt 38 1 Lebanon 38 1 Uruguay 36 2 Greece 35 4 Philippines 32 2 Mexico 30 3 Ethiopia 27 1 Kenya (Seychelles) 27 1 China 20 3 El Salvador 19 3 Colombia 13 1 Total 38

Conversely Australia, with an individualism score of 90, ranks second out of 53 countries on individualism (Hofstede, 1980), implying an emphasis on individual autonomy, emotional independence, assertiveness, self-expression and individual achievement. For most of the sample, therefore, the cultural contrast between country of

104 origin and Australia is substantial. It follows that most have had to make a big cultural adjustment if it is assumed that they are representative of their birth cultures (which is not necessarily the case). In the present study, however, men and women did not differ significantly in the individualism score of their countries of origin, t(36) = –.67, p = .51

(see Table 5.2 for means).

Sample representativeness. In requiring that participants possess “two separate cultural identities”, the sample made no claim to be statistically representative of migrants in general, and this is evident in the sample characteristics. Most participants were well educated (over 80% university educated) and all had a good command of

English (40% spoke it in the home, including three participants from English-speaking countries). However, the intention was not that the sample be representative but that it include migrants from a range of backgrounds. Tellingly, they were diverse in their responses, some finding the migrant experience positive, some not. This is appropriate to test the study’s research questions.

5.2 Self-report Measures of Psychological Distress and Well-being

Each participant completed a questionnaire that contained self-report measures of psychological distress and well-being (Appendix A). Descriptive statistics (including reliability coefficients) relating to the psychological scales are presented in Appendix C.

Psychological Adjustment Measures. Three measures were used. All were self- reports and are discussed below.

Social Avoidance and Distress Scale. The Social Avoidance and Distress Scale

(SAD; Watson & Friend, 1969) was employed as a cognitive and behavioural measurement of social adjustment. The SAD contains 28 items, 14 of which assess social avoidance and 14 of which measure social anxiety. It uses a True/False response

105 format, evenly divided between true and false items. Scale scores on the true/false scale range from 0 (lowest avoidance and distress) to 28 (highest avoidance and distress).

Social avoidance and distress as measured by the SAD refers to an individual’s tendency to avoid social situations and to feel anxious while in such situations. The authors distinguished between avoidance and the failure to approach, arguing that the opposite of avoidance is not social approach, but simply lack of avoidance. Further, they deliberately included only items that assessed subjective distress and behavioural avoidance and excluded items that dealt with physiological indices of anxiety and impaired behavioural performance (Leary, 1991). Patterson & Strauss (1972) used factor analysis to verify the subscale structure for avoidance and distress; however, the subscales have not been validated in this way, and most research using this scale has used the full-scale version (Leary, 1991). Watson and Friend (1969) reported three experimental and one correlational study of the SAD. The data indicated that people defined as high anxious (those with scores of 11 and above) on the SAD were less interested in participating in group activities and talked less when in an interpersonal situation than the low anxious (those with scores below 3). Watson and Friend (1969) reported a mean for university students of 9.1 (SD = 8.0), adequate internal consistency

(Cronbach’s alpha = .77), test-retest reliability (r = .68 to .79 over a 1-month period), and construct validity (Watson & Friend, 1969). More recently, Hofmann, Dibartolo,

Holaway, and Heimberg (2004), in a study conducted to check the psychometrics of the

SAD on a sample of 199 college students, reported a Cronbach’s alpha of .93.

Cronbach’s alpha for the present thesis was .93.

Fear of Negative Evaluation Scale. The Fear of Negative Evaluation Scale (FNE;

Watson & Friend, 1969) was employed as a cognitive measurement of sociocultural adaptation. The FNE is a 30-item measure of the degree to which people expect to be

106 negatively evaluated by others, to look foolish, and to make a bad impression. Watson and Friend (1969) defined fear of negative evaluation as “apprehension about others’ evaluations, distress over their negative evaluations, avoidance of evaluative situations, and the expectation that others would evaluate oneself negatively”. In elaboration, they further stated that “fear of loss of social approval would be identical to FNE” (p. 449), and that the opposite of high FNE is the lack of apprehension about other’s evaluations, but not necessarily a desire or need to be evaluated positively.

The FNE’s items are scored true or false, split roughly equally between positively and negatively scored items. True responses are scored as 1 point and scores range from

0 to 30, with higher scores indicating greater fear of negative evaluation. The FNE norms (and also the SAD norms) were derived from a pilot study by Watson and Friend based on an original sample of 205 college undergraduates that revealed a mean of 15.5

(SD = 8.6), with total scores of “18 or higher” and “9 or lower” on the FNE corresponding to cut-off scores for the top and bottom 25%, respectively. They reported that participants high on FNE became nervous in evaluative conditions, and were more likely to expect disapproval from a leader and in addition more likely to seek approval from others in order to avoid anticipated disapproval than low FNE participants. The authors reported high internal consistency (Cronbach’s alphas = .94 to .96), strong test- retest reliability (r = .78 to .94 over a 1-month period), and construct validity (Watson

& Friend, 1969).

Arkowitz, Lichtenstein, McGovern and Hines (1975), using a sample of 205 high and low frequency daters in a college population, reported a mean FNE score of 7.85 for high frequency daters, and 18.74 for low frequency daters. British FNE norms were established by Stopa and Clark (1991) in two studies using a large sample of students.

107

Research on individual differences on the FNE has found that, compared with low scorers, participants high in FNE avoid situations that could lead to potentially threatening comparisons with others (Friend & Gilbert, 1973), are more affected by negative evaluations about themselves (Smith & Saranson, 1975), worry more about how they come across to other people (Leary, 1983), are more socially anxious and more easily embarrassed (Heimberg, Hope, Rapee, & Bruch, 1988; Miller, 1995) are more sensitive to criticism (Atlas, 1994), are more dependent (Overholser & Freheit,

1994), and are inclined to interpret others’ facial expressions as negative (Winton,

Clark, & Edelmann, 1995) than low FNE participants. The FNE has been validated through significant positive correlations with selected personality measures (e.g., social approval, locus of control; Corcoran & Fischer, 2000) and a negative correlation with self-acceptance (Durm & Glaze, 2001). Cronbach’s alpha for the present thesis was .96.

Beck Depression Inventory, second edition (BDI–II; Beck, Steer, & Brown,

1996). The BDI-II was employed as a measure of psychological adjustment. The BDI-II is the 21-item Beck Depression Inventory (BDI; Beck & Steer, 1993), which assessed the severity of depression in normal and psychiatric populations of both adolescents and adults, updated in light of criteria from the Diagnostic and Statistical Manual of Mental

Disorders (DSM-IV; American Psychiatric Association, 1994) for major depressive disorder. The BDI–II comprises 21 items relating to symptoms and attitudes considered to reflect depth or severity of depressive symptomatology in individuals aged 13 years and older (Beck, Steer, & Brown, 1996). The BDI–II measures the cognitive, affective, somatic, and vegetative aspects of depressive experience. Participants indicate the extent to which they have experienced the symptoms over the last two weeks.

Individual items are rated on an ordered (0 to 3) scale, with higher scores indicating higher levels of depression. The maximum total score for all 21 items is 63. According

108 to the BDI-II manual, scores from 0 to 13 indicate minimal depression, 14 to 19 indicate mild depression, 20 to 28 indicate moderate depression, and 29 to 63 indicate severe depression. More typically, clinically depressed or maladaptively nonclinical populations score in the 14 to 28 range (Beck et al., 1996).

The BDI-II has demonstrated high internal consistency (alpha coefficients were

.93 among college students, and .92 among adult psychiatric outpatients (Beck et al.,

1996). The authors also demonstrated adequate validity (e.g. content, construct, factorial), and established diagnostic discrimination. In their psychometric evaluation of the BDI-II, Dozois, Dobson and Ahnberg (1998) established corresponding BDI-II cut- off scores for classifying groups of research participants (0 to 12, nondepressed; 13 to

19, dysphoric; 20 to 63, dysphoric-depressed) based on cut-offs developed by Kendall,

Hollon, Beck, Hammen, and Ingram (1987) for the BDI. Dozois et al. (1998) also reported a high level of internal consistency (Cronbach’s alpha of .91). More recent research by Steer, Rissmiller, and Beck (2000) found that the total BDI-II score was not significantly correlated with gender, age, or ethnicity of the participants. Cronbach’s alpha for the present thesis was .93.

Psychological Well-being Scale. The Psychological Well-being Scale (PWB;

Ryff, 1989a; Ryff & Essex, 1992; Ryff, Lee, Essex & Schmutte, 1994) was employed as a measurement of psychological adjustment. Derived from the literature on lifespan development, mental health, and personal growth (Ryff, 1989b), the PWB consist of 84 items rated by a response scale ranging from 1 (strongly disagree) to 6 (strongly agree).

The version of the PWB used here contains six subscales: environmental mastery, positive relations with others, autonomy, personal growth, purpose in life, and self- acceptance.

109

Each PWB subscale was operationalised with a 14-item scale of positively and negatively phrased items, which were mixed and, where appropriate, reversed to produce a single inventory. Each subscale has a range of 14 to 84, high scores indicating higher self-ratings of the dimension assessed. These six dimensions are found in Table

5.4 and describe psychological well-being in terms of the presence of positive attributes rather than the absence of mental illness.

110

Table 5.4 Dimensions of Ryff’s Psychological Well-being Scale (PWB)

Dimension Score Definition

Environmental High Has a sense of mastery and competence in managing the mastery environment; controls a complex array of external activities; makes effective use of surrounding opportunities; is able to choose or create contexts suitable to personal needs and values. Low Has difficulty managing everyday affairs; feels unable to change or improve surrounding context; is unaware of surrounding opportunities; lacks sense of control over external world. Positive High Has warm, satisfying, trusting relationship with others; is relations concerned about the welfare of others; is capable of strong with others empathy, affection, and intimacy; understands the give and take of human relationships. Low Has few close, trusting relationships with others; finds it difficult to be warm, open, and concerned about others; is isolated and frustrated in interpersonal relationships; is not willing to make compromises to sustain important ties with others. Autonomy High Is self-determining and independent; is able to resist social pressures to think and act on certain ways; regulates behaviour from within; evaluates self by personal standards. Low Is concerned about the expectations and evaluations of others; relies on judgments of others to make important decisions; conforms to social pressures to think and act in certain ways. Personal High Has a feeling of continued development; sees self as growing and growth expanding; is open to new experiences; has sense of realising his or her potential; sees improvement in self and behaviour over time; is changing in ways that reflect more self-knowledge and effectiveness. Low Has a sense of personal stagnation; lacks sense of improvement or expansion over time; feels bored and uninterested with life; feels unable to develop new attitudes or behaviours. Purpose High Has goals in life and a sense of directedness; feels there is meaning in life to present and past life; holds beliefs that give life purpose; has aims and objectives for living. Low Lacks a sense of meaning in life; has few goals or aims; lacks sense of direction; does not see purpose of past life; has no outlook or beliefs that give life meaning. Self- High Possesses a positive attitude towards the self; acknowledges ad acceptance accepts multiple aspects of self, including good and bad qualities, feels positive about past life. Low Feels dissatisfied with self; is disappointed with what has occurred in past life, is troubled about certain personal qualities; wishes to be different from what one is.

111

Ryff (1989a) demonstrated the convergent validity of the scale with positive and significant correlations with five prior measures of positive functioning: Affect Balance

Scale (Bradburn, 1969), Life Satisfaction Index (Neugarten, Havighurst & Tobin 1961),

Self-Esteem Scale (Rosenberg, 1965), Locus of Control – Internal (Levenson, 1974), and Revised Philadelphia Geriatric Center Morale Scale (Lawton, 1975). Coefficients ranged from .25 to .73.

Discriminant validity was shown via negative and significant correlations with three established prior measures of negative functioning: Locus of Control-Powerful

Others, and Chance Control (Levenson, 1974), and Zung Depression Scale (Zung,

1965). Coefficients ranged from –.30 to –.60.

Intercorrelations amongst the six dimensions were all positive and significant and ranged from .32 to .76 (Ryff, 1989a). The three highest intercorrelations (between self- acceptance and environmental mastery, r = –.76, between self-acceptance and purpose in life, r = .72, and between purpose in life and personal growth, r = .72) suggest that these dimensions might be linked at some underlying causal level. Ryff (1989a), however, found that they showed differential age profiles, and hence represent different facets of positive psychological well-being.

Ryff (1989a) reported a high level internal consistency for the original 20-item version scale, with Cronbach’s alpha ranging from .86 for autonomy to .93 for the self- acceptance. She also reported excellent test-retest reliability over a six-week period, with reliability coefficients ranging from .81 for environmental mastery and personal growth, to .88 for autonomy (Ryff, 1989a).

The PWB scales have shown replicable patterns of significant age and gender differences (Ryff, 1989a; 1991; 1995). Ryff (1989a) asked young, middle-aged, and old-aged adults to rate themselves on each of the dimensions of well-being. She found

112 that environmental mastery and autonomy, increased with age, particularly from young adulthood to midlife; personal growth and purpose in life decreased, especially from midlife to old age; and positive relations with others and self-acceptance did not change significantly (Ryff, 1989, 1995; Ryff & Keyes, 1995; Ryff et al., 2003). In a metastudy of gender differences, Ryff (1995) found that women consistently scored higher than men on positive relations with others and personal growth; but she found no significant differences between men and women on the other subscales (Keyes, Shmotkin, & Ryff,

2002; Ryff, 1989; Ryff & Keyes, 1995).

Other studies using the Ryff scales explored the effect of education, and socio- economic status on psychological well-being (Clarke, Marshall, Ryff, & Wheaton,

2001; Marmot, Ryff, Bumpass, Shipley, & Marks, 1997; Ryff, 1989; Ryff & Keyes,

1995; Ryff, Keyes, & Hughes, 2003). In respect of education, psychological well-being in general, and purpose in life and personal growth in particular, were associated with higher levels of education (Keyes & Ryff, 1998, chap. 7; Marmot et al., 1997; Ryff et al., 2003).

In a study investigating how a life experience such as community relocation relates to mental health, Ryff and Essex (1992) used the 14-item, shortened version of the original PWB (Ryff, 1989a). Results were encouraging for the robustness of the

PWB; internal consistency for the reduced scales ranged from .86 to .93, correlations with the 20-item original scales ranged from .97 to .98, and test-retest reliability over ranged from .91 to .88 (Ryff & Essex, 1992).

In a 1997 study, Schmutte and Ryff examined the common affective underpinnings between measures of personality and well-being, with 2 samples of midlife adults completing measures of personality (NEO Five-Factor Inventory; Costa

& McCrae, 1992) and the shortened (14-item scales) versions of the original Ryff’s

113

(1989a) PWB. They found consistent relationships between the Big Five personality factors and the six PWB subscales. Internal consistency (alpha) coefficients for the 14- item scales ranged from .82 to .90.

More recently, Diehl, Hastings and Stanton (2001) used the shortened (14-item scales) versions of Ryff’s (1989a) PWB in their study to investigate the extent to which persons’ self-representations are different for different social roles and contexts, across the adult life span, and their associations with psychological well-being. They obtained coefficients of internal consistency of .91, .89, .87, .88, .84, and .87 for self-acceptance, environmental mastery, purpose in life, positive relations with others, personal growth, and autonomy, respectively.

Ryff (1989a, b) argued that previous perspectives on operationalizing well-being

“have little theoretical grounding” (Ryff, 1989a, p.1069) and neglect important aspects of positive functioning. To address this shortcoming Ryff drew on the theories of positive mental health (Jahoda, 1968), personal growth (e.g., Maslow 1968; Rogers,

1961; Allport, 1961), life span developmental (e.g., Erikson, 1959). In her view, well- being is not simple the attainment of happiness but is characterised by “the striving for perfection that represents the realization of one’s true potential” (Ryff, 1995, p.100). In this view, happiness is “the by-product of a life that is well-lived” (Ryff & Singer, 1998, p.5), and is reflected in six ideals: autonomy, personal growth, self-acceptance, purpose in life, environmental mastery, and positive relations with others. And on that basis,

Ryff (1989a) developed a multidimensional instrument to measure them: Ryff’s

Psychological Well-Being (PWB).

The Ryff scales have also been used cross-culturally. For instance, Ryff (1995) used it to investigate the psychological well-being in middle-aged Americans and South

Koreans. Interestingly, findings were broadly consistent with known cultural

114 differences. For instance, age and gender profiles were similar, but the Americans had higher scores on all of the Ryff scales; Americans were more likely to attribute positive qualities to themselves than were South Koreans; gender differences were the same in both cultures, with women rating themselves significantly higher than men on positive relations with others and personal growth; South Koreans placed greater emphasis on the well-being of other people (e.g., children) in defining their own well-being than did the Americans.

These and other findings lend credence to the robustness and usefulness of the

Ryff scales, including in a cross-cultural context. Cronbach’s alphas in the present thesis were .91 for environmental mastery, .92 for positive relations with others, .88 for autonomy, .88 for personal growth, .87 for purpose in life, and .91 for self-acceptance.

5.3 Self-confrontation Method (SCM)

Overview and rationale. Valuations were accessed and measured by means of the self-confrontation method (SCM), which is an investigative tool used to access and explore the dialogical self (Hermans, 1987a, 1987b, 1988, 1989, 1992; Hermans &

Hermans-Jansen, 1995). It did so by accessing I-positions. According to Hermans, “the

I … continuously organizes and interprets experience …”; people adopt different I- positions depending on their life circumstances. If those circumstances are cultural, then cultural identity is a special case of an I-position.

In the SCM, each valuation is assigned four indices, one each for the self- enhancement motive, the union-with-the-other motive, positive affect, and negative affect—respectively termed the S, O, P and N indices (or simply S, O, P and N for short). Collectively, they represent the “theme” of the valuation. Although valuations may in principle exhibit any combination of indices, many (but not all) can be assigned

115 to one or other of six “major themes”, each denoted by a characteristic feeling (Hermans

& Hermans-Jansen, 1995, p.73). (The SCM is elaborated in Chapter 4.)

Two additional concepts, “general feeling” (Hermans, 1988) and “ideal feeling”

(Hermans & Oles, 1996), are now explained. “General feeling” measures how a particular valuation impinges on the general experience of the person; “ideal feeling” measures how a valuation impinges on the ideal experience of the person. The more generalising power a valuation has, the more influential is its affective component in colouring the way the person generally feels, and the more idealising power a valuation has, the more it says about the way the person would like things to be.

The SCM is administered in three stages: (a) valuation elicitation, (b) affective rating, and (c) evaluation and integration (Lyddon, Yowell & Hermans, 2006). As an investigative tool, the SCM is particularly suited to a study of this kind. First, it was expressly designed to access I-positions. Second, it allows access to cultural identities both before and after migration, so obviating a costly and difficult longitudinal experimental design. Third, it avoids the bias of self-report questionnaires because its purpose is not apparent to the participant (Berry, 1990, p.482). The present thesis used an adapted version of the SCM. The procedure had three steps: (a) identification of the cultural positions of the self that are current in the person’s life; (b) formulation of a valuation system pertaining to each cultural position; and (c) assessment of the affective implications of the valuations of the two positions.

Valuation elicitation. In the SCM, valuations are elicited by using a set of standard, open-ended questions (see Table 5.5).

116

Table 5.5 Questions of the Self-confrontation method

Set 1: The Past

These questions are intended to guide you in reviewing one or more aspects of your life which may have been of great importance to you.  Has there been anything of major significance in your past life which still continues to exert a strong influence on you? Example: “After high school I never managed to build up a group of friends; I felt left out and stuck between two worlds.”

 Was there in the past any person or persons, experience or circumstance which greatly influenced your life and still appreciably affects your present existence? Example: “My parents have always treated us as equals. They always tried to keep me stimulated. I could always talk very well with my parents”.

Set 2: The Present

This set again consists of two questions which will lead you, after a certain amount of reflection, to formulate a response:  Is there anything in your present existence which is of major importance to you or exerts a significant influence on you? Example: “Bodily complaints: Every time I find that I have a problem, my body signals it in one way or another”.

 Is there in your present existence any person or persons or circumstance which exerts a significant influence on you? Example: “Wendy is a good friend”.

Set 3: The Future

The following questions will again guide you to a response:  Do you foresee anything that will be of great importance for, or exert a major influence on your future life? Example: “I want to pursue a combination of studying and volunteer work”.

 Do you feel that a certain person, persons or circumstance will exert a significant influence on your future life? Example: “John and I want children in the future”.

 Is there any future goal or object which you expect to play an important role in your life? Example: “I would like to meet somebody who would teach me how to have a normal life”.

Two final questions:  How do you generally feel these days?

 How would you like to feel?

These questions are not answered by a formulation. Instead, you answer them by adding two extra rows in the matrix using the same list of affect terms.

117

The questions are read aloud by the interviewer while sitting next to the participant, and relate to people, experiences, circumstances and goals from the past, present and future, that are significant in the judgment of the participant. The questions invite participants to reflect on their life in such a way that they feel free to mention those concerns that are most relevant from the perspective of the present. Participants are free to interpret the questions how they want. They are also encouraged to phrase the valuation in their own terms. During the interview, each elicited valuation is written by the participant on a small card. The typical form of expression is a sentence or sentences, as long as it as felt to be one total experience, where the participant brings together those events that he or she feels belong together as elements of a personal unit of meaning. A quick response is not required, and there is no one-to-one relation between question and answer.

The participant is encouraged to mention all valuations that come to mind and, as is typical of the SCM, each question leads to more than one valuation. The aim of each interview is to arrive at an exhaustive assessment (Hermans, 1993) of the experiences relevant to the participant from each I-position. At the end of each interview participants are asked whether it covers all the experiences they consider (from their vantage point of their present I-position perspective) to be important. If something is missing, they are free to add this. At the end of the procedure the range of valuations elicited from each participant is in most cases between 20 to 40 valuations for each I-position. Also at the end of the valuation elicitation phase, participants are asked the following questions:

“How do you generally feel these days?” and “How would you like to feel?” Responses are used to gauge the extent of generalisation (G) and idealisation (I) of a valuation within the system. These questions do not ask for a specific valuation but instead assess the participant’s “general feeling” and “ideal feeling”, respectively.

118

Affective rating. Next, the participant is presented with a standard list of affect terms (see Table 4.6), is asked to return to each valuation and, after reflecting on the experience it related to, to rate on a 0-5 point scale (0 = not at all, 1 = a little bit, 2 = to some extent, 3 = rather much, and 5 = very much) to what extent he or she experiences each of the 24 affects in relation to each valuation. Participants use a matrix they are provided with in which the rows consist of valuations and each of the columns represents an affect term in order to rate each valuation with the same list of affect terms

(see Appendix A). Next, participants also use the same 0-5 point scale to answer each of the last two final questions (regarding their general and ideal feelings) directly by going over all the 24 affect terms that are used to rate the valuations. The last two rows of the matrix are used by the participant to answer the last two questions referring to their general and ideal feelings.

5.4 Procedure

Pre-testing. To pre-test the measures, the instruments were trialled on the first two respondents. Trialling took the form of analysing the resulting data to determine that they were credible and hence that the instruments were working as intended. In the event, no problems were encountered at the pre-test stage and the instruments were not modified as a result.

Self-report measures. In a preliminary telephone interview, potential participants who met the selection criteria were sent the research questionnaire with a return envelope. Eight persons did not meet the criteria; an explanation was given to them and they were thanked for their interest. Selected participants were told what was required of them, and informed of the time commitment expected, the general purpose of the study and the use to which the data would be put. Participants signed a consent form attached to each questionnaire and returned it to the researcher in the envelope provided.

119

Four people who originally agreed to participate later declined, stating that they did not want to relive the experience of migration. All ethical procedures were observed: participation in the study was voluntary and confidential. To ensure confidentiality in the statistical analysis an identification number was allocated to each participant, and participants’ names have been changed in the case studies. As recounting migration experiences might cause distress, participants were informed that they could stop at any time. One participant took advantage of this; the interview was discontinued and the participant removed from the study sample. After participants had completed both the questionnaire and the SCM, they were fully debriefed.

The questionnaire took approximately 45 minutes to complete and was returned to the researcher. The questionnaire included: a Demographics page, the Social Anxiety and Distress Scale (SADS), the Fear of Negative Evaluation Scale (FNES), the

Psychological Well-being Scale (PWS), and the Beck’s Depression Inventory-ii (BDI-

II). After completing the questionnaire containing the self-report measures, participants returned them to the researcher in a pre-paid envelope. Participants were then telephoned to schedule the first interview under the SCM protocol.

Self-confrontation method interview. The procedure described below closely follows that set out in the SCM Manual (see Hermans & Hermans-Jansen (1995),

Chapter 9).

Participants took part in two or more in-depth face-to-face interviews using the

SCM. All participants were interviewed by the researcher, who was trained in the use of the SCM at an international two-week summer course at the University of Nijmegen in the Netherlands. Interviews averaged an hour and a half in length depending on the participant; to avoid fatigue none exceeded two and a half hours. Thirty-two participants were interviewed at home and six at the psychology clinic at Swinburne University’s

120

Hawthorn campus. Participants were told they would be questioned about what they saw as important experiences in their past and present lives in their country of origin and in Australia. The response of the thirty-eight participants was enthusiastic. In fact, the main problem in the interviews was to keep them focused and for many the interview was therapeutic either because they found new insights because of the questions they were asked or because the interview setting afforded them an opportunity to relieve important stories of their migration experience.

Identification of cultural positions of the self. At the first interview, participants were invited to select two cultural identities of their personality, after confirming that the sense of two identities corresponded to what was required for the research. They were told: “Many migrants feel that they have developed two identities as a result of moving to a new country. One identity retains aspects of how they were in their country of origin; the other identity relates to their new country”. Participants were allowed time to reflect and describe their perception of themselves as having two distinct identities in their own terms.

Formulation of valuation systems from each I-position. Also at the first interview, participants were invited to focus on their cultural identity of their country of origin; to think about their past, present, and future, and relate important experiences from this perspective (I-position). This produced a valuation system for each participant from the viewpoint of their society of origin (Hermans, 2001). At the second interview, which usually took place a week after the first one, participants were invited to focus on their Australian cultural identity and to think of their past, present and future, and to relate important experiences from this perspective (I-position). This resulted in a second valuation system of each participant (Hermans, 2001).

121

By the end of the interview procedure the number of valuations elicited from each participant averaged about 20 for each I-position.

Affective ratings. After the interviews had taken place, participants were asked to return to each valuation and, after reflecting on the experience it related to, to rate on a

0-5 point scale to what extent he or she experienced each of the 24 affects in relation to each valuation. Participants used a matrix provided by the researcher (in which the rows consist of valuations and each of the columns represents an affect term—see Appendix

A) to rate each valuation with the same list of affect terms. Next, participants used the same 0–5 point scale to answer each of the last two final questions (regarding their general and ideal feelings) directly by going over all the 24 affect terms that were used to rate the valuations. The last two rows of the matrix were used by the participant to answer the last two questions referring to their general and ideal feelings. The matrices were then collected by the researcher.

5.5 Treatment of the Interview Data

Once the affective ratings for the different valuations were obtained, affect scores were aggregated into four indices, each representing a different motive or affect

(Table 5.6): the S index (range 0 to 20) obtained by summing the scores for the four affects indicating the strength of the self-enhancement motive; the O index (range 0 to

20) obtained by summing the scores for the four affects indicating the strength of the union-with-the-other motive; the P index (range 0 to 40) obtained by summing the scores for the eight affects indicating the strength of positive affect; and the N index

(range 0 to 40) obtained by summing the scores for the eight affects indicating the strength of negative affect. Table 5.6 shows the composition of the four indices

(Hermans &Hermans-Jansen, 1995). The indices represent the affective character of the

122 valuations and give information about the affective organisation of each participant’s valuation system in general.

Besides the four indices discussed, there is a further, derived, index that is sometimes useful: well-being (W), defined as P – N (Hermans &Hermans-Jansen, 1995,

27). In this study, well-being is used in Study 2 to reduce dimensionality of the dataset.

(This use of “well-being” is not to be confused with a participant’s score on the PWB scale, also termed “well-being” in the literature.)

Table 5.6 Composition of indices in the SCM

Self- Union-with- Positive Negative enhancement (S) the-other (O) affect (P) affect (N)

Self-esteem Caring Joy Powerlessness

Strength Love Satisfaction Anxiety

Self-confidence Tenderness Enjoyment Shame

Pride Intimacy Trust Self-alienation

Safety Guilt

Energy Loneliness

Inner calm Inferiority

Freedom Anger

“General” and “Ideal” feeling. Two additional indices—a general feeling index and an ideal feeling index were derived. The extent of generalisation of any particular valuation within an I-position system of valuations was found by computing the product-moment correlation between the affective profile that belongs to a specific valuation and the affective profile representing how the person generally feels. The more positive the correlation (between the two affective profiles), the more this valuation is supposed to generalise within a particular I-position system of valuations.

123

The scores of the general feelings index were also used to assess the general well-being.

This is represented by the P–N difference belonging to the general feelings.

To measure the extent of idealisation of any particular valuation, the same procedure was followed. The correlation between the affective profile belonging to a specific valuation and the affective profile belonging to the ideal feeling indicates the ideal quality of this valuation. The height of the correlation indicates the extent of the idealisation of the valuation within a particular I-position system of valuations. The more positive the correlation between the two, the higher the idealisation.

Reliability of indices. The present thesis used Cronbach’s alphas to establish the reliability of the S, O, P, and N indices, consistent with Hermans (1987b) who also used this method with a group of 43 students (20 men and 23 women) and a group of 40 clients (20 men and 20 women). Hermans reported that in the student sample, the reliabilities (coefficient alpha) of the S, O, P, and N indices were .83, .86, .90, and .88, respectively. The correlation between S and O was .27, and between P and N was –.80.

In the client group (who had identity problems and problems in social relationships), the reliabilities of the S, O, P, and N indices were .83, .89, .95, and .91, respectively. The correlation between S and O was .64, and between P and N was –.67. When the groups were compared, the client group showed lower scores for S (p < .001), lower scores for

O (p < .001), and higher scores for N (p < .001) than did the student group. No difference was found between men and women (Hermans, 1987b). More recently, the differences in patterning of the indices between the valuation systems of the college student sample and the clinical sample was cited as evidence for the discriminant validity of the SCM (Lyddon, Yowell, & Hermans, (2006).

The Cronbach’s alphas in the present thesis were excellent: .96, .96, .98, and .96, for the S, O, P, and N indices respectively.

124

Coding of major themes. Another way of comparing valuations in the SCM is to arrange them according to six “major themes” (Hermans and Hermans-Jansen, 1995).

The numerical criteria for assigning valuations to major themes vary between researchers: “… the choice of such minimum scores is quite arbitrary” (Hermans &

Hermans-Jansen, 1995, p.254). However, they must comply with the general directions set out in Table 5.7. The numerical rules that were applied in the current research are set out in the Results section.

Strength and unity. Valuations that refer to an experience of “strength and unity”

(labelled +HH by Hermans) are typically characterised by both high self-enhancement and high union-with-the-other motives, and because they are normally associated with positive feelings, they are also characterised by both high positive affect and a low negative affect. These are valuations in which the fulfilment of both basic motives coincide, that is, the fulfilment of one motive goes with the fulfilment of the other, thus creating high levels of well-being, as in “I made a puppet—something from myself for my girlfriend." (S = 16, O =14, P = 17, N = 0)” (Hermans & Hermans-Jansen, 1995, p.95). A valuation with this profile of scores refers to and is abbreviated as valuation of the +HH type.

125

Table 5.7 Composition of major themes

Composition

Description S O P N

+HH Strength and unity High High High Low

+O Unity and love Low High High Low

–O Unfulfilled longing Low High Low High

–LL Powerlessness and isolation Low Low Low High

–S Aggression and anger High Low Low High

+S Autonomy and success High Low High Low

Unity and love. Valuations that refer to experience of “unity and love” (labelled

+O by Hermans) are typically characterised by low self-enhancement and high union- with-the-other motives, and because the wish was fulfilled, they are also characterised by both high positive affect and a low negative affect. These valuations depict an increased contact and union with something else (a person, group, animal, object, or the surrounding world) which is satisfying as in “I enjoy singing with a group of people, being together without any pressure. (S = 4, O = 13, P = 19, N = 0)” (Hermans &

Hermans-Jansen, 1995, p.83). A valuation with this profile of scores refers to and is abbreviated as valuation of the +O type.

Unfulfilled longing. Valuations that refer to “unfulfilled longing” (labelled –O by

Hermans) are typically characterised by low self-enhancement and high union-with-the- other motives, and because the longing is unfulfilled, they are also characterised by both low positive affect and high negative affect. These valuations depict “the adoption or maintenance of a loving orientation toward another person or object that is, or is

126 imagined to be, unreachable”, as in “I regret that it is over with Alice. I want to hold on to this feeling, and I will not allow the clichés to prevail; I find her too valuable. (S = 6,

O = 16, P = 2, N = 13)” (Hermans & Hermans-Jansen, 1995, p.83). A valuation with this profile of scores refers to and is abbreviated as valuation of the –O type.

Powerlessness and isolation. Valuations that refer to an experience of

“powerlessness and isolation” are typically characterised by low self-enhancement and low union-with-the other motives, and because they are normally associated with negative feelings, they are also characterised by both low positive affect and high negative affect. These are valuations where both motives are unfulfilled at the same time, and where the individual typically feels that there is no way out, as in “I was raped about 8 weeks ago by a man; because of this I feel humiliated and sad. (S = 0, O = 0, P

= 0, N = 18)” (Hermans & Hermans-Jansen, 1995, p.91). A valuation with this profile of scores refers to and is abbreviated as a valuation of the –LL type.

Aggression and anger. Valuations that refer to an experience of “aggression and anger” (labelled –S by Hermans) are typically characterised by both high self- enhancement and low union-with-the-other motives, and because they are normally associated with negative feelings, they are also characterised by low positive affect and high negative affect. These valuations depict “some kind of opposition in which individuals feel opposed to somebody or something that is threatening their self- esteem”, as in “I am opposed to those people who want to decide for me what I must do.

(S = 15, O = 0, P = 3, N = 16)” (Hermans & Hermans-Jansen, 1995, p.70). A valuation with this profile of scores refers to and is abbreviated as valuation of the –S type.

Autonomy and success. Valuations that refer to an experience of “autonomy and success” (labelled +S by Hermans) are typically characterised by high self-enhancement and high union-with-the-other motives, and because they are generally experienced as

127 positive, they are also characterised by high positive affect and a low negative affect.

These valuations depict a high degree of autonomy and productive coping with the environment, as in “I hope to achieve a better position through my evening courses (S =

17, O = 4, P = 12, N = 4)” (Hermans & Hermans-Jansen, 1995, p.74). A valuation with this profile of scores refers to and is abbreviated as valuation of the +S type.

5.6 Comparison with Previous Studies

Table 5.8 lists the dimensions of previous studies. The current research covers 38 participants and some 1600 valuations. Though not particularly large in terms of respondents, it is possibly the largest of its kind in terms of valuations. This may be because most studies of the dialogical self to date have been case studies with an idiographic approach, and valuable as these are, they do not, nor are they intended to, generate big datasets, without which detailed statistical analysis is impossible.

Besides having a larger sample than previous studies, the present thesis also differs in sample composition. Table 5.9 compares the respondent sample for the present thesis with those of other studies. Each sample is drawn from a minority group

(mostly immigrants but in one case an indigenous culturally distinct population) within a majority group, and each is ranked on six demographic dimensions according to the similarity of the minority and majority groups. The table shows that sample used in the present thesis is demographically more similar to its majority culture (in this case mainstream Australia) than are the samples used by the other studies to which it is compared. Table 5.8 supports this conclusion in that it shows that most empirical analysis to date using the SCM has been based on student and clinical samples, which may be atypical of the wider population.

128

Table 5.8 Empirical studies using the Self Confrontation Method

Nature of Avge no. of Author (year) Name of study N participants valuations Hermans, The basic motives of human 43 Students n.a. Hermans-Jansen, & existence: Their expression in 40 Clinical Van Gilst (1985) personal valuation Alford, Lyddon, & Adult attachment and working 143 Students 5 (min.) Schreiber, (2006) models of emotion Hermans & Oles, Value crisis: Affective 48 Students 8 (1996) organization of personal meanings Hermans & Oles Midlife crisis in men: affective 104 Men in 9 (1999) organization of personal midlife crisis meanings Hermans & Oles The personal meaning of 53 Students 10 (1994) values in a rapidly changing society Rioux & Barresi Experiencing science and 40 Students 4 (1997) religion alone and in conflict Hermans (1989) The meaning of life as an 25 Non-severe n.a. organized process clinical

Table 5.9 Demographic difference between minority and majority cultures for selected studies

Dimension of demographic difference Author of study Volun- Socio- Lan- Overall Race Religion tariness economic guage score

Chandler and Lalonde      4 (1998; 2004; 2008)

Van Meijl (2012)      3

O’Sullivan-Lago and de      2.5 Abreu (2008)

Bhatia and Ram (2007)   2

This study   1

Legend:  = All members of minority culture differ on the stated dimension;  = Some members of minority culture differ on the stated dimension. Note: Overall score was computed by scoring 1 for each dimension of difference where all respondents differ and 0.5 where only some respondents differ, then totalling.

129

In surveying the state of research to date, several authors have found that empirical advances lagged behind theoretical ones. Concerning the SCM, Lyddon,

Yowell, and Hermans (2006) identified a need for more empirical work:

The Self Confrontation Method is gaining support as an effective approach to assessment in both case study research and in empirical studies involving larger samples …, though the latter are not as common. … Although the Self Confrontation Method has garnered promising qualitative and quantitative research support, research using the method with larger samples is warranted. (p. 40)

Hermans and Dimaggio (2007) concurred:

[T]o be recognized as a respected science, it is necessary to develop the dialogical field in a theory-guided, empirical direction, taking advantage of both quantitative and qualitative methods and of both experimental and experiential approaches …. [N]ew and challenging theories should be created that may profit from equally developed assessment methods and research procedures that are essential to revise and improve existing theoretical notions. (p. 55)

As did Hermans (2008):

… the literature in the dialogical self field … shows more theoretical advances than methodological and empirical elaborations. Therefore, future empirical work on the basis of this theory should be encouraged and stimulated so that theoretical progress goes hand in hand with empirical study and practical application. (p.186)

Stemplewska-Żakowicz, Walecka, Gabińska, Zalewski and Suszek (2005) also advocated more empirical work, but with a particular emphasis. Having conducted experiments on positioning that they describe as “exemplars of a successful dialogue between post-positivist and social-constructivist methodology”, they conclude “that the positivist tradition in methodology may be useful for studies within dialogical self theory.”

130

These conclusions all point to a relative insufficiency of empirical research, which the present study, with its quantitative focus, goes some way towards filling.

5.7 Personal Reflections on the Experience of Migration

Conventional wisdom has it that the scientist, including the social scientist, is an impartial observer, neither influencing what is observed nor filtering it through the lens of subjectivity. But if this view was ever widely held, it is less so now than ever, particularly in the field of narrative research which, being inherently dialogical, must have an author in dialogue with an audience. According to Bruner (1990): "The-story- of-a-life as told to a particular person is in some deep sense a joint product of the teller and the told" (p. 124). Stiles (1993) stated: “… good practice recommends disclosure by the investigator of his or her expectations for the study, preconceptions, values, and orientation…” (p. 602). Most recently, Lawrence-Lightfoot (2005) stated: “All researchers—whether working within the quantitative or qualitative methodological paradigm—are selective in defining and shaping the data they collect and the interpretations that flow from their findings…The shaping hand of the investigator is counterbalanced by the skepticism and scrutiny that is the signature of good research”

(p. 11).

If, then, the researcher inevitably influences the outcome of the research, how is this to be accommodated? Wortham (2000) and Potter and Hepburn (2005) have developed and advocate analytical tools and procedures for delving more deeply into the

“interactional” (as opposed to “representational”) aspect of dialog. But these tools, powerful as they are, are too time-consuming to use on the approximately 1600 narrative “valuations” elicited in the present research. However, many of these problems, namely those that Potter and Hepburn term “contingent”, can be addressed by

131 the interviewer’s appropriate positioning in the narrative dyad. This was elaborated by

Hermans (2000):

the interactional nature of the autobiographical interview requires the interviewer to develop a position repertory which helps him to tune in to the flux of experiences on the side of the narrator. A broad repertory which enables the interviewer to move flexibly from one to another position, in accordance with the needs of the narrator, is a precondition for studying the person’s autobiography in all its complexity and multifacetness.

Fortunately my experience as a migrant has equipped me with a broad repertoire

(to use Hermans’ terminology). In two migrant episodes at very different stages in my life I have been, or have attempted to be, variously “separated”, “assimilated”,

“integrated” and “marginalised”, (to use Berry’s terminology) and sometimes more than one at the same time, if in different domains of life—separated at home, and assimilated

(or attempting to be so) at work. This has enabled me to read, and empathise with, my interviewees. It is for this reason that I write this section, and moreover do so in the first person, to remind the reader that the present research was in some degree shaped by my experiences as a migrant myself. What follows is a brief autobiographical sketch.

I was born and raised in the Dominican Republic. My first migration (though at the time I had no idea how long-lasting it would be) was supposedly to further my education. But the real reason was that my parents wanted me out of the house. That suited me; I wanted to go. So they arranged for me stay with an aunt while I studied

English in New York.

My aunt’s household was Dominican from top to bottom—artefacts, music, food, everything. She was a political exile. Her husband, a prominent political and intellectual figure, had been assassinated by the former dictator. But by the time he was assassinated in his turn, and her return became possible, her children, my cousins, had

132 become Americanised and completely bilingual, while she still spoke very little English.

Torn as she was between children and homeland, she stayed. Lawson (2000) describes this existential migrant condition as “a state of in between-ness”, and I was part of it. I inhabited a Dominican “ghetto” where everybody knew who I was and what I was. In the words of Deaux (2000), I was “remoored”.

Coming from a privileged family, I had lived a materially comfortable, but sequestered, existence. I found New York liberating and transforming. Although living with family, I was the freest I had ever been. At first I struggled with English. I did not want to sound like an outsider. “Passing”, as Goffman (1968) put it, was my goal. Like any teenager, I just wanted to fit in (Camilleri and Malewska-Peyre 1997). Going back was always on the cards, and in fact central to my Dominican identity. But it was not what I chose—at least, not straight away. I was to spend nine years in the States—a marriage, a child, a divorce, by which time my country of birth had become haloed by nostalgia. Only then, in a way defeated, did I return. Looking back, my migrant identities—Dominican and American—reflected the ambivalence that makes a person long for a place when living in another, identifying with home when abroad, and with abroad when home (King, 1995). I knew then that migration changes one, forever, which is why it is so often associated with loss. Life became complicated and I became wistful, always lacking, never complete.

My second migration was to Australia, where I now live. It was to accompany my second partner, a migrant himself, to his adopted country. By now, I was fluent in

English, and although the my accent was never to leave me, I was comfortable with it.

In Australia I embraced my difference. I was Dominican in Australia and so an outsider, but by now I accepted my difference as a creative resource, not a blemish to be erased. I played with cultural repertoires, confident in acquiring a new cultural dimension—

133 becoming Australian. Migration gave me a new perspective, a “third space” where I could explore and extend my cultural repertoire, creatively reconstructing my

Dominican and Australian identities as “hybrid” (Bhabha 1994).

I’m now a “foreigner” who speaks fluent English and lives in Australia. I am in it but not of it. I am between cultures. It is a role that can be hard to sustain, and has on occasion been the cause of unhappiness. But it also offers something precious: I am able to observe Australian culture with a degree of detachment and in this way reflect all the more on my own life in time and space. At times I feel a bitter-sweet nostalgia for my country of birth, for what was and never will be. But at other times my nostalgia is like an illness, which is what Johannes Hofer, the doctor who coined the word ‘nostalgia’ in

1688, believed it to be. It is a rare migrant that has not felt the pain of letting go the life they once knew.

The migrant story is one of pain. In Australia this is only beginning to be widely discussed. “In Australia I will”, the posters once trumpeted. Well, some did and some didn’t; and those who did, often had a hard time of it. Some deplore all this remembering. Nostalgia would seem to be, after all, the opposite of progress. But

Australia’s migrant experience is not an unmixed march of progress. It also contains countless stories of heartache. The nation’s identity would be greatly enriched if both sides of the migrant story were told. In Carey’s (2002) words: “… we are a nation of immigrants…”.

Having thus disclosed my preconceptions, values, and orientation (following

Stiles, 1993) I now come to my expectations for the present research. As should by now be apparent, this thesis is autobiographical in two senses: it calls on my participants to divulge their autobiographical details; equally, it is strengthened by my candid disclosure of my own autobiographical experience of migration—in general terms if not

134 in detail—in that my participants were all aware that I shared this aspect of life experience with them. In short, I satisfied all the criteria that I required of my participants; they knew it, and the knowledge possibly credentialled me in their eyes.

My research took as a framework my own experience of migration; and sharing this experience with my participants was crucial to establishing the authenticity of my relationship with them. In this I followed Janesick’s (1998) precept of “researcher as instrument”, in which I was a participant observer using my own subjectivity to interpret data.

More specifically, I wanted to give my survey participants the psychological space to talk freely, and in this way I hoped to encourage them to reflect on, and make sense of, their bifurcated life stories and possibly disrupted sense of self. I hoped that it would offer participants constructive outcomes as a result of migration. Accordingly, my research did not harbour presuppositions about the migration experience, nor did it define a population of which my participants would constitute a representative sample.

On the contrary, I expected diversity in self-narratives, which made the reported similarities in emotional valence all the more striking.

135

CHAPTER 6: Study 1 – Identification and Comparison of Migrant Groups

This chapter discusses the first of the three studies conducted as part of this thesis.

All were based on the same data and were complementary in nature. Study 1 was largely exploratory. Its aim was not to test hypotheses but to reveal the existence of naturally forming groups among migrants, that is, to determine whether migration brings into being new I-positions, and if so, whether there exist any patterns among migrants in their I-positions in country of origin and Australia that can be discerned in terms of demographics, psychological self-reports and self-confrontation indices.

Four groups of participants were identified by means of cluster analysis. Because cluster analysis is an exploratory technique that lacks formal tests of statistical significance, it was only once the groups were identified that hypotheses were tested.

Significant differences were found among the groups, and these differences were found to be plausible in terms of theory. The four groups identified in Study 1 were corroborated by Studies 2 and 3. Statistical analyses were performed using SPSS 16.0.

6.1 Preliminary Analysis

Self-report measures of distress and well-being. The distributions of all measures were inspected for outliers. Preliminary data screening found a single high outlier on BDI–II, and single low outliers on PWB Personal growth and Purpose in life.

Three low outliers were found on PWB Self-acceptance, two of which were marginal.

Other than the two marginals, all other outliers related to the same participant, which suggests that they were not due to a coding error but represented a single participant with mild depression.

Table 6.1 shows the sample means, standard deviations, skewness and kurtosis for the self-report measures of distress and well-being. Kolmogorov-Smirnov tests showed

136 that normality assumptions were significantly violated at the 5% level for SAD, FNE,

BDI-II and, marginally, PWB Self-acceptance. Although this might cast doubt on later

MANOVAs with these variables, MANOVA is normally robust to such violations

(Tabachnik & Fidell, 2006).

Table 6.1 Means and standard deviations for the self-report measures

Psychological Distress Psychological Well-being Scale

SAD FNE BDI–II PWB1 PWB2 PWB3 PWB4 PWB5 PWB6

Mean 8.4 13.3 9.1 63.1 67.1 63.4 72.7 66.1 64.6

SD 7.2 9.5 9.3 12.1 12.1 11.9 8.0 10.8 12.4

Skewness .831 .374 1.784 -.656 -.568 -.638 -1.018 -1.024 -.876

Kurtosis -.607 -1.261 4.099 .703 -.063 -.672 1.810 2.995 .568

N = 38 Note: SAD = Social Avoidance and Distress Scale; FNE = Fear of Negative Evaluation Scale; BDI-II = Beck Depression Inventory–II; PWB1 = Environmental mastery; PWB2 = Positive relations with others; PWB3 = Autonomy; PWB4 = Personal growth; PWB5 = Purpose in life; PWB6 = Self-acceptance.

Table 6.2 presents bivariate correlations of the self-report measures of distress and well-being. All were significantly (p < .01) correlated. As can be seen, the psychological distress scales were significantly positively correlated with each other.

Participants who had a high SAD score tended to have a high FNE score (r = .63); participants who had a high SAD score also tended to have a high BDI-II score

(r = .45); and participants who had a high FNE score tended to have a high BDI-II score

(r = .57). The PWB subscales, all of which measure various components of psychological well-being, were strongly positively correlated with each other (r = .63 to r = .90). Lastly, the PWB subscales were strongly negatively correlated with the three psychological distress scales (r = –.49 to r = –.80).

137

Table 6.2 Correlations among self-report measures

Psychological Distress Psychological Well-being Scale (PWB)

SAD FNE BDI-II PWB1 PWB2 PWB3 PWB4 PWB5 PWB6

Social Avoidance and Distress Scale (SAD)

Fear of Negative Evaluation Scale (FNE) .628**

Beck Depression Inventory–II (BDI-II) .449** .569**

PWB Environmental mastery (PWB1) -.575** -.668** -.744**

PWB Positive relations with others (PWB2) -.610** -.586** -.592** .725**

PWB Autonomy (PWB3) -.570** -.734** -.561** .718** .638**

PWB Personal growth (PWB4) -.508** -.528** -.623** .664** .661** .725**

PWB Purpose in life (PWB5) -.478** -.544** -.796** .863** .698** .628** .731**

PWB Self-acceptance (PWB6) -.524** -.672** -.750** .903** .677** .820** .759** .815**

N = 38 Note: Values are Pearson’s r correlations, * p < .05 (two-tailed), ** p < .01 (two-tailed).

138

Self-confrontation data. Each participant in the study produced about 40 valuations on average, divided roughly evenly between those relating to country of origin and those relating to Australia. They were further split across valuations relating to the past, the present, and the future.

Preliminary data screening found no outliers in the self-confrontation indices (S,

O, P and N) as they describe participants’ valuations relating to country of origin and

Australia. Table 6.5 (Section 5.2, p.116) shows the sample means, standard deviations, skewness and kurtosis for self-confrontation indices at country of origin and Australia.

Kolmogorov-Smirnov tests showed that normality assumptions were significantly violated at the 5% level for O for country of origin, and N for Australia. Although this might cast doubt on later MANOVAs with these variables, MANOVA is normally robust to such violations (Tabachnik & Fidell, 2006).

Table 6.3 presents bivariate correlations of the indices. For each set of indices

(i.e., for country of origin, Australia, general feeling, and ideal feeling), S, O and P in the same set were all significantly correlated with each other (p < .01). For country-of- origin indices, participants who were high on S tended also to be high on O and P

(r = .84 to r = .95). For Australian indices, participants who were high on S tended also to be high on O and P (r = .84 to r = .95). For indices relating to general feeling, participants who were high on S tended also to be high on O and P (r = .75 to r = .91).

For indices relating to ideal feeling, participants who were high on S tended also to be high on O and P (r = .75 to r = .78). Overall, high S and O were statistically associated both with each other and with high P. The same was also generally true if the indices came from different sets (although the ideal feeling indices were less strongly correlated with those of other sets).

139

Table 6.3 Correlations among self-confrontation indices

Country of origin Australia General feeling Ideal feeling S O P N S O P N S O P N S O P N Country of S origin O .839** P .947** .885** N -.050 .056 -.074 Australia S .764** .691** .711** .017 O .640** .781** .657** .002 .836** P .760** .734** .771** -.045 .941** .884** N .047 .057 .065 .662** -.306 -.274 -.323* General S .461** .445** .433** -.116 .732** .522** .600** -.322* feeling O .461** .599** .505** -.192 .746** .757** .711** -.400** .753** P .428** .553** .470** -.103 .694** .615** .630** -.314 .910** .818** N -.139 -.218 -.145 .362* -.435** -.415** -.375** .598** -.657** -.629** -.732** Ideal S .320 .338 .294 .260 .397* .282 .252 .008 .452** .266 .357* -.130 feeling O .514** .601** .530** .121 .538** .496** .481** -.038 .562** .509** .538** -.197 .781** P .292 .413** .312 .202 .241 .253 .181 .124 .386* .290 .372* -.044 .746** .777** N .069 .172 .105 .192 .110 .215 .199 .150 -.245 .095 -.122 .322** -.311 -.193 -.197 N = 38 Note: Values are Pearson’s r correlations, * p < .05 (two-tailed), ** p < .01 (two-tailed). S = self-enhancement; O = union-with-the-other; P = positive affect; N = negative affect.

140

The relationships among S, O and P on the one hand, and N on the other, were less clear-cut. For country-of-origin indices, N was not significantly correlated with S, O and P (r = –.07 to r = .06). For Australian indices, N was not significantly correlated with S (r = –.31) or O (r = –.27), but was moderately negatively correlated with P (r = –

.32, p < .05). For indices relating to the ideal feeling, N was not significantly correlated with S, O or P (r = –.19 to r = .31). However, for indices relating to the general feeling,

N was strongly negatively correlated with S, O and P (r = –.63 to r = –.73).

In general, correlations were highest between indices within the same set. Lower

(but still significant) correlations were observed between general feeling indices and those in country of origin. The lowest (mostly not significant) correlations were between ideal feeling and the other sets.

Relationship between self-report measures and self-confrontation indices.

Table 6.4 presents bivariate correlations of the self-report measures and self- confrontation indices. In general, self-report measures were not significantly correlated with self-confrontation indices for the country of origin I-position, and hardly at all with ideal feeling indices, but were mostly significantly correlated with indices for the

Australian I-position, and even more so with general feeling indices.

As is evident from Table 6.4, for country of origin, self-confrontation indices were significantly correlated with measures of psychological distress in one instance: negative affect was positively correlated with BDI-II (r = –.49, p < .01). Likewise, self- confrontation indices were significantly correlated with measures of psychological well- being in one case: P was positively correlated with PWB Autonomy (r = –.36, p < .05).

141

Table 6.4 Correlations among self-reports and self-confrontation indices

Country of origin Australia General feeling Ideal feeling S O P N S O P N S O P N S O P N SAD -.049 -.063 -.101 .259 -.286 -.206 -.249 .347* -.399* -.449** -.312 .379 -.071 -.166 -.131 .132 FNE -.149 -.109 -.222 .222 -.368* -.133 -.287 .355* -.482** -.399* -.426** .421** -.237 -.254 -.176 .093 BDI-II -.087 -.132 -.160 .491** -.311 -.237 -.335 .647** -.467** -.538** -.517** .654** .047 -.133 .064 .120 PWB1 .267 .267 .289 -.286 .516** .363* .446** -.476** .662** .598** .656** -.608** .181 .328* .163 -.193 PWB2 .179 .233 .201 -.214 .532** .437** .499** -.561** .601** .630** .557** -.638** .249 .372* .227 -.174 PWB3 .291 .216 .358* -.284 .432** .228 .391* -.376* .622** .407* .510** -.377* .320* .448** .384* -.316 PWB4 .194 .247 .274 -.096 .497** .307 .457** -.297 .706** .485** .640** -.490** .293 .374* .173 -.198 PWB5 .263 .294 .315 -.277 .510** .391* .481** -.439** .653** .660** .648** -.585** .091 .272 .122 -.041 PWB6 .224 .160 .255 -.299 .476** .258 .410* -.483** .673** .521** .609** -.538** .158 .279 .103 -.311 N = 38 Note: Values are Pearson’s r correlations, * p < .05, two-tailed, ** p < .01, two-tailed. S = self-enhancement; O = union-with-the-other; P = positive affect; N = negative affect; PWB1 = Environmental mastery; PWB2 = Positive relations with others; PWB3 = Autonomy; PWB4 = Personal growth; PWB5 = Purpose in life; PWB6 = Self-acceptance.

142

In the case of the Australian I-position, self-confrontation indices were significantly correlated with measures of psychological distress in three instances: S was negatively correlated with FNE (r = –.37, p < .05); and N was positively correlated with both SAD (r = .35, p < .01) and FNE (r = .36, p < .01). Self-confrontation indices were also significantly correlated with measures of psychological well-being: S was positively correlated with all PWB subscales (r = .42 to .53, all at p < .05); O was positively correlated with PWB Environmental mastery (r = .36, p < .05), Positive relations with others (r = .44, p < .01), and Purpose in life (r =.39, p < .05); P was positively correlated with all PWB subscales (r = .39 to .50, most at p < .01) ); and N was negatively correlated with all PWB subscales except Personal growth (r = –.38 to –

.56, most at p < .01).

In the case of general feeling, self-confrontation indices were significantly correlated with all measures of psychological distress in all cases but two: S (r = –.40 to

–.48, mostly at p < .01); O (r = –.40 to –.54, p < .05); P (r = –.43 to –.52, p < .05); and

N (r = .42 to .65, p < .05). Self-confrontation indices were also significantly correlated with all measures of psychological well-being: positively with S (r = .60 to .71, all at p

< .05); positively with O (r = .41 to .66, most at p.< .05); positively with P (r = .51 to

.66, all at p < .05) ); and negatively with N (r = –.49 to –.64, most at p < .05).

In the case of ideal feeling, self-confrontation indices were not significantly correlated with measures of psychological distress. However, they were significantly correlated with measures of psychological well-being in several cases. S was positively correlated with PWB Autonomy (r = .32, p < .05); O was positively correlated with

PWB Environmental mastery (r = .33, p < .05), PWB Positive relations with others (r =

.37, p < .05), and PWB Autonomy (r = .45, p < .01); and P correlated with PWB

Autonomy (r = .38, p < .05).

143

6.2 Distinguishing I-positions

The theory of dialogical self implies that life-changing events such as migration can give rise to additional I-positions, and that these I-positions may be accessed through the SCM. Although not all migrants necessarily develop in this way, the participants in this study were required to meet the criterion that they perceived themselves as having “developed two identities as a result of moving to [Australia]”, where one identity retains aspects of how they were in their country of origin and the other identity relates to their new country. As part of the data exploration, therefore, analogous self-confrontation indices pertaining to each participant’s I-position in country of origin were compared with those pertaining to the I-position in Australia with the aim of determining if the group means differed significantly.

Comparing self-confrontation indices for country of origin and Australian I- positions. Paired-samples t-tests were conducted to compare self-confrontation indices for country of origin and Australian I-positions. Table 6.5 shows means and standard deviations of self-confrontation indices. Independent samples t-tests were conducted to compare the self-confrontation indices for country of origin and Australia. Results revealed that S was significantly greater for Australia than for country of origin, t(37) = –3.31, p = .002 (two-tailed); and P was significantly greater for Australia than for country of origin, t(37) = –2.37, p = .023 (two-tailed). No significant differences were found in the case of O or N. Mean O for Australia did not differ significantly from that for country of origin, t(37) = –1.12, p = .270 (two-tailed); nor did N for Australia differ significantly from that for country of origin, t(37) = –0.49, p = .625 (two-tailed).

However, the significance of the difference in at least some self-confrontation indices, in this case S and P, shows that a new I-position has come into existence, for some participants at least.

144

Table 6.5 Means and standard deviations of self-confrontation indices

Index Location S O P N

Country of origin Mean 10.02 9.54 19.11 10.03

SD 3.91 4.23 7.83 6.14

Skewness -.123 -.211 -.133 .503

Kurtosis -.729 -.997 -.884 -.787

Australia Mean 11.45 10.07 21.17 10.45

SD 3.86 4.52 7.96 6.40

Skewness -.036 -.169 -.014 .378

Kurtosis .561 -1.220 -.873 -.926

N = 38 Note: S and O are scored from 0 to 20; P and N from 0 to 40; S = self-enhancement; O = union- with-the-other; P = positive affect; N = negative affect.

What these results show is that the differences between means for country of origin and Australia were small (though nevertheless significant), and that the variability in self-confrontation indices within groups were large (as revealed in their standard deviations). There is a simple explanation for both these findings, namely that the sample contained disparate responses. For some the new I-position reflected a view of life in Australia as an unalloyed reward, so creating a new I-position with more positive affect and greater satisfaction of the S- and O-motives; for others, it was the very opposite; and for others there was little difference. Over the whole sample, then, the effects may cancel out, so that group mean self-confrontation indices differ little, when in fact the self-confrontation indices for individual participants differ greatly, but in opposite directions (a case of the ‘ecological fallacy’). This implies that significant

145 differences in self-confrontation indices may be found when the sample is examined in more detail, as was done next.

Cluster analysis. Cluster analysis was used as a data reduction method to classify participants into a small number of relatively homogeneous groups for the purpose of exploring possible relationships among variables that were masked at the aggregate level of the whole sample. Participants were characterised according to their I-positions in country of origin and Australia. Since it was considered that the country-of-origin I- position might differ from the Australian, two cluster analyses were conducted, one to classify participants into clusters determined by their self-confrontation indices for country of origin, the other to cluster participants by their self-confrontation indices for

Australia. Thus, each participant was characterised by membership of two clusters.

Two cluster methods were employed. Hierarchical clustering was selected as the principal clustering method as it requires no a priori assumptions as to the nature of the clusters, instead allowing optimal cluster centroids to be determined as part of the clustering procedure (Clatworthy, Buick, Hankins, Weinman & Horne, 2005). Self- confrontation indices were clustered according to the CLUSTER procedure in SPSS, using Ward’s minimum variance method with a squared Euclidean distance metric to represent dissimilarity between case-pair. This is an agglomerative hierarchical cluster analysis procedure that joins clusters of cases together in a way that minimises increases in the within-cluster variance (or fusion coefficient) (Clatworthy et al., 2005). Because index metrics differed markedly in range and mean, they were converted to z-scores for clustering purposes so as to avoid bias when computing squared Euclidean distance.

Since clustering techniques lack formal tests of statistical significance,

Beauchaine and Beauchaine (2002) recommend that the robustness of the hierarchical solution be assessed by re-clustering by k-means clustering. Unlike hierarchical

146 clustering, k-means clustering is an non-hierarchical iterative procedure that assigns and reassigns cases to the “closest” cluster as measured by Euclidean distance, and requires that the number and initial centroids of the clusters be specified in advance. When k- means clustering is used to corroborate hierarchical clustering, it is appropriate to

“seed” the k-means cluster means with those previously established by hierarchical clustering (Beauchaine and Beauchaine, 2002).

There are no hard-and-fast rules for determining the appropriate number of clusters, but the cluster dendrogram and agglomeration schedule are taken to be a good guide. Parsimony favours fewer clusters, while explanatory power favours more

(Aldenderfer & Blashfield, 1984). An abrupt lessening in the incremental reduction in within-cluster variance in the agglomeration schedule as the number of clusters grows, and a similar shortening in dendrogram branches, show where further disaggregation is likely to become unhelpful.

Clusters in country of origin

Hierarchical clustering. The agglomeration schedule (Figure 6.1) and dendrogram (Figure 6.2) were examined for clustering by self-confrontation indices for country of origin.

147

150

100

Two-cluster solution

50 Within-cluster variance (fusion coefficient) variance Within-cluster (fusion

0 0 5 10 15 20 25 30 35 Stage

Figure 6.1 Agglomeration schedule: cluster in country of origin

Various clustering solutions were tried with the aim of achieving a satisfactory trade-off between parsimony and explanatory power. A two-cluster solution reduced the within-cluster variance by about half, a three-cluster solution by about two-thirds, and a four-cluster solution by about three-quarters (see Figure 6.1). The two-cluster solution was selected because it produced cluster sizes (21 and 17 respectively) that were sufficiently large to permit MANOVA, which ideally requires a sample size of 20 in each cell to ensure “robustness” (Tabachnik & Fidell, 2001).

The dendrogram in Figure 6.2 shows how participants were assigned to clusters in country of origin. Cluster 1 contained 21 participants and cluster 2, 17 participants. (For reasons of confidentiality participants are identified by a unique number followed by their first initial.) The length of the dendrogram branches following the first split virtually guarantees the validity of at least a 2-cluster solution.

1 48

Rescaled Distance Cluster Combine C A S E 0 5 10 15 20 25 Label Num +------+------+------+------+------+

Cluster 1 18 E-- 18 ─┐ 37 C-- 35 ─┤ 15 E-- 15 ─┤ 11 D-- 11 ─┤ 39 R-- 36 ─┼─────────────┐ 30 G-- 29 ─┤ │ 02 D-- 2 ─┘ │ 06 D-- 6 ─┐ │ 34 E-- 32 ─┼─────┐ ├─────────────────────────────────┐ 07 F-- 7 ─┤ │ │ │ 22 J-- 22 ─┤ │ │ │ 21 D-- 21 ─┤ │ │ │ 42 K-- 38 ─┘ ├───────┘ │ 23 M-- 23 ─┐ │ │ 26 D-- 26 ─┼─┐ │ │ 01 A-- 1 ─┤ │ │ │ 28 G-- 28 ─┘ ├───┘ │ 03 Y-- 3 ─┐ │ │ 25 N-- 25 ─┤ │ │ 17 L-- 17 ─┼─┘ │ 13 I-- 13 ─┘ │ │ Cluster 2 │ 04 N-- 4 ─┐ │ 12 L-- 12 ─┤ │ 36 R-- 34 ─┼─────┐ │ 05 H-- 5 ─┤ │ │ 10 A-- 10 ─┤ │ │ 20 S-- 20 ─┤ │ │ 40 C-- 37 ─┤ │ │ 27 J-- 27 ─┤ │ │ 33 F-- 31 ─┘ ├─────────────────────────────────────────┘ 09 S-- 9 ─┐ │ 35 L-- 33 ─┤ │ 16 I-- 16 ─┤ │ 19 B-- 19 ─┤ │ 08 K-- 8 ─┤ │ 24 L-- 24 ─┼─────┘ 32 G-- 30 ─┤ 14 M-- 14 ─┘

Figure 6.2 Dendrogram of 2-cluster solution: cluster in country of origin

149

K-means clustering. Table 6.6 shows that there was a close match between cluster membership under both clustering methods. This largely corroborates the hierarchical clustering procedure.

Table 6.6 Number of participants by cluster: country of origin

K-means cluster Hierarchical cluster

Cluster 1 Cluster 2 Total

Cluster 1 19 – 19

Cluster 2 2 17 19

Total 21 17 38

Two participants were classified differently by the clustering procedures: participants 01 and 34 were both placed in hierarchical cluster 1 but in k-means cluster

2. It is possible that this was because they lay on the margins of their respective clusters; a change in the clustering algorithm therefore caused them to switch from one cluster to another. An examination of the sum of squared differences between each observation and cluster centroids supports this supposition. Figure 6.3 confirms that k-means clustering correctly assigned each participant to the cluster with the “nearest” centroid, as required by definition. However, in some cases participants were almost equidistant between pairs of clusters, and in such cases hierarchical clustering assigned the participant to the other cluster of the pair. This is of little consequence for the analysis.

It means that the affected participant has features of both clusters—its hierarchical cluster and its k-means cluster. This is expected in cluster analysis, where it is normal for some observations to lie near cluster boundaries (Aldenderfer & Blashfield, 1984).

What is important in this case is that there were few instances of this, which suggests that cluster identity was mostly unambiguous.

150

20 Cluster 1 18 Cluster 2

16 Hierarchical cluster conflicts with K-means Hierarchical cluster conflicts with K-means 14

6 hierarchical cluster centriod

Sum of squared from differences squared of Sum 4

0 13 I--- 16 I--- 27 J--- 22 J--- 12 L--- 24 L--- 35 L--- 17 L--- 07 F--- 33 F--- 03 Y--- 01 A--- 08 K--- 09 S--- 10 A--- 34 E--- 42 K--- 15 E--- 18 E--- 19 B--- 20 S--- 02 D--- 04 N--- 05 H--- 06 D--- 11 D--- 25 N--- 26 D--- 36 R--- 37 C--- 39 R--- 40 C--- 21 D--- 28 G--- 30 G--- 32 G--- 14 M--- 23 M--- Respondent

Figure 6.3 Sum of squares from centroid: cluster in country of origin

Cluster means. As the clusters were purposely created to be as different as possible on the selected variables, significant differences between the groups on those variables were to be expected (Aldenderfer & Blashfield, 1984). In the present study, the purpose of conducting significance t-tests or ANOVAS using the variables that were used to form the clusters was to get an indication of whether and how clusters differed on each clustering variable (Mulry, Kalichman, Kelly, Ostrow, & Heckman, 1997), that is, to clarify the relative contributions of the different variables to the cluster solution or the degree of separation of the clusters.

Table 6.7 and Figure 6.4 show mean self-confrontation indices for country of origin broken down by cluster in country of origin. Independent samples t-tests were conducted to compare the mean self-confrontation indices for both clusters. Cluster 1 had lower S than Cluster 2, t(36) = –8.74, p < .001 (two-tailed). Cluster 1 also had lower

O than Cluster 2, t(36) = –7.31, p < .001 (two-tailed). Cluster 1 also had lower P than

151

Cluster 2, t(36) = –9.21, p < .001 (two-tailed). However, the clusters did not differ significantly on N, t(36) = .88, p = .385 (two-tailed).

Table 6.7 Mean self-confrontation indices for country of origin I-position, by cluster in country of origin

Index for country of origin I-position Cluster in country of origin S O P N

Cluster 1 (n = 21) Mean 7.15 6.64 13.28 10.83

SD 2.55 3.22 4.94 6.79

Cluster 2 (n = 17) Mean 13.55 13.13 26.31 9.06

SD 1.78 1.93 3.43 5.27

Note: S and O are scored from 0 to 20; P and N from 0 to 40. S = self-enhancement; O = union- with-the-other; P = positive affect; N = negative affect.

20 Cluster 1

Cluster 2 15 Mean score Mean

0 S-index O-index P-index N-index Index

Figure 6.4 Mean self-confrontation indices for country of origin I-position, by cluster in country of origin

Note: S and O are scored from 0 to 20; P and N from 0 to 40.

152

Clusters in Australia

Hierarchical clustering. The agglomeration schedule (Figure 6.5) and dendrogram (Figure 6.6) were examined, and a two-cluster solution selected for the same reasons as applied to clustering by self-confrontation indices for the Australian I- position.

150

100

Two-cluster solution

50 Within-cluster variance (fusion coefficient) variance Within-cluster (fusion

0 0 5 10 15 20 25 30 35 Stage

Figure 6.5 Agglomeration schedule: cluster in Australia

The dendrogram shows how participants were assigned to clusters in country of origin. Cluster 1 contained 14 participants and cluster 2, 24 participants. The length of the dendrogram branches following the first split virtually guarantees the validity of at least a 2-cluster solution.

153

Rescaled Distance Cluster Combine C A S E 0 5 10 15 20 25 Label Num +------+------+------+------+------+

Cluster 1 14 M-- 14 ─┐ 35 L-- 33 ─┤ 05 H-- 5 ─┤ 01 A-- 1 ─┤ 24 L-- 24 ─┼───┐ 08 K-- 8 ─┤ │ 12 L-- 12 ─┤ │ 03 Y-- 3 ─┘ ├───────────────────────────────────────────┐ 10 A-- 10 ─┐ │ │ 27 J-- 27 ─┤ │ │ 33 F-- 31 ─┤ │ │ 34 E-- 32 ─┼───┘ │ 32 G-- 30 ─┤ │ 36 R-- 34 ─┘ │ │ Cluster 2 │ 26 D-- 26 ─┐ │ 28 G-- 28 ─┤ │ 19 B-- 19 ─┼───┐ │ 13 I-- 13 ─┤ │ │ 17 L-- 17 ─┤ │ │ 18 E-- 18 ─┘ ├───────┐ │ 11 D-- 11 ─┐ │ │ │ 39 R-- 36 ─┤ │ │ │ 37 C-- 35 ─┤ │ │ │ 25 N-- 25 ─┼───┘ ├───────────────────────────────────┘ 15 E-- 15 ─┘ │ 02 D-- 2 ─┬─────┐ │ 21 D-- 21 ─┘ │ │ 09 S-- 9 ─┐ │ │ 42 K-- 38 ─┼─┐ ├─────┘ 06 D-- 6 ─┤ │ │ 20 S-- 20 ─┤ │ │ 16 I-- 16 ─┘ ├───┘ 07 F-- 7 ─┐ │ 40 C-- 37 ─┤ │ 22 J-- 22 ─┼─┘ 30 G-- 29 ─┤ 04 N-- 4 ─┤ 23 M-- 23 ─┘

Figure 6.6 Dendrogram of 2-cluster solution: cluster in Australia

154

K-means clustering. Table 6.8 and Figure 6.7 show that there was a perfect match between cluster membership under both clustering methods. This corroborates the previous hierarchical clustering procedure.

Table 6.8 Number of participants by cluster: Australia

K-means cluster Hierarchical cluster

Cluster 1 Cluster 2 Total

Cluster 1 14 – 14

Cluster 2 – 24 24

Total 14 24 38

20 Cluster 1 18 Cluster 2

6 hierarchical cluster centriod

Sum of squared from differences squared of Sum 4

0 13 I--- 16 I--- 22 J--- 27 J--- 12 L--- 17 L--- 24 L--- 35 L--- 07 F--- 33 F--- 03 Y--- 01 A--- 08 K--- 09 S--- 10 A--- 18 E--- 19 B--- 20 S--- 15 E--- 34 E--- 42 K--- 02 D--- 04 N--- 05 H--- 06 D--- 11 D--- 21 D--- 25 N--- 26 D--- 36 R--- 37 C--- 39 R--- 40 C--- 28 G--- 30 G--- 32 G--- 14 M--- 23 M--- Respondent

Figure 6.7 Sum of squares from centroid: cluster in Australia

155

Cluster means. Table 6.9 and Figure 6.8 show mean self-confrontation indices for country of origin broken down by country-of-origin cluster. Independent samples t-tests were conducted to compare the mean self-confrontation indices for both clusters.

Cluster 1 had higher S than Cluster 2, t(36) = 8.95, p < .001 (two-tailed). Cluster 1 also had higher O than Cluster 2, t(36) = 7.79, p < .001 (two-tailed). Cluster 1 also had higher P than Cluster 2, t(36) = 9.69, p < .001 (two-tailed). Finally, Cluster 1 had lower

N than Cluster 2, t(36) = –2.53, p = .016 (two-tailed).

Table 6. 9 Mean self-confrontation indices for Australian I-position, by cluster in Australia

Index for country of origin I-position Cluster in Australia S O P N

Cluster 1 Mean 15.59 14.51 29.91 7.23

(n = 21) SD 1.79 2.24 3.08 4.77

Cluster 2 Mean 9.03 7.48 16.06 12.32

(n = 17) SD 2.37 3.31 4.78 6.56

Note: S and O are scored from 0 to 20; P and N from 0 to 40. S = self-enhancement; O = union- with-the-other; P = positive affect; N = negative affect.

156

20 Cluster 1

Cluster 2 15 Mean score Mean

0 S-index O-index P-index N-index Index

Figure 6.8 Mean self-confrontation indices for Australian I-position, by cluster in Australia

Note: S and O are scored from 0 to 20; P and N from 0 to 40.

157

6.3 Analyses of Participant Type

Definition of participant type. Having characterised participants by their cluster in country of origin and in Australia, a new variable, termed participant “type”, was defined as the combination of the two cluster values as follows: participant type 1 combines cluster 1 in country of origin with cluster 1 in Australia; participant type 2 combines cluster 1 in country of origin with cluster 2 in Australia; participant type 3 combines cluster 2 in country of origin with cluster 1 in Australia; and participant type 4 combines cluster 2 in country of origin with cluster 2 in Australia. Table 6.10 is a crosstabulation of participants by cluster in country of origin and in Australia, where each cell represents a different participant type.

Table 6.10 Number of participants, by cluster in country of origin and in Australia

Cluster in Cluster in Australia country of origin Cluster 1 Cluster 2 Total

Cluster 1 3 18 21

Cluster 2 11 6 17

Total 14 24 38

A significant association was found between the two sets of clusters, 2(1,

N = 38) = 10.3, p < .001 (two-tailed). Members of Cluster 1 in country of origin were likely to be in Cluster 2 in Australia and unlikely to be in Cluster 1, and members of

Cluster 2 in country of origin were likely to be in Cluster 1 in Australia and unlikely to be in Cluster 2. This carries implications for categorising participants. Were the statistical association complete, participants would only be found on the diagonal, implying that a mere two participant types would serve to categorise all participants. In

158 fact a quarter of participants lay in “off-diagonal” cells, implying that more participant types are needed.

Main analyses. Table 6.11 shows means and standard deviations for self- confrontation indices by I-position location and participant type. “Location” is a variable that captures the location of the I-position (country of origin or Australia) to which the self-confrontation indices relate.

A 2-by-4 mixed-design repeated-measures MANOVA was conducted on the self- confrontation indices as dependent variables, with location as the within-subject independent variable and participant type as the between-subject independent variable.

159

Table 6.11 Mean self-confrontation indices for the I-position, by location and participant type

Location

Index Participant type Country of origin Australia

Mean SD Mean SD

Self-enhancement Type 1 (n = 3) 9.19 3.62 14.92 1.88

Type 2 (n = 18) 6.82 2.29 8.52 2.46

Type 3 (n = 11) 13.83 2.09 15.77 1.81

Type 4 (n = 6) 13.03 0.96 10.57 1.27

Union-with-the-other Type 1 (n = 3) 10.47 3.23 14.49 1.01

Type 2 (n = 18) 6.00 2.82 6.82 3.32

Type 3 (n = 11) 13.24 2.30 14.52 2.52

Type 4 (n = 6) 12.92 1.09 9.47 2.55

Positive affect Type 1 (n = 3) 18.23 4.49 27.38 3.85

Type 2 (n = 18) 12.46 4.61 15.07 4.95

Type 3 (n = 11) 26.46 4.18 30.61 2.64

Type 4 (n = 6) 26.04 1.62 19.03 2.81

Negative affect Type 1 (n = 3) 13.25 7.07 6.52 7.67

Type 2 (n = 18) 10.42 6.87 11.01 6.55

Type 3 (n = 11) 9.01 5.94 7.43 4.20

Type 4 (n = 6) 9.14 4.28 16.25 5.25

160

Main effects of type and location on I-position. A MANOVA on self- confrontation indices revealed a significant multivariate effect of participant type,

Pillai’s Trace = .88, F(12,99) = 3.42, p < .001, partial η2 = .29, and I-position location,

Pillai’s Trace = .49, F(4,31) = 7.55, p = .001, partial η2 = .50.

Student-Newman-Keuls post hoc comparisons (p < .05) showed that Type 2 was significantly lower on S (M = 7.67) and P (M = 13.77) than Type 4 (M = 11.80 and

M = 22.54 respectively) and Type 1 (M = 12.05 and M = 22.80), and that both these types were significantly lower than Type 3 (M = 14.80 and M = 28.53). A post hoc comparison showed that Type 2 was significantly lower on O (M = 6.41) than Type 4

(M = 11.19), Type 1 (M = 12.48) and Type 3 (M = 13.88). A post hoc comparison showed no significant differences between types on N.

These main effects were qualified by a significant interaction between location and type, Pillai’s Trace = .82, F(12,99) = 3.13, p = .001, partial η2 = .28. Simple effects analyses were conducted comparing self-confrontation indices among the four types at each I-position location.

161

Interactions of type and location on I-position: self-enhancement. Univariate comparisons showed significant differences between participant types on self- enhancement (S). Figure 6.9 compares mean S for country of origin and Australia.

Although S differed significantly according to the location of the I-position,

F(3,34) = 15.10, p < .001, partial η2 = .57, the effect differed by participant type.

A Student-Newman-Keuls post hoc comparison (p < .05) showed that Type 2

(M = 6.82) and Type 1 (M = 9.19) were significantly lower on S for country of origin than Type 4 (M = 13.03) and Type 3 (M = 13.83). A second Student-Newman-Keuls post hoc comparison (p < .05) showed that Type 2 (M = 8.52) and Type 4 (M = 10.57) were significantly lower on S for Australia than Type 1 (M = 14.92) and Type 3

(M = 15.77).

Type 1 10 Type 2

S-index 8 Type 3 Type 4 6

0 Origin Australia Location

Figure 6.9 Mean self-enhancement for country of origin and Australian I-position, by participant type

162

Interactions of type and location on I-position: union-with-the-other.

Univariate comparisons showed significant differences between participant types on union-with-the-other (O). Figure 6.10 compares mean O for country of origin and

Australia. Although O differed significantly according to the location of the I-position,

F(3,34) = 9.06, p < .001, partial η2 = .44, the effect differed by participant type.

A Student-Newman-Keuls post hoc comparison (p < .05) showed that Type 2

(M = 6.00) was significantly lower on O for country of origin than Type 1 (M = 10.47),

Type 4 (M = 12.92) and Type 3 (M = 13.24). A second Student-Newman-Keuls post hoc comparison (p < .05) showed that Type 2 (M = 6.82) and Type 4 (M = 9.47) were significantly lower on O for Australia than Type 1 (M = 14.49) and Type 3 (M = 14.52).

10 Type 1

8 Type 2

O-index Type 3 6 Type 4

0 Origin Australia Location

Figure 6.10 Mean union-with-the-other for country of origin and Australian I-position, by participant type

163

Interactions of type and location on I-position: positive affect. Univariate comparisons showed significant differences between participant types for positive affect

(P). Figure 6.11 compares mean P for country of origin and Australia. Although P differed significantly according to the location of the I-position, F(3,34) = 21.89, p < .001, partial η2 = .66, the effect differed by participant type.

A Student-Newman-Keuls post hoc comparison (p < .05) showed that Type 2

(M = 12.46) was significantly lower on P for country of origin than Type 1 (M = 18.23), which was significantly lower on P for country of origin than Type 4 (M = 26.04) and

Type 3 (M = 26.46). A second Student-Newman-Keuls post hoc comparison (p < .05) showed that Type 2 (M = 15.07) and Type 4 (M = 19.03) were significantly lower on P for Australia than Type 1 (M = 27.38) and Type 3 (M = 30.61).

20 Type 1 Type 2

P-index Type 3 15 Type 4

0 Origin Australia Location Figure 6.11 Mean positive affect for country of origin and Australian I-position, by participant type

164

Interactions of type and location on I-position: negative affect. Univariate comparisons showed significant differences between participant types for negative affect (N). Figure 6.12 compares mean N for country of origin and Australia. Although

N differed significantly according to the location of the I-position, F(3,34) = 10.14, p < .001, partial η2 = .47, the effect differed by participant type.

A Student-Newman-Keuls post hoc comparison (p < .05) showed no significant differences on N for country of origin. A second Student-Newman-Keuls post hoc comparison (p < .05) showed that the subset comprising Type 1 (M = 6.52), Type 3

(M = 7.42) and Type 2 (M = 11.01) was significantly lower on N in Australia than the subset comprising Type 2 (M = 11.01) and Type 4 (M = 16.25).

Type 1 10 Type 2

N-index 8 Type 3 Type 4 6

0 Origin Australia Location

Figure 6.12 Mean negative affect for country of origin and Australian I-position, by participant type

165

Type by location interaction for self-confrontation indices. Figure 6.13 shows mean self-confrontation indices for both the country of origin and Australian I-positions broken down by participant type (“Difference in index” is defined as the index for the

Australian I-position less the analogous index for the country of origin I-position).

Paired t-tests were used to compare analogous self-confrontation indices for country of origin and Australia broken down by participant type. The purpose was to determine which means were significantly different, and if so, to what degree.

10 Solid markers show significant changes (p < .05) 8

2 Type 1 Type 2 0 Type 3 Type 4

Difference in index -2

-4

-6

-8 S-index O-index P-index N-index Index

Figure 6.13 Differences in self-confrontation indices for country of origin and Australian I-positions, by participant type

In the case of participant Type 1, S for Australia (M = 14.92, SD = 1.88) was significantly higher than for country of origin (M = 9.19, SD = 3.62), t(2) = 5.44, p = .032 (two-tailed); and P for Australia (M = 27.38, SD = 3.85) was also significantly higher than for country of origin (M = 18.23, SD = 4.49), t(2) = 7.21, p = .019 (two- tailed). O and N did not differ significantly by location.

166

In the case of participant Type 2, S for Australia (M = 8.52, SD = 2.46) was significantly higher than for country of origin (M = 6.82, SD = 2.29), t(17) = 3.94, p = .001 (two-tailed); and P for Australia (M = 15.07, SD = 4.95) was also significantly higher than for country of origin (M = 12.46, SD = 4.61), t(17) = 3.20, p < .01 (two- tailed). O and N did not differ significantly.

In the case of participant Type 3, S for Australia (M = 15.77, SD = 1.81) was significantly higher than for country of origin (M = 13.83, SD = 2.09), t(10) = 2.97, p = .014 (two-tailed); and P for Australia (M = 30.61, SD = 2.64) was also significantly higher than for country of origin (M = 26.46, SD = 4.18), t(10) = 3.88, p = .003 (two- tailed). O and S did not differ significantly.

In the case of participant Type 4, S for Australia (M = 10.57, SD = 1.27) was significantly lower than for country of origin (M = 13.03, SD = 0.96), t(5) = –9.32, p < .001 (two-tailed); O for Australia (M = 9.47, SD = 2.55) was significantly lower than for country of origin (M = 12.92, SD = 1.09), t(5) = –3.87, p = .012 (two-tailed); and P for Australia (M = 19.03, SD = 2.81) was also significantly lower than for the country of origin (M = 26.04, SD = 1.62), t(5) = –8.19, p < .001 (two-tailed).

Conversely, N for Australia (M = 16.25, SD = 5.25) was significantly higher than for country of origin (M = 9.14, SD = 4.28), t(5) = 5.17, p = .004 (two-tailed).

Naming of participant types. Figure 6.14 shows the self-confrontation indices both in country of origin and in Australia relating to each participant type. Each of the four participant types was accorded a mnemonic name on the basis of its pattern of self- confrontation indices and the differences between valuations pertaining to the country of origin I-position and to the Australian I-position. Since self-enhancement, union- with-the-other, and positive affect were highly correlated, and negatively correlated

167

with negative affect, participant types were described by whether their S, O, and P were

relatively higher or lower for country of origin as compared with Australia.

Cluster 1 in Australia Cluster 2 in Australia

Type 1 (Progressive) Type 2 (Low-stable)

30 30 Origin Origin Australia Australia 25 25

20 20

15 15

10 10

Cluster 1 in 1 Cluster 5 5 country of origin country 0 0 S index O index P index N index S index O index P index N index

n = 3 n = 18

Type 3 (High-stable) Type 4 (Regressive)

30 30 Origin Origin Australia Australia 25 25

20 20

15 15

10 10

Cluster 2 in 2 Cluster 5 5 country of origin country 0 0 S index O index P index N index S index O index P index N index

n = 11 n = 6

Note: Solid markers show significant differences (p < .05).

Figure 6.14 Mean self-confrontation indices, by participant type

Participant type 1: “Progressive”. For this group (n = 3; 8%) the country of

origin I-position was characterised by a low S, O and P, and high N; but the Australian

I-position by high S, O and P, and a low N. When the two I-positions were compared,

the Australian was found to have significantly higher S and P. In summary, these are

168 people whose Australian I-position is perceived in a more positive light than their country of origin I-position.

In Gergen & Gergen’s (1986) typology, a “progressive” narrative links events “in such a way that one steadily progresses towards a goal” (p. 27). Consequently this group is termed “Progressive” as their well-being is higher for their Australian I- position, suggesting greater fulfilment of their S and O motives through migration.

Participant type 2: “Low-stable”. For this group (n = 18; 47%) the country of origin I-position was characterised by low S, O and P, and high N; and the Australian I- position also by low S, O and P, and high N. When the two I-positions were compared, the Australian was found to have significantly higher S and P, though the differences were small. In summary, these are people who perceive both their I-positions more negatively than do the other participant types.

In Gergen & Gergen’s (1986) typology, a “stability” narrative links events “in such a way that the protagonist remains essentially unchanged with respect to evaluative position” (p. 27). Consequently this group is termed “Low-stable” as their well-being remains low for their Australian I-position, suggesting largely unchanged fulfilment of their S and O motives through migration.

Participant type 3: “High-stable”. For this group (n = 11; 29%) the country of origin I-position was characterised by high S, O and P, and low N; and the Australian I- position also by high S, O and P, and low N. When the two I-positions were compared, the Australian was found to have significantly higher self S and P, though the differences were small. In summary, these are people who perceive both their I- positions more positively than do the other participant types.

In Gergen & Gergen’s (1986) typology, a “stability” narrative links events “in such a way that the protagonist remains essentially unchanged with respect to evaluative

169 position” (p. 27). Consequently this group is termed “High-stable” as their well-being remains high for their Australian I-position, suggesting largely unchanged fulfilment of their S and O motives through migration.

Participant type 4: “Regressive”. For this group (n = 6; 16%) the country of origin I-position was characterised by high S, O and P, and low N; but the Australian I- position by low S, O and P, and high N. When the two I-positions were compared, the

Australian was found to have significantly lower S, O and P, and higher N. In summary, these are people whose Australian I-position is perceived in a more negative light than their country of origin I-position.

In Gergen & Gergen’s (1986) typology, a “regressive” narrative is one in which one “is continuously moving away from the valued state” (p. 27). Therefore their self- narrative should reflect relative decrease in well-being, not only by location but also through time. Consequently this group is termed “Regressive” as their well-being is lower for their Australian I-position, suggesting reduced fulfilment of their S and O motives through migration.

Patterns of change. Two overall patterns—one absolute, one relative—are observed when comparing country of origin and Australian I-positions. In absolute terms the Australian I-positions of most participant types—84% of the sample—had a higher S and P (though not O) than their country of origin I-positions; and one participant type, Type 4 (Regressive), showed the opposite. But in relative terms the

Australian I-positions of most participants—the two “stable” types, totalling 76% of the sample—stayed relatively the same; that is, the ranking of S and P for country of origin and Australia, as compared with those of other participant types, did not differ.

170

6.4 Comparison of Participant Types

This section examines how the four participant types differed on demographic characteristics, on measures of psychological distress and of well-being, and on the self- confrontation indices.

Demographic characteristics. Table 6.12 shows the nominal demographic variables cross-tabulated by participant type. Chi-square analysis revealed no differences across participant types for gender, χ2(3, N = 38) = .49, p = .92 (two-tailed), marital status, χ2(6, N = 38) = 2.47, p = .87 (two-tailed), education, χ2(12,

N = 38) = 8.46, p = .75 (two-tailed), or language, χ2(6, N = 38) = 9.46, p = .15 (two- tailed). However, participant types did differ on ethnic identification, χ2(3,

N = 38) = 13.11, p < .01 (two-tailed). Those who identified with their ethnic group were more likely to be found among Type 1 (Progressive) (38% as compared to an expected

8%) and less likely to be among Type 4 (Regressive) (0% as compared to an expected

16%). Thus it seems that ethnic identification predisposes to higher S and O, and higher

P for the Australian I-position (though it is noted that this finding is based on a very small sample).

171

Table 6.12 Tabulation of demographic characteristics, by participant type

Participant type Total

Type 1 Type 2 Type 3 Type 4 N % (Progressi (Low- (High- (Regressi ve) stable) stable) ve)

Gender Male 1 6 5 2 14 37%

Female 2 12 6 4 24 63%

Marital Single 5 3 2 10 26% status Married 3 10 6 3 22 58%

Divorced 3 2 1 6 16%

Education Before yr 10 1 1 2 5%

Year 10 1 1 3%

Year 12 2 1 1 4 11%

Undergraduate 5 3 1 9 24%

Postgraduate 2 9 7 4 22 58%

Language English only 1 8 5 1 15 39% at home Bilingual 2 1 2 1 6 16%

Mother tongue 9 4 4 17 45%

Ethnic Belongs to 3 3 2 8 21% identifica- group tion No belonging 15 9 6 30 79%

Total N 3 18 11 6 38

% 8% 47% 29% 16%

Table 6.13 shows the continuous demographic variables cross-tabulated by participant type. ANOVA revealed no differences across participant types for years in

Australia, F(3,34) = 2.34, p = .09, age, F(3,34) = 1.26, p = .30, and individualism,

F(3,34) = .31, p = .82.

172

Table 6.13 Mean and standard deviation of demographic characteristics, by participant type

Type

Type 1 Type 2 Type 3 Type 4 (Progressi (Low- (High- (Regressi All ve) stable) stable) ve)

Years in Australia Mean 31.0 16.4 19.6 24.3 19.8

Standard deviation 11.0 10.6 8.9 9.6 10.5

Age Mean 50.7 39.4 42.8 42.5 41.8

Standard deviation 7.5 9.9 8.8 11.3 9.8

Individualism Mean 42.3 44.3 39.8 35.0 41.4

Standard deviation 31.7 31.7 31.7 31.7 31.7

Self-report measures of distress and well-being.

Psychological distress. Table 6.14 shows means and standard deviations for the measures of psychological distress by participant type. Participant types did not differ in their psychological distress: a single-factor between-subjects MANOVA with participant type as the independent variable was non-significant, Wilks’ λ = .738,

F(3,32) = 1.150, p = .339. Univariate comparisons also showed no significant differences in SAD, F(3,32) = 1.02, p = .40; in FNE, F(3,32) = 1.76, p = .17; or in BDI,

F(3,32) = 1.02, p = .40.

173

Table 6.14 Mean psychological distress, by participant type

Psychological distress Participant type SAD FNE BDI–II

Type 1 (Progressive) Mean 2.00 10.00 11.33

(n = 3) SD 1.73 10.15 9.07

Type 2 (Low-stable) Mean 9.11 15.72 10.61

(n = 18) SD 7.66 10.45 11.65

Type 3 (High-stable) Mean 7.82 8.45 4.91

(n = 11) SD 6.15 6.39 4.74

Type 4 (Regressive) Mean 10.33 16.33 10.83

(n = 6) SD 8.73 9.33 7.33

N = 38

Planned comparisons were performed to compare each participant type with every other participant type in respect of the three psychological distress measures. In respect of SAD, Type 1 differed significantly from Type 2 (t = –3.45, df = 16, p < .01), and from Type 3 (t = –2.76, df = 12, p < .05). Type 1 achieved a borderline difference from

Type 4 (t = –2.25, df = 6, p = .067)—a consequence, no doubt, of these groups’ small sample sizes. No other planned comparisons were significant. This indicates that Type 1

(Progressive) participants scored generally lower on SAD than the other types. In respect of FNES, Type 2 differed significantly from Type 3 (t = 2.32, df = 27, p < .05):

Type 2 (Low-stable) participants scored higher than Type 3 (High-stable). No significant differences were found between participant types on BDI-II. In conclusion, the incongruent results between the mean comparisons with the t-tests for the Type 4

(Regressive) participants may well be attributed to the small sample size.

Psychological well-being. Table 6.15 shows means and standard deviations for the measures of psychological well-being by participant type. A single-factor between-

174 subjects MANOVA with participant type as the independent variable was conducted and found to be marginally significant, Wilks’ λ = .40, F(18,83) = 1.77, p = .044.

Univariate comparisons showed that the difference between participant types was confined to PWB Positive relations with others, F(3,32) = 4.64, p = .008.

A post hoc comparison showed that Type 4 (Regressive) scored significantly lower on PWB Positive relations with others than did Type 1 (Progressive) and Type 3

(High-stable) (Student-Newman-Keuls, p < .05). The mean score for participants in

Type 2 (Low-stable) fell between the means for Type 4 (Regressive) and Type 1

(Progressive), and between the means for Type 4 (Regressive) and Type 3 (High- stable), but was not significantly different from them.

Table 6.15 Mean psychological well-being, by participant type

Psychological Well-being Scale Participant type PWB1 PWB2 PWB3 PWB4 PWB5 PWB6

Type 1 (Progressive) Mean 60.00 75.00 64.33 76.33 63.67 62.33

(n = 3) SD 8.66 7.00 9.50 4.16 7.64 10.97

Type 2 (Low-stable) Mean 59.44 63.94 59.50 70.89 62.33 62.17

(n = 18) SD 13.26 12.32 12.66 9.00 11.60 12.68

Type 3 (High-stable) Mean 71.00 75.00 70.73 76.18 73.09 71.45

(n = 11) SD 10.71 8.92 10.48 6.91 9.12 12.47

Type 4 (Regressive) Mean 61.17 57.83 61.00 69.67 65.83 60.17

(n = 6) SD 5.49 8.47 8.05 6.44 7.14 9.11

N = 38. Note: PWB1 = Environmental mastery; PWB2 = Positive relations with others; PWB3 = Autonomy; PWB4 = Personal growth; PWB5 = Purpose in life; PWB6 = Self- acceptance.

175

Self-confrontation indices.

Country of origin. Table 6.16 shows means and standard deviations for the country of origin I-position by participant type. A single-factor between-subjects

MANOVA with participant type as the independent variable was significant, Wilks’

λ = .21, F(12,82) = 5.49, p < .001. Univariate comparisons revealed significant differences across participant types for S, F(3,34) = 27.73, p < .001); for O,

F(3,34) = 23.62, p < .001; and for P, F(3,34) = 32.25, p < .001; but not for N,

F(3,34) = .42, p = .74.

Table 6.16 Mean self-confrontation indices for country of origin I-position, by participant type

Self-confrontation indices Participant type S O P N

Type 1 (Progressive) Mean 9.19 10.47 18.23 13.25

(n = 3) SD 3.62 3.23 4.49 7.07

Type 2 (Low-stable) Mean 6.82 6.00 12.46 10.42

(n = 18) SD 2.29 2.82 4.61 6.87

Type 3 (High-stable) Mean 13.83 13.24 26.46 9.01

(n = 11) SD 2.09 2.30 4.18 5.94

Type 4 (Regressive) Mean 13.03 12.92 26.04 9.14

(n = 6) SD 0.96 1.09 1.62 4.28

A post hoc comparison showed that Type 1 (Progressive) and Type 2 (Low- stable) scored significantly lower on S than Type 3 (High-stable) and Type 4

(Regressive) (Student-Newman-Keuls, p < .05). A post hoc comparison showed that

Type 2 (Low-stable) was significantly lower on O than Type 1 (Progressive), Type 3

(High-stable) and Type 4 (Regressive) (Student-Newman-Keuls, p < .05). A post hoc

176 comparison showed that Type 2 (Low-stable) scored significantly lower on P than did

Type 1 (Progressive), and that both of these types scored significantly lower than

Type 3 (High-stable) and Type 4 (Regressive) (Student-Newman-Keuls, p < .05).

Australia. Table 6.17 shows means and standard deviations for the Australian I- position by participant type. A single-factor between-subjects MANOVA with participant type as the independent variable was significant, Wilks’ λ = .18,

F(12,82) = 6.24, p < .001. Univariate comparisons showed significant differences across participant types for S, F(3,34) = 30.22, p < .001; for O, F(3,34) = 18.63, p < .001; for

P, F(3,34) = 36.58, p < .001, and for N, F(3,34) = 3.46, p < .05.

Table 6.17 Mean self-confrontation indices for the Australian I-position, by participant type

Self-confrontation indices Participant type S O P N

Type 1 (Progressive) Mean 14.92 14.49 27.38 6.52

(n = 3) SD 1.88 1.01 3.85 7.67

Type 2 (Low-stable) Mean 8.52 6.82 15.07 11.01

(n = 18) SD 2.46 3.32 4.95 6.55

Type 3 (High-stable) Mean 15.77 14.52 30.61 7.43

(n = 11) SD 1.81 2.52 2.64 4.20

Type 4 (Regressive) Mean 10.57 9.47 19.03 16.25

(n = 6) SD 1.27 2.55 2.81 5.25

A post hoc comparison showed that Type 2 (Low-stable) and Type 4 (Regressive) scored significantly lower on S than did Type 1 (Progressive) and Type 3 (High-stable)

(Student-Newman-Keuls, p < .05). A post hoc comparison showed that Type 2 (Low- stable) and Type 4 (Regressive) were significantly lower on O than were Type 1

177

(Progressive) and Type 3 (High-stable) (Student-Newman-Keuls, p < .05). A post hoc comparison showed that Type 2 (Low-stable) and Type 4 (Regressive) scored significantly lower on P than did Type 1 (Progressive) and Type 3 (High-stable)

(Student-Newman-Keuls, p < .05). A post hoc comparison showed that Type 1

(Progressive) and Type 3 (High-stable) scored significantly lower on N than did Type 4

(Regressive) (Student-Newman-Keuls, p < .05). The mean score for participants in

Type 2 (Low-stable) fell between the means for Type 1 (Progressive) and Type 3 (High- stable) and the mean for Type 4 (Regressive) but was not significantly different.

6.5 Summary

In Study 1, participants were first characterised in terms of demographic descriptors, psychological self-reports and self-confrontation indices. Cluster analysis was then employed as a data reduction technique. The participants were grouped into four types: Progressive, Low-stable, High-stable and Regressive. Following that, mixed- design repeated-measures MANOVAs were conducted on the self-confrontation indices as dependent variables, with location of the I-position (country of origin or Australia) as the within-subject independent variable and participant type as the between-subject independent variable.

Because the main effects were qualified by significant interactions between location and participant type, separate paired t-tests were conducted to compare self- confrontation indices pertaining to each participant’s I-position in country of origin with those pertaining to the I-position in Australia. Having established that participants could be grouped in types, and that the types differed in terms of self-confrontation indices,

MANOVA was used to explore where and how the participant types differed in terms of demographics and psychological self-reports.

178

Perhaps the most significant finding concerned the stability of I-positions.

Participants belonging to Types 2 (Low-stable) and 3 (High-stable), together comprising

76% of the sample, evinced little difference between their Australian and country-of- origin I-positions. True, their self-enhancement (S) and positive affect (P) were significantly higher for Australia than for country of origin, but only marginally so.

Only Types 1 (Progressive) and 4 (Regressive), together comprising 14% of the sample, showed much difference.

Another finding concerned the self-confrontation indices of participant types.

Participants belonging to Types 1 (Progressive), 2 (Low-stable) and 3 (High-stable), together comprising 84% of the sample, showed on average more self-enhancement and positive affect for their Australian valuations than for their country-of-origin valuations.

Type 4 (Regressive), comprising 16% of the sample, was the only group not to benefit in this way: not only did they experience less self-enhancement and union-with- the-other motives but their well-being declined as a result of reduced positive affect and increased negative affect. Interestingly, for Types 1, 2 and 3 the self-enhancement motive (S) was significantly higher for Australia than for country of origin but the union-with-the-other motive (O) was not significantly different.

A third finding concerned differences in psychological distress and well-being. As measured by SAD, Type 1 (Progressive) participants were in general less socially anxious than the other types, though the difference from Type 4 (Regressive) was of borderline significance due to the smallness of the sample. As measured by FNE, Type

2 (Low-stable) participants were more fearful of negative evaluation than Type 3 (High- stable). Again, the smallness of the sample appears to have prevented significance differences being found where otherwise they would have been expected: between, say,

179

Type 1 (Progressive) and Type 4 (Regressive). Lastly, no significant differences were found between participant types on depression as measured by BDI-II.

Turning now to psychological well-being, significant differences were found in one of these: Positive relations with others. Type 4 (Regressive) participants had significantly worse ‘relations with others’ than did Type 1 (Progressive) and Type 3

(High-stable). Type 2 (Low-stable) participants on average rated above Type 4

(Regressive) and below Type 1 (Progressive) and Type 3 (High-stable), but the differences were non-significant.

In conclusion, this chapter documents the first of three studies performed as part of thesis. Like the second study, which is discussed next, it is nomothetic in nature, but differs from it in the way the data were organised at the case level.

180

CHAPTER 7: Study 2 – The Dialogical Self as a Probability Distribution

This chapter discusses the second study conducted as part of this thesis. (A version of this material has appeared in the International Journal for Dialogical Science

(Sanchez-Rockliffe & Symons, 2010.) It characterises the dialogical self as a hypothetical probability distribution of themes of valuations, wherein the self- confrontation method is a technique for generating a sample of themes from which the properties of this distribution can be inferred, as in standard sampling theory. Since the dialogical self emerges from a probability distribution of valuation themes, a sampling interpretation is a fruitful way to view this technique. It was in this context that hypotheses concerning stability of the self were developed and formally tested.

Study 2 corroborated Study 1 in a broad sense. Unlike Study 1, which used cluster analysis to group individuals, Study 2 classified them by comparing their valuations in their country of origin and Australia. But despite the difference in analytical approach,

Study 2 also found differences in migration experience that in key ways resembled those of Study 1. In particular, both studies identified broadly similar patterns of stability and change in I-positions.

7.1 The Dialogical Self as a Probability Distribution of Themes

This section presents a conceptual model of the dialogical self that is grounded in statistical sampling theory. Such an approach is uniquely appropriate because the dialogical self is by nature not one thing but many, and so demands measurement methods that recognise the fact. Unlike most psychological variables, which describe a single characteristic of the individual, the dialogical self describes a multitude of semi- autonomous selves and their inter-relationships (Hermans & Kempen, 1993; Hermans &

Hermans-Jansen, 1995). In this sense, therefore, it is more like a community than an

181 individual. Communities are conveniently described not by considering a single member but by considering a sample and expressing the results as frequency distributions and correlations. This approach makes clear the basis by which statistically testable hypotheses are generated.

In this model, every individual can generate a potentially infinite number of valuations, and it is through these valuations that the individual’s dialogical self is expressed. Each valuation furthermore has a “theme”, by which is meant its vector of S,

O, P, N variables as measured by the SCM. It follows therefore that the dialogical self is described by a hypothetical multidimensional probability distribution of valuation themes.

Graphical representation. Figure 7.1 shows the probability distribution of themes for a hypothetical individual. Each point on the horizontal plane denotes a particular theme, and the height of the graphical surface at each point represents the probability with which the individual will generate valuations having that theme.

Alternatively the figure can be regarded as showing the relative proportion of themes in the total set of valuations. The distribution of themes, considered as a complex but well- defined mathematical entity, is thus a characteristic of the dialogical self. As with any model, it is a simplification—for instance it does not capture inter-relationships between

I-positions—but it does provide an objective and useful description of the dialogical self by conveying the relative weight of various themes in the life of the individual.

182

Mode 1 Mode 1

Global mean

Mode 2

Mode 3 Mode 3 Frequency

Note: Because the figure is necessarily limited to three dimensions for illustrative purposes, the four thematic dimensions (S, O, Thematic dimension 2 P and N) have been reduced to the Thematic dimension 1 two horizontal dimensions, and the third (vertical) dimension represents the frequency (or alternatively probability) with which a given self- narrative theme occurs. Figure 7.1 Hypothetical probability distribution of valuation themes: oblique view and plan view (inset)

183

The properties of the distribution need to be inferred from a sample of valuation themes. This sample is provided by the SCM: participants generate a sample of valuations, which are then scored to establish their themes. The modes of the distribution are of particular interest as they correspond to recurrent themes in valuations. For a multi-modal distribution, the mean is less informative than the modes.

It is, however, easy to calculate, and the average thematic value of a set of valuations is clearly a useful statistic. In general, inferring the detailed structure of a multivariate, multimodal probability distribution is difficult with modest sample sizes, as here. Hence a variety of ad hoc procedures are employed in the current research to study these distributions.

I-positions. The I-position is conceived of as the hypothetical author of a valuation. This study focuses on the themes of valuations, as it is these that are directly measurable. Participants were required to perceive themselves as having two separate cultural identities—that is, they should think of themselves as having achieved bicultural identification, one identity belonging to their country of origin and the other identity belonging to Australia. Later, in the SCM, participants were invited to focus first on their cultural identity of their country of origin, then on their cultural identity in

Australia, and to tell about important experiences and circumstances from these particular perspectives.

Thematic modes. Themes that regularly occur in valuations are of particular interest, as they reflect psychological invariance over I-positions. These valuations will be revealed as modes of the probability distribution of themes (see Figure 6.18). In a sample of valuations, clusters of roughly similar theme appear as modes. Strongly recurrent themes will tend to produce larger clusters and thus larger hills around the

184 mode. Figure 7.1 is drawn with three modes. In this case three sets of similar themes would tend to re-occur in the valuations of the particular individual.

Themes and migration. This section explores what happens to the distribution of themes when a person experiences a dramatic change in life circumstances, as for example occurs after migration. A requirement of the study was that participants acknowledge both an Australian identity and a continued identity associated with the country of origin. This implies that living as a migrant brings into being new I- positions. However, it may be that these new I-positions are thematically similar to corresponding I-positions in the country of origin, at least in some cases. For instance, the themes of domestic I-positions (I-as-parent, I-as-spouse etc.) might change little in a migrant who brings his or her family, but those for work I-positions might change greatly. The probability surface would then shift, more in some places than others. The study gauged the extent to which migration changed the thematic distribution for this sample.

7.2 Hypotheses

The aim of Study 2 was to explore how the dialogical self is affected by a radical change in the social environment, in particular by migration to a new country. However, since the temporal stability of the self is one of its abiding characteristics (James, 1890), an explanation is required as to how such change might occur. According to Hermans

(1987a, b, 1988, 1992), valuations have both latent and manifest content. The latent content is represented by the S and O motives; the manifest, by the P and N affects. In this theoretical framework, the S and O motives represent deep psychological drives; the

P and N affects represent the extent to which these motives are gratified in a particular social environment, and hence manifest as affects.

185

For an individual placed in two different environments, one can therefore consider four possibilities.

Hypothesis 1: Latents constant, manifests constant. This is the case of complete thematic stability. It would occur when the individual’s fundamental motives are constant and each environment offers an equivalent opportunity for gratification.

Hypothesis 2: Latents constant, manifests change. This is perhaps the most natural case. The new environment offers different possibilities for gratification but the individual’s fundamental motives remain the same.

Hypothesis 3: Latents change, manifests constant. This could arise from psychological development over the life cycle. As an individual progresses through life, it is natural to expect fundamental motives to change. This may not be accompanied by affective change.

Hypothesis 4: Latents change, manifests change. In this case the fundamental motives of the individual change, as does the extent of gratification. This could come about if the new environment creates new I-positions on the one hand, or simply for developmental reasons on the other.

7.3 Analysis Based on Major Themes

Two similar but comparable analyses were performed in Study 2. In the first analysis, participants were grouped on the basis of the profiles of major themes manifested by their valuations. This was done separately for valuations pertaining to the country of origin and to Australia. A transition matrix was then generated to determine which participants were in the same group for both sets of valuations, and which participants changed groups. Formal tests of statistical significance were then performed on the transition matrices to test the null hypothesis that participants remained in the same groupings, that is, that migration produced no significant change in I-position.

186

Thematic nature of I-positions. In Study 1, valuations (and hence participants) have been characterised in terms of self-confrontation indices. In this study, valuations were classified according to the six “major themes” defined by Hermans and Hermans-

Jansen (1995) rather than by indices. The rules for classifying valuations are set out in

Table 7.1 and are largely those used by Hermans and Hermans-Jansen (1995, p.254). As some latitude is permitted, however, the definitions of “high” or “low” self- enhancement and union-with-the-other were set at 17 and 4 respectively for this study.

This typology is not exhaustive, as valuations can fail to meet the criteria for inclusion under any of the major themes; by setting the rules in this way, the 1565 valuations obtained from all migrants were reduced a more manageable 570. The analysis was based on these.

Table 7.1 Rules for assigning valuations to major themes

Criteria for assignment Major theme P – N S – O S O

Strength and unity (+HH) ≥ 12 … > 17 > 17

Unity and love (+O) ≥ 12 ≤ –6 … …

Unfulfilled longing (–O) ≤ –12 ≤ –6 … …

Powerlessness and isolation (–LL) ≤ –12 … < 4 < 4

Aggression and anger (–S) ≤ –12 ≥ 6 … …

Autonomy and success (+S) ≥ 12 ≥ 6 … …

Key: S = Self-enhancement; O = Union-with-the-other; P = Positive affect; N = Negative affect.

Participant types were characterised by the major themes that their valuations revealed, yielding a thematic profile of each type. Table 7.2 tabulates those valuations by participant type, by I-position location (that is, country of origin or Australia), and by major theme. Each row represents the profile of the stipulated I-position of each

187 participant type. Since the absolute number of valuations in each row is arbitrary, it is the proportion assigned to each theme that is meaningful.

Table 7.2 Number of valuations, by participant type and major theme

Major theme

Power- Auto- Participant Strength Unity Unfulfil- lessness Aggres- nomy type and and led and sion and and Total unity love longing isolation anger success (+HH) (+O) (–O) (–LL) (–S) (+S)

Type 1 Origin 6 4 6 7 3 26 (Progressive) Australia 11 1 1 5 18

Total 17 4 6 8 1 8 44

Type 2 Origin 4 24 15 53 4 53 153 (Low-stable) Australia 9 8 5 39 8 48 117

Total 13 32 20 92 12 101 270

Type 3 Origin 52 7 7 10 2 28 106 (High-stable) Australia 40 3 1 2 17 63

Total 92 10 8 12 2 45 169

Type 4 Origin 24 4 3 9 8 48 (Regressive) Australia 4 1 5 8 11 10 39

Total 28 5 8 17 11 18 87

Grand total 150 51 42 129 26 172 570

Procedure for grouping participants. Participants were classified according to the following procedure:

Participants were represented by the proportions of all valuations in each major theme. Two profiles were produced for each participant: one pertaining to all valuations, termed the “global profile”, and one for valuations at country of origin.

188

Proportions were regarded as “important” if they were 20% or more of the total number of valuations; otherwise they were regarded as unimportant.

Participants were assigned to a common group if they had important proportions in precisely the same categories.

For this purpose, an important proportion in strength and unity (+HH) was regarded as equivalent to important proportions in both autonomy and success (+S) and unity and love (+O).

Likewise, an important proportion in powerlessness and isolation (–LL) was regarded as equivalent to important proportions in both aggression and anger (–S) and unfulfilled longing (–O).

Table 7.3 shows the resulting classification (names of participants in this table and subsequently were changed to preserve anonymity). Each row of the table gives the proportion of each participant’s valuations expressing each major theme. Proportions pertaining “important” major themes are bolded and underlined. The resulting vector of proportions is termed the participant’s “thematic profile”.

Ideally the country-of-origin profile for each participant would have been compared with the corresponding Australian one. This was not done because some participants had too few Australian valuations with major themes (two or fewer) to render comparison meaningful; they would have had to be dropped from the sample. By comparing the country-of-origin profile with the global profile the problem was mitigated, but at the cost of creating another one: the null hypothesis was now harder to reject since the global profile was contaminated with country-of-origin valuations. In the event this did not matter as this null was still rejected, for rejection of the one hypothesis entails rejection of the other. Under this approach only nine of the 38 in the sample needed to be dropped, limiting the test sample to 29.

189

On the basis of this table, participants were classified into two groups: “univalent” and “multivalent”. Univalent participants are those who expressed only a single affect, either positive or negative (but not both) in their valuations. Multivalent participants were those who expressed a mix of both positive and negative affects. The distinction is crucial as it is the basis for determining the stability of thematic profiles. Multivalent participants are of particular interest because they have unambiguously multimodal major themes and hence exhibit instability in their thematic profiles. These individuals are thus much in line with Hermans’ view of the dialogical self. It is noteworthy that this instability is not merely an artefact of migrants’ differential states of mind at origin and destination: of the 15 multivalent participants, 12 were multivalent at country of origin.

Of the univalent respondents, one expressed only –O; three expressed both –S and

–O; six expressed only +S; and 13 expressed both +S and +O. Of the multivalent respondents, three expressed a mix of most major themes (+S, +O, –S and –O), two expressed +S but also –LL, two expressed +O but also –LL, one expressed +HH but also –O, one expressed +HH but also –S, and six expressed +S but also –LL. The three most common profiles are those whose valuations expressed only autonomy and success

(+S); those that expressed both motives (S and O), both with positive affect; and those that expressed autonomy and success (+S), but combined with powerlessness and isolation (–LL).

To judge whether country-of-origin profiles were importantly different from global profiles, participants were subjectively classed as either “stable” or “unstable” on the basis of the degree of difference between their respective profiles. Of the 29 participants classed in this way, 23 were considered stable and six unstable.

190

Table 7.3 Thematic profiles showing “important” major themes, by participant and location

All valuations Valuations at country of origin

Positive affect Negative affect Positive affect Negative affect Stability Distance

+HH +S +O –LL –S –O +HH +S +O –LL –S –O

Univalent

–O only 17 Ljubica 0.10 0.20 0.10 0.60 0.10 0.20 0.10 0.60 X n.a.

–S and –O 07 Frida 0.11 0.06 0.17 0.33 0.33 0.00 0.10 0.10 0.10 0.50 0.20 S 0.23

03 Yoana 0.20 0.10 0.10 0.40 0.20 0.17 0.50 0.33 U 0.29

02 Dragan 0.04 0.86 0.04 0.07 1.00 S 0.17

+S only 28 Gonzalo 0.11 0.50 0.07 0.18 0.14 0.05 0.53 0.11 0.11 0.21 S 0.12

15 Etsuko 1.00 1.00 X n.a.

09 Sohrab 1.00 1.00 X n.a.

39 Rhonda 0.05 0.76 0.19 0.92 0.08 S 0.19

30 Germaine 0.53 0.18 0.18 0.12 0.50 0.10 0.30 0.10 S 0.15

36 Robert 0.13 0.75 0.13 0.13 0.73 0.13 S 0.02

191

All valuations Valuations at country of origin

Positive affect Negative affect Positive affect Negative affect Stability Distance

+HH +S +O –LL –S –O +HH +S +O –LL –S –O

+S and +O 13 Irene 0.11 0.21 0.53 0.11 0.05 0.00 0.18 0.73 0.09 S 0.24

34 Emilia 0.50 0.13 0.17 0.04 0.17 0.35 0.18 0.24 0.24 S 0.19

20 Sancho 0.56 0.19 0.13 0.13 0.67 0.08 0.00 0.17 0.08 S 0.16

24 Leo 1.00 X n.a.

14 Maryam 0.93 0.03 0.03 1.00 S 0.08

32 Griselda 0.50 0.25 0.25 0.50 0.25 0.25 X n.a.

27 Janet 0.33 0.33 0.17 0.08 0.08 0.00 0.40 0.40 0.20 U 0.44

08 Kate 0.73 0.27 0.71 0.29 S 0.02

10 Alicia 0.67 0.22 0.11 0.50 0.50 X n.a.

12 Laura 0.57 0.22 0.17 0.00 0.04 0.47 0.26 0.21 0.05 S 0.11

05 Hendrik 0.56 0.22 0.11 0.11 0.60 0.20 0.20 U 0.26

35 Luisa 0.52 0.29 0.16 0.03 0.39 0.39 0.17 0.06 S 0.16

01 Azzam 0.30 0.70 1.00 U 0.42

192

All valuations Valuations at country of origin

Positive affect Negative affect Positive affect Negative affect Stability Distance

+HH +S +O –LL –S –O +HH +S +O –LL –S –O

Multivalent

S and O, and 19 Beatriz 0.25 0.31 0.13 0.25 0.06 0.29 0.29 0.14 0.29 S 0.09 both positive and negative affect 23 Marta 0.50 0.25 0.25 0.50 0.50 X n.a.

18 Eva 0.50 0.25 0.25 0.50 0.50 X n.a.

+S and -LL 42 Keira 0.43 0.14 0.36 0.07 0.00 0.22 0.22 0.56 U 0.31

04 Nadine 0.33 0.67 1.00 X n.a.

+O and -LL 25 Nicole 0.06 0.11 0.28 0.50 0.06 0.10 0.40 0.40 0.10 S 0.20

06 Danilo 0.15 0.08 0.23 0.31 0.08 0.15 0.33 0.33 0.11 0.22 S 0.22

+HH and -O 33 Farid 0.31 0.15 0.15 0.38 0.27 0.09 0.18 0.45 S 0.10

+HH and -S 40 Carla 0.37 0.11 0.14 0.29 0.09 0.71 0.06 0.18 0.06 X 0.45

+S and -LL 26 Damon 0.78 0.22 0.67 0.33 S 0.16

193

All valuations Valuations at country of origin

Positive affect Negative affect Positive affect Negative affect Stability Distance

+HH +S +O –LL –S –O +HH +S +O –LL –S –O

11 Damara 0.72 0.28 0.60 0.40 S 0.17

22 Juan 0.60 0.33 0.07 0.50 0.33 0.17 S 0.14

37 Cristiano 0.55 0.45 0.38 0.63 S 0.24

21 Diana 0.09 0.27 0.45 0.09 0.09 0.07 0.36 0.50 0.07 S 0.13

16 Irina 0.13 0.25 0.19 0.25 0.06 0.13 0.20 0.30 0.30 0.10 0.10 S 0.22

Note: Names have been changed to preserve anonymity. Key: Bold cells indicate “important” major themes. Stability is coded as: S = stable; U = unstable; X = unclassified.

194

Tests of statistical significance. The null hypothesis was that thematic profiles are stable as between country of origin and Australia. Tests of the null are based on the differences between the profiles measured separately with respect to country of origin and Australia. Due to occasional lack of classified data at origin or in Australia, five participants in the sample of 38 were perforce dropped from the analysis. Table 7.4 reports averages, standard deviations, and t-statistics for the remaining 33, for the null hypothesis that mean differences are zero.

For two-tailed tests, only –LL is significant at the 5% level: thus on average, participants told fewer –LL valuations of Australia. It is, however, notable (though of borderline significance) that they also told more +HH and +S valuations and fewer +O valuations of Australia.

Table 7.4 Average differences in proportions of major themes

Major theme M SD t df p

Strength and unity (+HH) 0.08 0.27 1.80 32 0.081

Unity and love (+O) 0.09 0.27 1.92 32 0.064

Unfulfilled longing (–O) –0.05 0.21 –1.54 32 0.133

Powerlessness and isolation (–LL) –0.11 0.29 –2.33 32 0.026

Aggression and anger (–S) 0.03 0.13 1.59 32 0.122

Autonomy and success (+S) –0.04 0.16 –1.42 32 0.165

The question is whether the six average differences are jointly different from zero.

This requires a test of a mean, complicated by the fact that the mean is here a vector in six dimensions. Such a test can be based on the statistic:

2   T NV )/( 1 

195

where μ is the measured vector of means (of differences), V the covariance matrix of the sample of differences, and N the sample-size.1

In computing this statistic, account must be taken of the fact that the underlying variables are proportions that sum to unity. The resulting redundancy is solved by arbitrarily eliminating one of the six categories: –O was chosen. If this is done the statistic is distributed as a chi-square with five degrees of freedom under the null of no differences in means (in large samples).

There is strong evidence against the hypothesis of complete stability of the thematic profile, 2(5, N = 33) = 26.2, p < .002. In other words, for some individuals at least, the themes of Australian valuations tended to differ significantly from those of their country of origin valuations.

7.4 Analysis Based on Self-confrontation Indices

The second analysis of Study 2 resembled the first except that participants were grouped on the basis of the self-confrontation indices of their valuations—whether above or below the mean. As with in the first analysis, this was done separately for valuations pertaining to the country of origin and to Australia. A transition matrix was then generated to determine which participants were in the same group for both sets of

1 If the five selected variates were normally and independently distributed, this statistic would be the sum of squares of the t-statistics in Table 3, which is chi-square with five degrees of freedom in large samples.

The presence of the term V/N in the formula is to correct for possible cross-tabulations between the variates. This approach has the advantage that it follows on naturally from the t-tests. It is thus preferable to a contingency table analysis, which would be in any case vitiated by low expected frequencies in many cells.

196 valuations, and which participants changed groups. Formal tests of statistical significance were then performed on the transition matrices to test the null hypothesis that participants remained in the same groupings, that is, that migration produced no significant change in I-position.

Procedure for grouping participants. The preceding analysis was based on

Hermans’ typology of valuation themes. His classification has the advantage of clinical experience but results in large attrition of valuations since only about one third of valuations are so classified; diagnostic certainty is gained at the cost of potential loss of information. The following analysis therefore considers the full set of valuations. A typology of participants exploiting the full set is required. To reduce dimensionality, positive and negative affect indices are represented by a single index, well-being (see

Chapter 4).

Next, a version of Hermans’ typology is constructed from each participant’s average score on the three indices S, O and W, both at country of origin and in Australia.

An index value is considered high (or low) if it is higher (or lower) than the full sample mean in the country of origin. This gives an eight-way classification of individuals at the origin and in Australia. Using this typology it is possible to construct an 8 × 8 transition matrix to compare the themes expressed by migrants concerning the country of origin and Australia. This procedure extends the Hermans classification of valuations to a classification of individuals.

Note that under this typology it is possible for an individual to score low on S and

O and high on W, as well as high on S and O and low on W. Such cases are ignored in the Hermans approach but the proposed classification procedure throws up a few examples, termed +E and –E, respectively. The transition matrix is given in Table 7.5.

Row-totals and column-totals give the numbers of participants classified in each of the

197 eight nominated categories for country-of-origin and Australian valuations respectively.

The values in each cell of the matrix give the number of people who pass from the corresponding category of country-of-origin valuation to the corresponding category of

Australian valuation. For example, in the top left cell, eight participants pass from +HH for country-of-origin to +HH for Australia.

Table 7.5 Thematic transitions

Major theme in Australia

Major theme at country of origin +HH +S +O +E -LL -S -O -E Total

Strength and unity (+HH) 8 1 1 1 1 2 14

Autonomy and success (+S) 1 2 3

Unity and love (+O) 1 1

Low S, O, high W (+E) 3 3

Powerlessness and isolation (–LL) 1 1 1 10 1 1 15

Aggression and anger (–S) 0

Unfulfilled longing (–O) 1 1 2

High S, O, low W (–E) 0

Total 13 3 2 1 11 2 2 4 38

Key: Shaded cells show stable participants. Boxed cells show participants whose affects changed from positive to negative well-being or vice versa.

In respect of their country of origin I-positions, 14/38 = 37% of individuals are classified as +HH and 15/38 =39 % as –LL, compared to 34% and 29% with regard to that of Australia, respectively; 21/38 = 55% show positive affect in the country of origin, 19/38 = 50% show positive affect in Australia. The entries along the diagonal of the matrix show individuals with stable average themes with respect to country of origin and Australia. Thus 22/38 = 55% exhibited similar themes in valuations about the country of origin as about Australia. The bottom left portion of the matrix indicates

198 individuals with negative affect at the country of origin but positive affect in Australia,

3/38 = 8% in total; 6/38 = 16% (the top right portion) were the opposite.

In summary, there was considerable thematic stability across valuations about the country of origin and Australia. There was some evidence of change, contained in the off-diagonal portion of the transition matrix. Five participants exhibited positive affect with regard to Australia and negative affect with regard to the country of origin; three participants exhibited the opposite. This confirms the general picture emerging from the preceding section of small but measureable changes in a context of considerable stability.

Tests of statistical significance. Analogously to Table 7.4, Table 7.6 reports statistics for the differences in numerical index scores between origin and Australia.

Notably, average theme is higher in Australia along all four dimensions. Thus, the themes in Australia on average showed stronger sense of self-enhancement (S), stronger sense of union with the other (O), stronger positive affect (P), and stronger negative affect (N). The average increase in S and P were significant at the 5% level, the other two are not. The ensemble test of the four parameters together yields a value of 42.1, which has a p-level of 1.59×10-8. This is compelling evidence against the null hypothesis of unchanged themes.

Table 7.6 Average differences in thematic scores (Australia minus country of origin)

Index M SD t df p

Self-enhancement (S) 1.43 2.67 3.31 32 0.002

Unity and love (O) 0.53 2.91 1.12 32 0.271

Unfulfilled longing (P) 2.05 5.35 2.37 32 0.024

Autonomy and success (N) 0.41 5.17 0.49 32 0.627

199

7.5 Summary

Two separate but comparable analyses were performed in Study 2. In both, participants were grouped on the basis of the types of valuations they produced, though according to different rules. One analysis grouped participants according to the combinations of major themes manifested by their valuations; the other grouped them according to the self-confrontation indices of their valuations. This was done separately for valuations pertaining to the country of origin and to Australia. Transition matrices were then produced to determine which participants were in the same group for both sets of valuations, and which participants changed groups. Finally, tests of statistical significance were performed on the transition matrices to test the null hypothesis that participants remained in the same groupings, that is, that migration produced no significant change in I-position.

There were two significant findings. Under both analyses, results were as hypothesised: that is, I-positions were found to differ as between those pertaining to country of origin and those pertaining to Australia. The other significant finding comes from Analysis 1 and relates to locational differences in the major themes of valuations.

On average, participants told significantly fewer –LL valuations for Australia than for country of origin.

In conclusion, this chapter documents the second of three studies performed as part of thesis. Like the second study, which was discussed in the preceding chapter, it is nomothetic in nature, but differs from it in the way the data were organised at the case level. The third study, which is discussed next, was idiographic, analysing case studies representing each of the participant types identified by the first study.

200

CHAPTER 8: Study 3 – Case Studies

In Study 3, an idiographic approach was adopted to complement the nomothetic approach of Studies 1 and 2. Four case studies are discussed, one of each migrant type:

Progressive, Low-stable High-stable, and Regressive. The aim was to exemplify and corroborate the nomothetic principles previously identified. The idiographic analyses in this chapter furnish a more complete understanding of the individual than do the nomothetic analyses of the preceding two chapters.

The chapter begins with a discussion of the Euclidean distance between each participant’s valuation profiles pertaining to country of origin and to Australia. This assisted in the selection of case studies by showing how they are positioned in relation to other study participants. By comparing and contrasting pairs of valuations, one from country of origin and the other from Australia, the case studies illustrate how I-positions differed, or conversely were stable, between the two locations. Importantly, the case studies demonstrate at an individual level that while most migrants are stable in terms of their I-positions, a minority are not.

8.1 Euclidean distance

An objective corroboration of stability of the self was provided by the Euclidean distance between valuation profiles, defined as the square root of the sum of squared differences between them (Deza & Deza, 2009). Figures 7.1 and 7.2 both rank participants by Euclidean distance, shown by the length of the bars along the horizontal axis. The figures differ in that Figure 8.1 uses self-confrontation indices as the measure of Euclidean difference, while Figure 8.2 uses major themes. The dark blue and orange bars in both charts denote participant Type 1 Progressive and Type 4 Regressive, that is, the two migrant types that exhibited most instability in I-position.

201

An interesting, but unsurprising, finding is that participants of the unstable types—that is, Types 1 and 4—also tended to exhibit the greatest Euclidean distance between their profiles for country of origin and Australia. The tendency is more pronounced in Figure 8.1 than 7.2 because the former is based on self-confrontation indices, which were directly used to derive the participant types, while the latter is based on major themes, which are only indirectly linked to self-confrontation indices, and thence to major themes.

202

03 Yoana 18 Eva 05 Hendrik 40 Carla 01 Azzam 16 Irina 19 Beatriz 20 Sancho 04 Nadine 34 Emilia 27 Janet 35 Luisa 12 Laura 10 Alicia 39 Rhonda 42 Keira 09 Sohrab 13 Irene Type 4 37 Cristiano Type 3 21 Diana Type 2 32 Griselda 11 Damara Type 1 22 Juan 06 Danilo 17 Ljubica 30 Germaine 33 Farid 25 Nicole 24 Leo 23 Marta 02 Dragan 36 Robert 26 Damon 15 Etsuko 14 Maryam 08 Kate 07 Frida 28 Gonzalo

,0 ,2 ,4 ,6 ,8 ,10 ,12

Figure 8.1 Euclidean distance between profiles for country of origin and Australia based on mean self-confrontation indices of valuations

203

37 Cristiano 40 Carla 42 Keira 27 Janet 03 Yoana 04 Nadine 23 Marta 18 Eva 06 Danilo 33 Farid 34 Emilia 20 Sancho 12 Laura 10 Alicia 01 Azzam 16 Irina Type 4 05 Hendrik Type 3 13 Irene Type 2 07 Frida Type 1 26 Damon 25 Nicole 39 Rhonda 35 Luisa 28 Gonzalo 21 Diana 30 Germaine 14 Maryam 02 Dragan 11 Damara 22 Juan 19 Beatriz 36 Robert 08 Kate

,0.0 ,0.2 ,0.4 ,0.6 ,0.8 ,1.0

Figure 8.2 Euclidean distance between profiles for country of origin and Australia based on proportions of major themes of valuations

204

8.2 Case study: The ‘Progressive’ Migrant (Type 1)

Yoana (this and other names are pseudonyms) is representative of the Progressive migrant type. She is a Colombian who came to Australia at 23. At the time of the interview she was married with two young children, a girl and a boy, both still in primary school. Although she had obtained a degree in journalism in her native country, she was not working at the time of the interview and she listed her occupation as “house duties”. Two key points come across from her valuations about her youth in Colombia.

Her parents divorced when she was seven and she felt abandoned by her much-admired father, as well as sympathy for her mother’s predicament. This sense of abandonment was compounded when her sister migrated to Spain. The second theme concerns her distress at the quality of life in Colombia, which she perceives as “a world full of materialism, alcohol, sex, drugs, violence, etc.” In Australia she has made a happy marriage. Her valuations emphasise her devotion to her children. She is a convert to the

Baha’i religion and aspires to “a very pure and spiritual life”.

Yoana’s measures of psychological distress and well-being corroborate this picture of a determined and caring person who does not let adversity get the better of her. At the time of the interview she was confident in social situations (SAD = 1, cf. sample mean of 8.4); had little concern at giving a negative impression to others (FNE =

1, cf. sample mean of 13.3); and was not depressed with respect to the sample mean

(BDI-II = 10, cf. sample mean of 9.1). Her psychological well-being was above average on all subscales except ‘Purpose in life’, where she was slightly below (62 cf. sample mean of 66.1).

Major themes. Figure 8.3 shows the mean thematic profile of Type 1 participants

(Panel 1) and for Yoana, the case-study selected to exemplify them (Panel 2). The mean profile of valuations of Type 1 participants differed significantly according to whether

205 they pertained to country of origin or Australia, χ2(5, N = 44) = 16.56, p < .01 (two- tailed). Valuations located in country of origin were high on the themes of Unfulfilled longing and Powerlessness and isolation, but those in Australia were high on the themes of Strength and unity and Autonomy and success.

Yoana’s thematic profiles correlated highly with the mean profiles of Type 1 participants both for country of origin ( r = .67) and for Australian valuations (r = .90), indicating that she is representative of the Progressive migrant type.

206

Strength and unity 50%

40%

30% Autonomy and success Unity and love 20%

10% Origin

0% Australia

Aggression and anger Unfulfilled longing

Powerlessness and isolation

Panel 1 All Progressive migrants

Strength and unity 50%

40%

30% Autonomy and success Unity and love 20%

10% Origin

0% Australia

Aggression and anger Unfulfilled longing

Powerlessness and isolation

Panel 2 Yoana

Figure 8.3 Proportion of valuations, by major theme and location: Progressive migrants and Yoana

207

Self-confrontation indices. Figure 8.4 shows Yoana’s mean self-confrontation indices. On self-enhancement, her valuations of country of origin scored significantly less (M = 5.2, SD = 5.5) than those of Australia (M = 13.0, SD = 5.2), t(22) = –3.57, p < .01 (two-tailed). On union-with-the-other, there was no significant difference between her valuations of country of origin (M = 9.1, SD = 6.6) and those of Australia

(M = 13.6, SD = 5.8), t(22) = –1.77, p = .09 (two-tailed). On positive affect, her valuations of country of origin were significantly less (M = 13.7, SD = 12.3) than those of Australia (M = 25.0, SD = 12.5), t(22) = –2.24, p < .05 (two-tailed). On negative affect, her valuations of country of origin scored significantly more (M = 13.5, SD =

8.5) than those of Australia (M = 2.5, SD = 4.0), t(22) = 4.07, p < .001 (two-tailed).

40 Origin Australia 35 General Ideal 30

0 S-index O-index P-index N-index

Figure 8.4 Mean indices of valuations, by location: Yoana

208

Yoana is representative of the Type 1 (Progressive) migrant. According to Gergen

& Gergen (1986) a “progressive” narrative links events “in such a way that one steadily progresses towards a goal” (p. 27). Corroboration of this is found in Figure 8.5, which shows the temporal dimension of her self-confrontation indices.

Yoana’s S , O and P rise, and her N falls, in her future-oriented country-of-origin valuations—a clear temporal progression. But no such temporal progression is observed in her Australian valuations, which are at a uniformly high level. This suggests that

Yoana is basically affectively positive both about Australia (in all periods) and about the future (in both locations). What makes her “progressive” is the fact that her past- and present-oriented valuations are relatively affectively negative for her country of origin but not for Australia.

20 S O P 15 N

0 Past Present Future Past Present Future

Origin Australia Figure 8.5 Mean indices of valuations, by location and period: Yoana

209

Self-narrative. Yoana’s self-narrative is characterised by progression towards a desired goal. She espouses the twin goals of mother and spiritual being. The following contrasting valuations show how she is attaining these goals.

She has turned away from her own mother’s goal of focusing on her husband

(Yoana’s father) and instead focuses on her children:

Country of origin: “… my mother used to see my father as the centre of her universe” (–LL). Australia: “In my present existence the most important thing are my children …” (+HH).

She has successfully repudiated the drinking culture of her country of origin:

Country of origin: “I look at alcohol as a destructive drug” (–O). Australia: “I was so impressed by the fact that the youth were not drinking alcohol …” (+HH).

... and has likewise decided to be a spiritual person, and has achieved that aim:

Country of origin: “I decided to live a very pure and spiritual life”. Australia: “… I closed my eyes and I experience God” (+HH).

... and has successfully fought a familial tendency towards depression:

Country of origin and Australia: “Another influence has been my aunty who suffered from depression when I was a little girl. When … I suffered my own depression I decided never to be like her, therefore I had the strength to fight it and overcome it” (–LL).

Yoana’s is a self-narrative of setting new and very different goals, and achieving them—truly a progression to better things.

210

Table 8.1 Valuations with major themes, by location and period: Yoana

Valuation S O P N G I Theme Country of origin: Past 1 A mark in my life is that at that time my mother used to see my father as the 0 0 0 19 -0.69 -0.78 -LL centre of her universe. Now that I am married I refuse to see my husband the same way. Instead, my kids are the centre of my life. 2 An episode I remember is my father visiting us. During that time my mother 2 2 3 24 -0.71 -0.78 -LL used to prepare us the best possible way, we had to wear our best clothes and our best behaviour before he arrived. But to our great disappointment he will come and sit there for the whole afternoon reading newspapers. He never talked to us, never took us to the park or cinema, etc. My mother made us sit there with him, just watching him. From that moment I always experience a sense of not being good enough. 4 When I was 14 I clearly remember a scene like a scene from a movie where 0 13 0 18 -0.30 -0.33 -O we were in a party with very close friends. The son of one of those friends was only 4 and I recall that alcohol was offered to him and was encouraged to taste the delicious flavour and to my amazement Leonardo took it. From that moment I look at alcohol as a destructive drug. Country of origin: Present 9 In my opinion my extended [Colombian] family is very significant. There 0 14 0 18 -0.27 -0.34 -O was an episode in my life that confronted me to this reality. My son was operated on as an unexpected emergency and my daughter was left at school with no one to look after her while I was with her brother. 10 My children were watching TV and I just learned on the phone that my mum 5 13 26 5 0.48 0.46 +O was flying to Australia to help us during my sickness. The nice surprise was that my sister told me on the phone that my mother was just one hour away before she landed in Melbourne. 11 The same month I happened to be diagnosed with three lumps in my breast. 3 0 2 14 -0.36 -0.46 -LL When I learned about it I was very upset and the doctor had to comfort me. There I felt more than ever the ongoing importance of having an extended family. Australia: Past 1 When I was 24 here in Melbourne I went to a Youth Baha’i party and there 13 6 29 0 0.78 0.77 +S were boys and girls all about the same age. I was so impressed by the fact that the youth were not drinking alcohol and yet they were enjoying themselves and having a good time. It impressed me a lot the clear environment that surrounded me that night. Australia: Present 5 Another influence has been my aunty who suffered from depression when I 0 0 0 13 -0.55 -0.61 -LL was a little girl. When [this year] I suffered my own depression I decided never to be like her, therefore I had the strength to fight it and overcome it. 6 Classical music plays a very important role in my life. One day I was 19 19 35 0 0.90 0.96 +HH watching my daughter in a concert. She was playing a beautiful piece by Bach and I closed my eyes and I experience God. 7 I have a sign on my fridge that said “the child’s destiny lies in her/his parents 16 19 35 0 0.85 0.93 +HH hands” and I have made a commitment to my children to try to develop all their talents and reach their full potential. 8 In my present existence the most important thing are my children—they are 19 19 35 1 0.95 0.98 +HH my centre, totally.

211

8.3 Case study: The “Low-stable” Migrant (Type 2)

Juan is representative of the “Low stable” migrant type. He was born and brought up in El Salvador and emigrated to Australia as a young adult. His childhood was in some ways difficult. “[N]aïve and trusting” as a child, he nevertheless felt rejected by his mother; and though his father was “fair”, he was also “inflexible”. Having “learned to be with people who feel ambivalent about me”, he would “accept blame for anything” and became obedient in order to please those around him. One salient event he remarked on: his sister’s apparent suicide, for which he “admired” her. At school he was “insecure”, “a loner” and “shy”, and his academic performance poor. Later he would drop out of university in Australia and suffer an extended bout of depression. He had married in El Salvador, but his marriage broke up in Australia. Having left El

Salvador to live in a “peaceful place”, has no desire to return. On the plus side, he is

“happy” in Australia and expresses guarded optimism about the future and what it holds.

Juan’s measures of psychological distress and well-being complete the picture of a well-adjusted person who accepts life as it comes. At the time of the interview he felt confident in social situations (SAD = 2, cf. sample mean of 8.4); had little concern at giving a negative impression to others (FNE = 2, cf. sample mean of 13.3); and was not depressed (BDI-II = 0, cf. sample mean of 9.1). His psychological well-being was about average on all subscales except Self-acceptance, where it was somewhat higher (PWB6

= 70, cf. sample mean of 64.6).

Major themes. Figure 8.6 shows the mean thematic profile of Type 2 participants

(Panel 1) and for Juan, the case-study selected to exemplify them (Panel 2). The mean profile of valuations of Type 1 participants differed significantly, but only slightly, between country of origin and Australia, χ2(5, N = 270) = 14.09, p = .015 (two-tailed).

212

Valuations located in country of origin were high on Autonomy and success,

Powerlessness and isolation, and Aggression and anger whereas those in Australia, though still high on Autonomy and success and Powerlessness and isolation, were higher on Aggression and anger and lower on Unity and love.

Juan’s thematic profiles correlated highly with the mean profiles of Type 2 participants both for country of origin ( r = .89) and for Australian valuations (r = .97), indicating that he is representative of the Type 2 migrant type.

213

Strength and unity 50%

40%

30% Autonomy and success Unity and love 20%

10% Origin

0% Australia

Aggression and anger Unfulfilled longing

Powerlessness and isolation

Panel 1 All Low-stable migrants

Strength and unity 50%

40%

30% Autonomy and success Unity and love 20%

10% Origin

0% Australia

Aggression and anger Unfulfilled longing

Powerlessness and isolation

Panel 2 Juan

Figure 8.6 Proportion of valuations, by major theme and location: Type 2 participants and Juan

214

Self-confrontation indices. Figure 8.7 shows Juan’s mean self-confrontation indices. On self-enhancement, there was no significant difference between his valuations of country of origin (M = 7.4, SD = 4.9) and those of Australia (M = 9.2, SD

= 5.2), t(33) = –1.07, p = .29 (two-tailed). On union-with-the-other, there was no significant difference between his valuations of country of origin (M = 6.4, SD = 3.6) and those of Australia (M = 4.8, SD = 3.2), t(34) = 1.36, p = .18 (two-tailed). On positive affect, there was no significant difference between his valuations of country of origin (M = 14.9.7, SD = 9.4) and those of Australia (M = 16.8, SD = 10.5), t(32) = –.58, p = .57 (two-tailed). On negative affect, there was no significant difference between his valuations of country of origin (M = 20.5, SD = 12.3) and those of Australia (M = 13.4,

SD = 11.5), t(34) = 1.79, p = .08 (two-tailed).

40 Origin Australia 35 General 30 Ideal

0 S-index O-index P-index N-index

Figure 8.7 Mean indices of valuations, by location: Juan

215

Juan is representative of the Type 2, “Low stable” migrant type. According to

Gergen & Gergen (1986) a “stability” narrative links events “in such a way that the protagonist remains essentially unchanged with respect to evaluative position” (p. 27).

Corroboration of this is found in Figure 8.8, which shows the temporal dimension of his self-confrontation indices.

Juan’s self-confrontation indices show a slight temporal progression, which is, however, unrelated to location. That is, his past-oriented valuations for country of origin resemble his past-oriented valuations for Australia, and likewise for present- and future- oriented valuations. This explains why his valuations appear to be stable when the temporal dimension is ignored: they are stable with respect to location. Juan does show a slight progression in a temporal sense—his future-oriented S and O are slightly higher, and well-being (defined as P minus N) is substantially higher; but that progression is not noticeably linked to the act of migration.

20 S O P 15 N

0 Past Present Future Past Present Future

Origin Australia Figure 8.8 Mean indices of valuations, by location and period: Juan

216

Self-narrative. Juan’s self-narrative is characterised by stasis. He struggles; his is not an easy life. The following paired valuations show this.

He was, and is, an outsider:

Country of origin: “At school I felt insecure. I was a loner …” (–LL). Australia: “I felt that I did not belong to this society because people did not know who I was”. “I felt that I did not belong even to the groups I was participating in (e.g. University) (–LL).

His reaction to what to most people would be a life-changing event, is almost laconic: his sister’s suicide did not even qualify for a major theme under his affective rating. Later, in Australia, he suffers from occasional depression and a negative self- image:

Country of origin: “When [my sister] apparently killed herself I admired her for it.” Australia: “I was very depressed and I had very low self-esteem.” (–LL).

But his life is not without its small victories. He is less shy now:

Country of origin: “I was shy and I learnt not to show off because I thought people would not like me” (–LL). Australia: “I am able to express my feelings and thoughts more easily now” (+S).

... and is learning, and relishing, some independence:

Country of origin: “I learnt from [my father] that rules were not to be bent or broken” (–LL). “I learnt to accept blame for anything and to be obedient …” Australia: “For the first time in my life I feel I am working for myself as an individual” (+S).

… though apparently not enough to prevent his being classified as “Low-stable”.

Juan’s is a self-narrative of mental struggle and survival—stable, but compromised by underachievement.

217

Table 8.2 Valuations with major themes, by location and period: Juan

Valuation S O P N G I Theme Country of origin: Past 1 Throughout my upbringing I became naïve and trusting. My family 3 12 10 24 -0.40 -0.57 -O emphasised on me being good. 2 When I was 11, my brother and I went to Los Angeles to study English for a 16 8 29 9 0.79 0.65 +S year. This event set me apart from my peers. 3 My father is fair but inflexible. I learnt from him that rules were not to be 2 2 12 35 -0.25 -0.82 -LL bent or broken. 8 At school I felt insecure. I was a loner and was constantly failing subjects. I 2 1 3 33 -0.34 -0.88 -LL was shy and I learnt not to show off because I thought people would not like me. 11 At university I discovered the humanistic subjects and for the first time I 20 6 33 5 0.68 0.77 +S enjoyed schooling and studying. Country of origin: Present 30 I don’t feel the need to communicate with my family. If I had a problem I 10 4 21 5 0.53 0.64 +S would talk to a friend here. Australia: Past 18 I felt that I did not belong even to the groups I was participating in (e.g. 2 0 1 30 -0.16 -0.82 -LL University). 19 During the first 6 year in Australia I did not do much for myself and 0 0 0 39 -0.37 -0.96 -LL therefore achieved very little. I felt I was shutting down one aspect of my life after another. 22 After 5 years of studying I was asked to withdraw from university because I 1 1 5 24 -0.41 -0.85 -LL was failing, and I withdrew from my course and gave up. I was very depressed and I had very low self-esteem. 24 Not having a partner I felt that I could make my own decision and I started to 15 6 21 8 0.56 0.70 +S live my own life and not a family group life. 25 Getting my first professional job was a self-esteem boost for me. I felt I 15 7 23 3 0.62 0.83 +S finally belonged to a group of people that were similar to me. Australia: Present 26 I have a job. I’m still learning but I’m where I want to be. 13 4 25 7 0.75 0.71 +S 27 For the first time in my life I feel I am working for myself as an individual. 15 6 26 6 0.71 0.82 +S 29 I am able to express my feelings and thoughts more easily now. 12 4 29 1 0.57 0.84 +S Australia: Future 33 I want to go back to study to expand my professional horizon. 13 4 24 4 0.66 0.78 +S

218

8.4 Case study: The “High-stable” Migrant (Type 3)

Luisa is representative of the “High stable”, migrant type. She was born in

Mexico and emigrated to Australia as a mature adult of 30. Her childhood was largely happy, but not without the frustration of “not being recognised … because [she] was a girl”, and because being “the only girl among four brothers suppressed [her] character.”.

Nevertheless, she saw herself very much as the equal of the men in her family, describing herself as “competitive” with her brother and the “soul mate” of her husband.

She was also adventurous, with a “thirst for independence” and a fascination for distant lands, and Australia in particular. Though her relationship with her father was “not very close”, she states, “I treasure the memory where my father went to my school to drop the book he had bought for me - which he covered with beautiful paper and labelled with my name. I felt so special.”

Luisa’s measures of psychological distress and well-being complete the picture of an optimistic personality who gets the most she can from life. At the time of the interview she felt confident in social situations (SAD = 1, cf. sample mean of 8.4); had some, but still below-average, concern at giving a negative impression to others (FNE =

7, cf. sample mean of 13.3); and was not depressed (BDI-II = 3, cf. sample mean of

9.1). Her psychological well-being was well above average on all subscales (PWB1 =

77, cf. sample mean of 63.1; PWB2 = 83, cf. sample mean of 67.1; PWB3 = 79, cf. sample mean of 63.4; PWB4 = 84, cf. sample mean of 72.7; PWB5 = 83, cf. sample mean of 66.1; and PWB6 = 70, cf. sample mean of 64.6).

Major themes. Figure 8.9 shows the mean thematic profile of Type 3 participants

(Panel 1) and for Luisa, the case-study selected to exemplify them (Panel 2). The mean profile of valuations of Type 3 participants did not differ significantly according to whether they pertained to country of origin or Australia, χ2(5, N = 169) = 7.21, p = .21

219

(two-tailed). Valuations located in both country of origin and Australia were high on the themes of Strength and unity and Autonomy and success; and, for Luisa but not Type 3 in general, moderately high on Unity and love.

220

Strength and unity 50%

40%

30% Autonomy and success Unity and love 20%

10% Origin

0% Australia

Aggression and anger Unfulfilled longing

Powerlessness and isolation

Panel 1 All High-stable migrants

Strength and unity 50%

40%

30% Autonomy and success Unity and love 20%

10% Origin

0% Australia

Aggression and anger Unfulfilled longing

Powerlessness and isolation

Panel 2 Luisa

Figure 8.9 Proportion of valuations, by major theme and location: Type 3 participants and Luisa

221

Luisa’s thematic profiles correlated highly with the mean profiles of Type 3 participants both for country of origin ( r = .85) and for Australian valuations (r = .96), indicating that she is representative of the Type 3 migrant type.

Self-confrontation indices. Figure 8.10 shows Luisa’s mean self-confrontation indices. On self-enhancement, there was no significant difference between her valuations of country of origin (M = 14.2, SD = 5.6) and those of Australia (M = 17.2,

SD = 3.2), t(39) = –2.25, p = .03 (two-tailed). On union-with-the-other, her valuations of country of origin scored significantly less (M = 12.8, SD = 7.3) than those of

Australia (M = 17.3, SD = 3.4), t(36) = –2.71, p < .01 (two-tailed). On positive affect, her valuations of country of origin scored significantly less (M = 28.5, SD = 12.2) than those of Australia (M = 34.8, SD = 5.7), t(36) = –2.27, p < .05 (two-tailed). On negative affect, there was no significant difference between her valuations of country of origin

(M = 5.3, SD = 6.1) and those of Australia (M = 4.6, SD = 4.9), t(43) = .45, p = .66

(two-tailed).

40 Origin Australia 35 General Ideal 30

0 S-index O-index P-index N-index

Figure 8.10 Mean indices of valuations, by location: Luisa

222

Luisa is representative of the Type 3, “High stable” migrant type. According to

Gergen & Gergen (1986) a “stability” narrative links events “in such a way that the protagonist remains essentially unchanged with respect to evaluative position” (p. 27).

Corroboration of this is found in Figure 8.11, which shows the temporal dimension of her self-confrontation indices.

Luisa’s self-confrontation indices show little temporal or other progression. For the most part, the self-confrontation indices of all her valuations are quite similar, whether differentiated by location, by period, or both.

20 S O P 15 N

0 Past Present Future Past Present Future

Origin Australia Figure 8.11 Mean indices of valuations, by location and period: Luisa

223

Self-narrative. Luisa’s self-narrative is characterised by stasis. She brims with self-confidence and the achievements, real and perceived, that such confidence engenders. The following paired valuations show this.

From the outset she feels special and valued—a sentiment that stands her in good stead later in life:

Country of origin: “I treasure the memory where my father went to my school … I felt so special” (+HH). Australia: “This provided me with an interest and the freedom to be with John when we had time off work” (+HH).

She anticipates, and plans for, a marriage of equals, and achieves it:

Country of origin: “I came back to Mexico. John followed me a few months after and we were married” (+HH). Australia: “This was an encounter with a soul mate; we felt so happy in each other’s company from the beginning and we fell in love” (+HH).

She has a childhood dream, and achieves that too:

Country of origin: “Since my childhood I felt a special fascination with anything to do with Australia”. Australia: “I had already become an Australia citizen and I felt totally at home” (+HH).

Her two dreams—of a soul-mate and a spiritual home—are captured again in these valuations:

Country of origin: “I finally came to visit Australia. I met my future husband during this trip. It was on the cards”. Australia: “My first arrival in Australia was almost like a homecoming. Everything was more than I ever expected” (+HH).

And she does it all by complementing her country-of-origin self, not by repudiating or changing it:

224

Country of origin: “The joy of participating closely with the family is hard to describe. The warmth of the people in general” (+HH). Australia: “The early years of my children were the happiest and the most rewarding of my life. I was completely fulfilled and did not want for anything more” (+HH).

Luisa’s is a self-narrative of building on a strong sense of herself and achieving more—stability crowned by emotional reward.

225

Table 8.3 Valuations with major themes, by location and period: Luisa

Valuation S O P N G I Theme Country of origin: Past 3 Whilst in the U.S. (for a period of 2 and a half years) I had to conform with 8 0 0 19 -0.44 -0.46 -S my grandmother’s ideas of discipline, and that involved performing a number of responsibilities in the household which did not befit a fourteen year old. 4 School life - year 7 and 8 was a fantastic experience, both academic and 15 8 32 0 0.76 0.76 +S social. 5 Childhood years- enjoyed school enormously and always excelled in my 15 9 36 1 0.79 0.80 +S grades. 6 Social life revolved around our cousins and our neighbourhood friends. It 9 16 31 10 0.69 0.69 +O was fun. 7 I treasure the memory where my father went to my school to drop the book 20 20 35 0 0.93 0.94 +HH he had bought for me- which he covered with beautiful paper and labelled with my name. I felt so special. 8 At the age of 15 I returned home to Mexico. Because of the education system 11 5 27 3 0.56 0.57 +S of the time I could not go to high school and therefore the opportunity of going to university was closed to me. My parents chose to send me to a secretarial course instead. This was quite disappointing for many years. 9 When I finished secretarial school I applied for work in the airline industry. I 19 9 38 0 0.86 0.87 +S was very lucky to get a job straight away with American Airlines. From there I moved on to a better position with Qantas airways. I was 19 at the time. This was to be an experience that opened my horizons and changed my life. 10 Being the only girl among four brothers suppressed my character. And this 16 2 25 4 0.49 0.50 +S promoted in me a great thirst for independence. I became very competitive with Eduardo, my brother, who was closest in age to me and I felt quite happy when I could earn a better salary [than him, working as a secretary]. 11 My job with Qantas lasted 10 years. It offered me a fabulous wide experience 19 12 37 0 0.88 0.89 +S of interesting work, very rewarding in every sense: intellectually, socially and financially. 12 During those same 10 years I was able to travel extensively to places I never 19 8 35 0 0.80 0.81 +S dreamed of. 13 After 10 years of working for Qantas I finally came to visit Australia. I met 16 14 37 0 0.93 0.94 my future husband during this trip. It was on the cards. 14 [After John’s marriage proposal] I came back to Mexico. John followed me a 20 20 39 3 0.99 0.99 +HH few months after and we were married. 17 On the other hand I had a close relationship with my mother. As I have got 19 20 30 14 0.52 0.54 +HH older I have discovered many of her traits in my personality. 18 As my children reached the age of 10 and 12 years we felt we would like to 10 20 32 6 0.76 0.73 +O move to Mexico for them to absorb the Mexican culture; for all of us to live close to our extended family and cement our family values along the Mexican traditions. 19 The 8 years we spent in Mexico allowed us to enjoy a very rich experience of 18 20 36 2 0.98 0.97 +HH family life. The joy of participating closely with the family is hard to describe. The warmth of the people in general. Country of origin: Present 21 My bonding with my mother and brothers has not diminished for the fact I 20 20 34 11 0.80 0.80 +HH live away. Our communication is rich and constant. 22 I share my Mexican culture with my children here in Australia. We always 19 20 35 1 0.92 0.93 +HH speak Spanish amongst ourselves, enjoy our music, food, drinks and the spirit of our connection with Mexico. Country of origin: Future 24 In the future there will always be a special bonding with my children which 19 20 38 0 0.98 0.99 +HH comes from sharing our Mexican culture. 25 One of my important goals in the future is to continue visiting Mexico and to 9 20 31 2 0.71 0.72 +O keep in contact with my family, friends and culture.

226

Australia: Past 2 The great feeling I always had for Australia was a great influence in the way 19 18 35 1 0.95 0.96 +HH I found myself so fulfilled when I came here. I had never felt that way before. 3 I met John, my husband, on my first visit to Australia. This was an encounter 20 20 39 1 1.00 1.00 +HH with a soul mate; we felt so happy in each other’s company from the beginning and we fell in love. 5 [After our marriage in Mexico] we came back to Australia and I felt I had 12 19 36 4 0.90 0.89 +O migrated to a new country. I had taken a big step. 6 Over a period of time I felt a combination of emotions. I soon discovered I 10 16 24 9 0.54 0.53 +O did not belong here. I had no friends or family to seek support from. 8 I left my job to start a business importing Mexican silver jewellery. This 18 19 38 1 0.97 0.98 +HH provided me with an interest and the freedom to be with John when we had time off work. 9 The arrival of our first child meant the end of my business. By this stage I 20 20 39 1 1.00 1.00 +HH had already become an Australia citizen and I felt totally at home. 10 The early years of my children were the happiest and the most rewarding of 20 20 39 2 0.99 1.00 +HH my life. I was completely fulfilled and did not want for anything more. 11 I became a freelance interpreter when the children started going to school. 20 11 39 2 0.86 0.87 +S This was a very interesting and rewarding experience, as it allowed me to become familiar with the issues of the migrant communities in Melbourne. 12 Almost at the same time I joined a refugee support group. It was wonderful 20 18 39 3 0.98 0.98 +HH to be able to help so many families to settle with a new life in Australia. Australia: Present 15 [Since] January 2000 we [are] back in Melbourne. It took me about a week to 18 20 36 4 0.96 0.95 +HH settle back in, it was like coming back to my old shoes. I knew we had challenges ahead, but I had great confidence because we knew we were in a place where there is law and order. 16 Our time here has given us the satisfaction to look back and know it was a 20 20 38 1 0.99 1.00 +HH good decision. Referring to the way in which our children responded. They feel very happy and fulfilled. 17 John and I bought a business which taught us a totally different industry— 20 20 39 1 1.00 1.00 +HH hospitality. We thoroughly enjoy our work and the stability it has given us once more.

227

8.5 Case study: The “Regressive” Migrant (Type 4)

Carla is representative of the “Regressive”, migrant type. She emigrated from

Uruguay with her family at seven years of age. She became very unhappy in Australia, missing her relatives in Uruguay and feeling isolated both at school and within her family. She and her mother returned to Uruguay when she was 14 where she re- established connection with her Uruguayan family. She began a relationship which she thought would lead to marriage, but her mother forced her to return to Australia, and her boyfriend subsequently jilted her. She married at 18 in Australia and had two children, but the marriage ended in a bitter divorce with shared custody. Because of the shared custody she was unable to return to Uruguay with her children; however, when her grandmother became terminally ill, she obtained a dispensation from the court ruling and was able to return. She was happy there, but felt and still feels she could not remain in Uruguay as her children have better chances in Australia. She has returned and still feels estranged from her family in Australia. Carla in effect migrated three times to

Australia, twice as a child, when she was not party to the decision and seemingly would have opposed it if she could; and once as an adult, when she chose to return: “…home was Australia if I wanted [my children] to have opportunities.” It is for this reason that she is treated as a “voluntary” migrant for the purpose of this thesis.

Carla’s measures of psychological distress and well-being complete the picture of a person who has encountered disappointment in life and has let it get her down. At the time of the interview she lacked confidence in social situations (SAD = 12, cf. sample mean of 8.4); was concerned at giving a negative impression to others (FNE = 15, cf. sample mean of 13.3); and was mildly depressed (BDI-II = 20, cf. sample mean of 9.1).

Her psychological well-being was average or below on all subscales, particularly Self- acceptance, where she was well below (50 cf. sample mean of 64.6).

228

Major themes. Figure 8.12 shows the mean thematic profile of Type 4 participants (Panel 1) and for Carla, the case-study selected to exemplify them (Panel

2). The mean profile of valuations of Type 4 participants differed significantly according to whether they pertained to country of origin or Australia, χ2(5,

N = 87) = 27.23, p < .001 (two-tailed). Valuations located in country of origin were high on Strength and unity and Powerlessness and isolation, whereas those in Australia were high on Autonomy and success, Powerlessness and isolation, and Aggression and anger.

Carla’s thematic profiles correlated highly with the mean profiles of Type 4 participants both for country of origin ( r = .97) and for Australian valuations (r = .77), indicating that she is representative of the Type 4, (Regressive), migrant type.

229

Strength and unity 50%

40%

30% Autonomy and success Unity and love 20%

10% Origin

0% Australia

Aggression and anger Unfulfilled longing

Powerlessness and isolation

Panel 1 All Regressive migrants

Strength and unity 50%

40%

30% Autonomy and success Unity and love 20%

10% Origin

0% Australia

Aggression and anger Unfulfilled longing

Powerlessness and isolation

Panel 2 Carla

Figure 8.12 Proportion of valuations, by major theme and location: Type 4 participants and Carla

230

Self-confrontation indices. Figure 8.13 shows Carla’s mean self-confrontation indices. On self-enhancement, there was no significant difference between her valuations of country of origin (M = 13.7, SD = 7.6) and those of Australia (M = 11.1,

SD = 6.4), t(53) = 1.43, p = .16 (two-tailed). On union-with-the-other, her valuations of country of origin scored significantly more (M = 13.5, SD = 6.6) than those of Australia

(M = 6.8, SD = 6.0), t(55) = 4.03, p < .001 (two-tailed). On positive affect, her valuations of country of origin scored significantly more (M = 26.9, SD = 15.3) than those of Australia (M = 16.7, SD = 15.5), t(57) = 2.56, p < .05 (two-tailed). On negative affect, her valuations of country of origin scored significantly less (M = 15.4, SD =

12.7) than those of Australia (M = 22.3, SD = 12.3), t(56) = –2.14, p < .05 (two-tailed).

40 Origin Australia 35 General 30 Ideal

0 S-index O-index P-index N-index

Figure 8.13 Mean self-confrontation indices, by location: Carla

231

Carla is representative of the Type 4 (Regressive) migrant type. In Gergen &

Gergen’s (1986) typology, a “regressive” narrative is one in which one “is continuously moving away from the valued state” (p. 27). Corroboration of this is found in Figure

8.14, which shows the temporal dimension of Carla’s self-confrontation indices.

Temporally, Carla presents a complex picture. No clear temporal progression is observable in S and O. Well-being (defined as P minus N) appears to rise through time for Australia but not country of origin. Carla’s regressive classification is based on the fact that her Australian valuations are on average more negative than her country-of- origin ones. However, unfortunate for her as this no doubt is, it is notable that in

Australia the trend is up. It was Carla’s extremely rocky start in Australia that apparently gave rise to her “regressive” classification.

20 S O P 15 N

0 Past Present Future Past Present Future

Origin Australia Figure 8.14 Proportion of valuations, by location and period: Carla

232

Self-narrative. Carla’s self-narrative is characterised loss and disappointment; her desires are thwarted. She was stripped of all she held dear, and when finally she regained her freedom, it was hollow. The following valuations show this.

From the outset migration was a let-down. Her disappointment at not going to

Sydney (about which as a mere child she surely knew little) may mask a deeper antipathy:

Country of origin: “I felt I had a place in the family…. I was doted on by all family members.” (+S). Australia: “[W]e came to Melbourne instead of Sydney. This was disappointing. My parents had not kept their promise … I remember the fear of not being understood” (–S).

... and she makes the depth of her feelings clear:

Country of origin: “I remember feeling special and having lots of self-esteem, feeling good about myself, being very out there, and being very affectionate towards my relatives” (+HH). Australia: “I hated Australia” (–O).

At first she delights in the return to her longed-for home:

Country of origin: “From the minute I stepped into Uruguay I felt whole, I felt I was someone …. I remember kissing, hugging and being very affectionate with my gran and all my family” (+HH). Australia: “My dad was physically abusive towards me” (–LL). “I felt my mum did not protect me from my dad ” (–LL).

... but as an adult she finds she has changed. Or maybe it is her childhood home that has changed. Either way, things are not what they are, and she must make the best of it:

Country of origin: “… I no longer have the urge to be in Uruguay with my family, as I have come to terms with the fact that I don’t belong there …”.

233

Australia: “Eventually connecting with someone romantically in the long term” (–S).

Carla’s self-narrative is one of distancing from all she holds dear. She clings to a receding past, and when finally she revisits it, mirage-like it is no longer there. This is a story of regression from a desired state.

8.6 Summary

This study presents four case studies chosen to represent the four migrant types—

Progressive, Low-stable, High-stable and regressive— identified in the analysis. They were found to corroborate the findings of Studies 1 and 2 in that good matches were found between the case studies and the participant types they represent.

Yoana’s self-narrative is of setting new and very different goals, and achieving them—truly a progression to better things. She exemplifies the Progressive migrant type. Juan’s self-narrative is of mental struggle and survival—stable, but compromised by underachievement. He exemplifies the Low-stable migrant type. Luisa’s self- narrative builds on a strong sense of herself and achieving more—stability crowned by emotional reward. She exemplifies the ‘High-stable’ migrant type. Carla’s self-narrative is of distancing from all she holds dear. She clings to a receding past, and when finally she revisits it, mirage-like it is no longer there. This is a story of regression from a desired state. She exemplifies the Regressive migrant type.

234

Table 8.4 Valuations with major themes, by location and period: Carla

Valuation S O P N G I Theme Country of origin: Past 1 My extended family was important. I felt I had a place in the family. That 20 13 35 14 0.54 0.57 +S place was validated. I was the first girl born in the family. I was doted on by all family members. 2 I remember being 5 and starting prep at what was an all-boys school. Being 20 18 40 5 0.62 0.89 +HH one of the first girls I was chosen as a leader at school assemblies and was chosen for main characters at school plays. I had lots of friends. I remember I was well liked by all the kids. 3 I don’t recall any particular relative standing out for any particular reason, or 20 18 40 5 0.62 0.89 +HH any relative being more important than any other. I remember having a strong bond with all my family, in particular with my mother’s side. 4 I remember feeling special and having lots of self-esteem, feeling good about 20 18 40 5 0.62 0.89 +HH myself, being very out there, and being very affectionate towards my relatives. 5 Going on lots of holidays with immediate family. 20 18 40 5 0.62 0.89 +HH 6 I have not got emotional attachment memories to either of my parents while 0 0 0 32 -0.57 -0.88 -LL in Uruguay. 9 My only other memories of my sister was on the plane trip to Australia 0 15 0 35 -0.61 -0.66 -O calling “socorro, socorro, llamen la policia” when the plane was falling. 29 I wanted to die If I could not go back to Uruguay at age 14. my mother took 0 0 0 20 -0.19 -0.63 -LL me to Uruguay. 31 From the minute I stepped into Uruguay I felt whole, I felt I was someone. I 20 18 40 5 0.37 0.89 +HH felt I was alive. I was 14. I lived with my maternal grandmother for 3 months. I remember kissing, hugging and being very affectionate with my gran and all my family. 32 Falling in love at 14, my grandma and mother did not speak to me. They were 20 20 40 5 0.32 0.91 +HH angry with me, however I still felt whole. I was happy. I felt I was important to someone. I was worth something. 35 The relationship lasted over 1 year until news came from overseas that he 0 0 0 35 -0.50 -0.91 -LL was in another relationship. Three months later he married. There was closure to our relationship. 46 I travelled to Uruguay in 1999 aged 32 with both children. There was a 20 20 40 5 0.32 0.91 +HH family reunion for my grandma, all her children and grandchildren were present. This trip was a celebration for my grandmother. For me it was a goodbye, but also a return to my childhood, being with everyone that I loved and loved me back. I belonged again, and this was now shared with my children, who finally got to experience the person I was. This was an amazing trip uniting my children to mine and their ancestry. During the trip I came to terms with the fact that we (my children and I) will always be Uruguayans. However, home was Australia if I wanted them to have opportunities. This trip highlighted that regardless of the difficulties I had faced in Australia, I was able to raise 2 children on my own, return to studies, obtain university qualifications, was employed in an excellent position, and was still able to be a full time mum. I am proud of what I have achieved as a sole parent. 47 After the trip I decided we would return to Uruguay for an extended period of 20 20 40 5 0.32 0.91 +HH 3 months, so the children could further experience being Uruguayan. This happened in 2002. 48 After this trip I again felt fulfilled. I felt complete and had achieved my, and 20 20 40 0 0.48 1.00 +HH my children’s identity. Country of origin: Present 49 I am proud of who I am and where I came from. I am at peace with who I am. 20 20 40 0 0.48 1.00 +HH 50 I speak Spanish as much as possible. On special occasions such as birthdays I 20 20 40 0 0.48 1.00 +HH cater Uruguayan food and Latin music. 51 I have lots of communication over the internet with family of origin. 20 20 40 0 0.48 1.00 +HH

235

Australia: Past 11 I remember we came to Melbourne instead of Sydney. This was 15 4 13 28 -0.08 -0.36 -S disappointing. My parents had not kept their promise. We arrived at a hostel (Springvale). I remember the fear of not being understood when my dad told me to go and buy a coke at the milk bar. 13 The 1st day at school I screamed and cried for the teacher to go get my 2 10 0 30 -0.46 -0.61 -O brother. I hated Australia. 14 I was responsible for taking my younger sister to kinder. One day I forgot to 2 10 0 30 -0.46 -0.61 -O pick her up. It was 5 pm when I remembered. The kinder was closed and I was banging on the kinder doors. I was 7. 17 Home: parents worked. I was responsible for myself, sister and brother to a 10 0 9 34 -0.09 -0.66 -S degree. I had a key to our home and was responsible until parents finished work. “latch key kids”, “just my job”. I still resented coming here as I did not have this as my job in Uruguay. 18 Academically, my teacher continuously reinforced that English was my 12 4 12 31 -0.25 -0.56 -S second language and I would need to improve to get ahead. This also reinforced my wog mentality, that I did not fit in, that I was not an “Aussie”. 19 I was becoming more and more self-conscious, not belonging (8 to 10 years 12 4 12 31 -0.25 -0.56 -S old). 20 I had physical fights with my entire grade (year 5). Boys would hit me for no 10 0 0 36 -0.40 -0.82 -S reason. I always fought back as I was told in the hostel that they (my parents) did not want me to come home crying because someone had hit me. If I came home crying I would be hit by them. 21 In grade 6 I became very popular. I was friends with most popular kids in the 17 11 31 7 0.68 0.80 +S school. I felt like I did in Uruguay. 22 In year 7 and 8 I hated school. I felt like I did when I arrived in Australia, out 10 0 0 36 -0.40 -0.82 -S of place. 24 In year 8 I hated school. I wagged and felt no connection with school. 10 0 0 36 -0.40 -0.82 -S 26 My dad was physically abusive towards me. 0 0 0 30 -0.54 -0.82 -LL 27 I felt isolated within my family. I felt singled out in a negative way. I had no 6 0 2 30 -0.30 -0.74 -S one to turn to. 28 I felt my mum did not protect me from my dad. 0 0 0 25 -0.37 -0.73 -LL 39 I focused on work and went back to school, and made a new group of friends 19 11 38 6 0.50 0.76 +S (17 years old). 43 The separation was very violent. I had no support from extended family. I felt 6 0 0 35 -0.42 -0.81 -S that if I was in Uruguay this would not have occurred this way as family members would have intervened. I had no protection. Australia: Present 55 I am currently enjoying my work, but am evaluating long term goals and 20 11 40 5 0.74 0.79 +S exploring future employment options. Australia: Future 59 Eventually connecting with someone romantically in the long term. 8 2 16 28 0.01 -0.54 -S 60 My goals: Continue to enjoy being me, travelling, purchasing a house. 20 18 38 0 0.53 0.99 +HH

236

CHAPTER 9: General Discussion

This chapter considers the findings from the three studies that constituted the empirical work for this thesis. It opens with restatement of each study and synopses of the research questions. Next, the findings are compared and contrasted with other research in the field. For each finding the relationships, generalisations and implications for the broader theory of the dialogical self are identified. The final section considers the strengths and limitations of the research, showing where they break new ground and suggesting profitable areas for future research.

9.1 Summary of the Empirical Work

The present thesis comprises three separate studies. All were based on data from the same sample of 38 migrants to Australia, analysed in different ways. The data were obtained by means of in-depth interviews using the self-confrontation method complemented by self-report measures.

9.1.1 Study 1

The analytical purpose of Study 1 was data reduction: it employed the statistical technique of cluster analysis to group participants into four types, and on that basis analysed their characteristics. Study 1 considered the full set of self-narratives of the 38 participants, both in country of origin and Australia in terms of valuation theory.

Average self-enhancement (S), union-with-the-other (O) and positive affect (P) were found to be higher for Australian I-positions than for country of origin, but significantly so only for self-enhancement (S), and positive affect (P). This indicates the formation of

I-positions with different thematic content. Cluster analysis was performed on the four indices for country of origin and for Australia. In both cases, two clusters were identified, explaining about half the within-cluster variance. One was characterised by

237 high self-enhancement (S), union-with-the-other (O) and positive affect (P), and low negative affect (N); the other by low self-enhancement (S), union-with-the-other (O) and positive affect (P), and high negative affect (N).

This procedure served to identify four different participant “types”. Progressive migrants increase self-enhancement (S), union-with-the-other (O) and positive affect

(P), and reduce negative affect (N) as between origin and Australia; Regressive migrants, the opposite. Stable migrants showed no important changes in these indices: these were classified as High-stable or Low-stable according to whether their S, O, and

P indices were high or low. Progressives comprised 8% of the sample, Regressives

16%, and Stable types 76%, of which 29% were High-stable and 47% Low-stable.

Besides differing in motives and affects, participant “types” also differed in their psychological distress and well-being. Progressive participants were in general less socially anxious than the other types, though the difference from Regressive was borderline. Low-stable participants were more fearful of negative evaluation than High- stable. However, the smallness of the sample appears to have prevented significance differences being found where otherwise they would have been expected: between, say,

Progressives and Regressives. No significant differences were found between participant types on depression. On psychological well-being, some significant differences were evident: Regressives had significantly worse ‘relations with others’ than did Progressives and High-stables.

9.1.2 Study 2

The analytical purpose of Study 2 was, like that of Study 1, data reduction, but of a different kind. By treating the dialogical self as a hypothetical probability distribution of themes of valuations, it permitted formal hypothesis testing of differences between participants’ self-narratives pertaining to their countries of origin, and their self-

238 narratives pertaining to Australia. Study 2 considered the set of self-narratives classifiable into Hermans’ six major themes of self-narratives. The entire set of 1,565 valuations was allocated to Hermans’ six major themes wherever possible (about a third of the full set, as most self-narratives had no major theme). Using this classification, a profile giving the proportion of each major theme in the full set of valuations was attached to each individual both globally and for country of origin. Participants were then classified into two groups: “univalent” and “multivalent”. Univalent migrants were those who expressed only a single affect, either positive or negative (but not both) in their valuations. Multivalent migrants were those who expressed a mix of both positive and negative affects. Thirty-nine per cent of migrants expressed multivalent major themes. These migrants had multi-modal thematic I-positions and thus correspond to the dialogical view of the self.

Participants were subjectively classed as either “stable” or “unstable” on the basis of the degree of difference between their profiles calculated for narratives in Australia and at country of origin. On this basis 82% of migrants had stable narratives between origin and destination. Parametric tests were conducted for the differences in proportions of major themes at origin and destination. Participants’ narratives contained fewer –LL valuations in Australia, more +HH valuations, more +S valuations and fewer

O+ valuations, but only the first of these was significant at the 5% level. A joint test of the null hypothesis of complete thematic stability was decisively rejected.

A further analysis was conducted on the full set of valuations to escape the attrition implicit in Hermans’ typology of major themes. An eight-way classification of participants was obtained by considering whether their average scores on S, O and W (=

P – N) were higher or lower than the sample average, both at origin and in Australia.

The resulting matrix showed how their narratives differed between origin and Australia,

239 with entries on the diagonal corresponding to stability. Considerable thematic stability was again found across valuations for country of origin and Australia, though there was some evidence of change, contained in the off-diagonal portion of the transition matrix.

9.1.3 Study 3

Study 3 was idiographic in approach; its purpose was to corroborate the previous two nomothetic studies by examining the self-narratives of specific participants. Four case studies were chosen to represent the four migrant types identified in Study 1—

Progressive, Low-stable, High-stable and Regressive. In all cases a good match was found between the case study and the participant type it represented. Idiographic and nomothetic analyses have complementary strengths. The idiographic content of these valuations from each of the four groups served to provide more information about the range and individual natures of the experiences examined in this thesis.

9.2 Main Findings

The present thesis explored four areas of research: stability and change in the dialogical self, the nature of migrant identity, the composition of I-positions, and the psychological state of migration. This section presents the salient findings of the thesis, ordered according to the specific research questions to which they relate, and places them in the context of existing research,

9.2.1 Migrant identity

Research question 1a: How is biculturalism reflected in I-positions? The participants were found to experience dual cultural identities. Specifically, both compatible (Low- and High-stable) and conflictual (Progressive and Regressive) biculturality were evident in a sample of migrants who claimed to be integrated into the

Australian society and perceived themselves as having two cultural identities.

240

This finding builds on work by Phinney and Devich-Navarro (1997) that found that biculturalism has two components, ethnic and national. To be bicultural is to identify strongly with both, and this implies integration into the larger society. As in their research, participants in the present study explored meaningful stories of their lives as part of their membership of their ethnic group and of mainstream Australia; their sense of being part of two cultures; whether or not it caused conflict in their lives; and how they dealt with the relationship between the two cultures. The use of Hermans’ classification of valuations in terms of their positive and negative valences was one of the features that differentiated the present migrant’s self-narratives and gave the overall affective qualities of their lives as depicted in their stories from home and from

Australia. This pattern is evident in other research: LaFromboise et al. (1993) and

Phinney & Devich-Navarro (1997) found similar differences.

Based on positive and negative valences of the valuations in their self-narratives, and the extent to which self-enhancement (S) or union-with-the-other (O) dominate the self-narratives, the present study generated a typology of bicultural identification. Low- stable and High-stable participants did not perceive the mainstream and ethnic cultures as mutually exclusive or conflicting. In their stories and self-reported motivation and affects, they showed equal representation in both. This finding concurs with Phinney and Devich-Navarro’s (1997) finding that people who identified equally with both cultures and dealt with them by combining them exhibited “blended” biculturalism. For example, when asked to say if they are ethnic or American, they tended to say “I am both” (Phinney & Devich-Navarro, 1997). By not providing a distinctive differentiation between the motives and affects belonging to the respective stories from the two I- positions, the participants in the Low- and High-Stable clusters exhibited a blend of the two cultures in each I-position.

241

On the other hand, those participants who identified more with one culture than with the other, despite being able to function in both settings, demonstrated Phinney &

Devich-Navarro’s (1997) “alternating” biculturalism. In their stories, Progressive participants represented their Australian identity more positively than they represented their country of origin identity; while Regressive participants identified more positively with their country of origin identity than with their Australian identity. Although they outwardly identify with both cultures, their stories show that their dual cultures differ greatly in terms of motives and affect. Regressives, not unexpectedly, tend to keep their cultural identities apart; for them, Australian and country of origin identities are perceived as distinct, separate, and oppositional (Phinney & Devich-Navarro, 1997;

Vivero & Jenkins, 1999). The present results provide a foundation for research in bicultural identity and suggest that Berry’s model of acculturation does not fully explain bicultural identity integration.

In sum, the present research corroborated previous research that identifies two modes of biculturalism: compatible and oppositional. Low-stable and High-stable migrants, having compatible bicultural identities, viewed their heritage and host cultures as complementary, whereas Progressive and Regressive migrants, having oppositional identities, viewed them as discrepant and conflicting. It is the latter view that fosters internal conflict and may compromise well-being (Vivero & Jenkins, 1999), which may explain why Regressive migrants had significantly lower well-being (specifically

‘relations with others’) than did Progressives and High-stables. Present research furthermore corroborates previous findings that compatibility does not depend on how long migrants have lived in the host culture (Phinney & Devich-Navarro, 1997), or their migrants’ attitudes toward biculturalism (Vivero & Jenkins, 1999).

242

The compatible/oppositional dichotomy has been elaborated further in the work of

Benet-Martinez, Leu, Lee, and Morris (2002), Benet-Martinez & Haritatos (2005) and

Benet-Martinez, Lee, & Leu (2006). Their findings of “harmonious” and “conflictual” biculturality can also be used to interpret the present findings. The two migrant types—

Low-stable and High-stable—correlate with “harmonious” biculturality in the sense that they exhibited “congruence” between the two cultural identities in terms of themes and affective tone in their self-narratives. These migrants exhibited their two identities as

“congruent”: when either was activated, both reflected a psychological consonance of themes and affective tone and corresponded to each other. However, the other two migrant types—Progressive and Regressive—correlate to “conflictual” biculturality.

For them, their two identities related conflictually and were experienced as a paradox, reflecting a psychological dissonance between cultures: the activated identity elicited a cognitive style in terms of self-enhancement (S) and union-with-the-other (O) that was characteristic of the alternative culture. This translated into two modes of integration.

While all participants were all able to switch between cultures and elicit self-narratives, only Progressive and Regressive migrants (the “conflictual biculturals”) experienced the two cultures in an antagonistic way. Moreover, because the antagonism between the two cultures for the Regressive participants’ self-narratives resides in their Australian identity, this translated to a detriment of their present psychological well-being.

This finding can also be understood in the context of work by Benet-Martinez and

Haritatos (2005; Benet-Martinez, Lee, and Leu (2006), which investigated how persons connected to more than one culture internalise potentially conflicting cultural identifications. They found that “biculturals can differ … in the extent to which they perceive their cultural identities as largely integrated and compatible … or dissociated and difficult to integrate …” (p. 395). Benet-Martinez, Leu, Leu, and Morris (2002)

243 conceptualised the tendency to perceive cultural identities as either compatible or oppositional as an individual difference, which they termed bicultural identity integration. They found that those high on BII do not feel their cultures have a difficult coexistence, and do not see them as oppositional. Compared to individuals low on BII, those high on BII identify more with the mainstream culture, even though identification with their culture of origin may be similar for both groups.

In summary, the present research identified two types of migrant, Progressive and

Regressive, that are similar to those with “conflictual” biculturality (Boski, 2008) in that they exhibited disharmony between the two cultural identities in terms of themes and affective tone in their self-narratives from their country of origin and from Australia.

Like Vivero and Jenkins (1999)’s notion that bicultural migrants with low integration between their identities may feel “culturally homeless”, the Progressive and Regressive migrants in the present study have “different pieces of home in different places” that can never be put together, because they “contradict or conflict with one another” (p. 7).

While the sense of cultural homelessness resides in the past for Progressive migrants, the opposite holds for Regressive: the Progressives found harmony in their Australian cultural identity; the Regressives found conflict.

As Phinney and Devich-Navarro’s (1997) observed: “biculturalism is a complex and multidimensional phenomenon; there is not just one way of being bicultural.” (p.

19). But although the importance of this complex and important field has been acknowledged (e.g., Hermans & Kempen, 1998; LaFromboise, Coleman, & Gerton,

1993), empirical research is sparse (Benet-Martinez & Haritatos, 2005). The present thesis helps to fill this gap.

Socio-cultural adaptation (Ward & Searle, 1990) concerns the relationship between migrants and their host society, and it includes their ability to cope with the

244 everyday issues that require social interactions. On the other hand, psychological adaptation includes intra-psychic aspects such as feelings of well-being and emotional satisfaction. In the present research psychological adaptation was assessed with measures of depression and well-being, and socio-cultural adaptation was assessed with two measures of social anxiety. Social anxiety as assessed by the SAD scale is by definition, an expression of discomfort in or about, interpersonal situations; and fear of negative evaluation as assessed by the FNE Scale was defined by the scale's authors as

“apprehension about other's evaluation, distress over their negative evaluation, and the expectation that others would evaluate one negatively”, and suggested that “fear of loss of social approval would be identical to FNE” (Watson & Friend, 1969, p.449). That

Regressive migrants scored highest on both measures suggests, therefore, that social relationships with host nationals (and perhaps co-nationals as well) may have reduced their sociocultural and psychological adaptation. This is corroborated by their significantly lower scores on the Positive relations with others subscale of the psychological well-being measure.

Research question 1b: How is ethnic identification reflected in I-position?

Participant types were found to differ little demographically. This is unsurprising as the study sample was limited in the range of socio-economic status (most were middle- class), age range (most were in mid-life), and education (most were educated to tertiary level). Most too were from middle-income countries such as Mexico and Malaysia. But one variable, ethnic identity, was found to vary significantly between participant types.

The question posed was: “Do you think of yourself as belonging to a particular ethnic group, for instance Italian, Greek, Chinese etc.?” Of the entire sample, 21% answered yes, 79% no.

245

This might seem to clash with the requirement that all participants possess “two identities” (see Chapter 4 Method). However, having an (ethnic) “identity” is different from “belonging to a particular ethnic group”; the latter requires a continuing active association with compatriots. It is unsurprising, therefore, to find most of the sample claiming to have “two identities” but not “belonging to a particular ethnic group”. It is hard, nevertheless, to glean much from the response to this question as cell-counts were too small to permit a reliable test of statistical association. That said, it is noteworthy that all Progressive migrants answered in the affirmative while all Regressive migrants answered in the negative (the other two, stable, types responded in roughly the same proportions as the sample in aggregate, and are therefore unexceptional).

This is an unexpected finding, for it suggests that those individuals with the greatest well-being in Australia are the most involved at a personal level with their own ethnic communities—and the converse, of course. This might seem counterintuitive, but as Berry and Sam (1997) point out, integration “appears to be a consistent predictor of more positive outcomes than the alternatives” (p. 318). If so, those who maintain a close personal connection with the communities from which they come, as well as acculturating to the host society, will have the highest well-being. So although all people in the sample of migrants studied in this thesis were required to be “integrated” in Berry’s sense, some were apparently more “integrated” than others.

Finally, the present finding of low ethnic identification despite an a-priori endorsement of the integration strategy by all participants, agrees with Liebkind’s

(2008) notion that while the strength and nature of a migrant’s actual identification with the in-group will determine much of his or her response to acculturation, it should not, therefore, be confused with endorsement of native culture and/or adoption of host culture.

246

Evidence from case studies. Biculturalism and ethnic identification were apparent in the four case studies that exemplify the four migrant types identified in this study.

The two “stable” migrants—Juan (Low-stable) and Luisa (High-stable)—show

“blended/harmonious” biculturalism. This is evident in the similarity of motive and affect, whether positive or negative, as between their I-positions for country of origin and for Australia. For instance, Juan found both cultures equally alienating: in his country of origin he felt “insecure”, and in Australia he felt “he did not belong” (Table

8.2, p. 206). By way of contrast, for Luisa both cultures were equally welcoming, and her arrival in Australia “was almost like a homecoming” (Table 8.3, p.221). The point here is that each person harboured similar feelings towards two cultures, in one case negative, in the other positive, but nevertheless similar.

The two “unstable” migrants—Yoana (Progressive) and Carla (Regressive)—also show biculturalism, but of a conflictual kind. Referring to alcohol, of which she disapproves, Yoana recalls how she recalls a child being encouraged to consume liquor in her country of origin, while in Australia she had observed the very opposite (Table

8.1, p. 206). Carla, however, “wanted to die If [she] could not go back to Uruguay”

(Table 8.4, p. 230). Both participants participated in the respective cultures of their countries of origin and Australia, but their biculturalism was of an

“alternating/conflictual” type because its parts could not be reconciled.

9.2.2 Stability and change in the dialogical self

Research question 2a: Are country-of-origin and Australian I-positions similar, and if so, how? Figure 9.1 summarises the findings as they relate to I-positions of the four respondent types. The main finding is one of modest change in I-positions in a context of overall stability of thematic content of valuations. For some participants

(Types 1 and 4), I-positions were unstable, but for most (Types 2 and 3), there was little

247

difference in country-of-origin and Australian I-positions. Specifically, in Study 1, 76%

of participants showed similar thematic profiles in valuations about Australia and

country of origin. Nevertheless, the hypothesis of complete thematic constancy was

rejected.

Type 1 (Progressive) Type 2 (Low-stable)

20 20

15 15

10 10 index index - - O O

5 Origin 5 Origin Australia Australia 0 0 0 5 10 15 20 0 5 10 15 20 S-index S-index

Type 3 (High-stable) Type 4 (Regressive)

20 20

15 15

10 10 index index - - O O

5 5 Origin Origin Australia Australia 0 0 0 5 10 15 20 0 5 10 15 20 S-index S-index

Figure 9.1 Average wellbeing (W), by average self-enhancement (S) and union-with- the-other (O), by participant type

Note: Circle area represents well-being.

248

This finding supports the theory of the dialogical self given that modest change in a context of overall stability is essential for any theory if it is to be useful. Hermans

(2006) recognises this when commenting in a therapeutic context:

… stability and change need each other; a certain degree of stability is needed in order to realize a significant change. The experience of safety and continuity provided by the stable parts of the valuation system help clients to face the discontinuities in their lives that otherwise would be disrupting. In other words, stability and change are both necessary parts of the therapeutic reorganization of the valuation system. (p. 12–13; italics added)

Researchers in the field of narrative theory (for instance McAdams, 2014) tend to stress the organising role of the narrative rather than its stability. In fact both are valid.

A story has a plot, which is to say the events in it happen for reasons that are both plausible and evident. In telling their stories, therefore, individuals are implicitly saying:

“This person I am today is who I have been years becoming” (Rosenwald & Ochberg,

1992, p.9). Thus stability that accommodates incremental and (usually) gradual change is a necessary attribute of the narrative. In short, the two attributes—coherence and stability—are one and the same. More generally, McAdams (2001; 2014), in his life story model of identity, exemplifies this unity when he stresses the integrative nature of stories, through which the self seeks temporal coherence. Since the self is changing, self-narratives are constructed to integrate past, present and anticipated future. The life thus rendered consists of coherent stories with beginnings, middles, and ends

(McAdams, 1985; Polkinghorne, 1988). Likewise Baumeister, Stillwell, & Wotman

(1990) and Gergen & Gergen (1988) regard narratives as personal accounts that make life intelligible and hence predictable, which they do by imposing a logical structure on events. Interpreting personal accounts in this way reveals how individuals make sense

249 of their lives: narrators select from a virtually limitless array those moments they deem significant, and arrange them in a coherent order. This fashioning of order is more than a chronology; it implies a causal chain.

As noted earlier, Gergen and Gergen (1988) suggested that each event in a narrative can be viewed as moving through a two-dimensional space. One dimension refers to its positive or negative evaluation, the other refers to stability or progression.

In their “stability” narrative, there is no change in self-evaluation. It “links incidents, images, or concepts in such a way that the protagonist remains essentially unchanged with respect to evaluative position” (1986, p.27). In this case the function of the self- narrative is to provide the self with unity, continuity, purpose and meaning to life events: “One’s present identity is thus not a sudden and mysterious event, but a sensible result of a life story” (p. 19). In this way the narrative form engenders a sense of continuity by linking past, present and future in a plausible causal chain (Gergen &

Gergen, 1988). Thus, Gergen and Gergen’s (1986) framework for the analysis of stability narrative structure accounts for coherence, where coherence concerns the elements in the narrative that relate to the valued end goal.

Research question 2b: Do country-of-origin and Australian I-positions differ, and if so, how? Where I-positions differed, it was found that they did so in an intelligible way. When all participants were treated as a single group, the mean values for self- enhancement (S) and positive affect (P) were significantly higher for Australia than for country of origin, while union-with-the-other (O) and negative affect (N) did not change significantly. When participants were furthermore classified into types, further patterns emerged. Progressive, Low-stable and High-stable migrants, together comprising 84% of the sample, showed on average more self-enhancement (S) and positive affect (P) in their Australian than in their country-of-origin valuations. This suggests that most found

250 migration to be rewarding: it apparently gave them scope for self-enhancement (S), and their well-being (W) rose as a result of increased positive affect (P), particularly so in the case of Progressive migrants. Regressive migrants, comprising 16% of the sample, did not benefit in this way: not only did their Australian I-position show less self- enhancement (S) and union-with-the-other (O), but well-being was lower as a result of reduced positive affect (P) and increased negative affect (N).

Sanchez-Rockliffe and Symons (2010) studied changes in themes in these data.

Examining self-narratives, they found that many participants seemed to be responding positively to the social conditions they find in Australia. Once these factors were controlled for, there were no significant changes in the means of any of the self- confrontation indices between country of origin and Australia. In other words, these migrants’ greater self-enhancement (S) and positive affect (P) could reflect a favourable response to the peaceful, liberal society described by Hirst (2014). Interestingly, for

Progressive and High- and Low-stable migrants, self-enhancement (S) was significantly higher for their Australian I-position than for their country of origin, but union-with-the- other (O) was not. A possible explanation for this disparity, which is discussed later, is that migration accommodates self-enhancement (S) proportionately more than union- with-the-other (O) since by definition it takes the migrant away from family and friends.

The findings have noteworthy implications for the underlying theory of the dialogical self. Hermans and Hermans-Jansen (1995) interpret self-enhancement (S) and union-with-the-other (O) as fundamental motives, as distinct from positive affect (P) and negative affect (N), which are, as their names imply, affects related to well-being. It is natural, therefore, to expect positive affect (P) and negative affect (N) to change in the new social environment offered by migration to another country, as the individual finds more or less scope for fulfilment. In this way theory explains the observed change

251 in positive affect (P). Changes in motives, such as the observed increase in the self- enhancement (S), demand a different explanation. They suggest that the social environment may bring about I-positions that differ in their motives, at least in some cases. This too is consistent with theory: Hermans and Hermans-Jansen (1995) argue that motives evolve throughout the lifespan and differ between I-positions (e.g., worker, spouse, parent).

Finding unity and purpose in life amid constant change—Gergen’s multiphrenia—has been questioned by theorists such as Gergen (1991). The saturated and fragmented self depicted by postmodern theorists places the individual at war with societal change, struggling to make sense of life. Gergen’s “saturated self” has lost coherence and is instead numbed by a deluge of diverse and opposing messages. In this postmodern understanding of selfhood, people move from one identity to another as situations change and life moves on (Lifton, 1993). The findings of the present thesis are, however, consistent with Gergen. Although most of the sample can be seen as having overcome multiphrenia, some—the Regressives—can be seen as particularly affected by it. These are people whose lives are afflicted by the disruption resulting from changing cultures, which is reflected in their I-positions. In short, multiphrenia may affect everyone, but not everyone can cope with it. And in terms of multiphrenia, migration’s acculturative burden is large.

Narrative research into personality development has focused either on story- telling in particular social contexts, or on the content of the stories told (McAdams,

2012). The present thesis belongs to the body of work that explores how story content makes meaning out of important life experiences and draws connections between them.

Hermans and McAdams both focus on the content of self-narratives, noting that self- narratives differ in the importance they give to Bakan’s (1966) themes of “agency” and

252

“communion” (Hermans, 1988, 1996, 2010; McAdams, 2003). Consistent with this approach, the present thesis examined the autobiographical events the participants considered important when thinking about their lives, and evaluated their content in terms of two motives—self-enhancement (S) and union-with-the-other (O)—as they differed across time and space. It sought to compare migrants’ self-narratives in terms of the degree to which self-enhancement (S) or union-with-the-other (O) dominated the narratives across time (self-narratives from the past, present and future) and space (self- narratives from country of origin and from Australia). The findings are consistent with

Hermans’ (1988, 1996, 2010) narrative theory of personality, and offer an alternative to the modern notion of the self. Unlike the saturated postmodern self, in which plurality translates into fragmentation or saturation, the dialogical self is fluid and multiple

(Hermans (2001). Although theorists of the dialogical self talk of multiple I-positions, there is a substantial self that persists unchanged (Richardson, Rogers, & McCarroll,

1998). According to Hermans (1996a, cited in McAdams, 2003):

One’s life becomes a story with a cast of self-characters who assume different positions in the narrative, take on different voices, represent different self-facets, personify significant trends during different developmental chapters—all is in the same evolving story, the same identity. (p. 193)

Evidence from case studies. The coexistence of “modest change in a context of overall stability” is evident in the four case studies chosen to exemplify the four migrant types identified in this study. Two migrants represent the stability of the dialogical self, though from diametrically opposed stand-points. Juan (Low-stable) experiences “mental struggle and survival—stable, but compromised by underachievement” (Section 7.3, p.188). Luisa (High-stable) experiences “building on a strong sense of herself and achieving more—stability crowned by emotional reward” (Section 7.3, p.196).

Together, these two migrant types account for most of the sample. Two other migrants

253 represent change in the dialogical self, also from diametrically opposed stand-points.

Yoana (Progressive) experiences “setting new and very different goals, and achieving them—truly a progression to better things” (Section 7.1, p.181). Carla (Regressive) experiences a “distancing from all she holds dear. She clings to a receding past, and when finally she revisits it, mirage-like it is no longer there. This is a story of regression from a desired state” (Section 7.4, p.205). Together, these two migrant types account for a minority of the sample.

9.2.3 Composition of I-positions

Research question 3a: Can individuals simultaneously possess I-positions that differ significantly in their major themes? This question can be addressed by demonstrating that some individuals have multivalent I-positions, meaning that they express significant major themes with both positive affect (P) and negative affect (N).

Such individuals cannot have a single coherent I-position but speak with multiple voices. About 39% of the sample (15 of 38 participants—see p. 185 and Table 7.3) were found to be multivalent in this sense. This shows that a large subset of the sample had I- positions with unambiguously different thematic content. They are thus evidence for

Hermans’ dialogical view of the self (Hermans & Hermans-Jansen, 1995) and are harder to explain under views that emphasise an underlying integrated self, such as that of McAdams (2014). Moreover, the multivalence of these individuals is not an artefact of different experiences in Australia and country of origin: of the 15 identified multivalents, all but three were also multivalent in narratives concerned with the country of origin.

Research question 3b: If individuals simultaneously possess I-positions that differ significantly in their major themes, as discussed above, do their I-positions correspond to those for country of origin and Australia? If so, in what way? Participants’ I-

254 positions were found to differ between country of origin and Australia. For all but

Regressive migrants, self-enhancement (S) and positive affect (P) were significantly higher for their Australian I-position. For the stable types—Low-stable and High- stable—the increase was small but nevertheless statistically significant; for Progressive migrants it was large. This is consistent with the view that many, though not all, migrants acquire I-positions pertaining to their country of settlement that differ motivationally so as to profit from its different environment and opportunities, and this benefit is manifest as increased well-being.

This still leaves several questions. First, why was the difference small for most of the sample? This can be explained by the stability of personal identity, a theme explored earlier. Motives are more enduring than temporary mood states (McAdams, 2009), more labile than traits, and reflect “major thematic goals, desires, and preoccupations during a given period of a particular person’s life” (p. 295). It is therefore to be expected that although motives can and do change, they normally do not change much or quickly.

A second question that emerges is: Why did Regressive migrants not profit from new circumstances, as evidenced by their Australian I-position, in the same way as the others? Instead their Australian I-position regressed both in motives and well-being. For one reason or another they were unable to adapt so as to profit from their new environment, which they found less congenial than the one they had left. Interestingly, the reduction was greater for union-with-the-other (O) than for self-enhancement (S).

This suggests that such migrants suffered particularly as a consequence of separation from friends and family, and their difficulty in relating to others in an unfamiliar culture—not unexpected for a migrant. Though only a small proportion of the sample, they clearly merit further study.

255

The third and final question is: Why did self-enhancement (S) differ significantly and not union-with-the-other (O)? This is answered in the next section, which explores the relative motivations of successful migrants.

Research question 3c: What motivates successful migrants: the desire for self- enhancement or union-with-the-other? Progressive participants showed significantly more self-enhancement (S) in their Australian than in their country of origin I-positions;

Low- and High-stable participants, marginally more; and Regressive participants significantly less. However, in no case did their union-with-the-other (O) differ significantly. A possible explanation is that migration accommodates self-enhancement

(S) proportionately more than union-with-the-other (O) since by definition it takes the migrant away from family, friends and a familiar culture. Another kind of life-change— one that offered scope for union-with-the-other (O), such as migration for the purpose of family unification—might have displayed a similar boost in affect for some individuals.

But that was not the case in the current study.

According to Ward and Kennedy (1994) psychological adjustment is influenced by social support and personality, with adjustment being predicted when there is a fit between the migrants’ traits and host nationals’ traits. In this line of research, a close fit between migrants’ self-construals in their country of origin and their self-construals predominant in the host society predicts migrants’ psychological adjustment.

Singelis (1994) argued that individuals have both independent and interdependent self-construals that are manifest in individualist and collectivist attitudes. The attitudes that predominate in a given social context depend on the nature of the relationship, whether with a spouse, parent, neighbour, co-worker, or other party (Matsumoto,

Weissman, Preston, Brown & Kupperbusch, 1997; Matsumoto et al., 1997).

Reinterpreting pre-existing measures of individualism and collectivism, Singelis (1994)

256 showed that Asian Americans scored higher on the interdependent and lower on the independent self-construal factor than did European Americans. His independent self- construal items were mostly concerned with individual agency, whereas his interdependent items mostly concerned self-ingroup relationships. His study suggests that differences in cultural self-conceptions may be primarily localised in the individualist self and secondarily in the collective self. Therefore, a key difference may be the extent to which the self is conceptualised as a goal-directed agent, or embedded in an in-group, rather than whether the self is conceptualised in relation to other individuals.

Yet another view is embodied in the “culture fit” hypothesis. Triandis et al.

(1985), defined ‘allocentrism’ (the willing subordination of personal goals to those of the collectivity) and ‘idiocentrism’ (the prioritising of personal goals above those of the collectivity) as personality traits that are analogous at the individual level to collectivism and individualism at the community level. The culture fit hypothesis holds that allocentrism and idiocentrism are linked to their community-level analogs. Ward &

Chang (1997) found that allocentrics are generally better adjusted in collectivist, and idiocentrics in individualist, cultures.

In elaboration, Triandis and Suh (2002) state: “The idiocentrics in collectivist cultures are more likely than the allocentrics to feel oppressed by their culture and to seek to leave it” (p.141, italics added). This may explain why Low- and High-stable migrants showed higher self-enhancement (S) in their Australian self-narratives, for as people who have chosen to migrate, they are predominantly idiocentrics; and as idiocentrics, they “emphasise self-reliance, competition, uniqueness, hedonism, and emotional distance from in-groups” (Triandis and Suh, 2002; p.140)—all traits characteristic of self-enhancement.

257

The results of the present thesis are relevant to the current understanding of the relationship between sociocultural and psychological adjustment factors among integrated migrants, that is, ones who identify as having two cultural identities. In a pluralistic society such as Australia, migrants who adopt Berry’s integration strategy desire to be in, not between, two cultures. But, in the study sample of integrated migrants, there were individual differences in the manner of relating to the dominant culture and to the native culture. Regressive migrants showed low involvement in positive relationships in Australia, which points to their preference for a separation strategy, at least in this domain. Their ambivalence about interpersonal relationships with members of mainstream society, as exhibited by higher measures of social anxiety and fear of negative evaluation, may be due to perceived rejection, prejudice or discrimination in Australia. Consequently they tend to avoid contact with members of the mainstream society. That they showed high involvement with members of their native culture in their country-of-origin narratives supports this view, as it is these narratives that offer comfort, security and self-respect.

Evidence from case studies. Multivalence of the dialogical self emerged in two of the four case studies that exemplify the four migrant types identified in this study. In a valuation pertaining to his country of origin Juan (Low-stable) states, “My father is fair but inflexible. I learnt from him that rules were not to be bent or broken.” Likewise when Carla (Regressive) states, “My extended family was important... I was doted on...”

(Table 7.4), she is both appreciative (shown by a high O motive) and at the same time perhaps feeling suffocated (shown by a high S motive). The fact that these valuation rate highly on both positive and negative affect strongly suggests that they were being viewed from different I-positions (though in this case not ones associated with country of residence).

258

Evidence of adjustment to migration and for differing motives for migration

(more S, same O) is found in the many valuations pertaining to the Australian I-position that both emphasise personal achievement and show high positive affect. Luisa (High- stable) is an example when she writes, “I knew we had challenges ahead [in Australia], but I had great confidence.”

9.2.4 Psychological distress and well-being

Research question 4: How do migrants, when classified by their I-positions pertaining to country of origin and Australia, vary in psychological distress and well- being? As part of Study 1, participants were scored for psychological distress and psychological well-being (not to be confused with well-being as computed from their self-confrontation indices). Psychological distress was measured using three scales:

SAD, FNE and BDI-II. On the SAD scale, Progressive migrants were in general less socially anxious than the other types, though the difference from Regressive migrants was of borderline significance, probably due to the smallness of the sample. On the

FNE scale, Low-stable migrants were more fearful of negative evaluation than the

High-stable. Again, the smallness of the sample may have prevented significant differences being found where otherwise they would have been expected: between, say,

Progressive and Regressive migrants. Lastly, on the BDI-II scale, no significant differences were found between migrant types on depression.

Participants also provided self-evaluations of psychological well-being using

Ryff’s Psychological Well-being Scale. Significant differences were found in only one of the six subscales: Positive relations with others. Regressive migrants had significantly worse relations with others than did Progressive and High-stable migrants.

Low-stable migrants on average rated above Regressive, and below Progressive and

High-stable migrants, but the differences were not significant.

259

In summary, migrants varied in their psychological states in a plausible and consistent way. In general, Progressive and High-stable migrants were lowest on distress and highest on well-being; Regressive migrants were the opposite; and the

Low-stable migrants were intermediate. Where no significant differences were discerned, it is probably explained by the smallness of the sample.

Work by Searle and Ward (1990; Ward & Kennedy, 1992) on the prediction of psychological and socio-cultural adjustment during cross-cultural transitions may help to explain these findings. To recap, socio-cultural adaptation (Ward & Searle, 1990) concerns the relationship between migrants and their host society, and it includes their ability to cope with the everyday issues that require social interactions. On the other hand, psychological adaptation includes intra-psychic aspects such as feelings of well- being and emotional satisfaction. In the present research, psychological adaptation was assessed by means of measures of depression and well-being, and socio-cultural adaptation was assessed by means of measures of social adjustment (the SAD scale) and sociocultural adaptation (the FNE scale). The SAD scale is, by definition, a measure of discomfort in, or about, interpersonal situations, and the FNE Scale is defined as

“apprehension about other's evaluation, distress over their negative evaluation, and the expectation that others would evaluate one negatively”, and the authors suggested that

“fear of loss of social approval would be identical to FNE” (Watson & Friend, 1969, p.449). That Regressive migrants scored highest on both measures suggests, therefore, that social relationships with host nationals (and perhaps co-nationals as well) may have reduced their sociocultural and psychological adaptation. This is corroborated by their significantly lower scores on the Positive relations with others subscale of the psychological well-being measure.

260

Ward and Searle (1991) and Ward and Kennedy (1994) support these findings.

They found that migrants who identify strongly with the culture of their country of origin tend to be psychologically well adjusted; and those who identify strongly with their host country tend to be socio-culturally well adjusted. Further support of this came from Ward and Rana-Deuba (1999), who found that identification with culture of origin is the most salient factor in the migrants’ psychological well-being and is associated with a decrease in depressive symptoms. They found that those who identify strongly with their origin culture experience most social difficulty in the host culture, which impedes sociocultural adaptation. Though identification with one’s origin culture is generally beneficial and is not a problem for socio-cultural adjustment, it becomes one when identification with the host culture is weak, as it may impede culturally appropriate behaviours and attitudes. Ward and Kennedy (1994) found that the greatest social difficulty was experienced by those who distanced themselves from the host society in a strategy of “separation”.

The notion of different preferences according to whether the domain is public

(functional or utilitarian) or private (emotional, value-oriented) may also explain the findings for the Regressive migrants. Their preference for sociocultural adaptation and cultural maintenance may vary across life domains, so that while emphasising the need for both native and host culture identities, their host culture identity integration is expressed in the public domain, where they may seek economic or work integration, while in their private domain they prefer to adopt and maintain a separation strategy.

Such a pattern of cultural alternation has been found to be most adaptive. For instance, in an investigation of how migrants adapted culturally in different life domains, Arends-

Toth and van de Vijver (2003) found that Turkish migrants in the Netherlands adapted more to Dutch culture in the public domain than in the private, but that cultural

261 maintenance was important in both—that is, Turkish culture was preferred at home and

Dutch elsewhere. In Berry’s terminology, these migrants favoured separation in the private domain but integration in the public domain. Likewise, Phalet and Andriessen

(cited in van de Vijver & Phalet, 2004) found that the acculturation profile of successful minority students appeared to alternate between separation in the family and integration at school.

In the present thesis, Regressive migrants were the least adapted in terms of psychological distress and well-being measures, as described earlier. Indeed, the importance of host society social networks has been noted by others. For example,

LaFromboise et al. (1993) proposed that successful bicultural (Integrated) migrants are those who have established a stable social network with host nationals in the new culture. They argued that it was the attainment of this social network that constituted an important outcome of second-culture acquisition and one that enhanced the migrant’s ability to cope with life in a bicultural environment. This interpretation may explain the

Regressives’ significantly lower score on the Positive relations with others subscale of the psychological well-being measure.

The literature regarding the importance of social support for psychological well- being is ambiguous as to the source of social support: some researchers emphasise good relationships with host nationals (Furnham & Bochner, 1986), others emphasise good relationships with co-nationals (Searle &Ward, 1990). Also, while the sample in some of this research was limited to a particular type of migrant such as sojourners (Ward &

Kennedy, 1994), who reside in a host society for a relatively brief period, the present sample consists exclusively of people who came to Australia intending to stay indefinitely. Such migrants must be concerned with their relationship to the host society

(Taft, 1977), and the present finding that at least some Integrated migrants who, despite

262 having lived in Australia for years still have difficulty in social adaptation, indicates that attaining social belongingness in a non-native culture is not automatic. There seems to be more to integration and adjustment than meets the eye, with different levels of adaptation within the one strategy. More research is needed to elucidate this.

The literature suggests that a strategy of integration is associated with fewest psychological difficulties (Berry & Sam, 1997; Ward & Kennedy, 1994). The present findings support the view that integration is insufficient for optimal adaptation, at least for some migrants. In contrast, migrants who had positive relationships with host nationals and felt accepted by them (Progressive and High-stable) were well adapted both socio-culturally and psychologically. Their positive attitudes towards interactions with host nationals may have helped them to receive social support and hence cope more effectively. The present findings agree with Arends-Toth and van de Vijver’s

(2004) notion that within integration—the combination of adaptation and cultural maintenance—there are many variations in how it is managed and negotiated between migrants and the host society as a result of the various possible combinations of cultures. It points to the need of having the term “integration” more precisely defined in further research and theory. As Arends-Toth and van de Vijver’s (2006) note, it is plausible that given migrants’ massive preference for integration, most variations occur within integration and not between acculturation orientations.

9.2.5 Idiographic Support

Research question 5: Are differences between individuals classified by their I- positions reflected in the life stories of specific case studies? Good matches were found between the case studies and the participant types they represent, and in this way Study

3 corroborated Studies 1 and 2. A single case study was chosen to represent each participant type, four in all, and the content of their valuations described. Figure 9.2

263 displays the patterns relating to each participant type. Yoana’s self-narrative is of setting new and different goals, and achieving them—truly a progression to better things. She exemplifies the Progressive migrant. Juan’s self-narrative is of mental struggle and survival—stable, but compromised by underachievement. He exemplifies the Low- stable migrant. Luisa’s self-narrative of building on a strong sense of herself and achieving more—stability crowned by emotional reward. She exemplifies the High- stable migrant. Carla’s self-narrative is of distancing from all she holds dear. She clings to a receding past, and when finally she revisits it, mirage-like it vanishes. This is a story of regression from a desired state. She exemplifies the Regressive migrant.

264

Type 1 (Progressive) Type 2 (Low-stable)

+HH +HH

+S +O +S +O

-S -O -S -O

Origin Origin Australia -LL Australia -LL

Type 3 (High-stable) Type 4 (Regressive)

+HH +HH

+S +O +S +O

-S -O -S -O

Origin Origin Australia Australia -LL -LL

Figure 9.2 Proportion of valuations, by major theme, location, and participant type

Note: Arrows represent direction and indicative magnitude of significant differences in the composite I-position of all participants of each specified type.

Gergen and Gergen’s (1988) narrative forms are evident in participants’ self- narratives. Their narrative forms echoed those in the literary tradition of the Western world, with Frye (1957) noting that positively toned stories often take the form of

“comedy” or “romance”—where the protagonist succeeds; and negatively toned stories often take the form of “tragedy” or “irony”—where the protagonist fails. Gergen and

Gergen establish the narrative plot and, within it, which incidents should be construed

265 as advances, retreats or digressions. They distinguish progressive, regressive, and stability narratives. Progressive narratives begin with negative affective evaluations of events but proceed over time to positive ones, such as the well-being and success of the protagonist. Regressive narratives begin with positive affective evaluations that become progressively more negative, ending with failure. Stability narratives may be either positive or negative in affective tone but do not change evaluative valence over time.

Any life story may draw on these evaluative trajectories, and so provide thematic coherence to a life: “My Life has been all uphill” or “My life has been a series of ups and downs” (Habermas & Bluck, 2000).

As noted earlier, in a redemption sequence the narrator relates how bad turns to good. Although there are likely to be multiple outcomes of given life event, the redemptive narrator dwells on the positive—as do Progressive migrants. By contrast, in a contamination sequence the narrator dwells on the negative consequences of a positive event—as do Regressive migrants. In both cases, the narrator chooses what to relate and what to omit. This has implications for how self-narratives are to be interpreted and why they are so revealing. For if it is accepted that in their self-narratives people are attempting to provide an accurate accounting of events (Gergen & Gergen, 1988), then individual differences in narrative style and content reflect psychosocial adaptation as much as what actually happened.

9.3 Directions for Future Research

The previous section showed how the findings of the present thesis addressed the research questions that were set for it, and how they did so in the context of existing research. This section looks towards future directions for research emanating from the present findings. First it critically examines the theoretical and methodological strengths and limitations of the three studies that make up the thesis. Then it suggests future

266 directions for research designed to exploit the identified strengths and to overcome the weaknesses.

9.3.1 Strengths and Limitations

Although its antecedents go back a century or more, dialogical self theory came to prominence with Hermans’ work published in the 1990s. It is now one of several leading personality theories. While not dominating or supplanting, other theories, it does offer additional insights, particularly in the field of cultural psychology.

Until recently Berry’s model of acculturation was arguably the most widely held.

However, globalisation has since given rise to cultural mixing on an unprecedented scale. No longer do migrants leave their countries of origin expecting never to return.

Instead they may maintain contact with frequent visits and even more frequent telecommunication. The result has been cultural hybridisation and a range of related issues that all demand an explanatory model. These include power imbalances between minority and majority cultures, and strategies for mitigating the uncertainty that results.

Acculturation theories such as Berry’s, that see cultures as essentially dichotomous, have been criticised as unable to account adequately for these phenomena.

Dialogical self theory, which accommodates the many paths by which individuals can negotiate their cultural identities, is more accommodating. As both a cause and a consequence of its wide acceptance, dialogical self theory has been applied by numerous practitioners and researchers to a range of circumstances besides acculturation. Nevertheless, work remains to be done on testing and refining the theory.

The present thesis contributes to that goal by contributing theoretical insights and by strengthening the empirical evidence base of the theory of the dialogical self.

The dialogical self as a frequency distribution of valuations. In representing a person’s I-positions as a frequency distribution of valuations categorised by major

267 theme, this thesis introduces a new way of describing the dialogical self. The modes of this distribution correspond to frequently occurring self-narratives that are generically similar. In fact, much of the empirical work in this thesis is interpretable as estimating modal themes from collections of self-narratives. A frequency distribution does not capture the full richness of the dialogical self nor the I-positions that compose it, since the theory concerns not just I-positions but the relationships between them; but it does offer valuable conceptual insights for exploring the changes in the dialogical self that arise when people encounter changed social environments. Moreover, once it is accepted that valuations are in effect ‘sampled’ from an underlying population that represents the self, then the appropriate techniques for the statistical analysis of data pertaining to valuations become clear, namely those of sampling theory.

Filling a research need. Several authors have acknowledged the lack of empirically based investigation into dialogical self theory (Lyddon et al., 2006;

Hermans and Dimaggio, 2007; Hermans, 2008; Stemplewska-Żakowicz et al., 2005).

One reason for this lack may simply be that the theory is still quite recent (Hermans,

2006). Though foreshadowed in Hermans’ writings (Hermans, 1987a, 1987b, 1988,

1989, and 1992), the term “dialogical self” was introduced as late as 1992 (Hermans,

Kempen & Van Loon, 1992). Another may be the very richness of the theory, which transcends the traditional confines and challenges the core assumptions of narrative psychology in a quite radical way (Hermans, 2006):

Whereas narrative psychology typically assumes the existence of one more or less integrated self-narrative, dialogical psychology assumes that dialogue is able to surpass any existing narrative. This implies that dialogue may disrupt or undermine any existing narrative and that the shift toward another position may lead to a different story, even about the “same” events. (pp. 22–23)

268

This research has helped to fill this research gap in two ways. First, it used a larger sample of valuations than did earlier studies (see Table 5.8), and so permits a more sophisticated quantitative analysis. Second, its sample is demographically more similar to its majority culture (see Tables 5.8 and 5.9) than are the samples used by the other studies to which it is compared, which tend to be drawn from students or clinical patients. Student samples assist in the development of empirical support, and clinical samples are relevant to the SCM as a therapeutic tool. However, the dialogical self is a general theory, so it is of value to test it with a general sample as well, as it reduces the risk that differences in I-position repertoires could be due to demographic differences and not cultural ones.

This thesis draws its data from a sample of participants who are neither university students nor in therapy. This study fills a gap in the literature by focusing on nonclinical participants. Moreover it considered approximately 1600 valuations, which may be the largest such set of data anywhere in the literature. Lastly, most studies of the dialogical self to date have been case studies with an idiographic approach (Lyddon,

Yowell, & Hermans, 2006; Dimaggio, Salvatore, Azzara & Catania, 2003; Hermans,

1996b; Hermans & Hermans-Jansen, (1995); Hermans, 1987). Valuable as these are, they do not (nor are intended to) generate datasets suitable for the kind of statistical analysis conducted for this thesis.

9.3.2 Scope for Further Work

Generalising the findings. Although the findings were limited by the relative homogeneity of the sample, this is not necessarily a limitation. The aim of the present thesis was not to construct a universal model of migration but to test specific hypotheses that did not of themselves demand a representative (hence large and diverse) sample; and for that purpose the sample was statistically adequate. It is nevertheless true that the

269 sample was insufficient to allow statistical hypothesis testing where cross-tabulation was required, which prevented inference about specific subgroups of migrants. Even were this possible, the scope for generalisation would have been limited by the narrowness of the participants’ demographics. Most were middle-class, in mid-life, and well educated, and many were from middle-income countries like Mexico and

Malaysia. This would have introduced a considerable bias, as according to a meta- analysis by Moyerman and Forman (1992), the socio-economic status of the migrants had the strongest and most consistent influence on the acculturation process, to the extent that they considered it a protective factor for acculturation. As for how education is related to acculturation, Phinney and Flores (2002) note that many migrants migrate in order to improve their social status; education is one means of achieving this. Hence higher levels of education are more likely to expose migrants to mainstream values and ideas and to lead to jobs in mainstream settings. Suinn, Ahuna, and Khoo (1992) demonstrated that higher levels of education boost host culture identification. So, had the present sample included blue-collar and older migrants, or those from poor backgrounds in low-income countries, the findings might have been different. Another causal factor is the voluntariness of migration. The present study restricted the sample to voluntary migrants. But it is impossible a priori to hold voluntariness completely constant. Some “voluntary” migrants—for instance spouses and, more so, adolescents— have little or no say in the matter, as they migrate as part of a family group: Carla, discussed previously as a case study, is an example.

The role of such variables is a promising area for further application of dialogical self theory. Research could be conducted with different types of migrants from different countries of origin and destination to explore the differences between migrant groups and countries.

270

Study design: longitudinal vs cross-sectional. The present research analyses a cross-section of individuals observed at a single point in time. This implies that temporal variation cannot be used to ascertain directions of causality. Roughly half of the valuations pertain to events that occurred in the past, often the distant past: such data may be termed quasi-longitudinal. Because lifespan development is not controlled for, little is known as to why I-positions differ as they do, nor why some migrants are stable and others not.

A longitudinal study would help to deepen the understanding of this aspect of the migration experience. Intending migrants could be interviewed in terms of their current sense of their dialogical self before they left, and then interviewed again some years after arrival in Australia. This would provide clearer idea of the change in, and possible integration of, their I-positions. Currently it is known where migrants stand in terms their I-positions, but not how they got there and why; that must be inferred.

Implications for policy. The findings of the thesis may furnish practical insights into the specific field of migration policy. According to König (2009), there is a pressing and growing need, and a dialogical approach is a promising way to address it.

In a study of migrants to the Netherlands, she found that “acculturation is facilitated if a dialogue is encouraged between personal cultural positions about how and why individuals make and define their personal cultural choices, and these decisions are explored in an inter- and/or intra-personal dialogue.” Instead, however, “most integration programmes for asylum seekers in Holland submerge the fresh immigrant under a layer of knowledge about the new environment in a classroom setting” (p. 117).

Dialogical self theory can in principle contribute in two ways: clinical and psychoeducational. But first, research is needed to enable interventions to be targeted to those who would benefit. For while this thesis has established that migrants differ in

271 their capacity to prosper in Australia, there is as yet no way to predict the path that each intending or recent migrant will take.

In some cases migrants’ evaluations of their country of origin is a reliable guide.

For instance, Low-stable migrants (see Figure 9.1) tend to harbour negative affective evaluations of their country of origin to a degree that distinguishes them from other types of migrant; and furthermore they tend not to prosper psychologically in Australia.

So on that basis it might be possible to target them for early intervention. But other migrants are harder to target. For instance, Regressive and High-stable migrants share positive affective evaluations of their country of origin, but the High-stable migrants prosper in Australia while the Regressive migrants do not (see Figure 9.1). Diagnostic procedures are needed to distinguish these groups.

Dialogical self theory, possibly through the application of content analysis of self- narratives, is a promising candidate, especially given its longstanding therapeutic role in helping individuals integrate negative experiences into their life story. In this it could be complemented by self-report measures of depression and anxiety, since these were found to be more prevalent among the target migrant types. Resources did not permit the present thesis to go into this level of detail, but the database of evaluations is available in electronic form for future researchers.

9.4 Conclusions

The research conducted in this thesis used Hermans’ theory of the dialogical self and the self-confrontation method to study the effect of migration on the self. A sample of 38 migrants to Australia was interviewed in depth and the data analysed in several ways. Specifically, it was established that (1) those migrants that see themselves as having two perspectives (I-positions of country of origin and Australia) do actually differ in the contents in their perspectives; (2) the differences between the two

272 perspectives, while significant, is not large; and (3) such differences as were observed are characterised by meaningful patterns that are related to wellbeing and adjustment.

Overall, the research found evidence for Hermans’ dialogical view of the self and against the hypothesis of an underlying integrated self.

Study 1 found that most migrants showed greater self-enhancement (S) in their

Australian than in their country of origin I-positions, whereas union-with-the-other (O) did not differ significantly, suggesting that migration accommodates self-enhancement

(S) proportionately more than union-with-the-other (O). Migrants furthermore were found to vary in their psychological states in a plausible and consistent way. Those who showed greater self-enhancement (S) in their Australian than in their country of origin I- positions were also lowest on distress and highest on well-being. Study 3 corroborated the findings of Study 1 in a set of case studies.

Study 2 found there to be overall stability in migrants’ motives and affect, and where differences occurred, they were intelligible. The research corroborates previous research that identifies two modes of biculturalism: compatible and oppositional: most migrants viewed their heritage and host cultures as complementary, whereas a minority viewed them as discrepant and conflicting.

Besides these findings, the present research has advanced the methodological basis of analysing data generated by the self-confrontation method (SCM). Much previous research using SCM data has taken a case-study approach, and has been largely qualitative in approach. The present thesis complements this with its largely quantitative emphasis, and contributes to filling an acknowledged research gap.

273

Glossary

Valuation A unit of meaning from an individual’s self-expressed life-narrative (Hermans and Hermans-Jansen, 1995, p.15).

“[A]nything that a person finds to be of importance when thinking about his or her life situation” (Hermans, 1988, p.792).

Self-narrative An individual’s account of his or her life containing all that individual’s valuations.

Valuation system People organise their valuations into self-narratives that situate them in time and space—a valuation system. The valuation system is a composite of changing valuations.

Self-confrontation Four indices (S, O, P, N) that are attached to valuations by means of the indices self-confrontation method (SCM).

Self-confrontation The prescribed interviewing procedure for eliciting self-confrontation method (SCM) indices (S, O, P, N) pertaining to valuations.

Theme The vector of self-confrontation indices (S, O, P, N) pertaining to a valuation, as measured by the self-confrontation method (SCM).

Major theme Each of six categories into which many (but not all) themes can be placed by means of a set of computable rules. Major themes are believed to be of psychological significance.

“[G]eneral types of valuation, which reflect the organizational structure of the person’s valuation system… can be conceived of as major themes…” (Hermans and Hermans-Jansen, 1995, p.73)

I-position The hypothetical author of a valuation. It will usually be the case that a single I-position is the author of many valuations. Typically these valuations will arise from similar contexts.

Dialogical self The dialogical self arises when the set of valuations has many I- positions.

274

Distribution of The frequency distribution of themes from the full set of self-narratives themes an individual might express.

Modal theme The modes of an individual’s distribution of themes.

275

References

World Migration in Figures A joint contribution by UN-DESA and the OECD to the United Nations High-Level Dialogue on Migration and Development, 3-4 October 2013

Adams G., & Markus H. R. (2001). Culture as patterns: An alternative approach to the problem of reification. Culture & Psychology 7, 283–296.

Aldenderfer, M. S., & Blashfield, R. K. (1984). Cluster analysis. Newbury Park, CA: Sage Publishing.

American Psychiatric Association (1994). Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition (DSM IV). Washington DC: Author.

Anderberg, M. R. (1973). Cluster analysis for applications. New York: Academic Press.

Andreouli, E. (2013). Identity and acculturation: The case of naturalised citizens in Britain. Culture & Psychology, 19, 165–183.

Arends-Toth, J., & van de Vijver, F. J. R. (2003). Multiculturalism and acculturation: Views of Dutch and Turkish-Dutch. European Journal of Social Psychology, 33, 249–266.

Arends-Toth, J., & van de Vijver, F. J. R. (2004). Domains and dimensions in acculturation: Implicit theories of Turkish-Dutch. International Journal of Intercultural Relations, 28, 19–35.

Argyle, M. (1982). Intercultural communication. In S. Bochner (Ed.) Cultures in contact: Studies in cross-cultural interaction (pp. 61–80). Oxford: Pergamon.

Arkowitz, H., Lichtenstein, E., McGovern, K., & Hines, P. (1975). The behavioural assessment of social competence in males. Behaviour Therapy, 6, 3–13.

Arnett, J. J. (2002). The psychology of globalization. American Psychologist, 57(10), 774–783.

Atlas, G. D. (1994). Sensitivity to criticism: A new measure of responses to everyday criticisms. Journal of Psychoeductional Assessment, 12, 241–253.

Australian Bureau of Statistics (2013, December). Migration, Australia (no. 3412.0). Retrieved from http//:www.abs.gov.au/

Aveling, E., & Gillespie, A. (2008). Negotiating multiplicity: Adaptive asymmetries within second-generation Turks’ “society of mind”. Journal of Constructivist Psychology, 21(3), 200–222.

2 76

Bakan, D. (1966). The duality of human existence. Chicago: Rand-McNally.

Bakhtin, M. (1973). Problems of Dostoevsky’s poetics (2nd ed.) (Translated by R. W. Rotsel.) Ann Arbor, MI: Ardis. (Original work published 1929).

Barcinski, M., & Kalia, V. (2005). Extending the boundaries of the dialogical self: Speaking from within the feminist perspective. Culture & Psychology, Vol. 11(1), 101–109.

Blasfield, R. K., & Aldenderfer, M. S. (1988). The methods and problems of cluster analysis. In J. R. Nesselroade & R. B. Cattell (Eds.), Handbook of multivariate experimental psychology (pp. 447–473). New York: Plenum.

Barresi, J. (2002). From `the thought is the thinker' to `the voice is the speaker': William James and the dialogical self. Theory & Psychology 12, 237–250.

Barresi, J. (2011). Time and the dialogical self. In H. J. M. Hermans & T. Gieser (Eds.), Handbook of dialogical self theory (pp. 46-62). Cambridge: Cambridge University Press.

Beauchaine, T. P., & Beauchaine, R. J. III (2002). A comparison of maximum covariance and k-means cluster analysis in classifying cases into known taxon groups. Psychological Methods, 7, 245–261.

Beck, A. T., & Steer, R. A. (1993). Manual for the Beck Depression Inventory, San Antonio, Texas: The Psychological Corporation. Hartcort Brace & Company.

Beck, A. T., Steer, R. A., & Brown, G. K. (1996). Manual for the Beck Depression Inventory-II. San Antonio, Texas: The Psychological Corporation. Harcourt Brace & Company.

Bell, N. J., & Das, A. (2011). Emergent organization in the dialogical self: Evolution of a ''both'' ethnic identity position. Culture & Psychology 17, 241– 262.

Bem, D. J. (1972). Self-perception theory. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 6, pp. 1-62). New York: Academic Press.

Benet-Martinez, V., & Haritatos, J. (2005). Bicultural identity integration (BII): Components and psychological antecedents. Journal of Personality, 73, 1015–1050.

Benet-Martinez, V., Lee, F., & Leu, J. (2006). Biculturalism and cognitive complexity: Expertise in cultural representations. Journal of Cross-Cultural Psychology, 37, 386–407.

277

Benet-Martinez, V., Leu, J., Lee, V., & Morris, M. (2002). Negotiating biculturalism: Cultural frame switching in biculturals with oppositional versus compatible cultural identities. Journal of Cross-Cultural Psychology, 33, 492- 516.

Berger, P. (1966). Identity as a problem in the sociology of knowledge. European Journal of Sociology, 7, 105–115.

Berger, P. L. (1963). Invitation to sociology. Garden City, NY: Doubleday.

Berry, J. W. (1980). Acculturation as varieties of adaptation. In A. M. Padilla (Ed.), Acculturation: Theory, models and some new findings (pp. 9–25). Boulder CO: Westview.

Berry, J. W. (1984). Cultural relations in plural societies: Alternatives to segregation and their sociopsychological implications. In N. Miller & M. Brewer (Eds.), Groups in contact. New York: Academic Press.

Berry, J. W. ( 1985). Psychological adaptation of foreign students in Canada. In R. Samuda, & A. Wolfgang (Eds.), Intercultural counselling (pp. 235–248). Toronto: Hogreffe.

Berry, J. W. (1990). Psychology of acculturation. In J. J. Berman (Ed.), Nebraska Symposium on Motivation, 1989: Vol. 37, Cross-cultural perspectives (pp. 201–234). Lincoln: University of Nebraska.

Berry, J. W. (1992). Acculturation and adaptation in a new society. International Migration, 30, 69–85.

Berry, J. W. (1997), Immigration, acculturation, and adaptation. Applied Psychology: An International Review, 46, 5–68.

Berry, J. W. (2001). A psychology of immigration. Journal of Social Issues, 57(3), 615–631.

Berry, J. W. (2002). Acculturation and intercultural relations. In J. W. Berry, Y. H. Poortinga, M. H. Segall, & P .R. Dasen (Eds.), Cross-cultural psychology: Research and applications, 2nd edition, (pp. 344-383). Cambridge: Cambridge University Press.

Berry, J. W. (2003). Conceptual approaches to acculturation. In K.M. Chun, P. Balls Organista, G. Marin (Eds.), Acculturation: Advances in theory, measurement, and applied research. Washington, D.C.: American Psychological Association.

Berry, J. W. (2005). Acculturation: Living successfully in two cultures. International Journal of Intercultural Relations, 29, 697–712.

278

Berry, J. W., & Annis, R. C. (1974). Acculturative stress: The role of ecology, culture and differentiation. Journal of Cross-Cultural Psychology, 5, 382– 406.

Berry, J. W., Kim, U., Minde, T., & Mok, D. (1987). Comparative studies of acculturative stress. International Migration Review, 21, 491–511.

Berry, J. W., Kim, U., Power, S., Young, M., & Bujaki, M. (1989). Acculturation attitudes in plural societies. Applied Psychology: An International Review, 38, 185–206.

Berry, J. W., Poortinga, Y. H., Segall, M. H., & Dasen, P. R. (1992). Cross- cultural psychology: Research and applications (2nd Ed.). New York: Cambridge University Press.

Berry, J. W., & Sam, D. (1997). Acculturation and adaptation. In J. W. Berry, M. H. Seagull, & C. Kagitcibasi (Eds.), Handbook of cross-cultural psychology: Social behaviour and applications (Vol. 3, pp. 291–326). Needham Heights, MA: Allyn and Bacon.

Berry, J. W., Trimble, J., & Olmedo, E. (1986). Assessment of acculturation. In W. Lonner & J. Berry (Eds.), Field methods in cross-cultural research (pp. 291– 334). Newbury Park: CA: Sage.

Betancourt, H.,& Lopez, S. R. (1993). The study of culture, ethnicity, and race in American psychology. American Psychologist, 48, 629–637.

Betts, K. (1996). Explaining Australian immigration. Journal of the Australian Population Association, 13(2), 195–229.

Bhatia, S. (2002). Acculturation, dialogical voices, and the construction of the diasporic self. Theory & Psychology, 12, 55–77.

Bhatia, S. (2003). Is ‘integration’ the developmental end goal for all immigrants? Redefining ‘acculturation strategies’ from a genetic-dramatistic perspective. In I. Josephs (Ed.), Dialogicality in development (pp. 197-216). Westport: Greenwood Publishing Group.

Bhatia, S. (2007). American Karma. New York: New York University Press.

Bhatia, S. (2014). Acculturation and dialogical formation of immigrant identity: Race and culture in diaspora spaces. In Hermans, H. J. M. & Gieser, T. (Eds.), Handbook of dialogical self theory (pp. 115–131). Cambridge, UK: Cambridge University Press.

Bhatia, S., & Ram, A. (2001a). Rethinking ‘acculturation’ in relation to diasporic cultures and postcolonial identities. Human Development, 44, 1–18.

279

Bhatia, S., & Ram, A. (2001b). Locating the dialogical self in the age of transnational migration, border crossings, and diasporas. Culture & Psychology, 7, 297–309.

Bhatia, S., & Ram, A. (2004). Culture, hybridity, and the dialogical self: Cases from the South Asian diaspora. Mind, Culture, and Activity, 1(3), 224–240.

Bhatia, S., & Ram, A. (2009). Theorizing identity in transnational and diaspora cultures: A critical approach to acculturation. International Journal of Intercultural Relations 33, 140–149.

Bhugra, D. (2004). Migration, distress, and cultural identity. British Medical Bulletin, 69, 129–141.

Bhugra, D., Bhui, K., Mallet, R., Desai, M., Singh, J., & Leff, J. (1999). Cultural identity and its measurement: A questionnaire for Asians. International Review of Psychiatry, 11, 244–249.

Birman, D. (1994). Acculturation and human diversity in a multicultural society. In E. J. Trickett, R. J. Watts, & D. Birman (Eds.), Human diversity: Perspectives on people in context (pp. 115–141). San Francisco, CA: Jossey-Bass.

Birman, D., & Trickett, E. J. (2001). Cultural transitions in first-generation immigrants: Acculturation of Soviet Jewish refugee adolescents and parents. Journal of Cross-Cultural Psychology, 32(4), 456–477.

Bochner, S. (1982). The social psychology of cross-cultural relations. In S. Bochner (Ed.), Cultures in contact: Studies in cross-cultural interaction (pp. 5–44). Oxford: Pergamon.

Bochner, S. (1994). Cross-cultural differences in the self-concept: A test of Hofstede’s individualism/collectivism distinction. Journal of Cross-cultural Psychology, 25, 273–283.

Bochner, S., & Hesketh, B. (1994). Power distance, individualism/collectivism, and job-related attitudes in a culturally diverse work group. Journal of Cross- Cultural Psychology, 25, 233–257.

Bonanno, G. A. (2004). Loss, trauma, and human resilience: Have we understood the human capacity to thrive after extremely adverse events? American Psychologist, 59, 20–28.

Boneva, B. S., & Frieze, I. H. (2001). Toward a concept of a migrant personality. Journal of Social Issues, 57, 477–491.

Boneva, B. S., Frieze, I. H., Ferligoj, A., Jarosova, E., Pauknerova, D., & Orgocka, A. (1998). Achievement, power, and affiliation motives as clues to (e)migration desires. A four-countries comparison. European psychologist, 3(4), 247–254.

280

Boski, P. (2008). Five meanings of integration in acculturation research. International Journal of Intercultural Relations, 32, 142–153.

Breakwell, G. M. (1983). Threatened identities. New York: John Wiley & Sons.

Brewer, M. B., & Gardner, W. (1996). Who is this “we”? Levels of collective identity and self-representations. Journal of Social and Personality Psychology, 71, 83–93.

Brown, J. D. (1998). The self. Boston: McGraw Hill.

Bruner, J. (1990). Acts of meaning: Four lectures on mind and culture. Cambridge, MA: Harvard University Press.

Buriel, R., Calzada, S., & Vasquez, R. (1982). Relationship of traditional Mexican American culture to adjustment and delinquency among three generations of Mexican-American male adolescents. Hispanic Journal of Behavioral Sciences, 1, 41–55.

Camilleri, C. & Malewska-Peyre, H. (1997) Socialization and identity strategies. In Berry, J.W., Dasen, P.R. and Saraswathi, T.S. (Eds.), Handbook of Cross- Cultural Psychology, Vol. 2: Basic Processes and Human Development (pp. 41–68). London: Allyn & Bacon.

Castles, S., Iredale, R., & Vasta, E. (1992). Australian immigration between globalization and recession. The International Migration Review, 28(2), 370– 384.

Castles, S., & Vasta, E. (2000). Ethnicity in Australian society. In J. Najman & J. Western (Eds.), A sociology of Australian society (3rd ed.). South Melbourne: McMillan Co. of Australia.

Caughey, J. L. (1980). Personal identity and social organization. Ethos, 8, 173– 203.

Chandler, M. J., & Lalonde. C. (1998). Cultural continuity as a hedge against suicide in Canada's first nations. Transcultural Psychiatry, 35(2): 191–219.

Cheek, J. M. (1989). Identity orientations and self-interpretation. In D. M. Buss & N. Cantor (Eds.), Personality psychology: Recent trend and emerging directions (pp. 275–285). New York: Springer-Verlag.

Christopher, J. C., & Bickhard, M. H. (2007). Culture, self and identity: Interactivist contributions to a metatheory for cultural psychology. Culture & Psychology 13, 259–295.

Church, A. T. (2000). Culture and personality: Toward an integrated cultural trait psychology. Journal of Personality, 68(4), 651–703.

281

Clatworthy, J., Buick, D., Hankins, M., Weinman, J., & Horne, R. (2005). The use and reporting of cluster analysis in health psychology: A review. British Journal of Health Psychology, 10, 329–358.

Cohler, B J. (1982). Personal narrative and life course. In P.B. Baltes & O.G. Brim, Jr. (Eds.), Life span development and behavior (Vol. 44, pp. 205–241). New York: Academic Press.

Cooley, C. H. (1902). Human nature and the social order. New York: Scribner.

Corcoran, K., & Fischer, J. (2000). Measures for clinical practice: a sourcebook. Vol 2. Adults (3rd ed.). New York: The Free Press.

Crites, S. (1971). The narrative quality of experience. Journal of the American Academy of Religion, 39(3), 291-311.

Damon, W., & Hart, D. (1988). Self-understanding in childhood and adolescence. New York: Cambridge University Press.

Deaux, K. (1993). Reconstructing social identity. Personality and Social Psychology Bulletin, 19(1), 4–12.

Deaux, K. (2000). Surveying the landscape of immigration: Social psychological perspectives, Journal of Community and Applied Social Psychology. 10, 421– 431.

Deaux, K., Reid, A., Mizrahi, K., & Ethier, K.A. (1995). Parameters of social identity. Journal of Personality and Social Psychology, 68, 280–291.

Deza, M. M., & Deza, E. (2009). Encyclopaedia of distances. Berlin: Springer- Verlag.

Diehl, M., Hastings, C. T., & Stanton, J. M. (2001). Self-concept differentiation across the adult life span. Psychology and Aging, 16, 643–654.

Dimaggio, G., Salvatore, G., Azzara, C., & Catania, D. (2003). Rewriting self- narratives: The therapeutic process. Journal of Constructivist Psychology, 16, 155–181.

Dona, G., & Berry, J. W. (1994). Acculturation attitudes and acculturative stress of Central American refugees. International Journal of Psychology, 29, 57–70.

Dos Santos, M. A., Gauer, G., & Gomes, W. B. (2010). Dialogical valence: A novel measure for the dialogical self and it implications for psychotherapy. International Journal for Dialogical Science, 4(2), 105-117.

Dozois, D. J. A., Dobson, K. S., & Ahnberg, J. L. (1998). A psychometric evaluation of the Beck Depression Inventory-II. Psychological Assessment, 10, 83–89.

282

Durm, M. W., & Glaze, P. E. (2001). Construct validity for self-acceptance and fear of negative evaluation. Psychological Reports, 89, 386.

Elias, N., & Blanton, J. (1987). Dimensions of ethnic identity in Israeli Jewish families living in the United States. Psychological Reports, 60, 367–375.

Erikson, E. H. (1950). Childhood and society. London: Vintage [1995].

Erikson, E. H. (1956). The problem of ego identity. Journal of the American Psychoanalytic Association, 4, 56–121.

Erikson, E. H. (1968). Identity, youth and crisis. New York: Norton.

Falmagne, R. J. (2004). On the constitution of ‘self’ and ‘mind’: The dialectic of the system and the person. Theory & Psychology, 14(6), 822–845.

Freud, S. (1900). The interpretation of dreams. London: Allen & Unwin [1915].

Friedman, T. L. (2005). The world is flat: A brief history of the twenty-first century. New York: Farrar, Straus and Giroux

Friend, R., & Gilbert, J. (1973). Threat and fear of negative evaluation as determinants of locus of social comparison. Journal of Personality, 41, 328– 340.

Frieze, I. H., Boneva, B. S., Sarlija, N., Horvat, J., Ferligoj, A., Kogovšek, T., Miluska, J., Popova, L., Korobanova, J., Sukhareva, N., Erokhina, L., & Jarosova, E. (2004). Psychological differences in stayers and leavers: Emigration desires in Central and Eastern European university students. European Psychologist, 9(1), 15–23.

Furnham, A., & Bochner, S. (1986). Culture shock: Psychological reactions to unfamiliar environment. London: Methuen.

Garza-Guerrero, A. C. (1974). Culture shock: Its mourning and the vicissitudes of identity. Journal of the American Psychoanalytic Association, 22, 408–429.

Gecas, V. (1973). Self-conceptions of migrant and settled Mexican Americans. Social Science Quarterly, 54, 579-595

Gecas, V. (1982). The self-concept. Annual Review of Sociology, 8, 1-33.

Gecas, V., & Schwalbe, M. L. (1983). Beyond the looking glass self: Social structure and efficacy-based self-esteem. Social Psychology Quarterly, 46(2), 77-88.

Gergen, K .J. (1968). Personal consistency and the presentation of self. In C. Gordon & J. Gergen (Eds.), The self in social interaction (pp. 299–308). New York: Wiley.

283

Gergen, K. J. (1971). The concept of self. New York: Holt, Rinehart, & Winston.

Gergen, K. J. (1972). Multiple identity: The healthy, happy human being wears many masks. Psychology Today, 5, 31–35, 64–66.

Gergen, K. J. (1982). From self to science: What is there to know? In J. Suls (Ed.), Psychological perspectives on the self (Vol. 1, pp. 129–149). Hillsdale, NJ: Lawrence Erlbaum Associates.

Gergen, K.J (1991). The saturated self. New York: Basic Books.

Gergen, K. J. & Gergen, M. (1986). Narrative form and the construction of psychological science. In T. R. Sarbin (Ed.), Narrative Psychology: The storied nature of human conduct. New York: Praeger.

Gergen, K. J. & Gergen, M. (1988). Narrative and self as relationship. In L. Berkowitz (Ed.), Advances in experimental social psychology (Vol. 21, pp. 17–56). San Diego: Academic Press.

Ghaffarian, S. (1987). The acculturation of Iranians in the United States. Journal of Social Psychology, 127, 565–571.

Gillespie, A., Howarth, C., & Cornish, F. (2012). Four problems for researchers using social categories. Culture and psychology, 18(3), 391-402.

Gillespie,A., Kadianaki, I., & O’Sullivan-Lago, R. (2012). Encountering alterity: Geographic and semantic movements. In J. Valsiner (Ed.), The Oxford handbook of culture and psychology. (pp. 695-709). New York: Oxford University Press.

Goffman, E. (1959). The presentation of the self in everyday life. New York; Anchor.

Goffman, E. (1968) Stigma. Notes on the Management of Spoiled Identity. London: Penguin.

Graafsma, T. L. G. (1994). Psychoanalysis. In H. Bosma, T. Graafsma. H. Grotevant, & D. de Levita (Eds.), Identity and development: An interdisciplinary approach (pp. 21–24). Thousand Oaks, CA: Sage Publications.

Graves, T. D. (1967). Psychological acculturation in a tri-ethnic community. South-western Journal of Anthropology, 23, 337–350.

Greenwald, A. G. (1988). A social-cognitive account of the self’s development. In D. Lapsley & F. C. Power (Eds.), Self, ego, and identity: Integrative approaches (pp. 30–42). New York: Springer-Verlag.

Habermas, T., & Bluck, S. (2000). Getting a life: The emergence of the life story in adolescence. Psychological Bulletin, 126, 748–769.

284

Harre, R. (1983). Identity projects. In G. Breakwell (Ed), Threatened identities. Chichester: John Wiley.

Hawkins, F. (1989). Critical Years in Immigration: Canada and Australia Compared. Sydney: New South Wales University Press.

Heimberg, R. G., Hope, D. A., Rapee, R. M., & Bruch, M. A. (1988). The validity of the Social Avoidance and Distress Scale and the Fear of Negative Evaluation Scale with social phobic patients. Behaviour Research and Therapy, 26, 407–410.

Hermans, H. J. M. (1987a). Self as organized system of valuations: Towards a dialogue with the person. Journal of Counseling Psychology, 34, 10–19.

Hermans, H. J. M. (1987b). The dream in the process of valuation: A method of interpretation. Journal of Personality and Social Psychology, 53, 163–175.

Hermans, H. J. M. (1988). On the integration of nomothetic and idiographic research methods in the study of personal meaning. Journal of Personality, 56, 785–812.

Hermans, H. J. M. (1989). The meaning of life as an organized process. Psychotherapy, 26, 11–22.

Hermans, H. J. M. (1991). The person as co-investigator in self-research: Valuation theory. European Journal of Personality, 5, 217–234.

Hermans, H. J. M. (1992). Unhappy self-esteem: A meaningful exception to the rule. Journal of Psychology, 126, 555–570.

Hermans, H. J. M. (1993). Moving opposites in the self: A Heraclitean approach. Journal of Analytical Psychology, 38, 437–462.

Hermans, H. J. M. (1995). From assessment to change: The personal meaning of clinical problems in the context of the self-narrative. In R .A. Neimeyer and M. J. Mahoney (Eds.), Constructivism in psychotherapy (pp. 247–272). Washington, DC: American Psychological Association.

Hermans, H. J. M. (1996a). Voicing the self: From information processing to dialogical interchange. Psychological Bulletin, 119, 31–50.

Hermans, H. J. M. (1996b). Opposites in a dialogical self: Constructs as characters. Journal of Constructivist Psychology, 9, 1–26.

Hermans, H. J. M. (1998). Moving cultures: The perilous problems of cultural dichotomies in a globalizing society. American Psychologist, 53(10), 1111- 1120.

Hermans, H. J. M. (1999). Dialogical thinking and self-innovation. Culture & Psychology, 5, 67–87.

285

Hermans, H. J. M. (2000). The coherence of incoherent narratives. Narrative Inquiry, 10(1), 223–227.

Hermans, H. J. M. (2000). The position repertory of interviewer and narrator. Narrative Inquiry, 10(1), 191-194.

Hermans, H. J. M. (2001a). The dialogical self: Toward a theory of personal and cultural positioning. Culture & Psychology, 7(3), 243–282.

Hermans, H. J. M. (2001b). Mixing and moving cultures require a dialogical self. Human Development, 44, 24–28.

Hermans, H. J. M. (2001c). The construction of a personal position repertoire: Method and practice. Culture & Psychology, 7(3), 323–365.

Hermans, H. J. M. (2002a). The dialogical self as a society of mind: Introduction. Theory & Psychology, 12(2), 147–160.

Hermans, H.J.M. (2002b) The dialogical self: One person, different stories. In Y. Kashima, M. Foddy & M. Platow (Eds.), Self and identity: Personal, social and symbolic (pp. 71–99). New Jersey: Lawrence Erlbaum Associates.

Hermans, H. J. M. (2003). The construction and reconstruction of a dialogical self. Journal of Constructivist Psychology, 16, 89–130.

Hermans, H. J. M. (2004). The innovation of self-narratives: A dialogical approach. In. L. Angus & J. McLeod (Eds.), The handbook of narrative and psychotherapy: Practice, theory, and research (pp. 175–191). Thousand Oaks, CA: Sage

Hermans, H. J. M. (2004b). The dialogical self: Between exchange and power. In H. J. M. Hermans & G. Dimaggio (Eds.), The dialogical self in psychotherapy: An introduction (pp.13-28). London: Brunner-Routledge.

Hermans, H. J. M. (2006). Moving through three paradigms, yet remaining the same thinker. Counseling Psychology Quarterly, 19(1), 5–25.

Hermans, H. J. M. (2008). How to perform research on the basis of dialogical self theory? Journal of Constructivist Psychology, 21,185–199.

Hermans, H. J. M., & Dimaggio, G. (2007). Self, identity and globalization in times of uncertainty: A dialogical analysis. Review of General Psychology, 11(1), 31-61,

Hermans, H. J. M, & Hermans-Jansen, E. (1995). Self-narratives: The construction of meaning in psychotherapy. New York: Guilford Press.

Hermans, H. J. M. & Hermans-Konopka, A. (2012). Dialogical self theory: Positioning and counter-positioning in a globalizing society. Cambridge, UK: Cambridge University Press.

286

Hermans, H. J. M., & Kempen, H .J. G. (1993). The dialogical self: Meaning as movement. San Diego: Academic Press.

Hermans, H. J. M., & Kempen, H. J. G. (1998). Moving cultures: The perilous problems of cultural dichotomies in a globalizing society. American Psychologist, 53, 1111–1120.

Hermans, H. J. M., Kempen, H. J. G., & Van Loon, R. J. P. (1992). The dialogical self: Beyond individualism and rationalism. American Psychologist, 47, 23– 33.

Hermans, H. J. M., & Oles, P. K. (1996). Value crisis: Affective organization of personal meanings. Journal of Research in Personality, 30, 457–482.

Hermans, H. J. M., Rijks, T. I., & Kempen, H .J. G. (1993). Imaginal dialogues in the self: Theory and method. Journal of Personality, 61, 207–236.

Hirst, J. B. (2010). Looking for Australia. Melbourne, Australia: Black Inc.

Hirst, J. B. (2014). Australian history in 7 questions. Melbourne, Australia: Black Inc.

Hofmann, S. G., DiBartolo, P. M., Holaway, R. M., & Heimberg, R. G. (2004). Scoring error of Social Avoidance and Distress Scale and its psychometric implications. Depression and Anxiety, 19, 197–198.

Hofstede, G. (1980). Culture’s consequences: International differences in work- related values. Beverly Hills, CA: Sage.

Hofstede, G. (1991). Cultures and organizations: Software of the mind. New York: McGraw-Hill.

Holstein, J. A., & Gubrium, J. F. (2000). The self we live by: Narrative identity in a post-modern world. Symbolic Interaction, 23(4), 407-409.

Hovey, J. D. (2000). Acculturative stress, depression, and suicidal ideation in Mexican immigrants. Cultural diversity & Ethnic Minority Psychology, 5(2), 134–151.

Hovey, J .D. & King, C. A. (1996). Acculturative stress, depression, and suicidal ideation among immigrant and second-generation Latino adolescents. Journal of the American Academy of Child and Adolescent Psychiatry, 35, 1183– 1192.

Imamoglou, E. O. (1998). Individualism and collectivism in a model and scale of balanced differentiation and integration. Journal of Psychology, 132, 95–105.

James, W. (1890). The Principles of psychology (Vol.1). London: Macmillan.

James, W. (1948). Psychology. New York: World Publishing Co.

287

James, W. (1961). Psychology: The briefer course. New York: Harper & Row.

Jung, C. G. (1957). The undiscovered self. New York: New American Library.

Kaplan, M. S., & Marks, J. (1990). Adverse effects of acculturation: Psychological distress among Mexican American young adults. Social Science and Medicine, 31, 1313–1319.

Kadianaki, I. (2009). Dramatic life courses: Migrants in the making. In J. Valsiner, P. C. M. Molenaar, M. C. D. P. Lyra, & N. Chaudhary (Eds.), Dynamic Process Methodology in the Social and Developmental Sciences (pp. 477- 492). New York: Springer.

Kashima, Y., Yamaguchi, S. Kim, U., Choi, S. C., Gelfand, M. J. & Yuki, M. (1995). Culture, gender, and self: A perspective from individualism- collectivism research. Journal of Personality and Social Psychology, 69, 925– 937.

Kashima, Y. (2001). Culture and social cognition: Towards a social psychology of cultural dynamics. In D. Matsumoto (Ed.), Handbook of culture and psychology (pp. 325-360). New York: Oxford University Press.

Kendall, P. C., Hollon, S. D., Beck, A. T., Hammen, C. L., & Ingram, R. E. (1987). Issues and recommendations regarding the use of the Beck Depression Inventory. Cognitive Therapy and Research, 11, 289–299.

King, R. (1995) “Migrations, globalisation and place”. In Massey, D. and Jess, P. (Eds.) A Place in the World? Places, Culture and Globalisation (Vol. 4, pp. 5-44). Oxford: Open University Press.

Kinnvall, C. (2004). Globalization and religious nationalism: Self, identity, and the search for ontological security. Political Psychology, 25(5), 741–767.

Kinnvall, C. & Nesbitt-Larking, P. (2011). The political psychology of globalisation: Muslims in the west. Oxford, UK: Oxford University Press.

König, J. (2009). Moving experience: Dialogues between personal cultural Positions. Culture & Psychology, 15, 97–119.

Kosmitzki, C. (1996). The reaffirmation of cultural identity in the context of cross- cultural encounters. Personality and Social Psychology Bulletin, 22, 239–429.

Kuhn, M. H., & McPartland, T. S. (1954). An empirical investigation of self- attitudes. American Sociological Review, 19, 68–76.

LaFromboise, T., Coleman, H. L. K., & Gerton, J. (1993). Psychological impact of biculturalism: Evidence and theory. Psychological Bulletin, 114, 395–412.

Lawrence-Lightfoot, S. (2005). Reflections on portraiture: A dialogue between art and science. Qualitative Inquiry, (11)3, 3-15.

288

Leary, M. R. (1983). A brief version of the Fear of Negative Evaluation Scale. Personality and Social Psychology Bulletin, 9, 371-376.

Leary, M. R. (1991). Social anxiety, shyness, and related constructs. In J. P. Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.), Measures of personality and social psychological attitudes (pp. 161–194). New York: Academic Press.

Lee, E. (1966). A theory of migration. Demography, 3, 47–57.

Lyddon, W. J., Yowell, D. R., & Hermans, H .J. M. (2006). The self-confrontation method: Theory, research, and practical utility. Counselling Psychology Quarterly, 19(1), 27–43.

Maalouf, A. (1998). On identity (B. Bray, Trans.). London: The Harvill Press.

Madden, R., & Young, S. (1993). Women and men immigrating to Australia: Their characteristics and immigration decisions. Canberra: Australian Government Publishing Service.

Mahalingam, R. (2008). Power, social marginality, and the cultural psychology of identities at the cultural contact zones: Commentary on O’Sulligan-Lago, de Abreu, and Burgess. Human Development, 51, 368-273.

Mahalingam, R. (2012). Model minority myth: Engendering cultural psychology of Asian immigrants. In E. L. Grigorenko (Ed.), U. S. immigration and education: Cultural and policy issues across the lifespan (pp. 119-136). New York: Springer.

Mahalingam, R., & Leu, J. (2005). Culture, essentialism, immigration and representations of gender. Theory & Psychology, 15(6), 839-860.

Malhi, R. L. Boon, S. D. & Rogers, T. B. (2009). Being Canadian and being Indian: Subject positions and discourses used in South Asian-Canadian women’s talk about ethnic identity. Culture & Psychology, 15(2), 255-283.

Markus, A., Jupp, J., & McDonald, P. (2009). Australia’s immigration revolution. Sydney, Australia: Allen & Unwin.

Markus, H. R., & Kitayama, S. (1991). Culture and the self: Implications for cognition, emotion, and motivation. Psychological Review, 98, 224–253.

Markus, H. R., & Kunda, Z. (1986). Stability and malleability of the self-concept. Journal of Personality and Social Psychology, 51, 858–866.

Martindale, C. (1980). Subselves: The internal representation of situational and personal dispositions. In Wheeler, L. (Ed.), Review of personality and social psychology (pp. 193–218). Beverly Hills: Sage.

289

Maslow, A. H. (1961). Peak-experiences as acute identity experiences. American Journal of Psychoanalysis, 21, 254–260.

Massey, D. S., Arango, J., Hugo, G., Kouaouci, A., Pellegrino, A,. & Taylor, E. (2006). Theories of international migration: A review and appraisal. In A. Messina & G. Lahav, (Eds.), The migration reader: Exploring politics and policies. Boulder, CA: Lynne Rienner Publishers.

Maydell-Stevens, E., Masgoret, A., & Ward (2007). Problems of psychological and sociocultural adaptation among Russian-speaking immigrants in New Zealand. Social Policy Journal of New Zealand, 30, 178–196.

McAdams, D. P. (1982). Experiences in intimacy and power: Relationships between social motives and autobiographical memory. Journal of Personality and Social Psychology, 42, 292–302.

McAdams, D. P. (1985). Power, intimacy, and the life story: Personological inquiries into identity. Chicago: Dorsey Press.

McAdams, D. P. (1989). The development of a narrative identity. In D. Buss & N. Cantor (Eds.), Personality psychology: Recent trends and emerging directions (pp. 160–174). New York: Springer-Verlag.

McAdams, D. P. (1990). Unity and purpose in human lives: The emergence of identity as a life story. In A. Rabin, R. Zucker, R. Emmons, & S. Frank (Eds.), Studying persons and lives (pp. 148–200). New York: Springer.

McAdams, D. P. (1993). The stories we live by: Personal myths and the making of the self. New York: Guilford Press.

McAdams, D. P. (1996). Personality, modernity, and the storied self: A contemporary framework for studying persons. Psychological Inquiry, 7 (4), 295–321.

McAdams, D. P. (1997). The case for unity in the (post)modern self: A modest proposal. In R. Ashmore & L. Jussim (Eds.), Self and identity: Fundamental issues (pp. 46–78). New York: Oxford University Press.

McAdams, D. P. (2003). Identity and the life story. In R. Fivush & C. A. Haden (Eds.), Autobiographical memory and the construction of a narrative self (pp. 187–208). Hoboken, NJ: Taylor & Francis.

McAdams, D. P. (2006). The person: An introduction to personality psychology (4th ed.). Hoboken, NJ: John Wylie & Sons.

McAdams, D. P. (2012). Exploring psychological themes through life-narrative accounts. In James Holstein and Jaber Gubrium (Eds.), Varieties of Narrative Analysis. pp. 15-32. Los Angeles, CA: Sage Publications, Inc.

290

McAdams, D. P. (2014). The life narrative at midlife. In B. Schiff (Ed.), Rereading Personal Narrative and the Life Course. New Directions for Child and Adolescent Development, 145, 57-69.

McAdams, D. P., & Pals, J. L. (2006). A new big five: Fundamental principles for an integrative science of personality. American Psychologist, 61(3), 204–217.

McAdams, D. P., Reynolds, J., Lewis, M., Patten, A. H., & Bowman, P. J. (2001). When bad things turn good and good things turn bad: Sequences of redemption and contamination in life narrative and their relation to psychosocial adaptation in midlife adults and in students. Personality and Social Psychology Bulletin, 27, 474–485.

Mead, G. H. (1934). Mind, self, and society. Chicago: The University of Chicago Press.

Miller, R. S. (1995). On the nature of embarrassability: Shyness, social evaluation, and social skill. Journal of Personality, 63, 315–339.

Moffit, K. H., & Singer, J. A. (1994). Continuity in the life story: Self-defining memories, affect, and approach/avoidance personal strivings. Journal of Personality, 62, 21–43.

Moghaddam. F. M. (2002). Cultural surplus and social mobility among minorities. Culture Psychology, 8, 401–407.

Moyerman, D. R,. & Forman, B. D. (1992). Acculturation and adjustment: A meta- analytic study. Hispanic Journal of Behavioral Sciences, 14, 163–200.

Mulry, G., Kalichman, S. C., Kelly, J. A., Ostrow, D. J., & Heckman, T. G. (1997). Grouping gay man on dimensions reflecting sexual behavior preferences: Implications for HIV–AIDS prevention. Psychology and Health, 12, 405–415.

Nesdale, D., & Mak, A. S. (2003). Ethnic identification, self-esteem ,and immigrant psychological health. International Journal of Intercultural Relations, 27, 23–40.

Nesdale, D., Rooney, R., & Smith. L. (1997). Migrant ethnic identity and psychological distress. Journal of Cross-Cultural Psychology, 28, 569–588.

Niedenthal, P. M., & Beike, D. R. (1997). Interrelated and isolated self-concepts. Personality and Social Psychology Review, Vol. 1, 106–128.

Norasakkunkit, V., & Kalick, S. M. (2002). Culture, ethnicity, and emotional distress measures: The role of self-construal and self-enhancement. Journal of Cross-Cultural Psychology, 33, 56–70.

Okazaki, S. (1997). Sources of ethnic differences between Asian-American and White American college students on measures of depression and social anxiety. Journal of Abnormal Psychology, 106, 52–60.

291

O'Sullivan-Lago, R. (2011a). “I think they’re just the same as us”: Building solidarity across the self/other divide. Papers on Social Representations, 20, 3.1-3.27) [http://www.psych.lse.ac.uk/psr/]

O'Sullivan-Lago, R. (2011b) The possibility of resolving conflict through dialogue?: Continuing from power. Psychology & Society, 4(1), 67-70.

O’Sullivan-Lago, R., & de Abreu, G. (2008). Going back to their roots: Maintaining continuity in the dialogical self through signs. Psychology & Society, 1(1), 42-53.

O'Sullivan-Lago, R., & de Abreu, G. (2010a). Maintaining continuity in a cultural contact zone: Identification strategies in the dialogical self. Culture & Psychology, 16(1), 73-92.

O'Sullivan-Lago, R., & de Abreu, G. (2010b). The dialogical self in a cultural contact zone: Exploring the perceived ‘cultural correction’ function of schooling. Journal of Community & Applied Social Psychology, 20, 275-287.

O’Sullivan-Lago, R., de Abreu, G., & Burgess, M. (2008). ‘I am a human being like you’: An identification strategy to maintain continuity in a cultural contact zone. Human Development, 51, 349-357.

Overholser, J. C., & Freheit, S. R. (1994). Assessment of interpersonal dependency using the Milton Clinical Multiaxial Inventory-II (MCMI-II) and the Depressive Experience Questionnaire. Personality and Individual Differences, 17, 71–78.

Park, R. E, (1928). Human migration and the marginal man. The American Journal of Sociology, 33, 881–893.

Patterson, M. L., & Strauss, M .E. (1972). An examination of the discriminant validity of the social avoidance and distress scale. Journal of Consulting and Clinical Psychology, 39, 169.

Phalet, K., & Hagendoorn, L. (1996). Personal adjustment to acculturative transitions: The Turkish experience. International Journal of Psychology, 31, 131–144.

Phinney, J. S. (1990). Ethnic identity in adolescents and adults: Review of research. Psychological Bulletin, 108, 499–514.

Phinney, J. S. (1996). When we talk about American ethnic groups, what do we mean? American Psychologist, 51(9), 918–927.

Phinney, J. S. (1999). An intercultural approach in psychology: Cultural contact and identity. Cross-Cultural Psychology Bulletin, 33, 24–31.

Phinney, J. S. (2000). Identity formation across cultures: The interaction of personal, societal and historical change. Human Development, 43, 27–31.

292

Phinney, J. S., & Flores, J. (2002). “Unpackaging” acculturation: Aspects of acculturation as predictors of traditional sex role attitudes. Journal of Cross- Cultural Psychology, 33, 320–331.

Potter, J., & Hepburn, A. (2005). Qualitative interviews in psychology: problems and possibilities. Qualitative Research in Psychology, 2, 281–307.

Prokopiou, E., Cline, T., & de Abreu, G. (2012). “Silent” monologues, “loud” dialogues, and the emergence of hibernated I-positions in the negotiation of multivoiced cultural identities. Culture & Psychology, 18(4), 494-509.

Raggatt, P. T. F. (2000). Mapping the dialogical self: Towards a rationale and method of assessment. European Journal of Personality, 14, 65-90.

Raggatt, P. T. F. (2007). Forms of positioning in the dialogical self: A system classification and the strange case of Dame Edna Everage. Theory & Psychology, 17(3), 355-382.

Raggatt, P. T. F. (2010). The self positioned in time and space: Dialogical paradigms. Theory & Psychology, 20(3), 451-460.

Raggatt, P. T. F. (2014). Positioning in the dialogical self: Recent advances in theory construction. In Hermans, H. J. M. & Gieser, T. (Eds.), Handbook of dialogical self theory (pp. 29–45). Cambridge, UK: Cambridge University Press.

Redfield, R., Linton, R., & Herskovits, M. J. (1936). Memorandum on the study of acculturation. American Anthropologist, 38, 149–152.

Richmond, A. H. (1988). Immigration and ethnic conflict. London: The Macmillan Press.

Richmond, A. H. (1993). Reactive migration: Sociological perspectives on refugee movement. Journal of Refugee Studies, 6, 7–24.

Robins, R. W., & John, O. P. (1997). The quest for self-insight: Theory and research on accuracy and bias in self-perception. In R. Hogan, J. Johnson, & S. Briggs (Eds.), Handbook of personality psychology (pp. 649–679). San Diego, CA: Academic Press.

Roccas, S., Horenczyk, G., & Schwartz S. H. (2000). Acculturation discrepancies and well-being: The moderating role of conformity. European Journal of Social Psychology, 30, 323–334.

Rosenberg, M. (1986). Conceiving the self. Melbourne, FL: Kreiger.

Rosenberg, S., & Gara, M.A. (1985). The multiplicity of personal identity. Review of Personality and Social Psychology, 6, 87–114.

293

Rosenwald, G. C., & Ochberg, R. L. (Eds.) (1992). Storied lives: The cultural politics of self-understanding. CT: Yale University Press.

Rudmin, F. W. (2003). Critical history of the acculturation psychology of assimilation, separation, integration, and marginalization. Review of General Psychology, 7(1), 3-37.

Rudmin, F. W. (2010a). The phenomenology of acculturation: Retrospective reports from the Philippines, Japan, Quebec and Norway. Culture & Psychology, 16(3), 313-332.

Rudmin, F. W. (2010b). Steps towards the renovation of acculturation research paradigms: What scientists’ personal experiences of migration might tell science. Culture & Psychology, 16(3), 299-312.

Rumbaut, R.G. (2004). Ages, life stages, and generational cohorts: Decomposing the immigrant first and second generations in the United States. The International Migration Review, 38(3), 1160–1205.

Ryder, A. G., lden, L. E., & Paulhus, D. L. (2000). Is acculturation unidimensional or bidimensional? A head-to-head comparison in the prediction of personality, self-identity, and adjustment. Journal of Personality and Social Psychology, 79, 49–65.

Ryff, C. D. (1989a). Happiness is everything, or is it? Explorations on the meaning of psychological well-being. Journal of Personality and Social Psychology, 6, 1069–1081.

Ryff, C. D. (1989b). Beyond Ponce de Leon and life satisfaction: New directions in quest of successful aging. International Journal of Behavioural Development, 12, 35–55.

Ryff, C. D. (1991) Possible selves in adulthood and old age: A tale of shifting horizons. Psychology and Aging, 6, 286–295.

Ryff, C. D. (1995). Psychological well-being in adult life. Current Directions in Psychological Science, 4, 99–104.

Ryff, C. D., & Essex, M. J. (1992). The interpretation of life experience and well- being: The sample case of relocation. Psychology and Aging, 7, 507–517.

Ryff, C. D., & Keyes, C. L. M. (1995). The structure of psychological well-being revisited. Journal of Personality and Social Psychology, 69, 719–727.

Ryff, C. D., & Singer, B. (1996). Psychological well-being: Meaning, measurement, and implications for psychotherapy research. Psychotherapy and Psychosomatics, 65, 14–23.

294

Ryff, C. D., Lee, Y. H., Essex, M. J., & Schmutte, P. M. (1994). My children and me: Midlife evaluations of grown children and of self. Psychology and Aging, 9, 195–205.

Sanchez-Rockliffe, M., & Symons, J. (2010). How migration affects the dialogical self. International Journal for Dialogical Science, 4(2), 5–34.

Sarbin, T. R., & Allen, V. L. (1968). Role theory. In G. Lindzey and E. Aronson (Eds.), The handbook of social psychology. Reading, MA: Addison-Wesley.

Schwartz, S. H. (1990). Individualism-collectivism: Critique and proposed refinements. Journal of Cross-Cultural Psychology, 21, 139–195.

Schwartz, S. J, Unger, J. B., Zamboanga, B. L., & Szapocznik, J. (2010). Rethinking the concept of acculturation Implications for theory and research. American Psychologist, 65(4), 237-251.

Searle, W., & Ward, C. (1990). The prediction of psychological and sociocultural adjustment during cross-cultural transitions. International Journal of Intercultural Relations, 14, 449–464.

Shotter, J., & Gergen, K. (Eds.). (1989). Tests of identity. London: Sage.

Singelis, T. M. (1994). The measurement of independent and interdependent self- construals, Personality and Social Psychology Bulletin. 20, 580–59.

Singer, J. A. (1990). Affective responses to autobiographical memories and their relationship to long term goals. Journal of Personality, 58, 535–563.

Singer, J. A., & Salovey, P. (1993). The remembered self: Emotion and memory in personality. New York: The Free Press.

Smith, R. E., & Sarason, I. G. (1975). Social anxiety and the evaluation of negative interpersonal feedback. Journal of Consulting and Clinical Psychology, 43, 429.

Steer, R. A., Rissmiller, D. J., & Beck, A. T. (2000). Use of the Beck Depression Inventory-II with depressed geriatric inpatients. Behaviour Research and Therapy, 38, 31–318.

Stemplewska-Żakowicz, K., Walecka, J., Gabińska, A., Zalewski, B., & Suszek, H. (2005). Experiments on positioning, positioning the experiments. In P. Oles & H. J. M. Hermans (Eds.), The dialogical self: Theory and research (pp. 183– 199). Lublin, Poland: Wydawnictwo KUL.

Strauss, A. (1956). The social psychology of George Herbert Mead. Chicago: University of Chicago Press.

Stryker, S. (1980). Symbolic interactionism: A social structural version. Menlo Park, CA: Benjamin/Cummings.

295

Stiles, W. B. (1993). Quality control in qualitative research. Clinical Psychology Review, 13, 593–618.

Stonequist, E. V. (1937). The marginal man. New York: Scribner.

Stopa, L., & Clark, D. M. (2001). Social phobia: Comments on the viability and validity of an analogue research strategy and British norms for the Fear of Negative Evaluation questionnaire. Behavioural and Cognitive Psychotherapy, 29, 423–440.

Suinn, R. M., Ahuna, C., & Khoo, G. (1992). The Suinn-Lew Asian Self-Identity Acculturation Scale: Concurrent and factorial validation. Educational and Psychological Measurement, 52, 1041–1046.

Surgan, S. ,& Abbey, E. (2014). Identity construction among transnational migrants. In Hermans, H.J.M. & Gieser, T. (Eds.), Handbook of dialogical self theory (pp. 151–168). Cambridge, UK: Cambridge University Press.

Tabachnik, B. G., & Fidell, L. S. (2001). Using multivariate statistics. (4th ed.). Boston, MA: Allyn & Bacon.

Taft, R. (1986). Methodological considerations in the study of immigrant adaptation in Australia. Australian Journal of Psychology, 38(3), 339–346.

Taylor, R. (1969). Migration and motivation: A study of determinants and types. In J.A. Jackson (Ed.), Migration. Cambridge, UK: Cambridge University Press.

Triandis, H .C., & Suh, E .M. (2002). Cultural influences on personality. Annual Review of Psychology, 53, 133–160.

Triandis, H .C., Leung, K., Villareal, M. J. & Clack, F. L. (1985). Allocentric versus idiocentric tendencies: convergent and discriminant validation. Journal of Research in Personality, 19, 395–415.

Triandis, H. C., Bontempo, R., Villareal, M. J., Asai, M., & Luca, N. (1988). Individualism and collectivism: Cross-cultural perspectives on self-group relationships. Journal of Personality and Social Psychology, 54, 323–338.

OECD (Organisation for Economic Co-operation and Development) and UNDESA (United Nations Department of Economic and Social Affairs) (2013). World Migration in Figures. Paris: OECD.

Valentine, G. (1997). Tell me about …: Using interviews as a research methodology. In R. Flowerdew & D. Martin (Eds.) Methods in Human Geography (pp. 110–136). Essex, UK: Addison Wesley Longman.

Valsiner, J. (2000). Culture and human development. London: Sage.

Valsiner, J. (2002). Forms of dialogical relations and semiotic autoregulation within the self. Theory & Psychology, 12(2), 251–265.

296

Van de Vijver, F. J. R., Helms-Lorenz, M., & Feltzer, M. F. (1999). Acculturation and cognitive performance of migrant children in the Netherlands. International Journal of Psychology, 34, 149–162.

Van de Vijver, F. J. R., & Phalet, K. (2004). Assessment in multicultural groups: the role of acculturation. Applied Psychology: An International Review, 53, 215–236.

Van Geel, R., & de Mey, H. R. A. (2004). A theory-guided hexagonal representation of single valuation systems for use with Hermans’ self- confrontation method. Journal of Constructivist Psychology, 17, 85–104.

Van Meijl, T. (2012). Multicultural adolescents between tradition and postmodernity: Dialogical self theory and the paradox of localization and globalization. In H. J. M. Hermans (Ed.), Applications of dialogical self theory: New directions for child and adolescent development, (137, 39-52). Hoboken, NJ: Wiley.

Van Meijl, T. (2014). Multiculturalism, multiple identifications and the dialogical self: Shifting paradigms of personhood in sociocultural anthropology. In Hermans, H. J. M. & Gieser, T. (Eds.), Handbook of dialogical self theory (pp. 98–114). Cambridge, UK: Cambridge University Press.

Van Oudenhoven, J. P., & Eisses, A. M. (1998). Integration and assimilation of Moroccan immigrants in Israel and the Netherlands. International Journal of Intercultural Relations, 22, 293–307.

Verkuyten, M., & de Wolf, A. (2002). Being, feeling and doing: Discourses and ethnic self-definitions among minority group members. Culture & Psychology, 8, 371-399.

Volkan, V. D. (2004). From hope for a better life to broken spirits: An introduction. In J. P. Wilson & B. Drozdek (Eds.), Broken Spirits: The treatment of traumatized asylum seekers, refugees, war and torture victims (pp. 7–12). New York: Brunner-Routledge.

Ward, C. (1999). Models and measurement of acculturation. In W. J. Lonner, D. L. Dinnel, D. K. Forgas, & S. Hays (Eds.), Merging past, present and future (pp. 221–230). Lisse, The Netherlands: Swets & Zeitlinger.

Ward, C. (2001). The A, B, C’s of acculturation. In D. Matsumoto (Ed.), The handbook of culture and psychology (pp. 411–445). Oxford: Oxford University Press.

Ward, C., Bochner, S., & Furnham, A. F. (2001). The psychology of culture shock. London: Routledge.

Ward, C., & Chang, W. C. (1997). “Cultural fit”: A new perspective on personality and sojourner adjustment. International Journal of Intercultural Relations, 21, 525–533.

297

Ward, C., & Kennedy, A. (1993a). Where’s the culture in cross-cultural transition? Comparative studies of sojourner adjustment. Journal of Cross-Cultural Psychology, 24(2), 221–249.

Ward, C., & Kennedy, A. (1993b). Psychological and sociocultural adjustment during cross-cultural transitions: A comparison of secondary students overseas and at home. International Journal of Psychology, 28, 129–147.

Ward, C., & Kennedy, A. (1996). Crossing cultures: The relationship between psychological and sociocultural dimensions of cross-cultural adjustment. In J. Pandey, D. P. S. Bhawuk & D. Sinha (Eds.), Asian contributions to cross- cultural psychology (pp. 289–306). New Delhi: Sage.

Ward, C., & Rana-Deuba, A. (1999). Acculturation and adaptation revisited. Journal of Cross-Cultural Psychology, 30, 372–392.

Ward, C., & Searle, W. (1991). The impact of value discrepancies and cultural identity on psychological and sociocultural adjustment of sojourners. International Journal of Intercultural Relations, 15, 209–225.

Watson, D., & Friend, R. (1969). Measurement of social evaluative anxiety. Journal of Consulting and Clinical Psychology, 33, 448–457.

Westen, D. (1995). A clinical-empirical model of personality: Life after the Mischelian ice age and the NEO-lithic era. Journal of Personality, 63, 495– 524.

Whitbourne, S. K. (1986). The me I know: A study of adult identity. New York: Springer.

Wortham, S. (2000). Interactional positioning and narrative self-construction. Narrative Inquiry, 10(1), 157–184.

Yamada, A. M., & Singelis, T. M. (1999). Biculturalism and self-construal. International journal of Intercultural Relations, 23, 697–709.

Young, S., & Madden, R. (1992). Immigration Selection Policies Affecting Women and Men. In Proceedings of Second National Immigration Outlook Conference (pp. 252–259). Melbourne, Australia: Bureau of Immigration Research.

Zheng, X., & Berry, J. W. (1991). Psychological adaptation of Chinese sojourners in Canada. International Journal of Psychology, 26, 451–470.

THE DIALOGICAL SELF OF MIGRANTS:

EXPLORING THE MIGRATION EXPERIENCE THROUGH SELF-NARRATIVES

Mariel Sanchez-Rockliffe

Faculty of Health, Arts and Design

Swinburne University of Technology

Australia

A Report of an investigation submitted

As partial requirement for the degree of

Doctor of Psychology (Counselling Psychology)

December 2015

Contents

APPENDIX A RESEARCH INSTRUMENTS 1 A.1 Notice for the Purpose of Recruiting Participants 2 A.2 Letter to Participants 3 A.3 Questionnaire 4 A.4 Questions of the Self-confrontation Method 22 A.5 Affect Terms for Rating Valuations 23 A.6 Matrix of Valuation / Affect 24

APPENDIX B STATISTICAL ANALYSIS FOR STUDY 1 25 B.1 Reliability of Psychological Self-reports and Self-confrontation Indices 27 B.2 Tables 32 B.3 Cluster Analysis 40 B.4 Repeated-measures MANOVA 52 B.5 Paired t-tests 64 B.6 Single-factor between-subjects MANOVA 69

APPENDIX C STATISTICAL ANALYSIS FOR STUDY 2 103 C.1 Analysis of Valuations by Theme 104 C.2 T-tests 108

APPENDIX D SELF-NARRATIVES OF CASE-STUDY PARTICIPANTS 114 D.1 Participant 03 Yoana 114 D.2 Participant 22 Juan 117 D.3 Participant 35 Luisa 119 D.4 Participant 40 Carla 122

APPENDIX A

RESEARCH INSTRUMENTS

 Notice for the purpose of recruiting participants

 Letter to participants

 Questionnaire

 Questions of the self-confrontation method

 Affect terms for rating participants’ valuations

 Matrix of valuation / affect

A.1 Notice for the Purpose of Recruiting Participants

Are you a migrant who has lived in Australia for five or more years?

Many migrants feel that they have developed two identities as a result of moving to a new country. One identity retains aspects of how they were in their country of origin; the other identity relates to their new country. If this is your experience I would like to hear from you.

My research project

My name is Mariel Sanchez-Rockliffe, and I am a doctoral student of counselling psychology at Swinburne University of Technology. I am researching the link between personal identity and the migrant experience. I am investigating the life stories of migrants using structured interviews designed to identify aspects of people’s life stories that reflect their identity.

Would you like to participate?

If you are a migrant and you wish to participate in my research, I want to hear from you! Please note that this study is an On-Campus Project and participation will count towards your course requirement.

To participate you will first complete a questionnaire about personal qualities and relationships with others. You will then participate in a research interview. You will be able to follow up with questions and comments, and I can provide feedback on your individual life story. Naturally, all information you provide will be confidential and you are free to withdraw at any time.

0413 537 282 0413 537 282 0413 537 282 0413 537 282 0413 537 282 0413 537 282 0413 537 282 0413 537 282 0413 537 282 0413 537 282 0413 537 282 0413 537 282 0413 537 282 0413 537 282 0413 537 282 0413 537 282

A.2 Letter to Participants

DATE

NAME ADDRESS

Dear NAME

Understanding Self-narratives and Emotional Reactions using the Self-confrontation method

Thank you for agreeing to participate in my research. I enclose a blank questionnaire. Please complete it in one sitting, preferably within the next two weeks. I estimate that it will take about 45 minutes. When you have finished, please mail it to me at Swinburne in the pre-addressed, stamped envelope provided.

When I receive the completed questionnaire I will contact you to arrange a time to administer the second and final questionnaire of this study. This will take the form of a structured face-to- face interview.

If you have any questions, or need to contact me for any reason, please call me on 9429 2903 or 0413 537 282, or email me at [email protected].

Sincerely

Mariel Sanchez-Rockliffe

Enclosure: Questionnaire

A.3 Questionnaire

Understanding the Influence of Migration on Personal Identity

Information Sheet

Conducted by Mariel Sanchez-Rockliffe

Supervisors: Dr Glen Bates and Dr Emiko Kashima

This study compares the self-narratives (or life stories) of migrants with those of other migrants. The primary aim is to help us understand how the migrant experience affects one’s sense of personal identity.

Those wishing to participate will first complete a questionnaire relevant to personal qualities and relationships with others. The person would then participate in a research interview designed to elicit the person’s current life story.

The questionnaire will take approximately 45 minutes to complete and the interview will take approximately 90 minutes. All participants will receive feedback on their life story, and will have an opportunity to follow up any comments or reactions they have after the interview.

Participants will be identified by a number, and their data will not contain their name on any part of the record. Please be assured that while the general results from the study may appear in publications, information by individual participants will be kept confidential. You are free to withdraw your consent and discontinue participating in the study at any time.

Please consider the purpose and time commitment of this study before you decide whether or not to participate. Retain this information sheet for your own records.

Questions about this project can be directed to me (9214-5554) or to my supervisors, Dr Glen Bates (9214-8100), and Dr Emiko Kashima (9214-8206), Psychology Discipline, School of Behavioural Sciences, Swinburne University of Technology. If you have queries or concerns which the senior investigaror was unable to satisfy, please contact:

The Chair, SBS Research Ethics Committee School of Social and Behavioural Sciences, Mail H24 Swinburrne University of Technology, Hawthorn, Victoria 3122

If you have a complaint about the way that you were treated during this study please write to:

The Chair, Human Research Ethics Committee Swinburrne University of Technology, Hawthorn, Victoria 3122

Thank you.

Understanding the Influence of Migration on

Personal Identity

An investigation into how the migrant experience affects one’s sense of personal identity.

I (FULL NAME)……………………………………………………………………………………………

OF (ADDRESS)….………….…………………………………………………………………………….

CONTACT PHONE NUMBER……………………………………………………………………………

HAVE READ AND UNDERSTOOD THE INFORMATION ABOVE.

I AGREE TO PARTICIPATE IN THIS STUDY, REALISING THAT I AM FREE TO WITHDRAW AT ANY TIME.

I AGREE THAT RESEARCH DATA COLLECTED FOR THE STUDY MAY BE PUBLISHED ON THE CONDITION THAT MY NAME IS NOT USED.

NAME OF PARTICIPANT…………………………………………………………………………………

SIGNATURE………………………………………………….……..DATE………………………………

Demographics

Age ...... years

Occupation ...... specify

Sex Circle one

Male ...... 1

Female ...... 2

Marital status Circle one number to show what best describes your marital circumstances at the moment.

Single ...... 1

Married or marital relationship ...... 2

Divorced or separated ...... 3

Widowed ...... 4

Highest level of education Circle one number to show which best describes the highest level of education you have achieved

Before Year 10 ...... 1

Year 10 ...... 2

Year 11 ...... 3

SPAI Scale

Below is a list of behaviours that may or may not be relevant to you. Based on your personal experience, please indicate how frequently you experience these feelings and thoughts in social situations. A social situation is defined as a gathering of two or more people, for example a meeting, a party, a bar or restaurant, conversing with one or more other people, etc. Feeling anxious is a measure of how tense, nervous or uncomfortable you are during social encounters. Please use the scale listed below and circle the number that best reflects how frequently you experience these responses.

Never Very infrequent Infrequent Sometimes Frequent frequentVery Always

1 I feel anxious when entering social situations where there is a small group...... 1 2 3 4 5 6 7 2 I feel anxious when entering social situations where there is a large group...... 1 2 3 4 5 6 7 3 I feel anxious when I am in a social situation and I become the centre of attention...... 1 2 3 4 5 6 7 4 I feel anxious when I am in a social situation and I am expected to engage in 1 2 3 4 5 6 7 some activity...... 5 I feel anxious when making a speech in front of an audience...... 1 2 3 4 5 6 7 6 I feel anxious when speaking in a small informal meeting...... 1 2 3 4 5 6 7 7 I feel so anxious about attending social gatherings that I avoid these situations...... 1 2 3 4 5 6 7 8 I feel so anxious in social situations that I leave the social gathering...... 1 2 3 4 5 6 7 9 I feel anxious when in a small gathering with… strangers...... 1 2 3 4 5 6 7 Authority figures...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7 10 I feel anxious when in a large gathering with… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7 11 I feel anxious when in a bar or restaurant with… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7

Never Very infrequent Infrequent Sometimes Frequent frequentVery Always

12 I feel anxious and I do not know what to do when in a new situation with… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7 13 I feel anxious and I do not know what to do when in a situation involving confrontation with… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7 14 I feel anxious and I do not know what to do when in an embarrassing situation with… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7 15 I feel anxious when discussing intimate feelings with… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7 16 I feel anxious when stating an opinion to… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7 17 I feel anxious when talking about business with… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7

Never Very infrequent Infrequent Sometimes Frequent frequentVery Always

18 I feel anxious when approaching and/or initiating a conversation with… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7 19 I feel anxious when having to interact for longer than a few minutes with… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7 20 I feel anxious when drinking (any type of beverage) and/or eating in front of… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7 21 I feel anxious when writing or typing in front of… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7 22 I feel anxious when speaking in front of… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7 23 I feel anxious when being criticised or rejected by… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7

Never Very infrequent Infrequent Sometimes Frequent frequentVery Always

24 I attempt to avoid social situations where there are… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7 25 I leave social situations where there are… strangers...... 1 2 3 4 5 6 7 authority figures ...... 1 2 3 4 5 6 7 the opposite sex...... 1 2 3 4 5 6 7 people in general...... 1 2 3 4 5 6 7 26 Before entering a social situation I think about all the things that can go wrong. The types of thoughts I experience are: Will I be dressed properly? ...... 1 2 3 4 5 6 7 I will probably make a mistake and look foolish...... 1 2 3 4 5 6 7 What will I do if no one speaks to me? ...... 1 2 3 4 5 6 7 If there is a lag in the conversation what can I talk about? ...... 1 2 3 4 5 6 7 People will notice how anxious I am...... 1 2 3 4 5 6 7 27 I feel anxious before entering a social situation...... 1 2 3 4 5 6 7 28 My voice leaves me or changes when I am talking in a social situation...... 1 2 3 4 5 6 7 29 I am not likely to speak to people until they speak to me...... 1 2 3 4 5 6 7 30 I experience troublesome thoughts when I am in a social setting. For example: I wish I could leave and avoid the whole situation...... 1 2 3 4 5 6 7 If I mess up again I will really lose my confidence...... 1 2 3 4 5 6 7 What kind of impression am I making? ...... 1 2 3 4 5 6 7 Whatever I say it will probably sound stupid...... 1 2 3 4 5 6 7 31 I experience the following prior to entering a social situation: sweating ...... 1 2 3 4 5 6 7 frequent urge to urinate ...... 1.. 2 3 4 5 6 7 heart palpitations ...... 1 2 3 4 5 6 7 32 I experience the following in a social situation: sweating ...... 1 2 3 4 5 6 7 frequent urge to urinate ...... 1.. 2 3 4 5 6 7 heart palpitations ...... 1 2 3 4 5 6 7 33 I feel anxious when I am home alone...... 1 2 3 4 5 6 7

Never Very infrequent Infrequent Sometimes Frequent frequentVery Always

34 I feel anxious when I am in a strange place...... 1. 2 3 4 5 6 7 35 I feel anxious when I am on any form of public transportation (ie bus, train, plane). 1 2 3 4 5 6 7 ...... 36 I feel anxious when crossing streets...... 1 2 3 4 5 6 7 37 I feel anxious when I am in crowded public places (ie shops, church, movies, 1 2 3 4 5 6 7 restaurants)...... 38 Being in large open spaces makes me feel anxious...... 1 2 3 4 5 6 7 39 I feel anxious when I am in enclosed spaces (lifts, tunnels, etc)...... 1. 2 3 4 5 6 7 40 Being in high places makes me feel anxious (for example, tall buildings)...... 1 2 3 4 5 6 7 41 I feel anxious when waiting in a long line...... 1 2 3 4 5 6 7 42 There are times when I feel like I have to hold on to things because I am afraid I 1 2 3 4 5 6 7 will fall...... 43 When I leave home and go to various public places, I go with a family member or 1 2 3 4 5 6 7 friend...... 44 I feel anxious when riding in a car...... 1 2 3 4 5 6 7 45 There are certain places I do not go to because I may feel trapped...... 1 2 3 4 5 6 7

SADS Scale

Please circle T (True) or F (False) alongside each statement.

True False 1 I feel relaxed even in unfamiliar social situations...... T F

2 I try to avoid situations which force me to be very sociable...... T F

3 It is easy for me to relax when I am with strangers ...... T F

4 I have no particular desire to avoid people ...... T F

5 I often find social occasions upsetting ...... T F

6 I usually feel calm and comfortable at social occasions...... T F

7 I am usually at ease when talking to someone of the opposite sex...... T F

8 I try to avoid talking to people unless I know them well...... T F

9 If the chance comes to meet new people, I often take it...... T F

10 I often feel nervous or tense in casual get-togethers in which both sexes are present...... T F

11 I am usually nervous with people unless I know them well...... T F

12 I usually feel relaxed when I am with a group of people...... T F

13 I often want to get away from people...... T F

14 I usually feel uncomfortable when I am in a group of people I don’t know...... T F

15 I usually feel relaxed when I meet someone for the first time...... T F

16 Being introduced to people makes me tense and nervous...... T F

17 Even though a room if full of strangers I may enter it anyway...... T F

18 I would avoid walking up and joining a large group of people...... T F

19 When my superiors want to talk with me, I talk willingly...... T F

20 I often feel on edge when I am with a group of people...... T F

FNES Scale

Please circle T (True) or F (False) alongside each statement.

True False 1 I rarely worry about seeming foolish to others...... T F 2 I worry about what people will think of me even when I know it doesn’t make any difference. T F 3 I become tense and jittery if I know someone is sizing me up...... T F 4 I am unconcerned even if I know people are forming an unfavourable impression of me. ... T F 5 I feel very upset when I commit some social error...... T F 6 The opinions that important people have of me cause me little concern...... T F 7 I am often afraid that I may look ridiculous or make a fool of myself...... T F 8 I react very little when other people disapprove of me...... T F 9 I am frequently afraid of other people noticing my shortcomings...... T F 10 The disapproval of others would have little effect on me...... T F 11 If someone is evaluating me I tend to expect the worst...... T F 12 I rarely worry about what kind of impression I am making on someone...... T F 13 I am afraid that others will not approve of me...... T F 14 I am afraid that people will find fault with me...... T F 15 Other people’s opinions of me do not bother me...... T F 16 I am not necessarily upset if I do not please someone...... T F 17 When I am talking to someone, I worry about what they may be thinking about me...... T F 18 I feel that you can’t help making social errors sometimes, so why worry about it...... T F 19 I am usually worried about what kind of impression I make...... T F 20 I worry a lot about what my superiors think of me...... T F 21 I know if someone is judging me, it has little effect on me...... T F 22 I worry that others will think I am not worthwhile...... T F 23 I worry very little about what others will think of me...... T F 24 Sometimes I think I am too concerned with what other people think of me...... T F 25 I often worry that I will say or do the wrong things...... T F 26 I am often indifferent to the opinions others have of me...... T F 27 I am usually confident that others will have a favourable impression of me...... T F 28 I often worry that people who are important to me won’t think much of me...... T F 29 I brood about the opinions my friends have about me...... T F 30 I become tense and jittery if I know I am being judged by my superiors...... T F

FNES Scale

Please circle T (True) or F (False) alongside each statement.

Personal Wellbeing Scale

The following statements relate to descriptions of personal views of life. Each statement has six possible answers, with the numbers 1 and 6 being extreme ends of a continuum. Please circle the number which best expresses your answer. If you completely disagree with the statement, circle “1”. If you completely agree with the statement, circle “6”. If you think differently, circle the number which best expresses your response. Please circle only one response for each statement.

Strongly disagreeStrongly Disagree disagree Mildly Mildly agree Agree Strongly agree

1 In general, I feel I am in charge of the situation in which I live...... 1 2 3 4 5 6 2 Most people see me as loving and affectionate...... 1 2 3 4 5 6 3 Sometimes I change the way I act or think to be more like those around me...... 1 2 3.. 4 5 6 4 I am not interested in activities that will expand my horizons...... 1 2 3 4 5 6 5 I feel good when I think of what I’ve done in the past and what I hope to do in the future...... 1 2 3 4 5 6 6 When I look at the story of my life, I am pleased with how things have turned out...... 1 2 3 4 5 6 7 The demands of everyday life often get me down...... 1 ...... 2 3 4 5 6 8 Maintaining close relationships has been difficult and frustrating for me...... 1 ...... 2 3 4 5 6 9 I am not afraid to voice my opinions, even when they are in opposition to the opinions of most people...... 1 2 3... 4 5 6 10 In general, I feel that I continue to learn more about myself as time goes by...... 1 2 3... 4 5 6 11 I live life one day at a time and don’t really think about the future...... 1 2 3 4 5 6 12 In general, I feel confident and positive about myself...... 1 2 ...... 3 4 5 6 13 I do not fit very well with the people and the community around me...... 1 2 3 4 5 6 14 I often feel lonely because I have few close friends with whom to share my concerns...... 1 2 3 4 5 6 15 My decisions are not usually influenced by what everyone else is doing...... 1 2...... 3 4 5 6 16 I am the kind of person who likes to give new things a try...... 1 2 3 4 5 6 17 I tend to focus on the present, because the future nearly always brings me problems...... 1 2 3 4 5 6 18 I feel like many of the people I know have got more out of life than I have...... 1 2 ...... 3 4 5 6 19 I am quite good at managing the many responsibilities of my daily life...... 1 ...... 2 3 4 5 6 20 I enjoy personal and mutual conversations with family members or friends...... 1 2 .....3 4 5 6 21 I tend to worry about what other people think of me...... 1 2...... 3 4 5 6 22 I don’t want to try new ways of doing things—my life if fine the way it is...... 1 ...... 2 3 4 5 6 23 I have a sense of direction and purpose in life...... 1 2 3 4 5 6

ongly agree Strongly disagreeStrongly Disagree disagree Mildly Mildly agree Agree Str

24 Given the opportunity, there are many things about myself that I would change...... 1 2 3 4 5 6 25 I often feel overwhelmed by my responsibilities...... 1 2 3 4 5 6 26 It is important to me to be a good listener when close friends talk to me about their problems...... 1 ...... 2 3 4 5 6 27 Being happy with myself is more important to me than having others approve of me...... 1 2 3 4 5 6 28 I think it is important to have new experiences that challenge how you think about yourself and the world...... 1 2 3 4 5 6 29 My daily activities often seem trivial and unimportant to me...... 1 2 3 4 5 6 30 I like most aspects of my personality...... 1 2 3 4 5 6 31 If I were unhappy with my living situation, I would take effective steps to change it...... 1 2 3 4 5 6 32 I don’t have many people who want to listen when I need to talk...... 1 2 3 4 5 6 33 I tend to be influenced by people with strong opinions...... 1 2 .....3 4 5 6 34 When I think about it, I haven’t really improved much as a person over the years...... 1 2 3 4 5 6 35 I don’t have a good sense of what it is I am trying to accomplish in life...... 1 ...... 2 3 4 5 6 36 I made some mistakes in the past, but I feel that all in all everything has worked out for the best...... 1 2...... 3 4 5 6 37 I generally do a good job of taking care of my personal finances and affairs...... 1 2 3... 4 5 6 38 I feel like I get a lot out of friendships...... 1 2 3 4 5 6 39 People rarely talk me into doing things I don’t want to do...... 1 2 3 . 4 5 6 40 In my view, people of every age are able to continue growing and developing...... 1 2 3 4 5 6 41 I used to set goals for myself, but that now seems like a waste of time...... 1 ...... 2 3 4 5 6 42 In many ways, I feel disappointed about my achievements in life...... 1 2 3 4 5 6 43 I find it stressful that I can’t keep up with all of the things I have to do each day...... 1 2 3 4 5 6 44 It seems to me that most other people have more friends than I do...... 1 2 3 4 5 6 45 It is more important to me to ‘fit in’ with others than to stand alone on my principles...... 1 2 3 4 5 6 46 With time, I have gained a lot of insight about life that has made me a stronger, more capable person...... 1 2 3.. 4 5 6 47 I enjoy making plans for the future and working to make them a reality...... 1 ...... 2 3 4 5 6 48 For the most part, I am proud of who I am and the life I lead...... 1 2 3 4 5 6 49 I am good at juggling my time so that I can fit everything in that needs to get done...... 1 2 3 4 5 6 50 People would describe me as a giving person, willing to share my time with others. 1 2 3 4 5 6 51 I have confidence in my own opinions, even if they are contrary to the general consensus...... 1 2...... 3 4 5 6

Strongly disagreeStrongly Disagree disagree Mildly Mildly agree Agree Strongly agree

52 I have the sense that I have developed a lot as a person over time...... 1 2 3 4 5 6 53 I am an active person in carrying out the plans I set for myself...... 1 2 3 4 5 6 54 I envy many people for the lives they lead...... 1 2 3 4 5 6 55 My daily life is busy, but I derive a sense of satisfaction from keeping up with everything...... 1 ...... 2 3 4 5 6 56 I have not experienced many warm and trusting relationships with others...... 1 2 ...... 3 4 5 6 57 It is difficult for me to voice my own opinions on controversial matters...... 1 ...... 2 3 4 5 6 58 I do not enjoy being in new situations that require me to change my old familiar ways of doing things...... 1 2 3... 4 5 6 59 Some people wander aimlessly through life, but I am not one of them...... 1 ...... 2 3 4 5 6 60 My attitude about myself is probably not as positive as most people feel about themselves...... 1 2...... 3 4 5 6 61 I get frustrated when trying to plan my daily activities because I never accomplish the things I set out to do...... 1 2 3 4 5 6 62 I often feel like I am on the outside looking in when it comes to friendship...... 1 2 ...... 3 4 5 6 63 I often change my mind about decisions if my friends or family disagree...... 1 2...... 3 4 5 6 64 For me, life has been a continuous process of learning, changing, and growth...... 1 2 3 4 5 6 65 I sometimes feel as if I have done all there is to do in life...... 1 2 3 . 4 5 6 66 Many days I wake up feeling discouraged about how I have lived my life...... 1 2...... 3 4 5 6 67 My efforts to find the kinds of activities and relationships that I need have been quite successful...... 1 ...... 2 3 4 5 6 68 I know that I can trust my friends, and they know they can trust me...... 1 2 3 4 5 6 69 I am not the kind of person who given in to social pressures to think or act in certain ways...... 1 2 3 4 5 6 70 I enjoy seeing how my views have changed and matured over the years...... 1 2...... 3 4 5 6 71 My aims in life have been more a source of satisfaction than frustration to me...... 1 2 3 4 5 6 72 The past had its up and downs, but in general I wouldn’t want to change it...... 1 2 .....3 4 5 6 73 I have difficulty arranging my life in a way that is satisfying to me...... 1 2 3 4 5 6 74 I find it difficult to really open up when I talk with others...... 1 2 3... 4 5 6 75 I am concerned about how other people evaluate the choices I have made in my life...... 1 2 3 4 5 6 76 I gave up trying to make big improvements or changes in my life a long time ago...... 1 2 3 4 5 6 77 I find it satisfying to think about what I have accomplished in life...... 1 2 3 4 5 6 78 When I compare myself to friends and acquaintances, it makes me feel good about who I am...... 1 ...... 2 3 4 5 6

Strongly disagreeStrongly Disagree disagree Mildly Mildly agree Agree Strongly agree

79 I have been able to build a home and a lifestyle for myself that is much to my liking...... 1 2 3 4 5 6 80 My friends and I sympathise with each other’s problems...... 1 2 3 . 4 5 6 81 I judge myself by what I think is important, not by the values of what others think is important...... 1 ...... 2 3 4 5 6 82 There is truth to the saying you can’t teach an old dog new tricks...... 1 2 3 4 5 6 83 In the final analysis, I am not so sure that my life adds up to much...... 1 2 3 4 5 6 84 Everyone has their weaknesses, but I seem to have more than my share...... 1 2 ...... 3 4 5 6

BDI-II Scale

This questionnaire consists of 21 groups of statements. Please read each group of statements carefully, and then pick out the one statement in each group that best describes the way you have been feeling during the past two weeks, including today. Circle the number beside the statement you have picked. If several statements in the groups seem to apply equally well, circle the highest number for that group. Be sure that you do not choose more than one statement for any group, including Item 16 (Changes in Sleeping Pattern) or Item 18 (Changes in Appetite).

1 Sadness I do not feel sad...... 0 ...... I feel sad much of the time...... 1 ...... I am sad all of the time...... 2 ...... I am so sad or unhappy that I can’t stand it...... 3 ......

2 Pessimism I am not discouraged about my future...... 0 ...... I feel more discouraged about my future than I used to be...... 1 ...... I do not expect things to work out for me...... 2 ...... I feel my future is hopeless and will only get worse...... 3 ..

3 Past failure I do not feel like a failure...... 0 ...... I have failed more than I should have...... 1 ...... As I look back, I see a lot of failures...... 2 ...... I feel I am a total failure as a person...... 3 ......

4 Loss of pleasure I get as much pleasure as I ever did from the things I enjoy...... 0 ...... I don’t enjoy things as much as I used to...... 1 ...... I get very little pleasure from the things I used to enjoy...... 2 ...... I can’t get any pleasure from the things I used to enjoy...... 3 ......

5 Guilty feelings I don’t feel particularly guilty...... 0 ..... I feel guilty over many things I have done or should have done...... 1 ...... I feel quite guilty most of the time...... 2 ...... I feel guilty all of the time...... 3 ......

6 Punishment I don’t feel I am being punished...... 0 ...... feelings I feel I may be punished...... 1 ...... I expect to be punished...... 2 ...... I feel I am being punished...... 3 ......

7 Self-dislike I feel the same about myself as ever...... 0 ...... I have lost confidence in myself ...... 1...... I am disappointed in myself ...... 2 ...... I dislike myself...... 3 ......

8 Self-criticism I don’t criticise or blame myself more than usual...... 0 ...... I am more critical of myself than I used to be...... 1 ...... I criticise myself for all of my faults...... 2 ...... I blame myself for everything bad that happens...... 3 ......

9 Suicidal thoughts I don’t have any thoughts of killing myself...... 0 ...... or wishes I have thoughts of killing myself, but I would not carry them out...... 1 ...... I would like to kill myself ...... 2 ...... I would kill myself if I had the chance...... 3 ......

10 Crying I don’t cry any more than I used to...... 0 ...... I cry more than I used to...... 1 ...... I cry over every little thing...... 2 ...... I feel like crying, but I can’t...... 3 ......

11 Agitation I am no more restless or wound up than usual...... 0 ...... I feel more restless or wound up than usual...... 1 ...... I am so restless or agitated that it’s hard to stay still...... 2...... I am so restless or agitated that I have to keep moving or doing 3 something......

12 Loss of interest I have not lost interest in other people or activities...... 0 ... I am less interested in other people or things than before...... 1 ...... I have lost most of my interest in other people or things...... 2 ...... It’s hard to get interested in anything...... 3 ......

13 Indecisiveness I make decisions about as well as ever...... 0 ...... I find it more difficult to make decisions than usual...... 1 ... I have much greater difficulty in making decisions than I used to...... 2 ...... I have trouble making any decisions...... 3 ......

14 Worthlessness I do not feel I am worthless...... 0 ...... I don’t consider myself as worthwhile and useful as I used to...... 1 ...... I feel more worthless as compared to other people...... 2 .. I feel utterly worthless...... 3 ......

15 Loss of energy I have as much energy as ever...... 0...... I have less energy than I used to have...... 1 ...... I don’t have enough energy to do very much...... 2 ...... I don’t have enough energy to do anything...... 3 ......

16 Changes in I have not experienced any change in my sleeping pattern...... 0 ...... sleeping pattern I sleep somewhat more than usual...... 1a ...... I sleep somewhat less than usual...... 1b ...... I sleep a lot more than usual...... 2a .... I sleep a lot less than usual...... 2b ...... I sleep most of the day...... 3a ...... I wake up 1–2 hours early and can’t get back to sleep...... 3b ......

17 Irritability I am no more irritable than usual...... 0 ...... I am more irritable than usual...... 1 ... I am much more irritable than usual...... 2 ...... I am irritable all of the time...... 3 ......

18 Changes in I have not experienced any change in my appetite...... 0 ... appetite My appetite is somewhat less than usual...... 1a ...... My appetite is somewhat greater than usual...... 1b ...... My appetite is much less than usual...... 2a ...... My appetite is much greater than usual...... 2b ...... I have no appetite at all...... 3a ...... I crave food all the time...... 3b ......

19 Concentration I can concentrate as well as ever...... 0 ...... difficulty I can’t concentrate as well as ever...... 1 ...... It’s hard to keep my mind on anything for very long...... 2 . I find I can’t concentrate on anything...... 3 ......

20 Tiredness or I am no more tired or fatigued than usual...... 0 ...... fatigue I get more tired or fatigued than usual...... 1 ...... I am too tired or fatigued to do a lot of the things I used to do...... 2 ...... I am too tired or fatigued to do most of the things I used to do...... 3 ......

21 Loss of interest I have not noticed any recent change in my interest in sex...... 0 ...... in sex I am less interested in sex than I used to be...... 1 ...... I am much less interested in sex now...... 2 ...... I have lost interest in sex completely...... 3 ......

A.4 Questions of the Self-confrontation Method

Questions of the Self-confrontation method

Set 1: The Past

These questions are intended to guide you in reviewing one or more aspects of your life which may have been of great importance to you.

■ Has there been anything of major significance in your past life which still continues to exert a strong influence on you? Example: ‘After high school I never managed to build up a group of friends; I felt left out and stuck between two worlds’. ■ Was there in the past any person or persons, experience or circumstance which greatly influenced your life and still appreciably affects your present existence? Example: ‘My parents have always treated us as equals. They always tried to keep me stimulated. I could always talk very well with my parents’.

Set 2: The Present

This set again consists of two questions which will lead you, after a certain amount of reflection, to formulate a response:

■ Is there anything in your present existence which is of major importance to you or exerts a significant influence on you? Example: ‘Bodily complaints: Every time I find that I have a problem, my body signals it in one way or another’. ■ Is there in your present existence any person or persons or circumstance which exerts a significant influence on you? Example: ‘Wendy is a good friend’.

Set 3: The Future

The following questions will again guide you to a response:

■ Do you foresee anything that will be of great importance for, or exert a major influence on your future life? Example: ‘I want to pursue a combination of studying and volunteer work’. ■ Do you feel that a certain person, persons or circumstance will exert a significant influence on your future life? Example: ‘John and I want children in the future’. ■ Is there any future goal or object which you expect to play an important role in your life? Example: ‘I would like to meet somebody who would teach me how to have a normal life’.

Two final questions:

■ How do you generally feel these days? ■ How would you like to feel?

These questions are not answered by a formulation. Instead, you answer them by adding two extra tows in th matrix using the same list of affect terms.

A.5 Affect Terms for Rating Valuations

Affect Terms To what extent do you experience the following feelings in relation to the specific valuation you have formulated?

When you experience the feeling very much, give a 5 When you experience the feeling much, give a 4 When you experience the feeling rather much, give a 3 When you experience the feeling to some extent, give a 2 When you experience the feeling a little bit, give a 1 When you experience the feeling not at all, give a 0 1 Joy 13 Guilt 2 Powerlessness 14 Self-confidence 3 Self-esteem 15 Loneliness 4 Anxiety 16 Trust 5 Satisfaction 17 Inferiority 6 Strength 18 Intimacy 7 Shame 19 Safety 8 Enjoyment 20 Anger 9 Caring 21 Pride 10 Love 22 Energy 11 Self-alienation 23 Inner calm 12 Tenderness 24 Freedom

A.6 Matrix of Valuation / Affect

Matrix of Valuation x Affect

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 pnsnpsnpoononsnpnopnspppSOPN

Val. no.

Note: Rows represent valuations and columns represent affect terms used for the indices S, O, P and N. Each cell in the matrix indicates on 0-5 scale the extent to which a particular affect applies to a valuation.

APPENDIX B

STATISTICAL ANALYSIS FOR STUDY 1

Variables in the dataset

Variable name Description Affect scores affe_01..24 Mean affect scores for 24 affects Self-confrontation indices 1 Country of origin orign_s Self-enhancement (in country of origin) orign_o Union-with-the-other (in country of origin) orign_p Positive affect (in country of origin) orign_n Negative affect (in country of origin) Australia austr_s Self-enhancement (in Australia) austr_o Union-with-the-other (in Australia) austr_p Positive affect (in Australia) austr_n Negative affect (in Australia) General genrl_s Self-enhancement (general) genrl_o Union-with-the-other (general) genrl_p Positive affect (general) genrl_n Negative affect (general) Ideal ideal_s Self-enhancement (ideal) ideal_o Union-with-the-other (ideal) ideal_p Positive affect (ideal) ideal_n Negative affect (ideal) Psychological self-reports SAD sads_01..28 Individual item responses sads 2 Composite scale FNE fnes_01..30 Individual item responses fnes 2 Composite scale BDI-II bdi2_01..21 Individual item responses bdi2 2 Composite scale PWB pws__01..84 Individual item responses pws1 2 Environmental mastery subscale pws2 2 Positive relations with others subscale pws3 2 Autonomy subscale pws4 2 Personal growth subscale pws5 2 Purpose in life subscale pws6 2 Self-acceptance subscale Demographics dm_aged Age dm_gend Sex dm_marr Marital status dm_educ Education level dm_year Years in Australia dm_lang Language at home dm_refu Refugee status dm_ethn Ethnic identification dm_indi Individualism

Cluster variables Hierarchical clustering clu2_13 Cluster in country of origin clu2_23 Cluster in Australia K-means clustering qcl_13 Cluster in country of origin qcl_23 Cluster in Australia Type id_typed Group Notes 1 Self-confrontation indices were computed from affect scores before being entered into the SPSS data set. Z-scores of indices, which are computed as part of the clustering procedure, are prefixed by “z”: thus, zorign_s is the z-score of orign_s. 2 Computed in the course of the analysis. 3 Computed as part of the clustering procedure. 4 Computed from clu2_1 and clu2_2.

B.1 Reliability of Psychological Self-reports and Self-confrontation Indices

Cronbach’s alpha and other reliability statistics for the psychological self-reports and self-confrontation indices.

Social Anxiety and Distress Scale (SADS)

Case Processing Summary N % Cases Valid 38 100.0 Excludeda 0 .0 Total 38 100.0 a. Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha N of Items .930 28

Scale Statistics Mean Variance Std. Deviation N of Items 8.37 51.915 7.205 28

Fear of Negative Evaluation Scale (FNES)

Case Processing Summary N % Cases Valid 38 100.0 Excludeda 0 .0 Total 38 100.0 a. Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha N of Items .955 30

Scale Statistics Mean Variance Std. Deviation N of Items 13.26 90.740 9.526 30

BDI

Case Processing Summary N % Cases Valid 38 100.0 Excludeda 0 .0 Total 38 100.0 a.Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha N of Items .934 21

Scale Statistics Mean Variance Std. Deviation N of Items 9.05 87.403 9.349 21

Psychological Well-being Scale (PWS): Subscale 1

Case Processing Summary N % Cases Valid 38 100.0 Excludeda 0 .0 Total 38 100.0 a.Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha N of Items .910 14

Scale Statistics Mean Variance Std. Deviation N of Items 63.11 146.421 12.100 14

Psychological Well-being Scale (PWS): Subscale 2

Case Processing Summary N % Cases Valid 38 100.0 Excludeda 0 .0 Total 38 100.0 a.Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha N of Items .919 14

Scale Statistics Mean Variance Std. Deviation N of Items 67.05 145.943 12.081 14

Psychological Well-being Scale (PWS): Subscale 3

Case Processing Summary N % Cases Valid 38 100.0 Excludeda 0 .0 Total 38 100.0 a.Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha N of Items .884 14

Scale Statistics Mean Variance Std. Deviation N of Items 63.37 141.266 11.886 14

Psychological Well-being Scale (PWS): Subscale 4

Case Processing Summary N % Cases Valid 38 100.0 Excludeda 0 .0 Total 38 100.0 a.Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha N of Items .875 14

Scale Statistics Mean Variance Std. Deviation N of Items 72.66 64.447 8.028 14

Psychological Well-being Scale (PWS): Subscale 5

Case Processing Summary N % Cases Valid 38 100.0 Excludeda 0 .0 Total 38 100.0 a.Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha N of Items .866 14

Scale Statistics Mean Variance Std. Deviation N of Items 66.11 116.205 10.780 14

Psychological Well-being Scale (PWS): Subscale 6

Case Processing Summary N % Cases Valid 38 100.0 Excludeda 0 .0 Total 38 100.0 a.Listwise deletion based on all variables in the procedure.

Reliability Statistics Cronbach's Alpha N of Items .909 14

Scale Statistics Mean Variance Std. Deviation N of Items 64.55 154.038 12.411 14

S-index

Case Processing Summary N % Cases Valid 38 100.0 Excludeda 0 .0 Total 38 100.0 a. Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha N of Items .955 4

Scale Statistics Mean Variance Std. Deviation N of Items 10.8200 13.342 3.65265 4

O-index

Case Processing Summary N % Cases Valid 38 100.0 Excludeda 0 .0 Total 38 100.0 a. Listwise deletion based on all variables in the procedure.

Reliability Statistics Cronbach's Alpha N of Items .955 4

Scale Statistics Mean Variance Std. Deviation N of Items 9.9343 16.997 4.12274 4

P-index

Case Processing Summary N % Cases Valid 38 100.0 Excludeda 0 .0 Total 38 100.0 a. Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha N of Items .976 8

Scale Statistics Mean Variance Std. Deviation N of Items 20.3448 54.996 7.41591 8

N-index

Case Processing Summary N % Cases Valid 38 100.0 Excludeda 0 .0 Total 38 100.0 a. Listwise deletion based on all variables in the procedure. Reliability Statistics Cronbach's Alpha N of Items .962 8

Scale Statistics Mean Variance Std. Deviation N of Items 10.0915 30.929 5.56142 8

B.2 Tables

Descriptive statistics and tabulations for all variables. Categorical demographic variables disaggregated by sex are characterised by tabulations, interval demographic variables by descriptive statistics. Self-confrontation indices and psychological self- reports, are characterised by descriptive statistics and correlation coefficients.

Participant “Type” is defined in the main report. For convenience the following names were assigned to each type as follows: Type 1 = “Progressive”, Type 2 = “Low- stable”, Type 3 = “High-stable”, and Type 4 = “Regressive”.

Demographics (nominal) by Sex and Type

Case Processing Summary Cases Valid Missing Total N Percent N Percent N Percent Marital status * Sex 38 100.0% 0 .0% 38 100.0% Education * Sex 38 100.0% 0 .0% 38 100.0% Language at home * Sex 38 100.0% 0 .0% 38 100.0% Ethnic identity * Sex 38 100.0% 0 .0% 38 100.0%

Marital status * Sex Crosstab Count Sex

1 2 Total Marital status 1 4 6 10 2 7 15 22 3 3 3 6 Total 14 24 38

Chi-Square Tests Asymp. Sig. (2-

Value df sided) Pearson Chi-Square .728a 2 .695 Likelihood Ratio .717 2 .699 Linear-by-Linear Association .060 1 .806 N of Valid Cases 38 a. 3 cells (50.0%) have expected count less than 5. The minimum expected count is 2.21.

Education * Sex Crosstab Count Sex

1 2 Total Education 1 0 2 2 2 0 1 1 4 1 3 4 5 5 4 9 6 8 14 22 Total 14 24 38

Chi-Square Tests Asymp. Sig. (2-

Value df sided) Pearson Chi-Square 3.348a 4 .501 Likelihood Ratio 4.311 4 .366 Linear-by-Linear Association 1.245 1 .265 N of Valid Cases 38 a. 7 cells (70.0%) have expected count less than 5. The minimum expected count is .37. Language at home * Sex Crosstab Count Sex

1 2 Total Language at home 1 7 8 15 2 0 6 6 3 7 10 17 Total 14 24 38

Chi-Square Tests Asymp. Sig. (2-

Value df sided) Pearson Chi-Square 4.259a 2 .119 Likelihood Ratio 6.254 2 .044 Linear-by-Linear Association .071 1 .790 N of Valid Cases 38 a. 2 cells (33.3%) have expected count less than 5. The minimum expected count is 2.21. Ethnic identity * Sex Crosstab Count Sex

1 2 Total Ethnic identity 1 1 7 8 2 13 17 30 Total 14 24 38

Chi-Square Tests Asymp. Sig. (2- Exact Sig. (2- Exact Sig. (1-

Value df sided) sided) sided) Pearson Chi-Square 2.580a 1 .108 Continuity Correctionb 1.425 1 .233 Likelihood Ratio 2.934 1 .087 Fisher's Exact Test .216 .114 Linear-by-Linear Association 2.513 1 .113 N of Valid Cases 38 a. 1 cells (25.0%) have expected count less than 5. The minimum expected count is 2.95. b. Computed only for a 2x2 table

Demographics (nominal variables) by type

Case Processing Summary Cases Valid Missing Total N Percent N Percent N Percent Sex * Type 38 100.0% 0 .0% 38 100.0% Marital status * Type 38 100.0% 0 .0% 38 100.0% Education * Type 38 100.0% 0 .0% 38 100.0% Language at home * Type 38 100.0% 0 .0% 38 100.0% Ethnic identity * Type 38 100.0% 0 .0% 38 100.0%

Sex * Type Crosstab Count Type

1 2 3 4 Total Sex 1 1 6 5 2 14 2 2 12 6 4 24 Total 3 18 11 6 38

Chi-Square Tests Asymp. Sig. (2-

Value df sided) Pearson Chi-Square .494a 3 .920 Likelihood Ratio .486 3 .922 Linear-by-Linear Association .061 1 .805 N of Valid Cases 38 a. 5 cells (62.5%) have expected count less than 5. The minimum expected count is 1.11. Marital status * Type Crosstab Count Type

1 2 3 4 Total Marital status 1 0 5 3 2 10 2 3 10 6 3 22 3 0 3 2 1 6 Total 3 18 11 6 38

Chi-Square Tests Asymp. Sig. (2-

Value df sided) Pearson Chi-Square 2.467a 6 .872 Likelihood Ratio 3.557 6 .736 Linear-by-Linear Association .069 1 .792 N of Valid Cases 38 a. 10 cells (83.3%) have expected count less than 5. The minimum expected count is .47. Education * Type Crosstab Count Type

1 2 3 4 Total Education 1 1 1 0 0 2 2 0 1 0 0 1 4 0 2 1 1 4 5 0 5 3 1 9 6 2 9 7 4 22 Total 3 18 11 6 38

Chi-Square Tests Asymp. Sig. (2-

Value df sided) Pearson Chi-Square 8.457a 12 .749 Likelihood Ratio 8.252 12 .765 Linear-by-Linear Association 2.181 1 .140 N of Valid Cases 38 a. 18 cells (90.0%) have expected count less than 5. The minimum expected count is .08. Language at home * Type Crosstab Count Type

1 2 3 4 Total Language at home 1 1 8 5 1 15 2 2 1 2 1 6 3 0 9 4 4 17 Total 3 18 11 6 38

Chi-Square Tests Value df Asymp. Sig. (2-sided) Pearson Chi-Square 9.457a 6 .149 Likelihood Ratio 9.126 6 .167 Linear-by-Linear Association 1.034 1 .309 N of Valid Cases 38 a. 10 cells (83.3%) have expected count less than 5. The minimum expected count is .47.

Ethnic identity * Type Crosstab Count Type

1 2 3 4 Total Ethnic identity 1 3 3 2 0 8 2 0 15 9 6 30 Total 3 18 11 6 38

Chi-Square Tests Asymp. Sig. (2-

Value df sided) Pearson Chi-Square 13.113a 3 .004 Likelihood Ratio 12.462 3 .006 Linear-by-Linear Association 5.789 1 .016 N of Valid Cases 38 a. 6 cells (75.0%) have expected count less than 5. The minimum expected count is .63.

Oneway anova: Demographics by Sex and Type

Descriptives N Mean Std. Deviation Std. Error Years in Australia 1 14 18.71 10.738 2.870 2 24 20.38 10.586 2.161 Total 38 19.76 10.528 1.708 Age 1 14 41.07 10.411 2.782 2 24 42.17 9.613 1.962 Total 38 41.76 9.788 1.588 Individualism 1 14 38.43 12.457 3.329 2 24 43.13 24.519 5.005 Total 38 41.39 20.820 3.378

Descriptives 95% Confidence Interval for Mean

Lower Bound Upper Bound Minimum Maximum Years in Australia 1 12.51 24.91 4 34 2 15.90 24.85 5 42 Total 16.30 23.22 4 42 Age 1 35.06 47.08 23 57 2 38.11 46.23 21 58 Total 38.55 44.98 21 58 Individualism 1 31.24 45.62 19 65 2 32.77 53.48 13 91 Total 34.55 48.24 13 91

ANOVA Sum of Squares df Mean Square Years in Australia Between Groups 24.386 1 24.386 Within Groups 4076.482 36 113.236 Total 4100.868 37 Age Between Groups 10.607 1 10.607 Within Groups 3534.262 36 98.174 Total 3544.868 37 Individualism Between Groups 195.025 1 195.025 Within Groups 15844.054 36 440.113 Total 16039.079 37

ANOVA F Sig. Years in Australia Between Groups .215 .645 Within Groups Total Age Between Groups .108 .744 Within Groups Total Individualism Between Groups .443 .510 Within Groups Total

Oneway

Descriptives N Mean Std. Deviation Std. Error Years in Australia 1 3 31.00 11.000 6.351 2 18 16.44 10.562 2.489 3 11 19.64 8.936 2.694 4 6 24.33 9.606 3.921 Total 38 19.76 10.528 1.708 Age 1 3 50.67 7.506 4.333 2 18 39.39 9.936 2.342 3 11 42.82 8.750 2.638 4 6 42.50 11.256 4.595 Total 38 41.76 9.788 1.588 Individualism 1 3 42.33 31.723 18.315 2 18 44.33 22.316 5.260 3 11 39.82 21.325 6.430 4 6 35.00 10.789 4.405 Total 38 41.39 20.820 3.378

Descriptives 95% Confidence Interval for Mean

Lower Bound Upper Bound Minimum Maximum Years in Australia 1 3.67 58.33 20 42 2 11.19 21.70 4 42 3 13.63 25.64 5 34 4 14.25 34.41 6 32 Total 16.30 23.22 4 42 Age 1 32.02 69.31 43 58 2 34.45 44.33 21 57 3 36.94 48.70 28 58 4 30.69 54.31 24 57 Total 38.55 44.98 21 58 Individualism 1 -36.47 121.14 13 76 2 33.24 55.43 19 89 3 25.49 54.14 19 91 4 23.68 46.32 20 51 Total 34.55 48.24 13 91

ANOVA Sum of Squares df Mean Square Years in Australia Between Groups 702.545 3 234.182 Within Groups 3398.323 34 99.951 Total 4100.868 37 Age Between Groups 354.788 3 118.263 Within Groups 3190.081 34 93.826 Total 3544.868 37 Individualism Between Groups 430.776 3 143.592 Within Groups 15608.303 34 459.068 Total 16039.079 37

ANOVA F Sig. Years in Australia Between Groups 2.343 .090 Within Groups Total Age Between Groups 1.260 .303 Within Groups Total Individualism Between Groups .313 .816 Within Groups Total

T-test: Demographics (scale variables) by Sex

T-Test Group Statistics Sex N Mean Std. Deviation Std. Error Mean Age 1 14 41.07 10.411 2.782 2 24 42.17 9.613 1.962 Years in Australia 1 14 18.71 10.738 2.870 2 24 20.38 10.586 2.161 Individualism 1 14 38.43 12.457 3.329 2 24 43.13 24.519 5.005

Independent Samples Test Levene's Test for Equality of Variances

F Sig. Age Equal variances assumed .072 .790 Equal variances not assumed Years in Australia Equal variances assumed .185 .670 Equal variances not assumed Individualism Equal variances assumed 7.358 .010 Equal variances not assumed

Independent Samples Test t-test for Equality of Means

Mean t df Sig. (2-tailed) Difference Age Equal variances assumed -.329 36 .744 -1.095 Equal variances not assumed -.322 25.572 .750 -1.095 Years in Australia Equal variances assumed -.464 36 .645 -1.661 Equal variances not assumed -.462 27.013 .648 -1.661 Individualism Equal variances assumed -.666 36 .510 -4.696 Equal variances not assumed -.781 35.545 .440 -4.696

Independent Samples Test t-test for Equality of Means 95% Confidence Interval of the Difference Std. Error Difference Lower Upper Age Equal variances assumed 3.332 -7.853 5.663 Equal variances not assumed 3.405 -8.099 5.909 Years in Australia Equal variances assumed 3.579 -8.918 5.597 Equal variances not assumed 3.592 -9.031 5.710 Individualism Equal variances assumed 7.055 -19.005 9.612 Equal variances not assumed 6.011 -16.893 7.500

B.3 Cluster Analysis

 Hierarchical cluster analysis using Ward’s method on standardised (z-score)

indices for the country of origin I-position; saves membership of the highest two

clusters as variable clu2_1.

 Cluster analysis as above for Australia; saves membership of the highest two

clusters as variable clu2_2.

 Calculates and saves standardised (z-score) means for use in K-means clustering.

 K-means cluster analysis on standardised (z-score) indices for the country of origin

I-position (where clusters are seeded with z-score means); saves membership of the

highest two clusters as variable qcl_1; cross-tabulates hierarchical cluster

membership with K-means cluster membership.

 Cluster analysis as above for Australia, and saves membership of the highest two

clusters as variable qcl_2; repeats cross-tabulation as above.

 Creates the variable id_type from clu2_1 and clu2_2 as follows:

clu2_2 = 1 clu2_2 = 2

clu2_1 = 1 id_type = 1 id_type = 2

clu2_1 = 2 id_type = 3 id_type = 4

Country of origin

Ward method Proximities Case Processing Summarya Cases Valid Missing Total N Percent N Percent N Percent 38 100.0% 0 .0% 38 100.0% a. Squared Euclidean Distance used Ward Linkage Agglomeration Schedule Cluster Combined Stage Cluster First Appears Stage Cluster 1 Cluster 2 Coefficients Cluster 1 Cluster 2 Next Stage 1 4 12 .040 0 0 20 2 18 35 .119 0 0 15 3 8 24 .208 0 0 18 4 9 33 .299 0 0 14 5 11 36 .402 0 0 16 6 16 19 .524 0 0 14 7 23 26 .662 0 0 26 8 5 10 .827 0 0 22 9 7 22 1.032 0 0 23 10 20 37 1.260 0 0 19 11 1 28 1.501 0 0 26 12 21 38 1.798 0 0 23 13 27 31 2.143 0 0 19 14 9 16 2.489 4 6 25 15 15 18 2.848 0 2 27 16 11 29 3.235 5 0 27 17 3 25 3.638 0 0 24 18 8 30 4.045 3 0 25 19 20 27 4.555 10 13 22 20 4 34 5.108 1 0 29 21 6 32 5.671 0 0 32 22 5 20 6.303 8 19 29 23 7 21 7.014 9 12 32 24 3 17 7.752 17 0 28 25 8 9 8.612 18 14 31 26 1 23 9.663 11 7 33 27 11 15 10.929 16 15 30 28 3 13 12.756 24 0 33 29 4 5 15.184 20 22 35 30 2 11 17.717 0 27 36 31 8 14 20.627 25 0 35 32 6 7 23.543 21 23 34 33 1 3 27.233 26 28 34 34 1 6 38.928 33 32 36 35 4 8 50.686 29 31 37 36 1 2 73.989 34 30 37 37 1 4 148.000 36 35 0

Means Case Processing Summary Cases Included Excluded Total N Percent N Percent N Percent S origin * Ward in Origin 38 100.0% 0 .0% 38 100.0% O origin * Ward in Origin 38 100.0% 0 .0% 38 100.0% P origin * Ward in Origin 38 100.0% 0 .0% 38 100.0% N origin * Ward in Origin 38 100.0% 0 .0% 38 100.0%

Report Ward in Origin S origin O origin P origin N origin 1 Mean 7.1540 6.6386 13.2868 10.8255 Std. Deviation 2.54982 3.21958 4.93917 6.79496 N 21 21 21 21 2 Mean 13.5497 13.1280 26.3121 9.0574 Std. Deviation 1.78377 1.92550 3.43412 5.26844 N 17 17 17 17 Total Mean 10.0153 9.5418 19.1139 10.0345 Std. Deviation 3.90851 4.23074 7.83357 6.14444 N 38 38 38 38

Means Case Processing Summary Cases Included Excluded Total N Percent N Percent N Percent Zscore: S origin * Ward in Origin 38 100.0% 0 .0% 38 100.0% Zscore: O origin * Ward in Origin 38 100.0% 0 .0% 38 100.0% Zscore: P origin * Ward in Origin 38 100.0% 0 .0% 38 100.0% Zscore: N origin * Ward in Origin 38 100.0% 0 .0% 38 100.0%

Report Mean Ward in Origin Zscore: S origin Zscore: O origin Zscore: P origin Zscore: N origin 1 -.7320484 -.6862045 -.7438639 .1287364 2 .9042951 .8476644 .9188907 -.1590274 Total .0000000 .0000000 .0000000 .0000000

Quick Cluster Initial Cluster Centers Cluster

1 2 Zscore: S origin -0.732048 0.904295 Zscore: O origin -0.686205 0.847664 Zscore: P origin -0.743864 0.918891 Zscore: N origin 0.128736 -0.159027 Input from INITIAL Subcommand

Iteration Historya Change in Cluster Centers Iteration 1 2 1 .189 .132 2 .000 .000 a.Convergence achieved due to no or small change in cluster centers. The maximum absolute coordinate change for any center is .000. The current iteration is 2. The minimum distance between initial centers is 2.807. Final Cluster Centers Cluster

1 2 Zscore: S origin -.84092 .84092 Zscore: O origin -.79893 .79893 Zscore: P origin -.84096 .84096 Zscore: N origin .08936 -.08936

ANOVA Cluster Error

Mean Square df Mean Square df F Sig. Zscore: S origin 26.871 1 .281 36 95.507 .000 Zscore: O origin 24.255 1 .354 36 68.512 .000 Zscore: P origin 26.874 1 .281 36 95.545 .000 Zscore: N origin .303 1 1.019 36 .298 .589 The F tests should be used only for descriptive purposes because the clusters have been chosen to maximize the differences among cases in different clusters. The observed significance levels are not corrected for this and thus cannot be interpreted as tests of the hypothesis that the cluster means are equal. Number of Cases in each Cluster Cluster 1 19.000 2 19.000 Valid 38.000 Missing .000

Crosstabs Case Processing Summary Cases Valid Missing Total N Percent N Percent N Percent Kmeans in Origin * Ward in 38 100.0% 0 .0% 38 100.0% Origin

Kmeans in Origin * Ward in Origin Crosstabulation Count Ward in Origin

1 2 Total Kmeans in Origin 1 19 0 19 2 2 17 19 Total 21 17 38

Chi-Square Tests Asymp. Sig. (2- Exact Sig. (2- Exact Sig. (1-

Value df sided) sided) sided) Pearson Chi-Square 30.762a 1 .000 Continuity Correctionb 27.249 1 .000 Likelihood Ratio 39.471 1 .000 Fisher's Exact Test .000 .000 Linear-by-Linear Association 29.952 1 .000 N of Valid Cases 38 a.0 cells (.0%) have expected count less than 5. The minimum expected count is 8.50. b.Computed only for a 2x2 table T-Test Group Statistics Ward in Origin N Mean Std. Deviation Std. Error Mean S origin 1 21 7.1540 2.54982 .55642 2 17 13.5497 1.78377 .43263 O origin 1 21 6.6386 3.21958 .70257 2 17 13.1280 1.92550 .46700 P origin 1 21 13.2868 4.93917 1.07782 2 17 26.3121 3.43412 .83290 N origin 1 21 10.8255 6.79496 1.48278 2 17 9.0574 5.26844 1.27778

Independent Samples Test Levene's Test for Equality of Variances

F Sig. S origin Equal variances assumed 3.811 .059 Equal variances not assumed O origin Equal variances assumed 5.443 .025 Equal variances not assumed P origin Equal variances assumed 3.408 .073 Equal variances not assumed N origin Equal variances assumed 2.218 .145 Equal variances not assumed

Independent Samples Test t-test for Equality of Means

t df Sig. (2-tailed) Mean Difference S origin Equal variances assumed -8.744 36 .000 -6.39566 Equal variances not assumed -9.074 35.344 .000 -6.39566 O origin Equal variances assumed -7.309 36 .000 -6.48940 Equal variances not assumed -7.692 33.422 .000 -6.48940 P origin Equal variances assumed -9.209 36 .000 -13.02530 Equal variances not assumed -9.562 35.289 .000 -13.02530 N origin Equal variances assumed .879 36 .385 1.76815 Equal variances not assumed .903 35.951 .372 1.76815

Independent Samples Test t-test for Equality of Means 95% Confidence Interval of the Difference Std. Error Difference Lower Upper S origin Equal variances assumed .73143 -7.87908 -4.91224 Equal variances not assumed .70482 -7.82602 -4.96531 O origin Equal variances assumed .88790 -8.29014 -4.68865 Equal variances not assumed .84362 -8.20493 -4.77386 P origin Equal variances assumed 1.41440 -15.89384 -10.15676 Equal variances not assumed 1.36213 -15.78977 -10.26083 N origin Equal variances assumed 2.01083 -2.31000 5.84630 Equal variances not assumed 1.95739 -2.20181 5.73811

Australia

Proximities Case Processing Summarya Cases Valid Missing Total N Percent N Percent N Percent 38 100.0% 0 .0% 38 100.0% a. Squared Euclidean Distance used Cluster Ward Linkage Agglomeration Schedule Cluster Combined Stage Cluster First Appears Stage Cluster 1 Cluster 2 Coefficients Cluster 1 Cluster 2 Next Stage 1 14 33 .028 0 0 11 2 1 24 .060 0 0 25 3 22 29 .143 0 0 9 4 9 38 .279 0 0 21 5 11 36 .418 0 0 19 6 26 28 .562 0 0 15 7 13 17 .712 0 0 18 8 8 12 .895 0 0 17 9 4 22 1.080 0 3 20 10 6 20 1.284 0 0 16 11 5 14 1.515 0 1 25 12 7 37 1.749 0 0 29 13 31 32 1.999 0 0 22 14 10 27 2.290 0 0 24 15 19 26 2.612 0 6 27 16 6 16 2.941 10 0 21 17 3 8 3.285 0 8 31 18 13 18 3.648 7 0 27 19 11 35 4.113 5 0 26 20 4 23 4.633 9 0 29 21 6 9 5.174 16 4 32 22 30 31 5.769 0 13 24 23 2 21 6.408 0 0 35 24 10 30 7.176 14 22 28 25 1 5 7.968 2 11 31 26 11 25 9.157 19 0 30 27 13 19 10.810 18 15 34 28 10 34 12.805 24 0 33 29 4 7 15.007 20 12 32 30 11 15 17.411 26 0 34 31 1 3 19.911 25 17 33 32 4 6 24.271 29 21 35 33 1 10 31.450 31 28 37 34 11 13 38.781 30 27 36 35 2 4 49.251 23 32 36 36 2 11 68.678 35 34 37 37 1 2 148.000 33 36 0

Means Case Processing Summary Cases Included Excluded Total N Percent N Percent N Percent S Australia * Ward in 38 100.0% 0 .0% 38 100.0% Australia O Australia * Ward in 38 100.0% 0 .0% 38 100.0% Australia P Australia * Ward in 38 100.0% 0 .0% 38 100.0% Australia N Australia * Ward in 38 100.0% 0 .0% 38 100.0% Australia

Report Ward in Australia S Australia O Australia P Australia N Australia 1 Mean 15.5908 14.5132 29.9133 7.2324 Std. Deviation 1.78843 2.24427 3.08485 4.77435 N 14 14 14 14 2 Mean 9.0305 7.4794 16.0628 12.3227 Std. Deviation 2.37237 3.30684 4.78490 6.56004 N 24 24 24 24 Total Mean 11.4474 10.0708 21.1656 10.4473 Std. Deviation 3.86103 4.51559 7.96366 6.39937 N 38 38 38 38

Means Case Processing Summary Cases Included Excluded Total N Percent N Percent N Percent Zscore: S Australia * Ward 38 100.0% 0 .0% 38 100.0% in Australia Zscore: O Australia * Ward 38 100.0% 0 .0% 38 100.0% in Australia Zscore: P Australia * Ward 38 100.0% 0 .0% 38 100.0% in Australia Zscore: N Australia * Ward 38 100.0% 0 .0% 38 100.0% in Australia

Report Mean Zscore: S Zscore: O Zscore: P Zscore: N Ward in Australia Australia Australia Australia Australia 1 1.0731323 .9837986 1.0984490 -.5023820 2 -.6259939 -.5738825 -.6407619 .2930562 Total .0000000 .0000000 .0000000 .0000000

Quick Cluster Initial Cluster Centers Cluster

1 2 Zscore: S Australia 1.073132 -0.625994 Zscore: O Australia 0.983799 -0.573883 Zscore: P Australia 1.098449 -0.640762 Zscore: N Australia -0.502382 0.293056 Input from INITIAL Subcommand Iteration Historya Change in Cluster Centers Iteration 1 2 1 5.175E-7 5.379E-7 2 .000 .000 a.Convergence achieved due to no or small change in cluster centers. The maximum absolute coordinate change for any center is .000. The current iteration is 2. The minimum distance between initial centers is 2.995. Final Cluster Centers Cluster

1 2 Zscore: S Australia 1.07313 -.62599 Zscore: O Australia .98380 -.57388 Zscore: P Australia 1.09845 -.64076 Zscore: N Australia -.50238 .29306

ANOVA Cluster Error

Mean Square df Mean Square df F Sig. Zscore: S Australia 25.527 1 .319 36 80.103 .000 Zscore: O Australia 21.454 1 .432 36 49.682 .000 Zscore: P Australia 26.746 1 .285 36 93.902 .000 Zscore: N Australia 5.595 1 .872 36 6.413 .016 The F tests should be used only for descriptive purposes because the clusters have been chosen to maximize the differences among cases in different clusters. The observed significance levels are not corrected for this and thus cannot be interpreted as tests of the hypothesis that the cluster means are equal. Number of Cases in each Cluster Cluster 1 14.000 2 24.000 Valid 38.000 Missing .000

Crosstabs Case Processing Summary Cases Valid Missing Total N Percent N Percent N Percent Kmeans in Australia * Ward 38 100.0% 0 .0% 38 100.0% in Australia

Kmeans in Australia * Ward in Australia Crosstabulation Count Ward in Australia

1 2 Total Kmeans in Australia 1 14 0 14 2 0 24 24 Total 14 24 38

Chi-Square Tests Asymp. Sig. (2- Exact Sig. (2- Exact Sig. (1-

Value df sided) sided) sided) Pearson Chi-Square 38.000a 1 .000 Continuity Correctionb 33.824 1 .000 Likelihood Ratio 50.016 1 .000 Fisher's Exact Test .000 .000 Linear-by-Linear Association 37.000 1 .000 N of Valid Cases 38 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.16. b. Computed only for a 2x2 table T-Test Group Statistics Ward in Australia N Mean Std. Deviation Std. Error Mean S Australia 1 14 15.5908 1.78843 .47798 2 24 9.0305 2.37237 .48426 O Australia 1 14 14.5132 2.24427 .59981 2 24 7.4794 3.30684 .67501 P Australia 1 14 29.9133 3.08485 .82446 2 24 16.0628 4.78490 .97671 N Australia 1 14 7.2324 4.77435 1.27600 2 24 12.3227 6.56004 1.33906

Independent Samples Test Levene's Test for Equality of Variances

F Sig. S Australia Equal variances assumed .376 .544 Equal variances not assumed O Australia Equal variances assumed 5.840 .021 Equal variances not assumed P Australia Equal variances assumed 1.654 .207 Equal variances not assumed N Australia Equal variances assumed 3.219 .081 Equal variances not assumed

Independent Samples Test t-test for Equality of Means

t df Sig. (2-tailed) Mean Difference S Australia Equal variances assumed 8.950 36 .000 6.56037 Equal variances not assumed 9.642 33.459 .000 6.56037 O Australia Equal variances assumed 7.049 36 .000 7.03385 Equal variances not assumed 7.789 35.026 .000 7.03385 P Australia Equal variances assumed 9.690 36 .000 13.85049 Equal variances not assumed 10.836 35.535 .000 13.85049 N Australia Equal variances assumed -2.532 36 .016 -5.09030 Equal variances not assumed -2.752 34.055 .009 -5.09030

Independent Samples Test t-test for Equality of Means 95% Confidence Interval of the Difference Std. Error Difference Lower Upper S Australia Equal variances assumed .73300 5.07378 8.04697 Equal variances not assumed .68042 5.17677 7.94397 O Australia Equal variances assumed .99791 5.00999 9.05771 Equal variances not assumed .90300 5.20072 8.86698 P Australia Equal variances assumed 1.42932 10.95170 16.74928 Equal variances not assumed 1.27816 11.25707 16.44391 N Australia Equal variances assumed 2.01007 -9.16691 -1.01370 Equal variances not assumed 1.84967 -8.84905 -1.33156

Both country of origin and Australia

Crosstabs Case Processing Summary Cases Valid Missing Total N Percent N Percent N Percent Ward in Origin * Ward in 38 100.0% 0 .0% 38 100.0% Australia

Ward in Origin * Ward in Australia Crosstabulation Count Ward in Australia

1 2 Total Ward in Origin 1 3 18 21 2 11 6 17 Total 14 24 38

Chi-Square Tests Asymp. Sig. (2- Exact Sig. (2- Exact Sig. (1-

Value df sided) sided) sided) Pearson Chi-Square 10.264a 1 .001 Continuity Correctionb 8.212 1 .004 Likelihood Ratio 10.717 1 .001 Fisher's Exact Test .002 .002 Linear-by-Linear Association 9.994 1 .002 N of Valid Cases 38 a. 0 cells (.0%) have expected count less than 5. The minimum expected count is 6.26. b. Computed only for a 2x2 table

B.4 Repeated-measures MANOVA

Repeated-measures MANOVA on indices distinguished by a categorical variable having two values: country of origin and Australia.

General Linear Model Within-Subjects Factors Dependent Measure place Variable index_s 1 austr_s 2 orign_s index_o 1 austr_o 2 orign_o index_p 1 austr_p 2 orign_p index_n 1 austr_n 2 orign_n

Between-Subjects Factors N Type 1 3 2 18 3 11 4 6

Multivariate Testsc Effect Value F Hypothesis df Between Subjects Intercept Pillai's Trace .967 230.709a 4.000 Wilks' Lambda .033 230.709a 4.000 Hotelling's Trace 29.769 230.709a 4.000 Roy's Largest Root 29.769 230.709a 4.000 id_type Pillai's Trace .879 3.417 12.000 Wilks' Lambda .196 5.837 12.000 Hotelling's Trace 3.726 9.213 12.000 Roy's Largest Root 3.627 29.921b 4.000 Within Subjects place Pillai's Trace .494 7.553a 4.000 Wilks' Lambda .506 7.553a 4.000 Hotelling's Trace .975 7.553a 4.000 Roy's Largest Root .975 7.553a 4.000 place * id_type Pillai's Trace .824 3.126 12.000 Wilks' Lambda .282 4.207 12.000 Hotelling's Trace 2.178 5.385 12.000 Roy's Largest Root 2.003 16.524b 4.000

Multivariate Testsc Partial Eta Effect Error df Sig. Squared Between Subjects Intercept Pillai's Trace 31.000 .000 .967 Wilks' Lambda 31.000 .000 .967 Hotelling's Trace 31.000 .000 .967 Roy's Largest Root 31.000 .000 .967 id_type Pillai's Trace 99.000 .000 .293 Wilks' Lambda 82.310 .000 .419 Hotelling's Trace 89.000 .000 .554 Roy's Largest Root 33.000 .000 .784 Within Subjects place Pillai's Trace 31.000 .000 .494 Wilks' Lambda 31.000 .000 .494 Hotelling's Trace 31.000 .000 .494 Roy's Largest Root 31.000 .000 .494 place * id_type Pillai's Trace 99.000 .001 .275 Wilks' Lambda 82.310 .000 .344 Hotelling's Trace 89.000 .000 .421 Roy's Largest Root 33.000 .000 .667 a. Exact statistic b. The statistic is an upper bound on F that yields a lower bound on the significance level. c. Design: Intercept + id_type Within Subjects Design: place Mauchly's Test of Sphericityb Approx. Chi- Within Subjects Effect Measure Mauchly's W Square df Sig. place index_s 1.000 .000 0 . index_o 1.000 .000 0 . index_p 1.000 .000 0 . index_n 1.000 .000 0 .

Mauchly's Test of Sphericityb Epsilona Greenhouse- Within Subjects Effect Measure Geisser Huynh-Feldt Lower-bound place index_s 1.000 1.000 1.000 index_o 1.000 1.000 1.000 index_p 1.000 1.000 1.000 index_n 1.000 1.000 1.000 Tests the null hypothesis that the error covariance matrix of the orthonormalized transformed dependent variables is proportional to an identity matrix. a. May be used to adjust the degrees of freedom for the averaged tests of significance. Corrected tests are displayed in the Tests of Within-Subjects Effects table. b. Design: Intercept + id_type Within Subjects Design: place

Tests of Within-Subjects Effects Multivariatecd Within Subjects Effect Value F Hypothesis df Error df place Pillai's Trace .494 7.553a 4.000 31.000 Wilks' Lambda .506 7.553a 4.000 31.000 Hotelling's Trace .975 7.553a 4.000 31.000 Roy's Largest Root .975 7.553a 4.000 31.000 place * id_type Pillai's Trace .824 3.126 12.000 99.000 Wilks' Lambda .282 4.207 12.000 82.310 Hotelling's Trace 2.178 5.385 12.000 89.000 Roy's Largest Root 2.003 16.524b 4.000 33.000

Multivariatecd Partial Eta Within Subjects Effect Sig. Squared place Pillai's Trace .000 .494 Wilks' Lambda .000 .494 Hotelling's Trace .000 .494 Roy's Largest Root .000 .494 place * id_type Pillai's Trace .001 .275 Wilks' Lambda .000 .344 Hotelling's Trace .000 .421 Roy's Largest Root .000 .667 a. Exact statistic b. The statistic is an upper bound on F that yields a lower bound on the significance level. c. Design: Intercept + id_type Within Subjects Design: place d. Tests are based on averaged variables.

Univariate Tests Type III Sum of Source Measure Squares df Mean Square F place index_s Sphericity Assumed 36.928 1 36.928 22.185 Greenhouse-Geisser 36.928 1.000 36.928 22.185 Huynh-Feldt 36.928 1.000 36.928 22.185 Lower-bound 36.928 1.000 36.928 22.185 index_o Sphericity Assumed 5.475 1 5.475 2.139 Greenhouse-Geisser 5.475 1.000 5.475 2.139 Huynh-Feldt 5.475 1.000 5.475 2.139 Lower-bound 5.475 1.000 5.475 2.139 index_p Sphericity Assumed 61.170 1 61.170 11.525 Greenhouse-Geisser 61.170 1.000 61.170 11.525 Huynh-Feldt 61.170 1.000 61.170 11.525 Lower-bound 61.170 1.000 61.170 11.525 index_n Sphericity Assumed .294 1 .294 .038 Greenhouse-Geisser .294 1.000 .294 .038 Huynh-Feldt .294 1.000 .294 .038 Lower-bound .294 1.000 .294 .038 place * id_type index_s Sphericity Assumed 75.412 3 25.137 15.101 Greenhouse-Geisser 75.412 3.000 25.137 15.101 Huynh-Feldt 75.412 3.000 25.137 15.101 Lower-bound 75.412 3.000 25.137 15.101 index_o Sphericity Assumed 69.595 3 23.198 9.062 Greenhouse-Geisser 69.595 3.000 23.198 9.062 Huynh-Feldt 69.595 3.000 23.198 9.062 Lower-bound 69.595 3.000 23.198 9.062 index_p Sphericity Assumed 348.591 3 116.197 21.892 Greenhouse-Geisser 348.591 3.000 116.197 21.892 Huynh-Feldt 348.591 3.000 116.197 21.892 Lower-bound 348.591 3.000 116.197 21.892 index_n Sphericity Assumed 233.149 3 77.716 10.143 Greenhouse-Geisser 233.149 3.000 77.716 10.143 Huynh-Feldt 233.149 3.000 77.716 10.143 Lower-bound 233.149 3.000 77.716 10.143 Error(place) index_s Sphericity Assumed 56.595 34 1.665 Greenhouse-Geisser 56.595 34.000 1.665 Huynh-Feldt 56.595 34.000 1.665 Lower-bound 56.595 34.000 1.665 index_o Sphericity Assumed 87.043 34 2.560 Greenhouse-Geisser 87.043 34.000 2.560 Huynh-Feldt 87.043 34.000 2.560 Lower-bound 87.043 34.000 2.560 index_p Sphericity Assumed 180.459 34 5.308 Greenhouse-Geisser 180.459 34.000 5.308 Huynh-Feldt 180.459 34.000 5.308 Lower-bound 180.459 34.000 5.308 index_n Sphericity Assumed 260.503 34 7.662 Greenhouse-Geisser 260.503 34.000 7.662 Huynh-Feldt 260.503 34.000 7.662 Lower-bound 260.503 34.000 7.662

Univariate Tests Partial Eta Source Measure Sig. Squared place index_s Sphericity Assumed .000 .395 Greenhouse-Geisser .000 .395 Huynh-Feldt .000 .395 Lower-bound .000 .395 index_o Sphericity Assumed .153 .059 Greenhouse-Geisser .153 .059 Huynh-Feldt .153 .059 Lower-bound .153 .059 index_p Sphericity Assumed .002 .253 Greenhouse-Geisser .002 .253 Huynh-Feldt .002 .253 Lower-bound .002 .253 index_n Sphericity Assumed .846 .001 Greenhouse-Geisser .846 .001 Huynh-Feldt .846 .001 Lower-bound .846 .001 place * id_type index_s Sphericity Assumed .000 .571 Greenhouse-Geisser .000 .571 Huynh-Feldt .000 .571 Lower-bound .000 .571 index_o Sphericity Assumed .000 .444 Greenhouse-Geisser .000 .444 Huynh-Feldt .000 .444 Lower-bound .000 .444 index_p Sphericity Assumed .000 .659 Greenhouse-Geisser .000 .659 Huynh-Feldt .000 .659 Lower-bound .000 .659 index_n Sphericity Assumed .000 .472 Greenhouse-Geisser .000 .472 Huynh-Feldt .000 .472 Lower-bound .000 .472 Error(place) index_s Sphericity Assumed Greenhouse-Geisser Huynh-Feldt Lower-bound index_o Sphericity Assumed Greenhouse-Geisser Huynh-Feldt Lower-bound index_p Sphericity Assumed Greenhouse-Geisser Huynh-Feldt Lower-bound index_n Sphericity Assumed Greenhouse-Geisser Huynh-Feldt Lower-bound

Tests of Within-Subjects Contrasts Type III Sum of Source Measure place Squares df Mean Square F place index_s Linear 36.928 1 36.928 22.185 index_o Linear 5.475 1 5.475 2.139 index_p Linear 61.170 1 61.170 11.525 index_n Linear .294 1 .294 .038 place * id_type index_s Linear 75.412 3 25.137 15.101 index_o Linear 69.595 3 23.198 9.062 index_p Linear 348.591 3 116.197 21.892 index_n Linear 233.149 3 77.716 10.143 Error(place) index_s Linear 56.595 34 1.665 index_o Linear 87.043 34 2.560 index_p Linear 180.459 34 5.308 index_n Linear 260.503 34 7.662

Tests of Within-Subjects Contrasts Partial Eta Source Measure place Sig. Squared place index_s Linear .000 .395 index_o Linear .153 .059 index_p Linear .002 .253 index_n Linear .846 .001 place * id_type index_s Linear .000 .571 index_o Linear .000 .444 index_p Linear .000 .659 index_n Linear .000 .472 Error(place) index_s Linear index_o Linear index_p Linear index_n Linear

Tests of Between-Subjects Effects Transformed Variable: Average Type III Sum of Source Measure Squares df Mean Square F Intercept index_s 6638.055 1 6638.055 875.315 index_o 5979.799 1 5979.799 492.201 index_p 23763.094 1 23763.094 837.239 index_n 5332.235 1 5332.235 80.671 id_type index_s 726.958 3 242.319 31.953 index_o 847.013 3 282.338 23.239 index_p 3122.973 3 1040.991 36.677 index_n 171.115 3 57.038 .863 Error index_s 257.843 34 7.584 index_o 413.069 34 12.149 index_p 965.011 34 28.383 index_n 2247.360 34 66.099

Tests of Between-Subjects Effects Transformed Variable: Average Partial Eta Source Measure Sig. Squared Intercept index_s .000 .963 index_o .000 .935 index_p .000 .961 index_n .000 .703 id_type index_s .000 .738 index_o .000 .672 index_p .000 .764 index_n .470 .071

Estimated Marginal Means

1. Grand Mean 95% Confidence Interval Measure Mean Std. Error Lower Bound Upper Bound index_s 11.580 .391 10.785 12.376 index_o 10.991 .495 9.984 11.998 index_p 21.910 .757 20.371 23.449 index_n 10.379 1.156 8.031 12.727

2. Type 95% Confidence Interval Measure Type Mean Std. Error Lower Bound Upper Bound index_s 1 12.052 1.124 9.767 14.336 2 7.667 .459 6.734 8.600 3 14.803 .587 13.610 15.996 4 11.799 .795 10.184 13.415 index_o 1 12.483 1.423 9.591 15.375 2 6.408 .581 5.227 7.589 3 13.881 .743 12.371 15.391 4 11.193 1.006 9.148 13.238 index_p 1 22.803 2.175 18.383 27.223 2 13.768 .888 11.963 15.572 3 28.533 1.136 26.224 30.841 4 22.538 1.538 19.412 25.663 index_n 1 9.884 3.319 3.139 16.629 2 10.718 1.355 7.964 13.472 3 8.219 1.733 4.697 11.742 4 12.694 2.347 7.925 17.464

3. place 95% Confidence Interval Measure place Mean Std. Error Lower Bound Upper Bound index_s 1 12.444 .423 11.585 13.303 2 10.717 .441 9.819 11.614 index_o 1 11.324 .582 10.140 12.507 2 10.659 .505 9.632 11.685 index_p 1 23.022 .812 21.371 24.673 2 20.799 .838 19.096 22.501 index_n 1 10.302 1.175 7.915 12.689 2 10.456 1.265 7.885 13.027

4. Type * place 95% Confidence Interval Measure Type place Mean Std. Error Lower Bound Upper Bound index_s 1 1 14.917 1.214 12.449 17.385 2 9.186 1.268 6.610 11.763 2 1 8.519 .496 7.511 9.526 2 6.815 .518 5.763 7.867 3 1 15.775 .634 14.486 17.064 2 13.832 .662 12.486 15.178 4 1 10.566 .859 8.821 12.311 2 13.033 .897 11.210 14.855 index_o 1 1 14.492 1.673 11.092 17.891 2 10.474 1.451 7.525 13.423 2 1 6.817 .683 5.429 8.204 2 5.999 .592 4.796 7.203 3 1 14.519 .874 12.744 16.294 2 13.243 .758 11.703 14.782 4 1 9.467 1.183 7.064 11.871 2 12.918 1.026 10.833 15.003 index_p 1 1 27.376 2.333 22.635 32.117 2 18.231 2.406 13.341 23.120 2 1 15.072 .952 13.137 17.008 2 12.463 .982 10.467 14.459 3 1 30.605 1.218 28.129 33.081 2 26.460 1.256 23.906 29.013 4 1 19.034 1.650 15.682 22.386 2 26.041 1.701 22.584 29.499 index_n 1 1 6.519 3.374 -.338 13.375 2 13.249 3.634 5.864 20.634 2 1 11.015 1.377 8.215 13.814 2 10.422 1.484 7.407 13.436 3 1 7.427 1.762 3.846 11.008 2 9.011 1.898 5.155 12.868 4 1 16.247 2.386 11.399 21.095 2 9.142 2.570 3.920 14.364

Post Hoc Tests

Type Homogeneous Subsets index_s Student-Newman-Keuls Subset Type N 1 2 3 N 1 2 3 2 18 7.6670 4 6 11.7992 1 3 12.0516 3 11 14.8032 Sig. 1.000 .821 1.000 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 3.792. index_o Student-Newman-Keuls Subset Type N 1 2 N 1 2 2 18 6.4081 4 6 11.1927 1 3 12.4828 3 11 13.8808 Sig. 1.000 .149 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 6.075. index_p Student-Newman-Keuls Subset Type N 1 2 3 N 1 2 3 2 18 13.7676 4 6 22.5377 1 3 22.8034 3 11 28.5326 Sig. 1.000 .902 1.000 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 14.191. index_n Student-Newman-Keuls Subset Type N 1 N 3 11 8.2192 1 3 9.8840 2 18 10.7181 4 6 12.6944 Sig. .527 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 33.049.

Oneway

ANOVA Sum of Squares df Mean Square F Sig. S origin Between Groups 401.224 3 133.741 27.726 .000 Within Groups 164.004 34 4.824 Total 565.228 37 O origin Between Groups 447.527 3 149.176 23.619 .000 Within Groups 214.741 34 6.316 Total 662.268 37 P origin Between Groups 1680.123 3 560.041 32.253 .000 Within Groups 590.374 34 17.364 Total 2270.496 37 N origin Between Groups 49.999 3 16.666 .421 .739 Within Groups 1346.905 34 39.615 Total 1396.904 37 S Australia Between Groups 401.145 3 133.715 30.221 .000 Within Groups 150.434 34 4.425 Total 551.579 37 O Australia Between Groups 469.080 3 156.360 18.629 .000 Within Groups 285.371 34 8.393 Total 754.451 37 P Australia Between Groups 1791.441 3 597.147 36.576 .000 Within Groups 555.097 34 16.326 Total 2346.538 37 N Australia Between Groups 354.265 3 118.088 3.458 .027 Within Groups 1160.957 34 34.146 Total 1515.222 37

Post Hoc Tests

Homogeneous Subsets S origin Student-Newman-Keuls Subset for alpha = .05 Type N 1 2 N 1 2 2 18 6.8153 1 3 9.1865 4 6 13.0326 3 11 13.8318 Sig. .066 .526 Means for groups in homogeneous subsets are displayed. O origin Student-Newman-Keuls Subset for alpha = .05 Type N 1 2 N 1 2 2 18 5.9994 1 3 10.4740 4 6 12.9181 3 11 13.2425 Sig. 1.000 .144 Means for groups in homogeneous subsets are displayed.

P origin Student-Newman-Keuls Subset for alpha = .05 Type N 1 2 3 N 1 2 3 2 18 12.4628 1 3 18.2307 4 6 26.0414 3 11 26.4598 Sig. 1.000 1.000 .861 Means for groups in homogeneous subsets are displayed. N origin Student-Newman-Keuls Subset for alpha = .05 Type N 1 N 3 11 9.0114 4 6 9.1417 2 18 10.4216 1 3 13.2493 Sig. .641 Means for groups in homogeneous subsets are displayed. S Australia Student-Newman-Keuls Subset for alpha = .05 Type N 1 2 N 1 2 2 18 8.5187 4 6 10.5658 1 3 14.9167 3 11 15.7747 Sig. .096 .478 Means for groups in homogeneous subsets are displayed. O Australia Student-Newman-Keuls Subset for alpha = .05 Type N 1 2 N 1 2 2 18 6.8167 4 6 9.4673 1 3 14.4916 3 11 14.5191 Sig. .117 .987 Means for groups in homogeneous subsets are displayed. P Australia Student-Newman-Keuls Subset for alpha = .05 Type N 1 2 N 1 2 2 18 15.0724 4 6 19.0341 1 3 27.3761 3 11 30.6053 Sig. .094 .169 Means for groups in homogeneous subsets are displayed.

N Australia Student-Newman-Keuls Subset for alpha = .05 Type N 1 2 N 1 2 1 3 6.5186 3 11 7.4270 2 18 11.0145 11.0145 4 6 16.2471 Sig. .376 .125 Means for groups in homogeneous subsets are displayed.

B.5 Paired t-tests

Paired t-tests for the entire sample and for each participant type separately, comparing indices for the country of origin I-position with those for the Australian I- position.

T-Test

Paired Samples Statistics Mean N Std. Deviation Std. Error Mean Pair 1 S Australia 11.4474 38 3.86103 .62634 S origin 10.0153 38 3.90851 .63404 Pair 2 O Australia 10.0708 38 4.51559 .73253 O origin 9.5418 38 4.23074 .68632 Pair 3 P Australia 21.1656 38 7.96366 1.29188 P origin 19.1139 38 7.83357 1.27077 Pair 4 N Australia 10.4473 38 6.39937 1.03812 N origin 10.0345 38 6.14444 .99676

Paired Samples Correlations N Correlation Sig. Pair 1 S Australia & S origin 38 .764 .000 Pair 2 O Australia & O origin 38 .781 .000 Pair 3 P Australia & P origin 38 .771 .000 Pair 4 N Australia & N origin 38 .662 .000

Paired Samples Test Paired Differences

Mean Std. Deviation Std. Error Mean Pair 1 S Australia - S origin 1.43220 2.67123 .43333 Pair 2 O Australia - O origin .52900 2.90979 .47203 Pair 3 P Australia - P origin 2.05172 5.34764 .86750 Pair 4 N Australia - N origin .41276 5.16564 .83798

Paired Samples Test Paired Differences 95% Confidence Interval of the

Difference Lower Upper t Pair 1 S Australia - S origin .55418 2.31021 3.305 Pair 2 O Australia - O origin -.42743 1.48542 1.121 Pair 3 P Australia - P origin .29399 3.80944 2.365 Pair 4 N Australia - N origin -1.28514 2.11067 .493

Paired Samples Test df Sig. (2-tailed) Pair 1 S Australia - S origin 37 .002 Pair 2 O Australia - O origin 37 .270 Pair 3 P Australia - P origin 37 .023 Pair 4 N Australia - N origin 37 .625

T-Test

Type = 1 Paired Samples Statisticsa Mean N Std. Deviation Std. Error Mean Pair 1 S Australia 14.9167 3 1.87639 1.08333 S origin 9.1865 3 3.61658 2.08803 Pair 2 O Australia 14.4916 3 1.00979 .58300 O origin 10.4740 3 3.22608 1.86258 Pair 3 P Australia 27.3761 3 3.84528 2.22007 P origin 18.2307 3 4.49202 2.59347 Pair 4 N Australia 6.5186 3 7.67489 4.43110 N origin 13.2493 3 7.06852 4.08101 a. Type = 1 Paired Samples Correlationsa N Correlation Sig. Pair 1 S Australia & S origin 3 .978 .134 Pair 2 O Australia & O origin 3 .873 .325 Pair 3 P Australia & P origin 3 .872 .325 Pair 4 N Australia & N origin 3 .877 .319 a. Type = 1 Paired Samples Testa Paired Differences

Mean Std. Deviation Std. Error Mean Pair 1 S Australia - S origin 5.73022 1.82438 1.05331 Pair 2 O Australia - O origin 4.01760 2.39632 1.38352 Pair 3 P Australia - P origin 9.14540 2.19803 1.26903 Pair 4 N Australia - N origin -6.73070 3.70415 2.13859

Paired Samples Testa Paired Differences 95% Confidence Interval of the Difference Lower Upper t Pair 1 S Australia - S origin 1.19819 10.26224 5.440 Pair 2 O Australia - O origin -1.93518 9.97039 2.904 Pair 3 P Australia - P origin 3.68519 14.60560 7.207 Pair 4 N Australia - N origin -15.93231 2.47092 -3.147

Paired Samples Testa df Sig. (2-tailed) Pair 1 S Australia - S origin 2 .032 Pair 2 O Australia - O origin 2 .101 Pair 3 P Australia - P origin 2 .019 Pair 4 N Australia - N origin 2 .088 a. Type = 1

Type = 2 Paired Samples Statisticsa Mean N Std. Deviation Std. Error Mean Pair 1 S Australia 8.5187 18 2.45534 .57873 S origin 6.8153 18 2.29341 .54056 Pair 2 O Australia 6.8167 18 3.32027 .78260 O origin 5.9994 18 2.81815 .66424 Pair 3 P Australia 15.0724 18 4.95009 1.16675 P origin 12.4628 18 4.61459 1.08767 Pair 4 N Australia 11.0145 18 6.54692 1.54312 N origin 10.4216 18 6.87254 1.61987 a. Type = 2 Paired Samples Correlationsa N Correlation Sig. Pair 1 S Australia & S origin 18 .703 .001 Pair 2 O Australia & O origin 18 .737 .000 Pair 3 P Australia & P origin 18 .740 .000 Pair 4 N Australia & N origin 18 .859 .000 a. Type = 2 Paired Samples Testa Paired Differences

Mean Std. Deviation Std. Error Mean Pair 1 S Australia - S origin 1.70338 1.83531 .43259 Pair 2 O Australia - O origin .81732 2.27649 .53657 Pair 3 P Australia - P origin 2.60958 3.45975 .81547 Pair 4 N Australia - N origin .59294 3.57427 .84246

Paired Samples Testa Paired Differences 95% Confidence Interval of the Difference Lower Upper t Pair 1 S Australia - S origin .79070 2.61606 3.938 Pair 2 O Australia - O origin -.31476 1.94939 1.523 Pair 3 P Australia - P origin .88908 4.33007 3.200 Pair 4 N Australia - N origin -1.18450 2.37038 .704

Paired Samples Testa df Sig. (2-tailed) Pair 1 S Australia - S origin 17 .001 Pair 2 O Australia - O origin 17 .146 Pair 3 P Australia - P origin 17 .005 Pair 4 N Australia - N origin 17 .491 a. Type = 2

Type = 3 Paired Samples Statisticsa Mean N Std. Deviation Std. Error Mean Pair 1 S Australia 15.7747 11 1.81116 .54609 S origin 13.8318 11 2.09383 .63131 Pair 2 O Australia 14.5191 11 2.51866 .75941 O origin 13.2425 11 2.30256 .69425 Pair 3 P Australia 30.6053 11 2.63742 .79521 P origin 26.4598 11 4.18274 1.26114 Pair 4 N Australia 7.4270 11 4.20208 1.26697 N origin 9.0114 11 5.93648 1.78992 a. Type = 3 Paired Samples Correlationsa N Correlation Sig. Pair 1 S Australia & S origin 11 .389 .238 Pair 2 O Australia & O origin 11 .567 .069 Pair 3 P Australia & P origin 11 .538 .088 Pair 4 N Australia & N origin 11 .620 .042 a. Type = 3 Paired Samples Testa Paired Differences

Mean Std. Deviation Std. Error Mean Pair 1 S Australia - S origin 1.94295 2.17183 .65483 Pair 2 O Australia - O origin 1.27658 2.25048 .67855 Pair 3 P Australia - P origin 4.14550 3.54665 1.06936 Pair 4 N Australia - N origin -1.58440 4.68734 1.41329

Paired Samples Testa Paired Differences 95% Confidence Interval of the Difference Lower Upper t Pair 1 S Australia - S origin .48390 3.40201 2.967 Pair 2 O Australia - O origin -.23531 2.78848 1.881 Pair 3 P Australia - P origin 1.76283 6.52817 3.877 Pair 4 N Australia - N origin -4.73340 1.56459 -1.121

Paired Samples Testa df Sig. (2-tailed) Pair 1 S Australia - S origin 10 .014 Pair 2 O Australia - O origin 10 .089 Pair 3 P Australia - P origin 10 .003 Pair 4 N Australia - N origin 10 .288 a. Type = 3

Type = 4 Paired Samples Statisticsa Mean N Std. Deviation Std. Error Mean Pair 1 S Australia 10.5658 6 1.27287 .51965 S origin 13.0326 6 .95798 .39109 Pair 2 O Australia 9.4673 6 2.54887 1.04057 O origin 12.9181 6 1.08573 .44325 Pair 3 P Australia 19.0341 6 2.80741 1.14612 P origin 26.0414 6 1.61607 .65976 Pair 4 N Australia 16.2471 6 5.25201 2.14413 N origin 9.1417 6 4.28065 1.74757 a. Type = 4 Paired Samples Correlationsa N Correlation Sig. Pair 1 S Australia & S origin 6 .868 .025 Pair 2 O Australia & O origin 6 .525 .285 Pair 3 P Australia & P origin 6 .672 .144 Pair 4 N Australia & N origin 6 .769 .074 a. Type = 4 Paired Samples Testa Paired Differences

Mean Std. Deviation Std. Error Mean Pair 1 S Australia - S origin -2.46676 .64841 .26471 Pair 2 O Australia - O origin -3.45084 2.18416 .89168 Pair 3 P Australia - P origin -7.00731 2.09669 .85597 Pair 4 N Australia - N origin 7.10543 3.36663 1.37442

Paired Samples Testa Paired Differences 95% Confidence Interval of the Difference Lower Upper t Pair 1 S Australia - S origin -3.14722 -1.78629 -9.319 Pair 2 O Australia - O origin -5.74298 -1.15870 -3.870 Pair 3 P Australia - P origin -9.20765 -4.80696 -8.186 Pair 4 N Australia - N origin 3.57237 10.63850 5.170

Paired Samples Testa df Sig. (2-tailed) Pair 1 S Australia - S origin 5 .000 Pair 2 O Australia - O origin 5 .012 Pair 3 P Australia - P origin 5 .000 Pair 4 N Australia - N origin 5 .004 a. Type = 4

B.6 Single-factor between-subjects MANOVA

MANOVAs with id_type as IV and the following DVs:

 Self-confrontation indices for the country of origin I-position

 Self-confrontation indices for the Australian I-position

 Self-confrontation indices for General feeling

 Self-confrontation indices for Ideal feeling

 Self-confrontation indices for the country of origin I-position (standardised)

 Self-confrontation indices for the Australian I-position (standardised)

 Psychological distress scales (BDI-II, FNES, SADS)

 Psychological Well-being Subscales (PWS1..6).

 Contrasts

Indices by Type

General Linear Model Between-Subjects Factors N Type 1 3 2 18 3 11 4 6

Descriptive Statistics Type Mean Std. Deviation N S origin 1 9.1865 3.61658 3 2 6.8153 2.29341 18 3 13.8318 2.09383 11 4 13.0326 .95798 6 Total 10.0153 3.90851 38 O origin 1 10.4740 3.22608 3 2 5.9994 2.81815 18 3 13.2425 2.30256 11 4 12.9181 1.08573 6 Total 9.5418 4.23074 38 P origin 1 18.2307 4.49202 3 2 12.4628 4.61459 18 3 26.4598 4.18274 11 4 26.0414 1.61607 6 Total 19.1139 7.83357 38 N origin 1 13.2493 7.06852 3 2 10.4216 6.87254 18 3 9.0114 5.93648 11 4 9.1417 4.28065 6 Total 10.0345 6.14444 38

Multivariate Testsc Effect Value F Hypothesis df Error df Intercept Pillai's Trace .956 168.113a 4.000 31.000 Wilks' Lambda .044 168.113a 4.000 31.000 Hotelling's Trace 21.692 168.113a 4.000 31.000 Roy's Largest Root 21.692 168.113a 4.000 31.000 id_type Pillai's Trace .857 3.298 12.000 99.000 Wilks' Lambda .211 5.486 12.000 82.310 Hotelling's Trace 3.418 8.449 12.000 89.000 Roy's Largest Root 3.324 27.426b 4.000 33.000

Multivariate Testsc Partial Eta Effect Sig. Squared Intercept Pillai's Trace .000 .956 Wilks' Lambda .000 .956 Hotelling's Trace .000 .956 Roy's Largest Root .000 .956 id_type Pillai's Trace .000 .286 Wilks' Lambda .000 .404 Hotelling's Trace .000 .533 Roy's Largest Root .000 .769 a. Exact statistic b. The statistic is an upper bound on F that yields a lower bound on the significance level. c. Design: Intercept + id_type Tests of Between-Subjects Effects Type III Sum of Source Dependent Variable Squares df Mean Square F Corrected Model S origin 401.224a 3 133.741 27.726 O origin 447.527b 3 149.176 23.619 P origin 1680.123c 3 560.041 32.253 N origin 49.999d 3 16.666 .421 Intercept S origin 2842.386 1 2842.386 589.260 O origin 2811.692 1 2811.692 445.176 P origin 10706.485 1 10706.485 616.593 N origin 2705.870 1 2705.870 68.304 id_type S origin 401.224 3 133.741 27.726 O origin 447.527 3 149.176 23.619 P origin 1680.123 3 560.041 32.253 N origin 49.999 3 16.666 .421 Error S origin 164.004 34 4.824 O origin 214.741 34 6.316 P origin 590.374 34 17.364 N origin 1346.905 34 39.615 Total S origin 4376.828 38 O origin 4121.998 38 P origin 16153.509 38 N origin 5223.195 38 Corrected Total S origin 565.228 37 O origin 662.268 37 P origin 2270.496 37 N origin 1396.904 37

Tests of Between-Subjects Effects Partial Eta Source Dependent Variable Sig. Squared Corrected Model S origin .000 .710 O origin .000 .676 P origin .000 .740 N origin .739 .036 Intercept S origin .000 .945 O origin .000 .929 P origin .000 .948 N origin .000 .668 id_type S origin .000 .710 O origin .000 .676 P origin .000 .740 N origin .739 .036 Error S origin O origin P origin N origin Total S origin O origin P origin N origin Corrected Total S origin O origin P origin N origin a. R Squared = .710 (Adjusted R Squared = .684) b. R Squared = .676 (Adjusted R Squared = .647) c. R Squared = .740 (Adjusted R Squared = .717) d. R Squared = .036 (Adjusted R Squared = -.049)

Post Hoc Tests

Type Homogeneous Subsets S origin Student-Newman-Keulsab Subset Type N 1 2 2 18 6.8153 1 3 9.1865 4 6 13.0326 3 11 13.8318 Sig. .066 .526 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 4.824. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05.

O origin Student-Newman-Keulsab Subset Type N 1 2 2 18 5.9994 1 3 10.4740 4 6 12.9181 3 11 13.2425 Sig. 1.000 .144 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 6.316. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. P origin Student-Newman-Keulsab Subset Type N 1 2 3 2 18 12.4628 1 3 18.2307 4 6 26.0414 3 11 26.4598 Sig. 1.000 1.000 .861 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 17.364. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. N origin Student-Newman-Keulsab Subset Type N 1 3 11 9.0114 4 6 9.1417 2 18 10.4216 1 3 13.2493 Sig. .641 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 39.615. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05.

General Linear Model

Between-Subjects Factors N Type 1 3 2 18 3 11 4 6

Descriptive Statistics Type Mean Std. Deviation N S Australia 1 14.9167 1.87639 3 2 8.5187 2.45534 18 3 15.7747 1.81116 11 4 10.5658 1.27287 6 Total 11.4474 3.86103 38 O Australia 1 14.4916 1.00979 3 2 6.8167 3.32027 18 3 14.5191 2.51866 11 4 9.4673 2.54887 6 Total 10.0708 4.51559 38 P Australia 1 27.3761 3.84528 3 2 15.0724 4.95009 18 3 30.6053 2.63742 11 4 19.0341 2.80741 6 Total 21.1656 7.96366 38 N Australia 1 6.5186 7.67489 3 2 11.0145 6.54692 18 3 7.4270 4.20208 11 4 16.2471 5.25201 6 Total 10.4473 6.39937 38

Multivariate Testsc Effect Value F Hypothesis df Error df Intercept Pillai's Trace .970 246.906a 4.000 31.000 Wilks' Lambda .030 246.906a 4.000 31.000 Hotelling's Trace 31.859 246.906a 4.000 31.000 Roy's Largest Root 31.859 246.906a 4.000 31.000 id_type Pillai's Trace .986 4.037 12.000 99.000 Wilks' Lambda .181 6.236 12.000 82.310 Hotelling's Trace 3.652 9.029 12.000 89.000 Roy's Largest Root 3.408 28.118b 4.000 33.000

Multivariate Testsc Partial Eta Effect Sig. Squared Intercept Pillai's Trace .000 .970 Wilks' Lambda .000 .970 Hotelling's Trace .000 .970 Roy's Largest Root .000 .970 id_type Pillai's Trace .000 .329 Wilks' Lambda .000 .435 Hotelling's Trace .000 .549 Roy's Largest Root .000 .773 a. Exact statistic b. The statistic is an upper bound on F that yields a lower bound on the significance level. c. Design: Intercept + id_type Tests of Between-Subjects Effects Type III Sum of Source Dependent Variable Squares df Mean Square F Corrected Model S Australia 401.145a 3 133.715 30.221 O Australia 469.080b 3 156.360 18.629 P Australia 1791.441c 3 597.147 36.576 N Australia 354.265d 3 118.088 3.458 Intercept S Australia 3832.597 1 3832.597 866.218 O Australia 3173.582 1 3173.582 378.111 P Australia 13117.780 1 13117.780 803.471 N Australia 2626.659 1 2626.659 76.925 id_type S Australia 401.145 3 133.715 30.221 O Australia 469.080 3 156.360 18.629 P Australia 1791.441 3 597.147 36.576 N Australia 354.265 3 118.088 3.458 Error S Australia 150.434 34 4.425 O Australia 285.371 34 8.393 P Australia 555.097 34 16.326 N Australia 1160.957 34 34.146 Total S Australia 5531.252 38 O Australia 4608.430 38 P Australia 19369.957 38 N Australia 5662.770 38 Corrected Total S Australia 551.579 37 O Australia 754.451 37 P Australia 2346.538 37 N Australia 1515.222 37

Tests of Between-Subjects Effects Partial Eta Source Dependent Variable Sig. Squared Corrected Model S Australia .000 .727 O Australia .000 .622 P Australia .000 .763 N Australia .027 .234 Intercept S Australia .000 .962 O Australia .000 .917 P Australia .000 .959 N Australia .000 .693 id_type S Australia .000 .727 O Australia .000 .622 P Australia .000 .763 N Australia .027 .234 Error S Australia O Australia P Australia N Australia Total S Australia O Australia P Australia N Australia Corrected Total S Australia O Australia P Australia N Australia a. R Squared = .727 (Adjusted R Squared = .703) b. R Squared = .622 (Adjusted R Squared = .588) c. R Squared = .763 (Adjusted R Squared = .743) d. R Squared = .234 (Adjusted R Squared = .166)

Post Hoc Tests

Type Homogeneous Subsets S Australia Student-Newman-Keulsab Subset Type N 1 2 2 18 8.5187 4 6 10.5658 1 3 14.9167 3 11 15.7747 Sig. .096 .478 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 4.425. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05.

O Australia Student-Newman-Keulsab Subset Type N 1 2 2 18 6.8167 4 6 9.4673 1 3 14.4916 3 11 14.5191 Sig. .117 .987 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 8.393. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. P Australia Student-Newman-Keulsab Subset Type N 1 2 2 18 15.0724 4 6 19.0341 1 3 27.3761 3 11 30.6053 Sig. .094 .169 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 16.326. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. N Australia Student-Newman-Keulsab Subset Type N 1 2 1 3 6.5186 3 11 7.4270 2 18 11.0145 11.0145 4 6 16.2471 Sig. .376 .125 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 34.146. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05.

General Linear Model

Between-Subjects Factors N Type 1 3 2 18 3 11 4 6

Descriptive Statistics Type Mean Std. Deviation N S general 1 18.3333 1.52753 3 2 10.2778 4.08448 18 3 16.9091 4.30011 11 4 12.3333 4.45720 6 Total 13.1579 5.09651 38 O general 1 17.6667 1.52753 3 2 9.7778 4.90565 18 3 17.3636 2.94186 11 4 12.6667 3.82971 6 Total 13.0526 5.29123 38 P general 1 34.3333 5.03322 3 2 19.8333 8.06773 18 3 33.5455 6.31449 11 4 24.1667 7.62671 6 Total 25.6316 9.55898 38 N general 1 4.3333 4.50925 3 2 14.2778 10.13907 18 3 6.7273 6.43570 11 4 13.8333 6.96898 6 Total 11.2368 9.03278 38

Multivariate Testsc Effect Value F Hypothesis df Error df Intercept Pillai's Trace .969 243.880a 4.000 31.000 Wilks' Lambda .031 243.880a 4.000 31.000 Hotelling's Trace 31.468 243.880a 4.000 31.000 Roy's Largest Root 31.468 243.880a 4.000 31.000 id_type Pillai's Trace .553 1.864 12.000 99.000 Wilks' Lambda .469 2.276 12.000 82.310 Hotelling's Trace 1.089 2.691 12.000 89.000 Roy's Largest Root 1.045 8.625b 4.000 33.000

Multivariate Testsc Partial Eta Effect Sig. Squared Intercept Pillai's Trace .000 .969 Wilks' Lambda .000 .969 Hotelling's Trace .000 .969 Roy's Largest Root .000 .969 id_type Pillai's Trace .048 .184 Wilks' Lambda .015 .223 Hotelling's Trace .004 .266 Roy's Largest Root .000 .511 a. Exact statistic b. The statistic is an upper bound on F that yields a lower bound on the significance level. c. Design: Intercept + id_type Tests of Between-Subjects Effects Type III Sum of Source Dependent Variable Squares df Mean Square F Corrected Model S general 388.532a 3 129.511 7.691 O general 462.238b 3 154.079 9.132 P general 1534.115c 3 511.372 9.415 N general 573.575d 3 191.192 2.658 Intercept S general 5177.439 1 5177.439 307.470 O general 5109.864 1 5109.864 302.856 P general 19362.023 1 19362.023 356.473 N general 2373.561 1 2373.561 33.003 id_type S general 388.532 3 129.511 7.691 O general 462.238 3 154.079 9.132 P general 1534.115 3 511.372 9.415 N general 573.575 3 191.192 2.658 Error S general 572.520 34 16.839 O general 573.657 34 16.872 P general 1846.727 34 54.316 N general 2445.293 34 71.920 Total S general 7540.000 38 O general 7510.000 38 P general 28346.000 38 N general 7817.000 38 Corrected Total S general 961.053 37 O general 1035.895 37 P general 3380.842 37 N general 3018.868 37

Tests of Between-Subjects Effects Partial Eta Source Dependent Variable Sig. Squared Corrected Model S general .000 .404 O general .000 .446 P general .000 .454 N general .064 .190 Intercept S general .000 .900 O general .000 .899 P general .000 .913 N general .000 .493 id_type S general .000 .404 O general .000 .446 P general .000 .454 N general .064 .190 Error S general O general P general N general Total S general O general P general N general Corrected Total S general O general P general N general a. R Squared = .404 (Adjusted R Squared = .352) b. R Squared = .446 (Adjusted R Squared = .397) c. R Squared = .454 (Adjusted R Squared = .406) d. R Squared = .190 (Adjusted R Squared = .119)

Post Hoc Tests

Type Homogeneous Subsets S general Student-Newman-Keulsab Subset Type N 1 2 3 2 18 10.2778 4 6 12.3333 12.3333 3 11 16.9091 16.9091 1 3 18.3333 Sig. .384 .058 .546 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 16.839. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05.

O general Student-Newman-Keulsab Subset Type N 1 2 2 18 9.7778 4 6 12.6667 12.6667 3 11 17.3636 1 3 17.6667 Sig. .225 .097 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 16.872. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. P general Student-Newman-Keulsab Subset Type N 1 2 2 18 19.8333 4 6 24.1667 24.1667 3 11 33.5455 1 3 34.3333 Sig. .308 .053 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 54.316. a.Uses Harmonic Mean Sample Size = 6.188. b.Alpha = .05. N general Student-Newman-Keulsab Subset Type N 1 1 3 4.3333 3 11 6.7273 4 6 13.8333 2 18 14.2778 Sig. .186 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 71.920. a.Uses Harmonic Mean Sample Size = 6.188. b.Alpha = .05.

General Linear Model

Between-Subjects Factors N Type 1 3 2 18 3 11 4 6

Descriptive Statistics Type Mean Std. Deviation N S ideal 1 19.3333 1.15470 3 2 16.1667 4.42254 18 3 18.6364 2.50091 11 4 18.0000 2.28035 6 Total 17.4211 3.60654 38 O ideal 1 19.6667 .57735 3 2 15.7222 3.64297 18 3 19.1818 1.60114 11 4 18.3333 1.96638 6 Total 17.4474 3.19372 38 P ideal 1 37.6667 4.04145 3 2 34.0000 6.65096 18 3 38.0000 2.60768 11 4 38.1667 2.56255 6 Total 36.1053 5.29553 38 N ideal 1 1.0000 1.73205 3 2 1.1667 1.75734 18 3 2.3636 4.98543 11 4 .5000 .83666 6 Total 1.3947 2.97321 38

Multivariate Testsc Effect Value F Hypothesis df Error df Intercept Pillai's Trace .978 349.658a 4.000 31.000 Wilks' Lambda .022 349.658a 4.000 31.000 Hotelling's Trace 45.117 349.658a 4.000 31.000 Roy's Largest Root 45.117 349.658a 4.000 31.000 id_type Pillai's Trace .405 1.289 12.000 99.000 Wilks' Lambda .619 1.362 12.000 82.310 Hotelling's Trace .575 1.423 12.000 89.000 Roy's Largest Root .501 4.130b 4.000 33.000

Multivariate Testsc Partial Eta Effect Sig. Squared Intercept Pillai's Trace .000 .978 Wilks' Lambda .000 .978 Hotelling's Trace .000 .978 Roy's Largest Root .000 .978 id_type Pillai's Trace .237 .135 Wilks' Lambda .201 .148 Hotelling's Trace .171 .161 Roy's Largest Root .008 .334 a. Exact statistic b. The statistic is an upper bound on F that yields a lower bound on the significance level. c. Design: Intercept + id_type Tests of Between-Subjects Effects Type III Sum of Source Dependent Variable Squares df Mean Square F Corrected Model S ideal 57.551a 3 19.184 1.539 O ideal 106.147b 3 35.382 4.435 P ideal 152.079c 3 50.693 1.946 N ideal 16.533d 3 5.511 .603 Intercept S ideal 8049.404 1 8049.404 645.910 O ideal 8221.639 1 8221.639 1030.556 P ideal 33806.480 1 33806.480 1298.047 N ideal 39.142 1 39.142 4.285 id_type S ideal 57.551 3 19.184 1.539 O ideal 106.147 3 35.382 4.435 P ideal 152.079 3 50.693 1.946 N ideal 16.533 3 5.511 .603 Error S ideal 423.712 34 12.462 O ideal 271.247 34 7.978 P ideal 885.500 34 26.044 N ideal 310.545 34 9.134 Total S ideal 12014.000 38 O ideal 11945.000 38 P ideal 50574.000 38 N ideal 401.000 38 Corrected Total S ideal 481.263 37 O ideal 377.395 37 P ideal 1037.579 37 N ideal 327.079 37

Tests of Between-Subjects Effects Partial Eta Source Dependent Variable Sig. Squared Corrected Model S ideal .222 .120 O ideal .010 .281 P ideal .141 .147 N ideal .617 .051 Intercept S ideal .000 .950 O ideal .000 .968 P ideal .000 .974 N ideal .046 .112 id_type S ideal .222 .120 O ideal .010 .281 P ideal .141 .147 N ideal .617 .051 Error S ideal O ideal P ideal N ideal Total S ideal O ideal P ideal N ideal Corrected Total S ideal O ideal P ideal N ideal a. R Squared = .120 (Adjusted R Squared = .042) b. R Squared = .281 (Adjusted R Squared = .218) c. R Squared = .147 (Adjusted R Squared = .071) d. R Squared = .051 (Adjusted R Squared = -.033)

Post Hoc Tests

Type Homogeneous Subsets S ideal Student-Newman-Keulsab Subset Type N 1 2 18 16.1667 4 6 18.0000 3 11 18.6364 1 3 19.3333 Sig. .404 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 12.462. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05.

O ideal Student-Newman-Keulsab Subset Type N 1 2 18 15.7222 4 6 18.3333 3 11 19.1818 1 3 19.6667 Sig. .086 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 7.978. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. P ideal Student-Newman-Keulsab Subset Type N 1 2 18 34.0000 1 3 37.6667 3 11 38.0000 4 6 38.1667 Sig. .486 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 26.044. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. N ideal Student-Newman-Keulsab Subset Type N 1 4 6 .5000 1 3 1.0000 2 18 1.1667 3 11 2.3636 Sig. .701 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 9.134. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05.

General Linear Model

Between-Subjects Factors N Type 1 3 2 18 3 11 4 6

Descriptive Statistics Type Mean Std. Deviation N Zscore: S origin 1 -.2120501 .92530949 3 2 -.8187148 .58677293 18 3 .9764599 .53571039 11 4 .7719928 .24510004 6 Total .0000000 1.00000000 38 Zscore: O origin 1 .2203424 .76253277 3 2 -.8372956 .66611364 18 3 .8747274 .54424664 11 4 .7980487 .25663006 6 Total .0000000 1.00000000 38 Zscore: P origin 1 -.1127496 .57343191 3 2 -.8490497 .58907932 18 3 .9377424 .53395063 11 4 .8843293 .20630115 6 Total .0000000 1.00000000 38 Zscore: N origin 1 .5232053 1.15039201 3 2 .0629916 1.11849666 18 3 -.1665101 .96615409 11 4 -.1453090 .69666974 6 Total .0000000 1.00000000 38

Multivariate Testsc Effect Value F Hypothesis df Error df Intercept Pillai's Trace .140 1.262a 4.000 31.000 Wilks' Lambda .860 1.262a 4.000 31.000 Hotelling's Trace .163 1.262a 4.000 31.000 Roy's Largest Root .163 1.262a 4.000 31.000 id_type Pillai's Trace .857 3.298 12.000 99.000 Wilks' Lambda .211 5.486 12.000 82.310 Hotelling's Trace 3.418 8.449 12.000 89.000 Roy's Largest Root 3.324 27.426b 4.000 33.000

Multivariate Testsc Partial Eta Effect Sig. Squared Intercept Pillai's Trace .306 .140 Wilks' Lambda .306 .140 Hotelling's Trace .306 .140 Roy's Largest Root .306 .140 id_type Pillai's Trace .000 .286 Wilks' Lambda .000 .404 Hotelling's Trace .000 .533 Roy's Largest Root .000 .769 a. Exact statistic b. The statistic is an upper bound on F that yields a lower bound on the significance level. c. Design: Intercept + id_type Tests of Between-Subjects Effects Type III Sum of Source Dependent Variable Squares df Mean Square F Corrected Model Zscore: S origin 26.264a 3 8.755 27.726 Zscore: O origin 25.003b 3 8.334 23.619 Zscore: P origin 27.379c 3 9.126 32.253 Zscore: N origin 1.324d 3 .441 .421 Intercept Zscore: S origin .797 1 .797 2.523 Zscore: O origin 1.724 1 1.724 4.887 Zscore: P origin 1.145 1 1.145 4.046 Zscore: N origin .116 1 .116 .111 id_type Zscore: S origin 26.264 3 8.755 27.726 Zscore: O origin 25.003 3 8.334 23.619 Zscore: P origin 27.379 3 9.126 32.253 Zscore: N origin 1.324 3 .441 .421 Error Zscore: S origin 10.736 34 .316 Zscore: O origin 11.997 34 .353 Zscore: P origin 9.621 34 .283 Zscore: N origin 35.676 34 1.049 Total Zscore: S origin 37.000 38 Zscore: O origin 37.000 38 Zscore: P origin 37.000 38 Zscore: N origin 37.000 38 Corrected Total Zscore: S origin 37.000 37 Zscore: O origin 37.000 37 Zscore: P origin 37.000 37 Zscore: N origin 37.000 37

Tests of Between-Subjects Effects Partial Eta Source Dependent Variable Sig. Squared Corrected Model Zscore: S origin .000 .710 Zscore: O origin .000 .676 Zscore: P origin .000 .740 Zscore: N origin .739 .036 Intercept Zscore: S origin .121 .069 Zscore: O origin .034 .126 Zscore: P origin .052 .106 Zscore: N origin .741 .003 id_type Zscore: S origin .000 .710 Zscore: O origin .000 .676 Zscore: P origin .000 .740 Zscore: N origin .739 .036 Error Zscore: S origin Zscore: O origin Zscore: P origin Zscore: N origin Total Zscore: S origin Zscore: O origin Zscore: P origin Zscore: N origin Corrected Total Zscore: S origin Zscore: O origin Zscore: P origin Zscore: N origin a. R Squared = .710 (Adjusted R Squared = .684) b. R Squared = .676 (Adjusted R Squared = .647) c. R Squared = .740 (Adjusted R Squared = .717) d. R Squared = .036 (Adjusted R Squared = -.049)

Post Hoc Tests

Type Homogeneous Subsets Zscore: S origin Student-Newman-Keulsab Subset Type N 1 2 2 18 -.8187148 1 3 -.2120501 4 6 .7719928 3 11 .9764599 Sig. .066 .526 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = .316. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05.

Zscore: O origin Student-Newman-Keulsab Subset Type N 1 2 2 18 -.8372956 1 3 .2203424 4 6 .7980487 3 11 .8747274 Sig. 1.000 .144 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = .353. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. Zscore: P origin Student-Newman-Keulsab Subset Type N 1 2 3 2 18 -.8490497 1 3 -.1127496 4 6 .8843293 3 11 .9377424 Sig. 1.000 1.000 .861 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = .283. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. Zscore: N origin Student-Newman-Keulsab Subset Type N 1 3 11 -.1665101 4 6 -.1453090 2 18 .0629916 1 3 .5232053 Sig. .641 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 1.049. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05.

General Linear Model

Between-Subjects Factors N Type 1 3 2 18 3 11 4 6

Descriptive Statistics Type Mean Std. Deviation N Zscore: S Australia 1 .8985225 .48598152 3 2 -.7585465 .63592945 18 3 1.1207532 .46908834 11 4 -.2283361 .32967020 6 Total .0000000 1.00000000 38 Zscore: O Australia 1 .9790122 .22362287 3 2 -.7206274 .73529027 18 3 .9851040 .55777003 11 4 -.1336480 .56445881 6 Total .0000000 1.00000000 38 Zscore: P Australia 1 .7798480 .48285331 3 2 -.7651285 .62158397 18 3 1.1853402 .33118218 11 4 -.2676623 .35252701 6 Total .0000000 1.00000000 38 Zscore: N Australia 1 -.6139128 1.19932008 3 2 .0886375 1.02305649 18 3 -.4719645 .65663877 11 4 .9063124 .82070773 6 Total .0000000 1.00000000 38

Multivariate Testsc Effect Value F Hypothesis df Error df Intercept Pillai's Trace .167 1.550a 4.000 31.000 Wilks' Lambda .833 1.550a 4.000 31.000 Hotelling's Trace .200 1.550a 4.000 31.000 Roy's Largest Root .200 1.550a 4.000 31.000 id_type Pillai's Trace .986 4.037 12.000 99.000 Wilks' Lambda .181 6.236 12.000 82.310 Hotelling's Trace 3.652 9.029 12.000 89.000 Roy's Largest Root 3.408 28.118b 4.000 33.000

Multivariate Testsc Partial Eta Effect Sig. Squared Intercept Pillai's Trace .212 .167 Wilks' Lambda .212 .167 Hotelling's Trace .212 .167 Roy's Largest Root .212 .167 id_type Pillai's Trace .000 .329 Wilks' Lambda .000 .435 Hotelling's Trace .000 .549 Roy's Largest Root .000 .773 a. Exact statistic b. The statistic is an upper bound on F that yields a lower bound on the significance level. c. Design: Intercept + id_type Tests of Between-Subjects Effects Type III Sum of Source Dependent Variable Squares df Mean Square F Corrected Model Zscore: S Australia 26.909a 3 8.970 30.221 Zscore: O Australia 23.005b 3 7.668 18.629 Zscore: P Australia 28.247c 3 9.416 36.576 Zscore: N Australia 8.651d 3 2.884 3.458 Intercept Zscore: S Australia 1.649 1 1.649 5.555 Zscore: O Australia 1.905 1 1.905 4.629 Zscore: P Australia 1.345 1 1.345 5.224 Zscore: N Australia .013 1 .013 .015 id_type Zscore: S Australia 26.909 3 8.970 30.221 Zscore: O Australia 23.005 3 7.668 18.629 Zscore: P Australia 28.247 3 9.416 36.576 Zscore: N Australia 8.651 3 2.884 3.458 Error Zscore: S Australia 10.091 34 .297 Zscore: O Australia 13.995 34 .412 Zscore: P Australia 8.753 34 .257 Zscore: N Australia 28.349 34 .834 Total Zscore: S Australia 37.000 38 Zscore: O Australia 37.000 38 Zscore: P Australia 37.000 38 Zscore: N Australia 37.000 38 Corrected Total Zscore: S Australia 37.000 37 Zscore: O Australia 37.000 37 Zscore: P Australia 37.000 37 Zscore: N Australia 37.000 37

Tests of Between-Subjects Effects Partial Eta Source Dependent Variable Sig. Squared Corrected Model Zscore: S Australia .000 .727 Zscore: O Australia .000 .622 Zscore: P Australia .000 .763 Zscore: N Australia .027 .234 Intercept Zscore: S Australia .024 .140 Zscore: O Australia .039 .120 Zscore: P Australia .029 .133 Zscore: N Australia .902 .000 id_type Zscore: S Australia .000 .727 Zscore: O Australia .000 .622 Zscore: P Australia .000 .763 Zscore: N Australia .027 .234 Error Zscore: S Australia Zscore: O Australia Zscore: P Australia Zscore: N Australia Total Zscore: S Australia Zscore: O Australia Zscore: P Australia Zscore: N Australia Corrected Total Zscore: S Australia Zscore: O Australia Zscore: P Australia Zscore: N Australia a. R Squared = .727 (Adjusted R Squared = .703) b. R Squared = .622 (Adjusted R Squared = .588) c. R Squared = .763 (Adjusted R Squared = .743) d. R Squared = .234 (Adjusted R Squared = .166)

Post Hoc Tests

Type Homogeneous Subsets Zscore: S Australia Student-Newman-Keulsab Subset Type N 1 2 2 18 -.7585465 4 6 -.2283361 1 3 .8985225 3 11 1.1207532 Sig. .096 .478 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = .297. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05.

Zscore: O Australia Student-Newman-Keulsab Subset Type N 1 2 2 18 -.7206274 4 6 -.1336480 1 3 .9790122 3 11 .9851040 Sig. .117 .987 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = .412. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. Zscore: P Australia Student-Newman-Keulsab Subset Type N 1 2 2 18 -.7651285 4 6 -.2676623 1 3 .7798480 3 11 1.1853402 Sig. .094 .169 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = .257. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. Zscore: N Australia Student-Newman-Keulsab Subset Type N 1 2 1 3 -.6139128 3 11 -.4719645 2 18 .0886375 .0886375 4 6 .9063124 Sig. .376 .125 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = .834. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05.

General Linear Model

Between-Subjects Factors N Type 1 3 2 18 3 11 4 6

Descriptive Statistics Type Mean Std. Deviation N SADS 1 2.00 1.732 3 2 9.11 7.661 18 3 7.82 6.145 11 4 10.33 8.733 6 Total 8.37 7.205 38 FNES 1 10.00 10.149 3 2 15.72 10.448 18 3 8.45 6.393 11 4 16.33 9.331 6 Total 13.26 9.526 38 BDI2 1 11.33 9.074 3 2 10.61 11.653 18 3 4.91 4.742 11 4 10.83 7.333 6 Total 9.05 9.349 38

Multivariate Testsc Effect Value F Hypothesis df Error df Intercept Pillai's Trace .591 15.444a 3.000 32.000 Wilks' Lambda .409 15.444a 3.000 32.000 Hotelling's Trace 1.448 15.444a 3.000 32.000 Roy's Largest Root 1.448 15.444a 3.000 32.000 id_type Pillai's Trace .280 1.167 9.000 102.000 Wilks' Lambda .738 1.150 9.000 78.030 Hotelling's Trace .329 1.122 9.000 92.000 Roy's Largest Root .212 2.400b 3.000 34.000

Multivariate Testsc Partial Eta Effect Sig. Squared Intercept Pillai's Trace .000 .591 Wilks' Lambda .000 .591 Hotelling's Trace .000 .591 Roy's Largest Root .000 .591 id_type Pillai's Trace .324 .093 Wilks' Lambda .339 .096 Hotelling's Trace .355 .099 Roy's Largest Root .085 .175 a. Exact statistic b. The statistic is an upper bound on F that yields a lower bound on the significance level. c. Design: Intercept + id_type

Tests of Between-Subjects Effects Type III Sum of Source Dependent Variable Squares df Mean Square F Corrected Model SADS 158.095a 3 52.698 1.016 FNES 451.697b 3 150.566 1.762 BDI2 267.208c 3 89.069 1.021 Intercept SADS 1324.591 1 1324.591 25.549 FNES 3946.496 1 3946.496 46.179 BDI2 2197.027 1 2197.027 25.179 id_type SADS 158.095 3 52.698 1.016 FNES 451.697 3 150.566 1.762 BDI2 267.208 3 89.069 1.021 Error SADS 1762.747 34 51.846 FNES 2905.672 34 85.461 BDI2 2966.687 34 87.255 Total SADS 4582.000 38 FNES 10042.000 38 BDI2 6348.000 38 Corrected Total SADS 1920.842 37 FNES 3357.368 37 BDI2 3233.895 37

Tests of Between-Subjects Effects Partial Eta Source Dependent Variable Sig. Squared Corrected Model SADS .397 .082 FNES .173 .135 BDI2 .396 .083 Intercept SADS .000 .429 FNES .000 .576 BDI2 .000 .425 id_type SADS .397 .082 FNES .173 .135 BDI2 .396 .083 Error SADS FNES BDI2 Total SADS FNES BDI2 Corrected Total SADS FNES BDI2 a. R Squared = .082 (Adjusted R Squared = .001) b. R Squared = .135 (Adjusted R Squared = .058) c. R Squared = .083 (Adjusted R Squared = .002)

Post Hoc Tests

Type Homogeneous Subsets SADS Student-Newman-Keulsab Subset Type N 1 1 3 2.00 3 11 7.82 2 18 9.11 4 6 10.33 Sig. .195 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 51.846. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. FNES Student-Newman-Keulsab Subset Type N 1 3 11 8.45 1 3 10.00 2 18 15.72 4 6 16.33 Sig. .449 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 85.461. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. BDI2 Student-Newman-Keulsab Subset Type N 1 3 11 4.91 2 18 10.61 4 6 10.83 1 3 11.33 Sig. .625 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 87.255. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05.

General Linear Model

Between-Subjects Factors N Type 1 3 2 18 3 11 4 6

Descriptive Statistics Type Mean Std. Deviation N PWS1 1 60.00 8.660 3 2 59.44 13.263 18 3 71.00 10.714 11 4 61.17 5.492 6 Total 63.11 12.100 38 PWS2 1 75.00 7.000 3 2 63.94 12.317 18 3 75.00 8.922 11 4 57.83 8.472 6 Total 67.05 12.081 38 PWS3 1 64.33 9.504 3 2 59.50 12.655 18 3 70.73 10.479 11 4 61.00 8.050 6 Total 63.37 11.886 38 PWS4 1 76.33 4.163 3 2 70.89 9.003 18 3 76.18 6.911 11 4 69.67 6.439 6 Total 72.66 8.028 38 PWS5 1 63.67 7.638 3 2 62.33 11.596 18 3 73.09 9.115 11 4 65.83 7.139 6 Total 66.11 10.780 38 PWS6 1 62.33 10.970 3 2 62.17 12.678 18 3 71.45 12.469 11 4 60.17 9.109 6 Total 64.55 12.411 38

Multivariate Testsc Effect Value F Hypothesis df Error df Intercept Pillai's Trace .986 352.240a 6.000 29.000 Wilks' Lambda .014 352.240a 6.000 29.000 Hotelling's Trace 72.877 352.240a 6.000 29.000 Roy's Largest Root 72.877 352.240a 6.000 29.000 id_type Pillai's Trace .759 1.750 18.000 93.000 Wilks' Lambda .398 1.768 18.000 82.510 Hotelling's Trace 1.142 1.756 18.000 83.000 Roy's Largest Root .647 3.344b 6.000 31.000

Multivariate Testsc Partial Eta Effect Sig. Squared Intercept Pillai's Trace .000 .986 Wilks' Lambda .000 .986 Hotelling's Trace .000 .986 Roy's Largest Root .000 .986 id_type Pillai's Trace .044 .253 Wilks' Lambda .044 .265 Hotelling's Trace .045 .276 Roy's Largest Root .012 .393 a. Exact statistic b. The statistic is an upper bound on F that yields a lower bound on the significance level. c. Design: Intercept + id_type

Tests of Between-Subjects Effects Type III Sum of Source Dependent Variable Squares df Mean Square F Corrected Model PWS1 978.301a 3 326.100 2.498 PWS2 1568.117b 3 522.706 4.638 PWS3 901.494c 3 300.498 2.362 PWS4 287.138d 3 95.713 1.552 PWS5 811.170e 3 270.390 2.635 PWS6 756.667f 3 252.222 1.735 Intercept PWS1 97929.796 1 97929.796 750.035 PWS2 114257.076 1 114257.076 1013.822 PWS3 101028.299 1 101028.299 794.147 PWS4 132861.773 1 132861.773 2153.747 PWS5 108567.196 1 108567.196 1058.157 PWS6 101472.023 1 101472.023 698.005 id_type PWS1 978.301 3 326.100 2.498 PWS2 1568.117 3 522.706 4.638 PWS3 901.494 3 300.498 2.362 PWS4 287.138 3 95.713 1.552 PWS5 811.170 3 270.390 2.635 PWS6 756.667 3 252.222 1.735 Error PWS1 4439.278 34 130.567 PWS2 3831.778 34 112.699 PWS3 4325.348 34 127.216 PWS4 2097.414 34 61.689 PWS5 3488.409 34 102.600 PWS6 4942.727 34 145.374 Total PWS1 156744.000 38 PWS2 176250.000 38 PWS3 157818.000 38 PWS4 202993.000 38 PWS5 170356.000 38 PWS6 164047.000 38 Corrected Total PWS1 5417.579 37 PWS2 5399.895 37 PWS3 5226.842 37 PWS4 2384.553 37 PWS5 4299.579 37 PWS6 5699.395 37

100

Tests of Between-Subjects Effects Partial Eta Source Dependent Variable Sig. Squared Corrected Model PWS1 .076 .181 PWS2 .008 .290 PWS3 .089 .172 PWS4 .219 .120 PWS5 .066 .189 PWS6 .178 .133 Intercept PWS1 .000 .957 PWS2 .000 .968 PWS3 .000 .959 PWS4 .000 .984 PWS5 .000 .969 PWS6 .000 .954 id_type PWS1 .076 .181 PWS2 .008 .290 PWS3 .089 .172 PWS4 .219 .120 PWS5 .066 .189 PWS6 .178 .133 Error PWS1 PWS2 PWS3 PWS4 PWS5 PWS6 Total PWS1 PWS2 PWS3 PWS4 PWS5 PWS6 Corrected Total PWS1 PWS2 PWS3 PWS4 PWS5 PWS6 a. R Squared = .181 (Adjusted R Squared = .108) b. R Squared = .290 (Adjusted R Squared = .228) c. R Squared = .172 (Adjusted R Squared = .099) d. R Squared = .120 (Adjusted R Squared = .043) e. R Squared = .189 (Adjusted R Squared = .117) f. R Squared = .133 (Adjusted R Squared = .056)

101

Post Hoc Tests

Type Homogeneous Subsets PWS1 Student-Newman-Keulsab Subset Type N 1 2 18 59.44 1 3 60.00 4 6 61.17 3 11 71.00 Sig. .301 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 130.567. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. PWS2 Student-Newman-Keulsab Subset Type N 1 2 4 6 57.83 2 18 63.94 63.94 1 3 75.00 3 11 75.00 Sig. .318 .175 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 112.699. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. PWS3 Student-Newman-Keulsab Subset Type N 1 2 18 59.50 4 6 61.00 1 3 64.33 3 11 70.73 Sig. .314 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 127.216. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05.

102

PWS4 Student-Newman-Keulsab Subset Type N 1 4 6 69.67 2 18 70.89 3 11 76.18 1 3 76.33 Sig. .453 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 61.689. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. PWS5 Student-Newman-Keulsab Subset Type N 1 2 18 62.33 1 3 63.67 4 6 65.83 3 11 73.09 Sig. .261 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 102.600. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05. PWS6 Student-Newman-Keulsab Subset Type N 1 4 6 60.17 2 18 62.17 1 3 62.33 3 11 71.45 Sig. .367 Means for groups in homogeneous subsets are displayed. Based on observed means. The error term is Mean Square(Error) = 145.374. a. Uses Harmonic Mean Sample Size = 6.188. b. Alpha = .05.

103

APPENDIX C

STATISTICAL ANALYSIS FOR STUDY 2

Variables in the dataset

Variable name Description

Descriptors id Unique identification number of valuation name Name of participant to which valuation relates type Type of participant to which valuation relates location 1 I-position to which valuation relates period 2 Period to which valuation relates

Affect scores aff01..24 Mean affect scores for 24 affects

Self-confrontation indexS Self-enhancement index indices 3 indexO Union-with-the-other index indexP Positive affect index indexN Negative affect index

Correlation coefficients corrgen Correlation with General feeling affect scores corride Correlation with Ideal feeling affect scores

Valuation details story Text of valuation mandala Theme of valuation Notes 1 Coded as: O = pertaining to I-position in country of origin; A = pertaining to I-position in Australia; G = pertaining to general feeling; I = pertaining to ideal feeling. 2 Coded as: 1 = pertaining to the past; 2 = pertaining to the present; 3 = pertaining to the future. 3 Computed from affect scores.

104

C.1 Analysis of Valuations by Theme

Cross-tabulations of valuations by theme, location and participant type.

Crosstabs

Case Processing Summary Cases Valid Missing Total N Percent N Percent N Percent Location * Mandala * Type 570 100.0% 0 .0% 570 100.0%

105

Location * Mandala * Type Crosstabulation Mandala Type +HH +O +S -LL -O 1 Location A Count 11 0 5 1 0 Expected Count 7.0 1.6 3.3 3.3 2.5 % within Location 61.1% .0% 27.8% 5.6% .0% % within Mandala 64.7% .0% 62.5% 12.5% .0% O Count 6 4 3 7 6 Expected Count 10.0 2.4 4.7 4.7 3.5 % within Location 23.1% 15.4% 11.5% 26.9% 23.1% % within Mandala 35.3% 100.0% 37.5% 87.5% 100.0% Total Count 17 4 8 8 6 Expected Count 17.0 4.0 8.0 8.0 6.0 % within Location 38.6% 9.1% 18.2% 18.2% 13.6% % within Mandala 100.0% 100.0% 100.0% 100.0% 100.0% 2 Location A Count 9 8 48 39 5 Expected Count 5.6 13.9 43.8 39.9 8.7 % within Location 7.7% 6.8% 41.0% 33.3% 4.3% % within Mandala 69.2% 25.0% 47.5% 42.4% 25.0% O Count 4 24 53 53 15 Expected Count 7.4 18.1 57.2 52.1 11.3 % within Location 2.6% 15.7% 34.6% 34.6% 9.8% % within Mandala 30.8% 75.0% 52.5% 57.6% 75.0% Total Count 13 32 101 92 20 Expected Count 13.0 32.0 101.0 92.0 20.0 % within Location 4.8% 11.9% 37.4% 34.1% 7.4% % within Mandala 100.0% 100.0% 100.0% 100.0% 100.0% 3 Location A Count 40 3 17 2 1 Expected Count 34.3 3.7 16.8 4.5 3.0 % within Location 63.5% 4.8% 27.0% 3.2% 1.6% % within Mandala 43.5% 30.0% 37.8% 16.7% 12.5% O Count 52 7 28 10 7 Expected Count 57.7 6.3 28.2 7.5 5.0 % within Location 49.1% 6.6% 26.4% 9.4% 6.6% % within Mandala 56.5% 70.0% 62.2% 83.3% 87.5% Total Count 92 10 45 12 8 Expected Count 92.0 10.0 45.0 12.0 8.0 % within Location 54.4% 5.9% 26.6% 7.1% 4.7% % within Mandala 100.0% 100.0% 100.0% 100.0% 100.0% 4 Location A Count 4 1 10 8 5 Expected Count 12.6 2.2 8.1 7.6 3.6 % within Location 10.3% 2.6% 25.6% 20.5% 12.8% % within Mandala 14.3% 20.0% 55.6% 47.1% 62.5% O Count 24 4 8 9 3 Expected Count 15.4 2.8 9.9 9.4 4.4 % within Location 50.0% 8.3% 16.7% 18.8% 6.3% % within Mandala 85.7% 80.0% 44.4% 52.9% 37.5% Total Count 28 5 18 17 8 Expected Count 28.0 5.0 18.0 17.0 8.0 % within Location 32.2% 5.7% 20.7% 19.5% 9.2% % within Mandala 100.0% 100.0% 100.0% 100.0% 100.0%

106

Location * Mandala * Type Crosstabulation Mandala Type -S Total 1 Location A Count 1 18 Expected Count .4 18.0 % within Location 5.6% 100.0% % within Mandala 100.0% 40.9% O Count 0 26 Expected Count .6 26.0 % within Location .0% 100.0% % within Mandala .0% 59.1% Total Count 1 44 Expected Count 1.0 44.0 % within Location 2.3% 100.0% % within Mandala 100.0% 100.0% 2 Location A Count 8 117 Expected Count 5.2 117.0 % within Location 6.8% 100.0% % within Mandala 66.7% 43.3% O Count 4 153 Expected Count 6.8 153.0 % within Location 2.6% 100.0% % within Mandala 33.3% 56.7% Total Count 12 270 Expected Count 12.0 270.0 % within Location 4.4% 100.0% % within Mandala 100.0% 100.0% 3 Location A Count 0 63 Expected Count .7 63.0 % within Location .0% 100.0% % within Mandala .0% 37.3% O Count 2 106 Expected Count 1.3 106.0 % within Location 1.9% 100.0% % within Mandala 100.0% 62.7% Total Count 2 169 Expected Count 2.0 169.0 % within Location 1.2% 100.0% % within Mandala 100.0% 100.0% 4 Location A Count 11 39 Expected Count 4.9 39.0 % within Location 28.2% 100.0% % within Mandala 100.0% 44.8% O Count 0 48 Expected Count 6.1 48.0 % within Location .0% 100.0% % within Mandala .0% 55.2% Total Count 11 87 Expected Count 11.0 87.0 % within Location 12.6% 100.0% % within Mandala 100.0% 100.0%

107

Chi-Square Tests Asymp. Sig. (2- Exact Sig. (2- Type Value df sided) sided) 1 Pearson Chi-Square 16.564a 5 .005 .002 Likelihood Ratio 20.847 5 .001 .001 Fisher's Exact Test 15.788 .002 N of Valid Cases 44 2 Pearson Chi-Square 14.085b 5 .015 .013 Likelihood Ratio 14.509 5 .013 .015 Fisher's Exact Test 13.901 .015 N of Valid Cases 270 3 Pearson Chi-Square 7.214c 5 .205 .199 Likelihood Ratio 8.526 5 .130 .166 Fisher's Exact Test 6.443 .243 N of Valid Cases 169 4 Pearson Chi-Square 27.227d 5 .000 .000 Likelihood Ratio 32.881 5 .000 .000 Fisher's Exact Test 29.188 .000 N of Valid Cases 87 a. 10 cells (83.3%) have expected count less than 5. The minimum expected count is .41. b. 0 cells (.0%) have expected count less than 5. The minimum expected count is 5.20. c. 5 cells (41.7%) have expected count less than 5. The minimum expected count is .75. d. 5 cells (41.7%) have expected count less than 5. The minimum expected count is 2.24.

108

C.2 T-tests

Performs t-tests for each participant separately, comparing indices for the country of origin I-position with those for the Australian I-position.

Name = 03 Yoana

Group Statisticsa Location N Mean Std. Deviation Std. Error Mean IndexS O 12 5.17 5.524 1.595 A 12 13.00 5.222 1.508 IndexO O 12 9.08 6.640 1.917 A 12 13.58 5.775 1.667 IndexP O 12 13.67 12.280 3.545 A 12 25.00 12.519 3.614 IndexN O 12 13.50 8.469 2.445 A 12 2.50 4.011 1.158 a. Name = 03 Yoana Independent Samples Testa Levene's Test for Equality of Variances

F Sig. IndexS Equal variances assumed .376 .546 Equal variances not assumed IndexO Equal variances assumed 1.783 .195 Equal variances not assumed IndexP Equal variances assumed .113 .740 Equal variances not assumed IndexN Equal variances assumed 5.016 .036 Equal variances not assumed

Independent Samples Testa t-test for Equality of Means

t df Sig. (2-tailed) Mean Difference IndexS Equal variances assumed -3.570 22 .002 -7.833 Equal variances not assumed -3.570 21.931 .002 -7.833 IndexO Equal variances assumed -1.771 22 .090 -4.500 Equal variances not assumed -1.771 21.586 .091 -4.500 IndexP Equal variances assumed -2.239 22 .036 -11.333 Equal variances not assumed -2.239 21.992 .036 -11.333 IndexN Equal variances assumed 4.066 22 .001 11.000 Equal variances not assumed 4.066 15.699 .001 11.000

109

Independent Samples Testa t-test for Equality of Means 95% Confidence Interval of the Difference Std. Error Difference Lower Upper IndexS Equal variances assumed 2.194 -12.384 -3.282 Equal variances not assumed 2.194 -12.385 -3.281 IndexO Equal variances assumed 2.540 -9.768 .768 Equal variances not assumed 2.540 -9.774 .774 IndexP Equal variances assumed 5.062 -21.832 -.835 Equal variances not assumed 5.062 -21.832 -.835 IndexN Equal variances assumed 2.705 5.390 16.610 Equal variances not assumed 2.705 5.256 16.744 a. Name = 03 Yoana

Name = 22 Juan

Group Statisticsa Location N Mean Std. Deviation Std. Error Mean IndexS O 19 7.42 4.880 1.120 A 17 9.24 5.238 1.271 IndexO O 19 6.37 3.562 .817 A 17 4.82 3.245 .787 IndexP O 19 14.89 9.398 2.156 A 17 16.82 10.460 2.537 IndexN O 19 20.47 12.267 2.814 A 17 13.41 11.457 2.779 a. Name = 22 Juan Independent Samples Testa Levene's Test for Equality of Variances

F Sig. IndexS Equal variances assumed .623 .436 Equal variances not assumed IndexO Equal variances assumed .003 .957 Equal variances not assumed IndexP Equal variances assumed .493 .488 Equal variances not assumed IndexN Equal variances assumed .976 .330 Equal variances not assumed

110

Independent Samples Testa t-test for Equality of Means

t df Sig. (2-tailed) Mean Difference IndexS Equal variances assumed -1.076 34 .290 -1.814 Equal variances not assumed -1.071 32.875 .292 -1.814 IndexO Equal variances assumed 1.354 34 .185 1.545 Equal variances not assumed 1.362 33.985 .182 1.545 IndexP Equal variances assumed -.583 34 .564 -1.929 Equal variances not assumed -.579 32.424 .566 -1.929 IndexN Equal variances assumed 1.779 34 .084 7.062 Equal variances not assumed 1.786 33.928 .083 7.062

Independent Samples Testa t-test for Equality of Means 95% Confidence Interval of the Difference Std. Error Difference Lower Upper IndexS Equal variances assumed 1.687 -5.242 1.613 Equal variances not assumed 1.693 -5.260 1.631 IndexO Equal variances assumed 1.141 -.773 3.863 Equal variances not assumed 1.135 -.761 3.851 IndexP Equal variances assumed 3.309 -8.654 4.796 Equal variances not assumed 3.329 -8.707 4.849 IndexN Equal variances assumed 3.970 -1.007 15.131 Equal variances not assumed 3.955 -.976 15.100 a. Name = 22 Juan

Name = 35 Luisa

Group Statisticsa Location N Mean Std. Deviation Std. Error Mean IndexS O 25 14.16 5.603 1.121 A 20 17.15 3.183 .712 IndexO O 25 12.80 7.263 1.453 A 20 17.25 3.416 .764 IndexP O 25 28.52 12.159 2.432 A 20 34.75 5.674 1.269 IndexN O 25 5.28 6.120 1.224 A 20 4.55 4.872 1.089 a. Name = 35 Luisa

111

Independent Samples Testa Levene's Test for Equality of Variances

F Sig. IndexS Equal variances assumed 7.446 .009 Equal variances not assumed IndexO Equal variances assumed 15.997 .000 Equal variances not assumed IndexP Equal variances assumed 5.818 .020 Equal variances not assumed IndexN Equal variances assumed 1.634 .208 Equal variances not assumed

Independent Samples Testa t-test for Equality of Means

t df Sig. (2-tailed) Mean Difference IndexS Equal variances assumed -2.125 43 .039 -2.990 Equal variances not assumed -2.252 39.213 .030 -2.990 IndexO Equal variances assumed -2.522 43 .015 -4.450 Equal variances not assumed -2.711 35.665 .010 -4.450 IndexP Equal variances assumed -2.111 43 .041 -6.230 Equal variances not assumed -2.271 35.520 .029 -6.230 IndexN Equal variances assumed .434 43 .666 .730 Equal variances not assumed .445 43.000 .658 .730

Independent Samples Testa t-test for Equality of Means 95% Confidence Interval of the Difference Std. Error Difference Lower Upper IndexS Equal variances assumed 1.407 -5.828 -.152 Equal variances not assumed 1.328 -5.675 -.305 IndexO Equal variances assumed 1.765 -8.009 -.891 Equal variances not assumed 1.641 -7.780 -1.120 IndexP Equal variances assumed 2.951 -12.181 -.279 Equal variances not assumed 2.743 -11.796 -.664 IndexN Equal variances assumed 1.681 -2.660 4.120 Equal variances not assumed 1.639 -2.575 4.035 a. Name = 35 Luisa

112

Name = 40 Carla

Group Statisticsa Location N Mean Std. Deviation Std. Error Mean IndexS O 28 13.71 7.644 1.445 A 32 11.09 6.438 1.138 IndexO O 28 13.50 6.619 1.251 A 32 6.84 5.984 1.058 IndexP O 28 26.93 15.299 2.891 A 32 16.72 15.544 2.748 IndexN O 28 15.36 12.708 2.402 A 32 22.28 12.290 2.173 a. Name = 40 Carla

Independent Samples Testa Levene's Test for Equality of Variances

F Sig. IndexS Equal variances assumed 1.589 .213 Equal variances not assumed IndexO Equal variances assumed .045 .834 Equal variances not assumed IndexP Equal variances assumed .284 .596 Equal variances not assumed IndexN Equal variances assumed .336 .564 Equal variances not assumed

Independent Samples Testa t-test for Equality of Means

t df Sig. (2-tailed) Mean Difference IndexS Equal variances assumed 1.442 58 .155 2.621 Equal variances not assumed 1.425 53.099 .160 2.621 IndexO Equal variances assumed 4.091 58 .000 6.656 Equal variances not assumed 4.063 54.951 .000 6.656 IndexP Equal variances assumed 2.557 58 .013 10.210 Equal variances not assumed 2.560 57.176 .013 10.210 IndexN Equal variances assumed -2.143 58 .036 -6.924 Equal variances not assumed -2.138 56.385 .037 -6.924

113

Independent Samples Testa t-test for Equality of Means 95% Confidence Interval of the Difference Std. Error Difference Lower Upper IndexS Equal variances assumed 1.818 -1.018 6.260 Equal variances not assumed 1.839 -1.068 6.309 IndexO Equal variances assumed 1.627 3.399 9.913 Equal variances not assumed 1.638 3.373 9.940 IndexP Equal variances assumed 3.993 2.217 18.203 Equal variances not assumed 3.989 2.223 18.197 IndexN Equal variances assumed 3.231 -13.392 -.456 Equal variances not assumed 3.239 -13.411 -.438 a. Name = 40 Carla

114

APPENDIX D

SELF-NARRATIVES OF CASE-STUDY PARTICIPANTS

Indices are denoted as follows: S = Self-enhancement; O = Union-with-the-other;

P = Positive affect; N = Negative affect; G = Generalisation; I = Idealisation.

Themes are denoted as follows: Strength and unity = +HH, Unity and love = +O,

Unfulfilled longing = –O, Powerlessness and isolation = –LL, Aggression and anger = –

S, Autonomy and success = +S.

D.1 Participant 03 Yoana

Valuations in country of origin, by location and period

Self-narrative S O P N G I Theme Past 1 A mark in my life is that at that time my mother used to see my father 0 0 0 19 -0.69 -0.78 -LL as the centre of her universe. Now that I am married I refuse to see my husband the same way. Instead, my kids are the centre of my life. 2 An episode I remember is my father visiting us. During that time my 2 2 3 24 -0.71 -0.78 -LL mother used to prepare us the best possible way, we had to wear our best clothes and our best behaviour before he arrived. But to our great disappointment he will come and sit there for the whole afternoon reading newspapers. He never talked to us, never took us to the park or cinema, etc. My mother made us sit there with him, just watching him. From that moment I always experience a sense of not being good enough. 3 Even though my parents divorced when I was 7 I still remember 6 8 31 0 0.62 0.61 going to his apartment and admiring his tidy, clean and impeccable place where you will find everything in order and the fragrance in the environment was so delicious to breathe. Today I do the same at home. Perhaps I inherited it from my father. 4 When I was 14 I clearly remember a scene like a scene from a movie 0 13 0 18 -0.30 -0.33 -O where we were in a party with very close friends. The son of one of those friends was only 4 and I recall that alcohol was offered to him and was encouraged to taste the delicious flavour and to my amazement Leonardo took it. From that moment I look at alcohol as a destructive drug. 5 When I was 19 my sister and I went to our father because she wanted 16 15 18 10 0.33 0.37 his permission to go to live in Spain, but my father said to her what about your sister? So she answered: I do not care about her. This made me very sad. 6 When I was 21 I reached the point that I did not understand why I 0 0 17 26 -0.42 -0.47 was living in a world full of materialism, alcohol, sex, drugs, violence, etc. And one day I was sitting down at the entrance of the building I used to live in and I asked God whether that good life or

115

world ever existed? And my tears came down and then I started a new spiritual commitment with myself. I decided to live a very pure and spiritual life. 7 When I was 16 my only sister was my idol. We lived together away 7 14 18 15 0.19 0.12 from home and she would make every single decision about me. One day she was getting ready for a party and I complimented her dress and the way she looked. One of the friends present told me I was full of complexes because I always complimented my sister and was so proud of her. But another older friend responded that I was not a “complex” girl. I was rather a pure and innocent person and my sister was the complex one. This episode made me very sad and unappreciated. 8 When my sister left I was 20 and I felt desolate and abandoned. I had 9 14 17 13 0.32 0.21 to learn everything from scratch to be on my own and standing on my own feet. Present 9 In my opinion my extended [Colombian] family is very significant. 0 14 0 18 -0.27 -0.34 -O There was an episode in my life that confronted me to this reality. My son was operated on as an unexpected emergency and my daughter was left at school with no one to look after her while I was with her brother. 10 My children were watching TV and I just learned on the phone that 5 13 26 5 0.48 0.46 +O my mum was flying to Australia to help us during my sickness. The nice surprise was that my sister told me on the phone that my mother was just one hour away before she landed in Melbourne. 11 The same month I happened to be diagnosed with three lumps in my 3 0 2 14 -0.36 -0.46 -LL breast. When I learned about it I was very upset and the doctor had to comfort me. There I felt more than ever the ongoing importance of having an extended family. Future 12 My mother has influenced me a lot. Her strength and patience have 14 16 32 0 0.77 0.83 made a mark in my life. Also I hold onto her and won’t allow her to go because I want my kids to experience not only a grandmother but somehow an extended family.

Valuations in Australia, by location and period

Self-narrative S O P N G I Theme Past 1 When I was 24 here in Melbourne I went to a Youth Baha’i party and 13 6 29 0 0.78 0.77 +S there were boys and girls all about the same age. I was so impressed by the fact that the youth were not drinking alcohol and yet they were enjoying themselves and having a good time. It impressed me a lot the clear environment that surrounded me that night. 2 Last year in December my children were to compete in swimming but 12 14 14 4 0.46 0.48 my son Habib was sick after the hard training because he suffers from asthma. Close to the date of competition I suggested to him to think about it and perhaps leave it till next year. He insisted he wanted to compete, even though … he was still not breathing well. Yet the next morning he was up, ready to compete. I talked to him and suggested he should not go ahead… When it was his time to get in [to the water] I was sitting down on the beach and when he walked towards me I saw a very week face and dark rings around his eyes, but I also noticed a great determination in those eyes—it was like a spark of

116

courage. He went ahead and completed [the task] successfully. He swam, but the last 5 minutes were an agony to my heart. 3 The first time my kids asked me what communion meant (and this 15 17 28 0 0.80 0.83 because they were taken to the church by their school) I explained to them the real meaning but also I taught them that religion can be wrongly interpreted and that in that particular case it has been taken only as a ritual. Present 4 My experience with my sister has made me aware of my situation 12 14 26 4 0.56 0.57 right now with my kids. I am very careful how to handle their personalities and being extra-cautious of allowing each of them to develop their own personalities. I can see now when my daughter sometimes belittles my son I immediately intervene and explain to them why she knows more and why he should not worry about it … they are at different ages and in different levels of their lives… 5 Another influence has been my aunty who suffered from depression 0 0 0 13 -0.55 -0.61 -LL when I was a little girl. When [this year] I suffered my own depression I decided never to be like her, therefore I had the strength to fight it and overcome it. 6 Classical music plays a very important role in my life. One day I was 19 19 35 0 0.90 0.96 +HH watching my daughter in a concert. She was playing a beautiful piece by Bach and I closed my eyes and I experience God. 7 I have a sign on my fridge that said “the child’s destiny lies in her/his 16 19 35 0 0.85 0.93 parents hands” and I have made a commitment to my children to try to develop all their talents and reach their full potential. 8 In my present existence the most important thing are my children— 19 19 35 1 0.95 0.98 +HH they are my centre, totally. 9 In November 2000 my husband went to have his 7th operation that 7 12 3 7 0.19 0.14 year. Just before he entered the theatre he was very weak and scared. I assured him that everything was going to be okay. Next I went to a little chapel in the hospital and I cried and asked Future 10 A goal in my future is to write book about motherhood. 15 12 27 1 0.80 0.80 11 Another goal is for me and my children to create a Music Fund or 15 13 33 0 0.89 0.89 institution that will help to nurture music abilities in children. I believe through music you can bring peace to the world. 12 My kids are my future because they are still very young and even 13 18 35 0 0.85 0.90 though they get older I am happy to be around for them. General feelings 17 16 35 4 Ideal feelings 20 20 40 0

117

D.2 Participant 22 Juan

Valuations in country of origin, by location and period

Self-narrative S O P N G I Theme Past 1 Throughout my upbringing I became naïve and trusting. My family 3 12 10 24 -0.40 -0.57 -O emphasised on me being good. 2 When I was 11, my brother and I went to Los Angeles to study 16 8 29 9 0.79 0.65 +S English for a year. This event set me apart from my peers. 3 My father is fair but inflexible. I learnt from him that rules were not 2 2 12 35 -0.25 -0.82 -LL to be bent or broken. 4 I was not wanted by my mother and somehow I learnt to be with 3 8 5 26 -0.48 -0.72 people who feel ambivalent about me. 5 With my mother I learnt to accept blame for anything and to be 5 6 8 34 -0.41 -0.82 obedient in order to please her. 6 I was very close to my brother. Sometimes I preferred to be on my 8 11 19 7 0.16 0.62 own, but still felt part of a pair. 7 My sister was very conflictive. She acted as if she was angry with the 5 4 5 27 -0.41 -0.71 world. When she apparently killed herself I admired her for it. 8 At school I felt insecure. I was a loner and was constantly failing 2 1 3 33 -0.34 -0.88 -LL subjects. I was shy and I learnt not to show off because I thought people would not like me. 9 I was never a competitive person and I gave up as soon as I saw a 6 9 13 32 -0.33 -0.85 competitor. 10 I admired my father. People admired him and I wanted to be like him 9 6 19 29 -0.04 -0.57 but I wanted to do it by myself. 11 At university I discovered the humanistic subjects and for the first 20 6 33 5 0.68 0.77 +S time I enjoyed schooling and studying. 12 My first relationship was about being attached to someone and trying 4 7 14 31 -0.32 -0.76 to please her. 13 I wanted to keep my marriage together no matter what because not 6 4 5 34 -0.29 -0.79 doing so would be failing society on my duty. I knew the relationship was wrong from the beginning, but I felt I had to do my duty. 14 After my first son was born I felt strong negative feelings towards 4 9 3 28 -0.40 -0.71 pregnancy and to a lesser extent towards my son. 15 I left my country of origin because I wanted to live in a peaceful 12 13 28 7 0.43 0.75 place and I realised that was not going to happen in my country in my lifetime. 20 I gave up on having a wife. I feel that I shed a part of me that was not 5 0 12 7 0.28 0.16 working. 23 Finishing the marriage was a turning point in my life because I started 9 4 17 13 0.30 0.17 achieving. Present 30 I don’t feel the need to communicate with my family. If t had a 10 4 21 5 0.53 0.64 +S problem I would talk to a friend here. Future 32 I don’t think I will go back to my home country to live. 12 7 27 3 0.74 0.83

118

Valuations in Australia, by location and period

Self-narrative S O P N G I Theme Past 16 The first three years was a real struggle. I felt I was going down- not 4 2 3 25 -0.27 -0.79 achieving academically or socially and with getting a proper job. 17 I felt that I did not belong to this society because people did not know 4 0 1 19 -0.06 -0.58 who I was. 18 I felt that I did not belong even to the groups I was participating in 2 0 1 30 -0.16 -0.82 -LL (e.g. University). 19 During the first 6 year in Australia I did not do much for myself and 0 0 0 39 -0.37 -0.96 -LL therefore achieved very little. I felt I was shutting down one aspect of my life after another. 21 I have always felt very close to my children. 6 8 15 27 -0.32 -0.75 22 After 5 years of studying I was asked to withdraw from university 1 1 5 24 -0.41 -0.85 -LL because I was failing, and I withdrew from my course and gave up. I was very depressed and I had very low self-esteem. 24 Not having a partner I felt that I could make my own decision and I 15 6 21 8 0.56 0.70 +S started to live my own life and not a family group life. 25 Getting my first professional job was a self-esteem boost for me. I 15 7 23 3 0.62 0.83 +S felt I finally belonged to a group of people that were similar to me. Present 26 I have a job. I’m still learning but I’m where I want to be. 13 4 25 7 0.75 0.71 +S 27 For the first time in my life I feel I am working for myself as an 15 6 26 6 0.71 0.82 +S individual. 28 I am still working on being assertive at work. 11 8 18 8 0.38 0.75 29 I am able to express my feelings and thoughts more easily now. 12 4 29 1 0.57 0.84 +S 31 I am happy living in Australia for now. 13 8 28 7 0.67 0.79 Future 33 I want to go back to study to expand my professional horizon. 13 4 24 4 0.66 0.78 +S 34 I feel that I have the rest of my life in front of me, and I need to 12 9 24 7 0.47 0.74 discover it. 35 I want to remain unattached for a couple of years and then find 9 6 20 10 0.62 0.47 someone with whom I can be myself. 36 I would like to relate to my children as individuals. 12 9 23 3 0.55 0.89

119

D.3 Participant 35 Luisa

Valuations in country of origin, by location and period

Self-narrative S O P N G I Theme Past 1 The frustration of not being recognised amongst the family members 0 0 0 10 -0.72 -0.71 (of 4 brothers) because I was a girl. Up to the age of 13. 2 At the age of 13 I went to live in the U.S. I had to learn to live away 3 0 0 6 -0.17 -0.22 from home and the loving care of my parents and brothers. 3 Whilst in the U.S. (for a period of 2 and a half years) I had to 8 0 0 19 -0.44 -0.46 -S conform with my grandmother’s ideas of discipline, and that involved performing a number of responsibilities in the household which did not befit a fourteen year old. 4 School life- year 7 and 8 was a fantastic experience, both academic 15 8 32 0 0.76 0.76 +S and social. 5 Childhood years- enjoyed school enormously and always excelled in 15 9 36 1 0.79 0.80 +S my grades. 6 Social life revolved around our cousins and our neighbourhood 9 16 31 10 0.69 0.69 +O friends. It was fun. 7 I treasure the memory where my father went to my school to drop the 20 20 35 0 0.93 0.94 +HH book he had bought for me- which he covered with beautiful paper and labelled with my name. I felt so special. 8 At the age of 15 I returned home to Mexico. Because of the education 11 5 27 3 0.56 0.57 +S system of the time I could not go to high school and therefore the opportunity of going to university was closed to me. My parents chose to send me to a secretarial course instead. This was quite disappointing for many years. 9 When I finished secretarial school I applied for work in the airline 19 9 38 0 0.86 0.87 +S industry. I was very lucky to get a job straight away with American Airlines. From there I moved on to a better position with Qantas airways. I was 19 at the time. This was to be an experience that opened my horizons and changed my life. 10 Being the only girl among four brothers suppressed my character. 16 2 25 4 0.49 0.50 +S And this promoted in me a great thirst for independence. I became very competitive with Eduardo, my brother, who was closest in age to me and I felt quite happy when I could earn a better salary [than him, working as a secretary]. 11 My job with Qantas lasted 10 years. It offered me a fabulous wide 19 12 37 0 0.88 0.89 +S experience of interesting work, very rewarding in every sense: intellectually, socially and financially. 12 During those same 10 years I was able to travel extensively to places 19 8 35 0 0.80 0.81 +S I never dreamed of. 13 After 10 years of working for Qantas I finally came to visit Australia. 16 14 37 0 0.93 0.94 I met my future husband during this trip. It was on the cards. 14 [After John’s marriage proposal] I came back to Mexico. John 20 20 39 3 0.99 0.99 +HH followed me a few months after and we were married. 15 Since my childhood I felt a special fascination with anything to do 17 17 37 0 0.95 0.96 with Australia. As if it was inside my system. And during the course of my first job I met the first person from Australia- that was such an exciting moment. So, 10 years later, and having met and got to know lots of Australians whilst still in Mexico, I finally arrived Down- under. From day one, I felt such familiarity and contentment with

120

everything here! 16 My relationship with my father was not a very close one due mainly 10 13 26 9 0.60 0.59 to his very strict upbringing- I never really related to him. 17 On the other hand I had a close relationship with my mother. As I 19 20 30 14 0.52 0.54 +HH have got older I have discovered many of her traits in my personality. 18 As my children reached the age of 10 and 12 years we felt we would 10 20 32 6 0.76 0.73 +O like to move to Mexico for them to absorb the Mexican culture; for all of us to live close to our extended family and cement our family values along the Mexican traditions. 19 The 8 years we spent in Mexico allowed us to enjoy a very rich 18 20 36 2 0.98 0.97 +HH experience of family life. The joy of participating closely with the family is hard to describe. The warmth of the people in general. 20 However, on the other hand, only 3 years after we arrived [in 10 10 12 20 -0.07 -0.09 Mexico] we encountered a series of very adverse circumstances which had a very negative effect on our stability, financially and emotionally. For my husband, a huge disappointment with the people. The lack of principles and honesty. We had to be alert and defensive at all times. So many circumstances totally beyond our control weighed down the joy of being there. Present 21 My bonding with my mother and brothers has not diminished for the 20 20 34 11 0.80 0.80 +HH fact I live away. Our communication is rich and constant. 22 I share my Mexican culture with my children here in Australia. We 19 20 35 1 0.92 0.93 +HH always speak Spanish amongst ourselves, enjoy our music, food, drinks and the spirit of our connection with Mexico. 23 My mother is still the uniting force of my extended family. I regret at 13 17 30 11 0.49 0.51 times the physical distance. Future 24 In the future there will always be a special bonding with my children 19 20 38 0 0.98 0.99 +HH which comes from sharing our Mexican culture. 25 One of my important goals in the future is to continue visiting 9 20 31 2 0.71 0.72 +O Mexico and to keep in contact with my family, friends and culture.

Valuations in Australia, by location and period

Self-narrative S O P N G I Theme Past 1 My first arrival in Australia was almost like a homecoming. 13 14 31 14 0.43 0.43 Everything was more than I ever expected. 2 The great feeling I always had for Australia was a great influence in 19 18 35 1 0.95 0.96 +HH the way I found myself so fulfilled when I came here. I had never felt that way before. 3 I met John, my husband, on my first visit to Australia. This was an 20 20 39 1 1.00 1.00 +HH encounter with a soul mate; we felt so happy in each other’s company from the beginning and we fell in love. 4 John proposed marriage a few months later after that first visit. This 19 16 39 2 0.93 0.93 prompted a second trip to Australia. I met his friends and his family and we in fact were getting to know each other better. All this while I felt I was living an adventure. 5 [After our marriage in Mexico] we came back to Australia and I felt I 12 19 36 4 0.90 0.89 +O had migrated to a new country. I had taken a big step.

121

6 Over a period of time I felt a combination of emotions. I soon 10 16 24 9 0.54 0.53 +O discovered I did not belong here. I had no friends or family to seek support from. 7 I tried a couple of jobs. My boss in the second job was someone who 15 17 32 8 0.84 0.83 understood my situation and gave me a lot of support. At the same time I started to attend the Alliance Francaise- little by little getting a new pattern of life here. 8 I left my job to start a business importing Mexican silver jewellery. 18 19 38 1 0.97 0.98 +HH This provided me with an interest and the freedom to be with John when we had time off work. 9 The arrival of our first child meant the end of my business. By this 20 20 39 1 1.00 1.00 +HH stage I had already become an Australia citizen and I felt totally at home. 10 The early years of my children were the happiest and the most 20 20 39 2 0.99 1.00 +HH rewarding of my life. I was completely fulfilled and did not want for anything more. 11 I became a freelance interpreter when the children started going to 20 11 39 2 0.86 0.87 +S school. This was a very interesting and rewarding experience, as it allowed me to become familiar with the issues of the migrant communities in Melbourne. 12 Almost at the same time I joined a refugee support group. It was 20 18 39 3 0.98 0.98 +HH wonderful to be able to help so many families to settle with a new life in Australia. 13 When my husband and I were trying hard to keep our enthusiasm 14 18 17 13 0.47 0.44 [while in Mexico] and our heads above water we happened to see the Australian film “Shine”. This gave us a great nostalgia for Australia and the goodness of this country. 14 Later on, we had dear friends from Melbourne spending Christmas 17 19 34 6 0.91 0.88 with us. We were so delighted in their company, catching up with all things Australian, the Aussie accent etc. when they went back our hearts sank, and shortly after, we made the decision to come back [to Australia]. Then it was a matter of time. Present 15 [Since] January 2000 we [are] back in Melbourne. It took me about a 18 20 36 4 0.96 0.95 +HH week to settle back in, it was like coming back to my old shoes. I knew we had challenges ahead, but I had great confidence because we knew we were in a place where there is law and order. 16 Our time here has given us the satisfaction to look back and know it 20 20 38 1 0.99 1.00 +HH was a good decision. Referring to the way in which our children responded. They fill very happy and fulfilled. 17 John and I bought a business which taught us a totally different 20 20 39 1 1.00 1.00 +HH industry—hospitality. We thoroughly enjoy our work and the stability it has given us once more. Future 18 Of great importance in the future in Australia is to see how my 14 10 32 16 0.29 0.30 children will take off with their own expectations and fulfilment of their lives. 19 As a goal for the immediate future is focusing on the relationship 15 20 34 0 0.87 0.88 with my husband as we contemplate the stage of life where it will only be he and I again. 20 Another very important goal is to remain active, interested and 19 10 35 2 0.80 0.79 +S involved in my business until I drop. General feelings 20 20 39 2 Ideal feelings 20 20 39 1

122

D.4 Participant 40 Carla

Valuations in country of origin, by location and period

Self-narrative S O P N G I Theme Past 1 My extended family was important. I felt I had a place in the family. 20 13 35 14 0.54 0.57 +S That place was validated. I was the first girl born in the family. I was doted on by all family members. 2 I remember being 5 and starting prep at what was an all boys school. 20 18 40 5 0.62 0.89 +HH Being one of the first girls I was chosen as a leader at school assemblies and was chosen for main characters at school plays. I had lots of friends. I remember I was well liked by all the kids. 3 I don’t recall any particular relative standing out for any particular 20 18 40 5 0.62 0.89 +HH reason, or any relative being more important than any other. I remember having a strong bond with all my family, in particular with my mother’s side. 4 I remember feeling special and having lots of self-esteem, feeling 20 18 40 5 0.62 0.89 +HH good about myself, being very out there, and being very affectionate towards my relatives. 5 Going on lots of holidays with immediate family. 20 18 40 5 0.62 0.89 +HH 6 I have not got emotional attachment memories to either of my parents 0 0 0 32 -0.57 -0.88 -LL while in Uruguay. 7 My brother was very close and we would walk to school together, and 9 8 27 18 0.24 0.15 play together after school (16 months apart in age). 8 My memories of my sister are when she was 8 months old and my 10 16 22 18 0.36 0.27 dad holding her on the frame of the new house. 9 My only other memories of my sister was on the plane trip to 0 15 0 35 -0.61 -0.66 -O Australia calling “socorro, socorro, llamen la policia” when the plane was falling. 10 I recall my dad saying we would go to live in Australia and I would 15 10 12 15 -0.11 0.09 love it as we would go to school on a ferry and not wear school uniform. 29 I wanted to die If I could not go back to Uruguay at age 14. my 0 0 0 20 -0.19 -0.63 -LL mother took me to Uruguay. 31 From the minute I stepped into Uruguay I felt whole, I felt I was 20 18 40 5 0.37 0.89 +HH someone. I felt I was alive. I was 14. I lived with my maternal grandmother for 3 months. I remember kissing, hugging and being very affectionate with my gran and all my family. 32 Falling in love at 14, my grandma and mother did not speak to me. 20 20 40 5 0.32 0.91 +HH They were angry with me, however I still felt whole. I was happy. I felt I was important to someone. I was worth something. 33 At 15 I was again taken from Uruguay against my wishes. But I felt I 1 5 1 35 -0.65 -0.80 was loved and felt he would follow me to Australia and that the love would be strong to conquer Australia. The plan was that when I was old enough we would return to Uruguay. 34 I was preparing to save, to marry Freddy when he arrived in 12 10 30 35 -0.16 -0.24 Australia. 35 The relationship lasted over 1 year until news came from overseas 0 0 0 35 -0.50 -0.91 -LL that he was in another relationship. Three months later he married. There was closure to our relationship. 37 When he (Freddy) got married I made a conscious decision to stop 8 13 17 24 -0.30 -0.21 my connection with the Latin community and let go of my Latin self.

123

38 My loss of my ultimate Latin relationship had ended, and I again felt I 12 12 24 24 #DIV/ #DIV/ did not belong in Uruguay either. 0! 0! 45 In 1999 my grandma became terminally ill. I felt it was my last 20 15 35 5 0.34 0.73 chance to be with her, and for the children to meet their grandmother. I went to the family court and begged for them to give me permission to travel with the children. Permission was granted. 46 I travelled to Uruguay in 1999 aged 32 with both children. There was 20 20 40 5 0.32 0.91 +HH a family reunion for my grandma, all her children and grandchildren were present. This trip was a celebration for my grandmother. For me it was a goodbye, but also a return to my childhood, being with everyone that I loved and loved me back. I belonged again, and this was now shared with my children, who finally got to experience the person I was. This was an amazing trip uniting my children to mine and their ancestry. During the trip I came to terms with the fact that we (my children and I) will always be Uruguayans. However, home was Australia if I wanted them to have opportunities. This trip highlighted that regardless of the difficulties I had faced in Australia, I was able to raise 2 children on my own, return to studies, obtain university qualifications, was employed in am excellent position, and was still able to be a full time mum. I am proud of what I have achieved as a sole parent. 47 After the trip I decided we would return to Uruguay for an extended 20 20 40 5 0.32 0.91 +HH period of 3 months, so the children could further experience being Uruguayan. This happened in 2002. 48 After this trip I again felt fulfilled. I felt complete and had achieved 20 20 40 0 0.48 1.00 +HH my, and my children’s identity. Present 49 I am proud of who I am and where I came from. I am at peace with 20 20 40 0 0.48 1.00 +HH who I am. 50 I speak Spanish as much as possible. On special occasions such as 20 20 40 0 0.48 1.00 +HH birthdays I cater Uruguayan food and Latin music. 51 I have lots of communication over the internet with family of origin. 20 20 40 0 0.48 1.00 +HH 57 At present my relationship with my mother and sister is “just there”. 8 6 13 33 -0.46 -0.83 It is somewhat of a façade. Future 52 Future goals—Uruguayan identity. My dilemma for future romantic 15 8 29 24 0.49 0.08 relationships is whether I want a future with a Latin family in Uruguay. Or will I not care what my future partners background is? This is because of the comfort that I find when sharing my feelings of my place of origin. 53 If I were to return to Uruguay, it would be to visit the country, and 14 17 29 23 -0.04 0.38 family if possible. But I no longer have the urge to be in Uruguay with my family, as I have come to terms with the fact that I don’t belong there, and have not been part of their day to day life for 30 years.

Valuations in Australia, by location and period

Self-narrative S O P N G I Theme Past 11 I remember we came to Melbourne instead of Sydney. This was 15 4 13 28 -0.08 -0.36 -S disappointing. My parents had not kept their promise. We arrived at a hostel (Springvale). I remember the fear of not being understood

124

when my dad told me to go and buy a coke at the milk bar. 12 I remember 2 days after arriving I had to organise my self, brother 0 5 0 28 -0.58 -0.69 and sister to go to breakfast in the large eating lounges, grab our own food, take my sister to kinder and then my brother and I had to go on a bus to school. I had to grow up and be independent in 2 days. 13 The 1rst day at school I screamed and cried for the teacher to go get 2 10 0 30 -0.46 -0.61 -O my brother. I hated Australia. 14 I was responsible for taking my younger sister to kinder. One day I 2 10 0 30 -0.46 -0.61 -O forgot to pick her up. It was 5 pm when I remembered. The kinder was closed and I was banging on the kinder doors. I was 7. 15 We moved to South Yarra. All community members at hostel move 16 12 36 8 0.45 0.75 to South Yarra after a year too. All children went to the same school so hostel friends were maintained. 16 School was difficult due to society identifying me as a wog. There 11 10 8 20 -0.20 -0.18 was lots of violence and lots of punch ups. 17 Home: parents worked. I was responsible for myself, sister and 10 0 9 34 -0.09 -0.66 -S brother to a degree. I had a key to our home and was responsible until parents finished work. “latch key kids”, “just my job”. I still resented coming here as I did not have this as my job in Uruguay. 18 Academically, my teacher continuously reinforced that English was 12 4 12 31 -0.25 -0.56 -S my second language and I would need to improve to get ahead. This also reinforced my wog mentality, that I did not fit in, that I was not an “Aussie”. 19 I was becoming more and more self-conscious, not belonging (8 to 10 12 4 12 31 -0.25 -0.56 -S years old). 20 I had physical fights with my entire grade (year 5). Boys would hit 10 0 0 36 -0.40 -0.82 -S me for no reason. I always fought back as I was told in the hostel that they (my parents) did not want me to come home crying because someone had hit me. If I came home crying I would be hit by them. 21 In grade 6 I became very popular. I was friends with most popular 17 11 31 7 0.68 0.80 +S kids in the school. I felt like I did in Uruguay. 22 In year 7 and 8 I hated school. I felt like I did when I arrived in 10 0 0 36 -0.40 -0.82 -S Australia, out of place. 23 Socially I had lots of friends. I had a best friend Virginia, and I had a 9 6 18 22 0.04 -0.22 second best friend. We would do everything together as a group. 24 In year 8 I hated school. I wagged and felt no connection with school. 10 0 0 36 -0.40 -0.82 -S 25 I had a feeling of belonging only with a group of Spanish speaking 15 11 34 13 0.70 0.64 kids. 26 My dad was physically abusive towards me. 0 0 0 30 -0.54 -0.82 -LL 27 I felt isolated within my family. I felt singled out in a negative way. I 6 0 2 30 -0.30 -0.74 -S had no one to turn to. 28 I felt my mum did not protect me from my dad. 0 0 0 25 -0.37 -0.73 -LL 30 From the age of 7 to 14 I felt no sense of self. Other than year 6, I 5 0 7 22 -0.19 -0.55 didn’t belong anywhere. I was in limbo. 36 My job was very successful. I was very happy. I had a purpose. I felt 20 15 40 0 0.57 0.91 important. I was financially independent and was saving my money. I bought our clothes, food etc. I no longer had to rely on my parents other than for a roof over my head. 39 I focused on work and went back to school, and made a new group of 19 11 38 6 0.50 0.76 +S friends (17yearsold). 40 At 17 and a half I entered my second major relationship. I felt I had to 13 12 24 23 0.00 0.18 prove that I could be loved and was worthy of that love. I had to prove this to my parents. 41 I had 2 children from this relationship. With the 1rst child I was very 20 15 37 3 0.56 0.84

125

happy in the relationship and excited about the birth. 42 The second child was planned by me, even though I was aware that I 10 10 25 39 -0.05 -0.49 no longer wanted to be in the relationship. My main concern was for my first born to have a full brother or sister (for a sense of belonging). 43 The separation was very violent. I had no support from extended 6 0 0 35 -0.42 -0.81 -S family. I felt that if I was in Uruguay this would not have occurred this way as family members would have intervened. I had no protection. 44 Family court—share guardianship. Rob did not allow me to travel to 5 0 0 35 -0.42 -0.81 Uruguay with the children; therefore, in total I did not return to Uruguay for 17 years. Present 54 Watching my children turn into young adults, watch them achieve. 17 15 34 19 0.34 0.48 55 I am currently enjoying my work, but am evaluating long term goals 20 11 40 5 0.74 0.79 +S and exploring future employment options. 56 Still coming to terms with friendships. I have lots of friends, but at 15 6 23 21 0.61 0.04 the moment I have not got a friend that I would consider my soul mate. Future 58 Goals: (1) I am looking forward to seeing Europe for my fortieth 20 17 38 2 0.50 0.96 birthday. Goal (2): 2: Helping my children to be the best people they can be. Support them in reaching for the star. 59 Eventually connecting with someone romantically in the long term. 8 2 16 28 0.01 -0.54 -S 60 My goals: Continue to enjoy being me, travelling, purchasing a 20 18 38 0 0.53 0.99 +HH house.