The Impact of Race, Class and Gender on Second-Generation Caribbean Immigrants’ Assimilation Patterns Into the United States

Home , Dominican Americans, Sociology of immigration

The Impact of Race, Class and Gender on Second-Generation Caribbean Immigrants’ Assimilation Patterns into the United States

Dissertation

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in The Graduate School of The Ohio State University.

Mauricia Alissa John, M.Sc.

Graduate Program in Rural Sociology

The Ohio State University

2012

Dissertation Committee

Dr. Linda Lobao, Advisor

Dr. Joseph Donnermeyer

Dr. Kristi Lekies

Dr. Lazarus Adua

Mauricia Alissa John

2012

ABSTRACT

The premise of this research rests on the idea that race, class and gender are all central to the immigrants’ experience and that assimilation into the dominant culture is influenced by the immigrants’ national origin, the immigrants’ gender and his or her family’s socioeconomic status.

I draw from the classical assimilation model proposed by (Gordon, 1964) to construct measures of assimilation and I also employ the contemporary assimilation model proposed by Gans (1992) and Portes and Rumbaut (1996) to examine the ways in which race, class and gender affect the immigrants’ registration to vote, proficiency in

English, educational achievement and the decision to out-marry/partner. I employ the

Children of Immigrants’ Longitudinal Study to determine the assimilation patterns of second-generation immigrants from Cuba, Dominican Republic, Haiti, Jamaica and the

West Indies to the United States. I examine the effects of parental social class on the subsequent assimilation patterns of their children in young adulthood at an average age of

24.

For race and ethnicity, I hypothesize that there should be no racial and ethnic differences relative to language proficiency. I hypothesize that racial and ethnic groups

ii with weaker ties to their countries of origin should be more politically active in the host society. I hypothesize that those immigrants belonging to larger ethnic enclaves should be less likely to out-marry/partner. Lastly I hypothesize that Afro-Caribbeans should be likely to attain higher levels of education because they are economically motivated to migrate.

With regard to social class, I hypothesize that parents with higher levels of economic resources and education should rear children that are more proficient in English and who attain higher levels of education. I also hypothesize that immigrants belonging to higher socio-economic groups should be more likely to out-marry/partner because they are likely to form ties and stronger relationships with the dominant culture. Lastly, I expect that immigrants with higher levels of education should be more politically active because they understand and value the importance of civic involvement.

In terms of gender, I hypothesize that girls should be more proficient in English compared to boys due to the gender dynamic in child rearing in the Caribbean. I hypothesize that women should be more likely to out-marry/partner compared to men due to the transformation in gender roles. Lastly, I expect that there should be no differences between males and females relative to civic engagement and educational attainment.

Results indicate that race, class and gender are all significantly related to the assimilation of second-generation Caribbean immigrants. However, class trumps race for this population sample—parents’ socioeconomic status is most significantly related to the four measures of assimilation. Even though both classical and contemporary models of

iii assimilation were used in the dissertation to guide the research, findings suggest that neither of these models fully explain the integration of second-generation Caribbean immigrants. This study shows that second-generation Caribbean immigrants are more representative of a linear path as opposed to a segmented path towards assimilation.

DEDICATION

Dedicated to my parents Irmin and John-Albert, my sister Marsha John and my grandmother, Gloria John.

ACKNOWLEGEMENTS

I am thankful to all of my advisors for their selfless amounts of time and supervision, and for being my mentors throughout my graduate career in the rural sociology program. I am indebted to Dr. Linda Lobao because she has been instrumental in my admittance into the rural sociology program and she has continued to guide and shape me into the academic scholar that I am today. I am grateful to her for teaching me to treat my dissertation not just as the end of my graduate work, but the beginning of a wonderful career and future ahead of me. Her goal has always been for me to mature scholastically, capitalize on my strengths and to create the most meaningful and competitive work in my field of study. I am obliged to Linda for her consistently brilliant ideas for improving my research --thank you Linda for helping me to see the “forest and not just the trees.” Without you I would not be the rural sociologist that I am today.

Dr. Joe Donnermeyer has been a great mentor; his advice to me has always been direct, intelligent and practical. On many occasions Joe has helped me to see the humor in situations by teaching me to be less intense, which is my usual approach to my work and career. Joe has also taught me the art of making light of situations that would

vi otherwise seem devastating to me. Thank you Joe for your continued support, advice and for breaking up the monotony with your eclectic sense of humor and stories that always put a smile on my face. I am forever appreciative to you for molding me into becoming a great lecturer.

Dr. Cathy Rakowski has also been a great source of knowledge and guidance throughout my journey; she has been instrumental in assisting me in honing in on my research questions and she always goes above and beyond in any task of which I have asked her to be a part. Thank you Cathy for giving of your time and effort selflessly during this process. Moreover, I am truly thankful to Linda, Joe and Cathy for the endless letters of recommendations they wrote on my behalf; you have all been instrumental in helping me to secure my job at Kutztown University of Pennsylvania; thank you for your patience and willingness to help me succeed in my career.

Dr. Kristi Lekies, though I’ve only recently gotten to know her, has always been willing, enthusiastic and has generously rendered her time and assistance to me. Thank you Kristi for your words of wisdom, for your support and patience throughout this process. Dr. Lazarus Adua has also been a great inspiration to me throughout the months leading up to my defense. He has given of his time, knowledge and expertise in statistical modeling unselfishly. Thank you Lazarus for your time, advice and all of your hard work in helping me realize my goals.

My dissertation and my journey over the past four years would not have been possible without the assistance, support and encouragement from my parents. I will be

vii forever grateful for the emotional and financial support they provided me throughout this process. My mother Irmin, has always been for me, a voice of reason, calm and reassurance; she taught me the value and importance of prayer and always provided that vote of confidence when I encountered obstacles throughout my academic career. My mother has always been my mentor, friend and supporter. My father, John though he never says much has always been my pillar of strength, his foremost hope for me is regardless of the path I take in life, his support, love and provision for whatever I needed would be unending. My dad remains my silent motivator and voice of reason. My sister

Marsha continues to believe in my dreams and her advice has also been invaluable along the way—she has been my best friend and exemplary throughout this process. My grandmother Gloria has been a tremendous source of strength, her prayers, well wishes and positivity has kept me motivated, inspired and optimistic about my future and direction in life. My uncle Clinton has also been instrumental in my graduate career; I am indebted to him for his generosity, support and eagerness to help me succeed.

I am also thankful to Brian Soller and Dmitry Tumin, who have both assisted me tremendously with the statistical modeling of my research. Their advice, knowledge and expertise have been crucial in the successful completion of my dissertation and words cannot express how thankful I am for their assistance, proficiency, and your generosity of time.

I have also to thank, my friends and colleagues in the rural sociology program for their support, positivity and encouragement over the past four years. I am grateful to have

viii met such intelligent, industrious and committed rural sociologists. Thank you to, Joe

Campbell, Corey Anderson, Rebecca Som Castellano, Godwin Apaliyah, Molly Bean,

Danielle Deemer, Hong Anh Nguyen, Angela Thatcher and Anjel Stough-Hunter for your support throughout my journey. Lastly, I am obliged to the administrative staff here in the school for their efficiency and willingness to get things done in a timely manner—thank you to Amy Schmidt, India Fuller and Mary Capoccia for your organization and proficient service to me as a graduate student.

VITA

May, 2005…………………………………………………B.S. Business Economics

June, 2007……………………………………………M.Sc. Development Economics

September 2008 to Present…………………………..Graduate Teaching Associate,

School of Environmental and Natural Resources, The Ohio State University

Fields of Study

Major Field: Rural Sociology

TABLE OF CONTENTS

Abstract..…………………………………………………………………………………..ii

Dedication …………………………………………...... ………………………………….v

Acknowledgements……………………………………………………………………. ...vi

Vita………………………………………………………………………………………..x

List of Tables.....………………………………………...... xii

List of Figures……………………………………………………………………………xv

Chapter 1: Introduction………………………...……..…………………………………...1

Chapter 2: History of Caribbean Immigrants…………………………………………....29

Chapter 3: Review of the Literature...... 57

Chapter 4: Methodology ……………………………………………………………….110

Chapter 5: Results and Analysis ……………………………………………………….156

Chapter 6: Conclusion…………………………………………..……………………...199

References………………………………………………………………………………220

Appendix A……………………………………………………………………………..243

Appendix B…………………………………………………………………………...... 245

LIST OF TABLES

Table 4.1. English Proficiency………………………………………………………….127

Table 4.2. Registration to Vote……………………………………………………… 128

Table 4.3. Respondents in the Study that are Single, Out-married/partnered and In- married/partnered……………………………………………………………………….130

Table 4.4. Educational Attainment…………………………………………………… 132

Table 4.5. Distribution of Males and Females in the Sample………………………….134

Table 4.6: Demographics for the countries used in the Study………………………….136

Table 4.7 Racial and Ethnic Self-Identification of Respondents……………………….137

Table 4.8. Data Used to Create the Five-Proposed Racial and Ethnic Categories……..138

Table 4.9. The Distribution of Respondents from Five Different Caribbean Islands… 139

Table 4.10. Family’s Economic Situation………………………………………………143

Table 4.11. Homeownership Status…………………………………………………… 144

Table 4.12. Family Structure………………………………………………………… 145

Table 4.13. Mother Currently Employed in Regular Occupation………………………146

Table 4.14. Correlation Matrix for Parents’ Education and Job Prestige………………146

Table 4.15. Correlation Matrix for all Five Measures of Social Class…………………146

xii

Table 4.16. The Highest Number of School Years Completed………………………...147

Table 4.17. Respondent’s Education Expectation……………………………………...149

Table 4.18. Respondent’s Exposure to Discrimination………………………………...150

Table 4.19. Respondent Currently Residing with Mother……………………………...151

Table 4.20. Respondents’ Citizenship Status…………………………………………...152

Table 4.21. Length of Stay in the U.S…………………………………………………..153

Table 4.22. Expected Signs for all Four Measures of Assimilation……………………155

Table 5.1. Language Proficiency Logistic Regression…………………………………164

Table 5.2. Language Proficiency Racial and Ethnic Comparisons…………..…………165

Table 5.3. Political Participation Logistic Regression………………………………….169

Table 5.4. Political Participation Racial and Ethnic Comparisons……………………..170

Table 5.5. Out-marriage/partnership Logistic Regression……………………………...176

Table 5.6. Educational Attainment Logistic Regression……………………………….181

Table 5.7. Educational Attainment Racial and Ethnic Comparisons…………………...182

Table 5.8. Language Proficiency-Nationality Logistic Regression…………………….187

Table 5.9. Language Proficiency-Nationality Comparisons……………………………188

Table 5.10. Political Participation-Nationality Logistic Regression…………………...190

Table 5.11. Political Participation- Nationality Comparisons………………………….191

Table 5.12. Educational Attainment-Nationality Logistic Regression…………………194

Table 5.13: Educational Attainment- Nationality Comparisons………………………..195

xiii

Table 5.14. Overall Effects of Race, Class and Gender on Four Measures of

Assimilation…………………………………………………………………………….198

A 1: Mean and Standard Deviation for all Variables Used in the Study……………….243

B 1: Correlation Matrix for English Proficiency Using Nationality……………………246

B 2: Correlation Matrix for English Proficiency Using Race………………………….246

B 3: Correlation Matrix for Educational Attainment Using Nationality……………….247

B 4: Correlation Matrix for Educational Attainment Using Race…………………...…247

B 5: Correlation Matrix for Out-Marriage/Partnership Using Race……………………248

B 6: Correlation Matrix for Political Participation Using Nationality…………………248

B 7: Correlation Matrix for Political Participation Using Race………………………..249

xiv

LIST OF FIGURES

Figure 2.1. The Caribbean Islands Used in the Research, Cuba, Dominican Republic,

Haiti, Jamaica and the English speaking West Indies…………………………………....30

Figure 2.2. The Location of the Gulf of Mexico, Caribbean Sea, Atlantic Ocean and the

Greater and Lesser Antilles………………………………………………………………31

Figure 2.3. The Mariel Boatlift…………………………………………………………..43

2.4: Sugarcane Workers in Montego Bay, Jamaica, 1900……………………………….45

Figure 3.1. The Number of Legal Immigrants Arriving to the U.S. between 1820 and

2010………………………………………………………………………………………63

Figure 3.2. The Number of Legal Immigrants Arriving to the U.S. from Various Regions between 1820 and 2010………………………………………………………………….63

CHAPTER ONE

INTRODUCTION

General Background

Every year immigrant men, women and children leave the developing world and enter developed nations in search of economic success, political freedom, to escape religious persecution and to reunite with friends, families and loved ones. The proposed research focuses not on the push factors of migration, but rather, what happens to the immigrant after he or she has arrived into the dominant culture. Very often in developed countries there are political debates and issues about the ways in which governments of various receiving countries should address the trend of increased immigration and cultural complexity both for legal and illegal immigrants. Proponents of immigration argue that immigrants should be given a pathway to naturalization and citizenship and they are also against the dismantling of the family structure through the deportation of undocumented immigrants. Alternatively opponents of both documented and undocumented immigration are not in favor of local jobs being taken away from citizens

1 and they argue that undocumented immigrants are evading taxes and are not contributing economically to the well being of the economy.

Since 1965, the passage of the landmark Immigration Act facilitated a period of mass immigration to the U.S. Previously earlier immigrants were primarily of European descent. However, today’s immigrants are more ethnically and culturally diverse, composing of immigrants from Latin America and the Caribbean, Asia and Africa. This new wave of immigrants is often referred to as “new immigrants.”

Much scholarly research has debated the differences between the experiences of

European immigrants and their descendants and new immigrants and their children. Alba and Nee (1997, 2003) argue that if the earlier waves of European immigrants and their descendants can be characterized as successfully assimilated into the American mainstream, then we should not automatically expect the same or similar paths of assimilation among new immigrants and their children.

Migration and assimilation patterns continue to be of grave concern in the U.S. and many other first world nations today. Various racial and ethnic groups from around the world continuously seek improved living conditions, greater job stability and improved life chances for themselves and their families. Immigration is not restricted to race, class, gender or socio economic status. People from all walks of life migrate both legally and illegally in search of better opportunities and to support their families, who are living in less than suitable conditions in many developing nations. Migration is not just economically driven, but also politically, socially and religiously motivated.

Both economists and sociologists are interested in the role of migrants in the labor market. Sociological theories of migration have attempted to overcome the neo classical approaches, which focus almost exclusively on factors associated with the labor market.

Sociologists have extended the scope of migration theory from a rational choice (micro- level) and systems (macro-level) approach to include a social component by explaining the inter-personal decision making process based on such factors as social ties, social networks, social capital and assimilation into the dominant society.

The topic of my research focuses on the impact of race/ethnicity, class and gender on the assimilation and adaptation processes of second-generation Caribbean immigrants into the United States. More specifically I examine the extent to which the race/ethnicity, class, gender, and other social factors such as discrimination, family structure, aspirations and citizenship status all influence the assimilation trajectory into the U.S. as the destination country. I employ four proxies, which will act as measures of assimilation because the adaptation process is not directly observable. The four central models to my research are language proficiency, out-marriage, political participation and educational attainment. These measures of assimilation will be further discussed in my methodology.

The relationship between assimilation and the well being of immigrant children has been the focus of debate in recent sociological literature. Much of this work has questioned whether classical theories of immigrant adaptation, which assumed assimilation to be an integral part of the process of upward mobility for immigrants, are still applicable to today’s immigrant children. The economic, social, and cultural effects

3 of assimilation on the second-generation have also been widely debated. Accordingly, many researchers and scholars believe that the long-run implications of this wave of immigration will be primarily determined not by what happens to the immigrants themselves, but by the outcomes of their children. While only 11% of the total population is foreign-born (Malone et al. 2003), a full 20% of children under age 18 are part of immigrant families (Hernandez, 1999) either as immigrants themselves, or as the U.S.- born children of immigrants.

The proposed research focuses on those immigrants who were either born in the

United States or brought to the U.S. at a very young age. Baseline information has been collected on their background, aspirations, educational attainment, and family’s socio economic characteristics. I employ a cross-sectional analysis of the data to fully examine the extent to which race/ethnicity, class and gender affect the ways in which these immigrants assimilate differently or similarly into the dominant culture. My goal is to shed light on whether or not race, class and gender still matter in the assimilation process for a more homogeneous group of Caribbean immigrants.

MOTIVATION OF THE STUDY

During the periods before World War I, international immigration was involuntary because slaves were deployed from West Africa into the plantations of the

‘New World’; a form of forced migration. However, in recent years, international immigration has become more voluntary with developed nations implementing immigration policies to encourage voluntary migration from developing to developed

4 nations. Migration remains a controversial topic historically and even today. Politicians have heated debates on whether policies should be enacted, eradicated or amended in order to facilitate the movement of immigrants into the U.S.

As an immigrant from Trinidad and Tobago, my interest has always been in migration patterns and the various ways in which immigrants assimilate into the host country. More specifically, it has always been fascinating to me how immigrants are able to integrate into the dominant culture and simultaneously retain elements of their own culture, which in turn creates a hybrid identity and lifestyle. Not only are immigrants faced with decisions to assimilate but also, they are restricted and sometimes forced to assimilate into certain social groups due to their race and ethnicity, overt forms of discrimination as well as the absence of mobility ladders, which would enable them to move ahead in the host country. Much of the assimilation literature has focused on Latin

American countries and has failed to take into account the integration of Afro-Caribbean natives into the United States. Many studies have focused on the upward mobility of particularly Cuban immigrants and their patterns of assimilation into the U.S. The proposed research extends the literature beyond Latin America and further employs a sensitivity analysis, which compares racial and ethnic differences among Cubans,

Haitians, Dominicans, Jamaicans and other West Indians.

Both linear and segmented models of assimilation have been proposed when positioning immigrants into the dominant culture. More recent researchers like Alejandro

Portes and Ming Zhou (1993) argue against proponents of the classical linear assimilation

5 model like Gordon (1964), who contends that immigrants assimilate solely through a linear process, where over time they completely abandon their native traditions and fully adopt the behaviors and traditions of the dominant culture, hence becoming fully assimilated within the mainstream culture. As a counter argument to linear assimilation,

Portes and Zhou (1993; 2006) introduced the segmented assimilation model in which they show that assimilation among immigrants can take one of two trajectories whereby immigrants either assimilate upwardly into the middle class or downwardly into the underclass. This dissertation integrates elements of both the segmented assimilation model and the classical model of assimilation to examine the adaptations of second- generation Caribbean immigrants into the United States based on their race, class and gender.

THEORETICAL FRAMEWORK

Migration Theory

In sociology there have been broad conceptual approaches to the study of labor migration from developing to developed countries. These approaches include functionalism, structuralism, dual labor market and middle range frameworks including social capital.

The functionalist approach focuses on micro economic processes, particularly the individual decision making processes. The individual cost-benefit model has been the most widely used and therefore the most influential framework with which to study migration. Also known as the neoclassical economic theory, the individual cost-benefit

6 model considers migration to be the outcome of rational individual decision-making when the existing wage differential between two countries is substantially large (Massey et al., 1993, 433). Additional variables that are important are migration costs and the probability of gaining employment. Therefore, the individual will choose to migrate if the expected net return is significantly greater than that from his existing employment plus migration costs (Massey et al., 1993, 434). High levels of individual capital, i.e. education, language skills, connections in destination countries, increase the net gain to be achieved and thus the probability to migrate as well. Under neoclassical economic theory, Sjaastad (1962) and Todaro (1969) suggest that international migration is related to the global supply and demand for labor. Nations with scarce labor supply and high demand will have high wages that pull immigrants in from nations with a surplus of labor.

The structural approach focuses on the macroeconomic processes of migration.

The approach argues that international migration is caused by socio-spatial inequalities that constrain the life chances of individuals and members of specific social classes in particular places (Koser, 1997). For the stucturalists, migration is a result of socio-spatial inequalities systematically reproduced within national and global economies.

The dual labor market theory argues that international migration is caused by a permanent demand for immigrant labor that is inherent to the economic structure of industrialized countries. According to Massey et al., (1993) international immigration is not caused by push factors such as unemployment in developing countries, but by

7 increasing need for foreign workers to work in industries of the developed countries. This school of thought assumes two markets: a relatively stable high-income market and a coexisting sub-economy with low incomes, low job security, and often dangerous working conditions. Migrant labor, although affecting both of these markets, is essentially a defining characteristic of the latter.

Wages in the secondary economy are relative to the level of the work force willing to work in such conditions. Since the secondary economy has low returns for social capital, i.e. education, skill, and past work experience, native workers shy away from working in these menial jobs. Immigrants fill this void, yet the higher the level of immigration the lower relative wages will be. Thus, wages are caught in a vicious cycle of immigration and labor needs. Little immigration will raise wages but also the need for labor, which in turn will lead to increased immigration and lower wages (Massey et al.,

1993, 433).

The aforementioned sociological theories have been criticized due to numerous shortcomings and caveats, primarily their failure to address pertinent issues regarding migration. The functionalist model fails to recognize social categories such as gender, ethnicity and social class of immigrants, and the structural theory does not account for political and non-economic barriers to mobility (Goss and Lindquist, 1995). The Dual

Labor Market Theory is victim to its own assumptions. The division of the market into two sectors does not always hold, and it is difficult to empirically prove that immigration is demand-driven. In response to these shortcomings, alternative theories have been used

8 to explain international migrations movement and how they adapt and adjust in their destinations (Koser, 1997). The social capital approach has been proposed as one possible appropriate middle range framework to apply to immigrants’ adjustment strategies.

Migrant networks, previously not addressed, both facilitate migration in sending countries and help change attitudes toward migration in receiving countries. Essentially, each act of migration is considered separately as each has a separate social context, but they do have certain common variables. Moreover, each act of migration makes it easier for other migrations to occur, as it alters the social context toward more flexibility. Cumulative causation specifically refers to the self-reinforcing cycle that perpetuates and increases migration (Massey et al., 1993). Under the social capital approach, migrant networks are defined as social ties that bind former and future migrants. The more developed these networks, the higher the propensity toward migration. These self-reinforcing cycles both reduce costs and risks in migration, and increase the net benefits gained from it.

Though my research does not directly focus on decisions and motivations to migrate, it is equally important to understand the theoretical underpinnings of migration before we can fully analyze and draw conclusions on the process of assimilation within the dominant culture. Since the motivation to migrate varies among racial and ethnic groups, migration theory will be instrumental in guiding the research and further explaining the ways in which second-generation immigrants assimilate after they have settled into the mainstream society and culture.

CLASSICAL SOCIOLOGICAL THEORY AND STRATIFICATION IN

MIGRATION

Migration has concerned sociologists over the course of many years. August

Comte, Emile Durkheim and Karl Marx have all been interested in the movement of people and the effects of migration. International migration is a powerful symbol of global inequality and these inequalities are reflected in the form of economic, political, social and spatial differences. Millions of families move each year across borders and continents seeking to reduce what they see as a gap between their own position and that of people in wealthier places. In turn there is a growing consensus in the development field, that migration represents an important livelihood diversification strategy for many in the world’s poorest nations (Ammassari et al., 2004).

International migration is a powerful symbol of global inequality; this uneven distribution can be seen in wages, labor market opportunities, and lifestyles. Migration and social inequality are related by the position or status that is often assigned to immigrants when they enter a new country of residence. The immigrants are offered lower job positions than the average citizen and they are usually underpaid. This situation leads to them swelling the number of poor and unemployed, with a small number of chances to change their social condition and frequently not benefiting from human rights guaranteed to others.

Social stratification and inequality refers to the hierarchical arrangement of individuals into divisions of power and wealth within a society. Various theoretical

10 perspectives of sociology interpret concepts of stratification differently. Advocates of functionalism have suggested that since stratification is generally found in developed societies, hierarchy must be needed in order to maintain stability.

Durkheim (1893) in ‘The Division of Labor’ believed that society is conceived of as a system of interrelated parts in which no part can be understood in isolation from the whole. A change in any part is seen as leading to a certain degree of imbalance, which in turn results in changes in other parts of the system and to some extent to a reorganization of the system as a whole. Durkheim stated that stratification and inequality was inevitable in every society and it was necessary in order for society to function. He saw competition as allowing those who had the most resources and talents to maintain their positions and assume high rank positions, whereas the less fit seek alternative specialties to mitigate competition.

Linking functionalist sociological theory to the migration research, one can deduce that the global labor market is a system; therefore the demand and supply of labor will adjust accordingly for the entire system to maintain equilibrium and function cohesively. Since migration between developing and developed nations is related to the global demand and supply of labor, nations with scarce labor supply and high labor demand will have high wages that will then pull immigrant workers in from nations with a surplus of labor. Immigration is consequently a response to inequality in wages and labor supply between developing and developed nations, and helps to maintain the functioning of the global labor market system.

Unlike Durkheim and Parsons who believed that stratification and inequality is essential in order to stabilize social structure, conflict theorists such as Karl Marx point to the inaccessibility of resources and the lack of social mobility found in stratified societies. Karl Marx addressed the inroads of capitalism and its corruption of other societies. In “The German Ideology” Marx writes, in reference to capitalists and their goods, “in place of old wants satisfied by the production of the country, we find new wants requiring for satisfaction in the production of distant lands and culture.” He also writes, “In place of the old local and national seclusion and self-sufficiency we have intercourse in every direction, universal interdependence of nations” (Marx, 1845). From a structural perspective, migration is a “natural outgrowth of disruptions and dislocations”, all part of capitalist development in sending/receiving countries; flows are generated as land, raw materials, and labor within “peripheral regions” come under the influence and control of markets. Thus migration is a by-product of global capitalism.

Contemporary patterns of international migration tend to be from the periphery (poor nations) to the core (rich nations) because factors associated with industrial development in the First World generated structural economic problems, and thus push factors, in the

Third World.

Under the classical sociological theory of inequality, one can argue from a functionalist perspective that migration is a result of rational choice and it occurs in order to maintain a well functioning global economy. From a structural perspective, migration occurs due to global inequalities and the process is as a result of global capitalism where

12 poor nations migrate towards rich nations due to push factors in the periphery brought about by core countries. Under this process the power dynamic remained between those who owned the means of production and those who do not.

ASSIMILATION THEORY

Both migration theory and stratification in migration as previously discussed are relevant in understanding how assimilation takes place once the immigrants begin to settle within the host country. Both theories are useful in understanding the behaviors of immigrants prior to entering the host society and their adaptations upon arrival. The following section briefly outlines the classical and segmented approaches to assimilation; a more in depth analysis of these perspectives is presented in chapter three.

Classical assimilation theory portrayed assimilation as an integral part of the movement of immigrant groups into the American middle class (Warner and Srole, 1945;

Gordon, 1964). Some scholars have argued that classical assimilation theory is no longer applicable for current Asian, African, Caribbean and Latin American immigrants, since their experiences are quite different from earlier waves of European immigrants. Gans

(1992) contends that not all scholars agree that these circumstances are sufficiently unique or significant to render classical assimilation theory inapplicable. The counter argument to this is that the differences between the two waves of immigrants are tremendously overstated (Alba and Nee, 1997, 2003; Perlmann and Waldinger, 1997). As reviewed by Alba and Nee (1997, 2003), thus assimilation theory had undergone many revisions and refinements before it began to face fundamental challenges in the 1990s.

While classical assimilation theory primarily seeks to explain the process of assimilation, why one immigrant’s family may be more or less assimilated than another, segmented assimilation theory explicitly considers both the process and the outcomes of assimilation. Segmented assimilation is primarily based on three major differentiated dimensions. First, some scholars have argued that assimilation outcomes may differ by immigrants’ characteristics such as ethnicity, socioeconomic status, social capital, family cohesion, and perhaps gender (Farley and Alba, 2002; Hirschman, 2001; Nagasawa,

Qian, and Wong, 2001; Portes and Rumbaut, 2001; Rong and Brown, 2001; St-Hilaire,

2003; Waldinger and Feliciano, 2004). Second, assimilation outcomes may differ by the characteristics of natives to whom immigrants assimilate (Gans, 1992; Rumbaut, 1994,

1997; Bankston and Zhou, 1997). Correspondingly, immigrants can assimilate into the middle-class that is representative of White natives, or they can assimilate into the inner city, representative of underclass minorities struggling with poverty, crime, and joblessness. Third, assimilation outcomes may depend on whether assimilation is

“wholesale” or “selective”, with the implication that limited assimilation is beneficial to the immigrant (Bankston and Zhou, 1995; Portes and Rumbaut, 2001; Portes and

DeWind, 2004; Portes and Schauffler, 1996). Selective acculturation is perhaps the most common interpretation of segmented assimilation theory.

This study will build from large social science literatures with respect to theoretical framing and study design. From a theoretical standpoint, the proposed research speaks to long-standing research on assimilation. While scholars and researchers have debated the

14 relevance of the classical model of assimilation, the empirical work done and the populations examined further prove this theory inadequate. This study provides a broad and systematic empirical assessment of the relationship between race, class, gender and second-generation Caribbean immigrants patterns of assimilation.

CHILDREN OF IMMIGRANTS

What happens to minority immigrants is of vital importance as the U.S. population becomes increasingly diverse. The proposed research also examines the parent-child relationship with regard to assimilation and further explains the degree to which parents’ socioeconomic status may have an influence on the adaptations of second-generation immigrants. The following section provides an overview of the children of first generation minority immigrants and the ways in which their ethnicity affects their behaviors and academic achievements, as well as their aspirations for themselves.

For visible minorities who are racially different from Whites, race and ethnicity have true ramifications on their educational outcomes, which eventually translate into ethnic differences in socioeconomic status. Many researchers who study racial and ethnic parenting styles (Schneider and Lee, 1990; Dornbusch, 1989; Sue and Okazaki, 1990;

Valenzuela and Dornbusch, 1994) have argued that the cultural difference in the ways in which families function may in fact account for the racial and ethnic disparities in educational achievement. There is racial and ethnic variation in the style of parent-child relationships as well as differences in how children respond to various types of parenting techniques. Black and White families are most similar in the relative influence of parental

15 communication on educational outcomes. There is clear evidence of distinct family relations among Hispanic and Asian youth. Hispanic youth value familistic ties, however such attitudes are only marginally related to school outcomes, except for those with well- educated parents. In Asian households, family interactions are clearly different from those of Whites, but it is unclear how these diverse family dynamics influence educational outcomes. This research looks at the parent-child relationship and further investigates the extent to which the socioeconomic status of the parent subsequently impacts the integration of the child into the dominant culture.

Earlier studies with adolescents reveal that those who arrived in the U.S. before school age or in the first year or two of formal school (sometimes referred to as 1.5 generation) are likely to perform better on academic tests than those who arrived after school was well under way (Glick and White, 2003). According to Fernandez-Kelly and

Schauffler (1994), children of immigrants today are inserted into the racial/ethnic hierarchy of the United States and they are likely to be influenced by the context of reception faced by their family based on this identification. Zhou (1997) argues that the children of immigrants from those groups that have been historically disadvantaged in the

United States are likely to face different opportunities and barriers than other children from immigrant families. “Regardless of their class origin or knowledge of English, non-

White immigrants face greater obstacles in gaining access to the White middle class mainstream” (Portes and Rumbaut, 2001). The proposed research examines the impact of family’s social class, more specifically the respondents’ own education and the

16 socioeconomic status of their parents on assimilation. I further investigate the degree to which immigrants are gaining upward mobility, based on their civic involvement, educational attainment, language proficiency and decision to out-marry/partner.

Discussions of neighborhood disadvantages are often rooted in social disorganization theory (Wilson, 1987) or epidemic theory (Crane, 1991) whereas explanations for the advantages of living in higher status neighborhoods usually follow social capital theory (Coleman, 1988; Sampson, Morenoff and Earls, 1999) and the notion of concentrated wealth (Massey and Denton, 1999). In school, peer pressure is much stronger among members of the anti-social groups (the ‘druggies’ and the ‘toughs’) than among members of the pro-social groups (the ‘populars’ and the ‘jocks’) (Clasen and Brown, 1985). As children reach adolescence, they increasingly pull away from their parents and develop close peer relations; as a result adolescents are more susceptible to peer influence. As children of immigrants live in neighborhoods with other immigrant children who possess limited levels of English proficiency skills, they too do not develop the proficiency needed to help them to succeed academically.

Jencks and Mayer (1990) argue that there are negative consequences associated with poor immigrant children living in affluent neighborhoods. Poor children who attend school alongside affluent children from the neighborhood may feel inferior and develop low self-esteem; they are also likely to form deviant sub cultures that downgrade school performance. Researchers found that poor children studying with other poor children are not likely to develop such negative feelings and insecurities (Coleman, 1988; Sampson,

Morenoff and Earls, 1999). By contrast, there is in fact another possibility that poor children going to school with high socio economic status neighbors, particularly co- ethnic neighbors, may not necessarily feel deprived but rather they may feel competitive and want to perform better. Jencks and Mayer (1990) further state that this may be especially relevant for immigrant adults and their children since most of them desire and believe in the American Dream of meritocracy.

Researchers have also suggested that the cultural values of many immigrant groups may in fact be responsible for immigrant children attaining higher educational accomplishments than their native counterparts. According to Rumbaut (1997), many immigrant children have significantly better education outcomes than would be predicted by their family socioeconomic status. Some of the variation in outcomes among immigrant children is attributable to systematic differences by national origin but there is still considerable heterogeneity within ethnic groups. It is therefore important to explain why some of these children do so well while others fall behind. To further examine the effects of nationality on assimilation, the proposed research examines the differences in both educational attainment and language proficiency of immigrants from five distinct

Caribbean countries (Cuba, Dominican Republic, Haiti, Jamaica and the English West

Indies) I will conduct a comparative analysis to determine if there are any differences in the outcomes and subsequent assimilation of these immigrants.

The vast majority of children of immigrants, both one and a half and second- generation are coming of age in a limited number of traditional gateway states, most

18 notable California, New York, Texas and Florida. Despite literature suggesting that the children of immigrants decline into the urban underclass (Gans, 1992; Massey, 1995;

Portes and Zhou, 1995), more recent investigations have argued that they will in fact surpass the educational and occupational attainments of their parents (Hirshman, 2001;

Zhou, 2001; Farley and Alba, 2002). There is continuing reference to the segmented assimilation hypothesis with the claim that immigrant adaptation is contingent upon race and context of settlement (Rosenbaum and Friedman, 2001).

This dissertation examines both parent-child relationships, primarily of immigrants in south Florida, and the ways in which the child’s own educational attainment impacts integration into the host society. I investigate the degree to which family’s social class remains significant even among a homogenous segment of immigrants. I further explore the degree to which social class affects assimilation outcomes politically, linguistically, educationally and in relation to out- marriage/partnership. I also incorporate supplementary measures of social class such as family structure, homeownership, family’s economic situation, and mother’s regular employment to determine if these factors together with parents’ socioeconomic status are positively related to assimilation.

The above review of the literature points to a number of factors such as race/ethnicity, class, neighborhood, family structure and age of entry which have various influences on the ways in immigrant children assimilate into the dominant culture. The proposed research focuses on Caribbean immigrants from the areas of Miami and Fort

Lauderdale. My contribution to the literature is the examination of the extent to which race, class and gender are all significant factors in predicting the assimilation process of

Cuban, Afro-Caribbean and mixed Caribbean immigrants into the US.

MODES OF ASSIMILATION

In the immigration literature there are three main modes of immigrants’ integration into the new society: assimilation; differential exclusion or chronic marginalization; and transnationalism.

As Gordon (1964) defined it, cultural assimilation entails a process of acculturation on the part of the immigrants, of becoming “like” in cultural patterns, such as language, behavior, customs, dress, beliefs, and values. Structural assimilation resulted only when the immigrants had been “taken up and incorporated” and entailed the full integration of the immigrants and their descendants into the major institutions of the society

(educational, occupational, political) and into the social cliques, clubs, and institutions of the core society that lead to intimate primary relationships, including intermarriage. For purposes of my research I focus on the two types of assimilation, the classical linear model as well as the contemporary segmented model.

“Differential exclusion” maintains that not all immigrants have been considered as

“assimilable” by the assimilation promoters (Castles, 2002) thus as a way of trying to control the growing ethnic diversity, the receiving countries have differentiated between immigrants according to their race (until recently), and of social and cultural background

(Castles, 2002) and (Samatar, 2004). “Chronic Marginality” model describes the

20 geographical enclaves that attract new comers to be close to people of the same background and identity (Castles, 2000). Ethnic-clustering areas can provide new comers with comfort and support. Not only can individuals benefit by mental and psychological support to survive the intense stresses associated with the early stages of immigration to a new society, but also in some cases, the ethnic group might be the source of better socioeconomic opportunities for ethnic entrepreneurs (Alba and Nee, 2003).

Nevertheless, many of these enclaves could become isolated ghettos characterized by impoverishment, public insecurity, and systematic social and political exclusion (Castles,

2000).

Transnationalism is the process by which immigrants “forge and sustain multi- stranded social relations that link together their societies of origin and settlement” (Basch et al., 1997). Foner (1985) argues that many transnational patterns such as keeping communication alive with loved ones back home, sending remittances, supporting particular political causes and return migration have a long history among immigrant groups in the U.S. Transnationalism may have effects on the extent to which immigrants can assimilate, both culturally and structurally, in the United States. For example, participation in transnational practices and the exercise of a diasporic citizenship has consequences on the degree to which immigrants can engage in domestic politics, particularly ethnic politics in American life. A growing number of immigrants live dual lives, speak two languages, have homes in two countries, lead lives that continuously cross two cultural domains and forming distinctive types of communities (Portes et al.,

1999). Even though, the proposed research does not examine the direct effects of transnationalism, the study incorporates a measure of civic involvement as an indicator of assimilation; I further examine the extent to which ties to country of origin limit the political participation of second-generation Caribbean immigrants into the dominant culture.

Heterolocalism is a socio-spatial, non-traditional model in the immigration literature developed by Wilbur Zelinsky (1998). Heterolocalism is linked to transnationalism, another recent development, which occurs as millions of migrants begin shaping a novel form of action-space, whereby ethnic groups can inhabit two or more countries simultaneously. Such behavior has begun to redefine the meaning of ethnic identity (Zelinsky, 2001).

Transmigration and transnational social spaces and practices have emerged as important topics in sociological and cultural studies since the 1900s (See Portes 1996;

Smith and Guarnizo 1998). The term transnational suggests those immigrants' behavior and activities span both tangible and nonmaterial borders. Trans-migrants take action, make decisions, and develop subjectivities and identities embedded in networks of relationships that connect them simultaneously to two or more nations cultures (Pessar,

2007). With the help of global media, ethnic tourism, and religious or secular festivals, they also include multiple exchanges of monetary and non-monetary resources, material and symbolic objects, commodities and cultural values (Guarnizo and Smith, 1998).

Though the assimilation literature points to various processes whereby an immigrant can acculturate into the dominant society, I focus mainly on the segmented and classical linear models of assimilation to describe the process among a subcategory of second-generation Caribbean immigrants.

RESEARCH QUESTIONS

Previous literature and empirical assessment of assimilation suggest that race, class and gender are all salient factors on an immigrants’ integration and adaptation. The study examines the extent to which race/ethnicity, social class and gender remain significant determinants of assimilation for a select subset of second-generation

Caribbean immigrants.

The sample of Caribbean immigrants chosen for the study is representative of an upwardly mobile population that is characteristic of high levels of education and generally high socioeconomic status. The proposed research investigates the extent to which class remains significant even among a highly educated group of immigrants. I also examine the parent-child relationship on assimilation. Parents’ socioeconomic status is used as a measure of social class; I investigate the degree to which family’s social class affects the successive outcomes of their children based on four different measures of assimilation: language proficiency; political participation; educational attainment and subsequent marriage and/or partnership. Social class is also measured in the study by the respondents’ own educational attainment, parents’ economic situation, mother’s regular employment, family structure and homeownership status.

Race and ethnicity is another central component to the study. I will conduct a comparative analysis to illustrate the differences in assimilation among Caribbean immigrants based on their nationality and also racial and ethnic self-identification. The use of both national origin and racial and ethnic self-identification allows me to tease out the effects of both nationality and race on the patterns of assimilation. I will examine the patterns of assimilation for those self-identifying as Caribbean Black, White, Hispanic and ‘Other’ and compare the results to the overall integration patterns of Cubans,

Dominicans, Haitians, Jamaicans and West Indians.

The assimilation literature has inadequately addressed the ways in which gender has impacted the immigrants adaptation; much of the focus has been on women’s entry into the labor force, their treatment relative to men and their overall patterns of labor force participation. The proposed research looks at the ways in which gender affects the integration of women into the dominant culture, more specifically the ways in which males and females assimilate differently based of their civic involvement, language proficiency, educational attainment and decisions to out-marry or partner.

Taking a conservative approach to the research, I will investigate the extent to which race/ethnicity, class and gender are all salient factors even amongst a small sample of second-generation Caribbean immigrants. More specifically, the research aims at exploring if the trends in assimilation, which are significant for broader groups of immigrants, still apply to this subset of second-generation Caribbean immigrants. The following are the three main research questions analyzed in the study.

(1) How does race and ethnicity impact second-generation Caribbean immigrants’ language proficiency, political participation, out-marriage/partnership, educational attainment and subsequent assimilation patterns?

(2) How does social class affect second-generation Caribbean immigrants’ language proficiency, political participation, out-marriage/partnership, educational attainment and subsequent assimilation patterns?

(3) How does gender impact second-generation Caribbean immigrants’ language proficiency, political participation, out-marriage/partnership, educational attainment and subsequent assimilation patterns?

DATA AND METHODOLOGY

The Children of Immigrants Longitudinal Study is used for my proposed research.

The data were collected to examine the adaptation process of second-generation children of immigrants with at least one foreign born parent as well as children who were born abroad but were brought to the U.S. at a very early age. The children attend the eighth and ninth grades in public and private schools in the metropolitan areas of Miami and

Fort Lauderdale in Florida. In the first survey, baseline information on immigrant families was collected such as children’s own demographic characteristics, language use, self-identities and academic attainment.

The second survey was conducted three years later, which corresponded with the time that the students were about to graduate from high school. Its purpose was to examine the evolution of key adaptation outcomes, including language knowledge and

25 preferences, ethnic identity, self-esteem and academic attainment over the adolescent years.

A third survey was conducted when respondents had reached early adulthood, at average age 24. A parental survey was also conducted. This survey targeted half of the total second-generation children immigrant parents, selecting them on a random basis.

Unlike the student surveys which were conducted mostly via self-administered questionnaires in school, the parental interviews were conducted face-to-face and mostly at home. Since many immigrant parents did not understand English, the questionnaire was translated and administered in six different foreign languages. The purpose of the interview was to establish directly characteristics of immigrant parents and families and their outlooks for the future, including aspirations and plans for the children. In total,

2,442 parents or 46 percent of the original student sample were interviewed. Their national origins closely resemble, in proportional terms, those of the student sample.

ORGANIZATION OF DISSERTATION

This chapter has presented an introduction to the study, and motivation for the research. I presented the theoretical framework for migration and assimilation. I explained the conceptual theories, which are instrumental in guiding my study by giving some insight into the contemporary and classical theories of assimilation. To further demonstrate the parent-child relationship analyzed in the study, I provided an overview of the children of immigrants. I also outlined the various modes of integration used in the

26 literature. The chapter ended with an outline of the research questions as well as a brief description of the dataset and sample for the study.

In the second chapter I provide an overview and historical analysis of Caribbean immigrants. I present the colonial histories of the islands used in the study and the factors that fueled migration from the Caribbean to the United States both pre and post World

War II. The chapter ends with an analysis of racial labeling, self-identification and a justification of the five racial and ethnic categories utilized in the research.

The third chapter provides a more detailed overview of theory and literature used in the dissertation. I particularly draw on the implications of linear and segmented assimilation theories for the case of second-generation immigrants integration into U.S. society. I conduct a conceptual analysis of race, class and gender relative to assimilation and introduce the four measures of assimilation proposed in the research. I also examine the limitations of both the classical and contemporary approaches to assimilation and explain my conceptual and empirical contribution to the literature.

Chapter four explains the methodological framework for the research. I present a detailed description of the dataset, the sub-sample used in the study and the limitations of the dataset. This chapter also provides a detailed description of the dependent and independent variables and a justification for the quantitative technique employed.

The fifth chapter presents the results of the comparative analysis conducted using both national origin and racial and ethnic self-identification. I demonstrate the

27 assimilation outcomes for the sample population using both racial and ethnic groups and the five distinct nationalities. I also discuss the implications of the research findings.

The final chapter presents the conclusions of the study and I address the ways in which my research has contributed to the larger literature on immigration.

Recommendations are also made for future research in order to fully understand the phenomenon of assimilation and how second-generation Caribbean immigrants have integrated and adapted into the dominant culture.

CHAPTER TWO

HISTORY OF CARIBBEAN IMMIGRANTS

Introduction

In order to understand the migration and subsequent assimilation patterns of

Caribbean immigrants it is crucial to grasp the history of each island, their legacies of colonialism and the manner in which different waves of immigrants arrived to the U.S.

The following chapter gives an overview of Caribbean immigrants and the ways in which their assimilation patterns are different from other waves of immigrants who settled in the

U.S. The chapter also outlines the arrival of English, Spanish and French speaking

Caribbean immigrants into the United States: I discuss the push factors of the sending countries that fueled the movement into the United States. I also review the periods in which the U.S. government restricted the entry of Caribbean immigrants, the re-entry of these immigrants and their adaptations into the dominant culture. The chapter ends with an overview of racial and ethnic self-identification and a justification of the five racial and ethnic groups proposed in the study. The proposed research is an investigation of the degree to which Cuban, Afro-Caribbean and mixed Caribbean immigrants assimilate differently or similarly into the dominant culture based on influences such as race, class

29 and gender. I further explore the extent to which these factors remain salient even among a relatively homogenous sub-sample of the Caribbean population.

Figure 2.1: The map shows the Caribbean islands used in the research, Cuba, Dominican Republic, Haiti, Jamaica and the English speaking West Indies. Source: http://www.bareboatsbvi.com/map_caribbean.html.

The Caribbean region represents the archipelago of islands spanning from Florida all the way to South America. These islands separate the Gulf of Mexico and the

Caribbean Sea to the west and are south of the Atlantic Ocean. The Caribbean is made up of the West Indies, which comprises the Greater Antilles to the north and Lesser Antilles to the south and east. The Bahamas and the Turks and Caicos are not part of the Antilles and do not border the Caribbean Sea. This study focuses on immigrants from Cuba,

Dominican Republic, Haiti, Jamaica and the English West Indies. Those English speaking Caribbean islands that were represented by a very small number of West

Indians were grouped in the study and labeled as the West Indies. West Indians are also referred to in the study as the British West Indies, as these islands were all colonies of

Britain.

Figure 2.2: The map shows the location of the Gulf of Mexico, Caribbean Sea, Atlantic Ocean and the Greater and Lesser Antilles. Source: http://mappery.com/map-of/Central-American-and- Caribbean-Islands-Map.

RACE AND ETHNICITY AS THEY RELATE TO CARIBBEAN IMMIGRANTS

I chose to focus on Caribbean immigrants primarily because the topic of assimilation and adaptation for this region has been understudied and even when research has focused on this region there seems to be a bias towards Latin American countries as

31 opposed to the English and French speaking islands. This region is also very unique in a variety of ways ranging from the economic and political histories of each island, the racial and ethnic composition of the people and the languages spoken by Caribbean immigrants. My goal is to examine the impact of race, class and gender on the assimilation pattern of Caribbean immigrants who have migrated from Jamaica, Haiti, the

West Indies, Cuba and Dominican Republic.

Caribbean immigrants present a formidable and necessary challenge to the existing models of ethnic adaptation in race relations’ research: the most dominant of which is the assimilation perspective (Hirschman, 1983). With the recent increase in the foreign-born Black population, sociologists can no longer understand ethnicity exclusively in the framework of White immigrants; nor can we continue to think of race as the most essential characteristic of Blacks in the United States. "Black" does not necessarily equal "African-American" anymore. Many authors have tried to explain the socioeconomic success of Caribbean immigrants in the United States. Caribbean Blacks are noted as having a greater motivation for achievement and a stronger work ethic than native-born Blacks. Sowell (1978) argues that because slaves in the Caribbean did not experience strong economic competition from a large White lower class as slaves did in the American south, they therefore had more opportunity for economic initiative and were socialized into a ‘spirit of capitalism’ early on. Thus Sowell (1978) attributes differences in work ethic to differences in the American and Caribbean systems of slavery. This assumption is somewhat difficult to prove or refute and therefore deemed a

32 less practical explanation, since it implies that a lack of achievement orientation, rather than racism or discrimination on the basis of color, is the cause of racial inequality in

American society.

A more practical and obvious explanation about the motivation and work ethic of

Caribbean immigrants focuses on differences between American and Caribbean society in modern times. According to Wilson (1987), although many American Blacks have entered the middle class in the last decades, there are some structural changes which have taken place in American society, such as the deterioration of the manufacturing industry and the flight of the middle class to the suburbs which have deprived many young urban

Blacks of successful models for achievement in their community. Alternatively, more positive examples of Black achievement and less rigid social classification of racial groups in the Caribbean region have led to a stronger orientation toward socioeconomic achievement among Caribbean Blacks. Wilson (1987) contends that this argument is more applicable to the British Caribbean where Blacks form the majority than it is to the

Hispanic Caribbean where they generally constitute a minority. I expect that Afro-

Caribbeans will have higher levels of educational attainment, compared to Cuban and mixed Caribbeans due to their exposure to more positive examples of Black achievement and less rigid social classification of racial groups in the Caribbean region. Not only is there a greater representation of Black success in the Caribbean but there is a substantial representation of middle class Blacks in many Caribbean islands. I use the term “mixed Caribbeans” in the study to represent those immigrants from Haiti,

Dominican Republic, Jamaica and the West Indies who self identify as either multi-racial,

Hispanic, or ‘other’. This racial and ethnic classification is further discussed later in this chapter and in chapter four.

Caribbean immigrants may also have a stronger motivation for achievement due to the process of migration itself. Those who voluntarily come to the U.S. do so primarily for economic reasons. Chiswick (1979) argues that economically motivated migration is considered to be more selective with respect to economic skills and abilities than politically driven migration. Palmer (1974) argues that this hypothesis is particularly plausible for immigrants for the English speaking Caribbean, almost all of whom are motivated by the desire to escape poor domestic economies. I expect that Black

Caribbeans will be more likely to be upwardly mobile and attain higher levels of assimilation compared to mixed Caribbeans because their decision to migrate is usually economically rather than politically driven. It has long been argued that the main reason why Caribbean Blacks are more successful in the American economy than native-born

Blacks is that they suffer less from racial discrimination by Whites (Reid, 1939). Though these arguments have not been proven systematically, Foner (1985) argues that anecdotally, evidence confirms that Caribbean Blacks are perceived more favorable than

American Blacks by Whites; some assume that they are favored because of their British accents, others believe that they present themselves to prospective or current employers as culturally distinct from, and possibly superior to Native American Blacks.

Kalmijin (1996) contends that since the children and grandchildren of Caribbean immigrants have not attended schools that are heavily based on British or other European traditions, they are going to be less likely to be exposed to the Caribbean culture and their language and accents will not resemble that of their parents. As a result of this, if White employers are less able to distinguish the second-generation Blacks from African

Americans, American born Caribbeans will have less of an advantage over African

Americans than Caribbean immigrants. Accordingly, this assimilation perspective posits that the children of Caribbean immigrants are becoming more similar to African

Americans than their parents. As a result, Caribbean Blacks represent a scenario where cultural assimilation may hamper as oppose to improve their socioeconomic achievement.

Waters (1999) study of West Indian immigrants provides evidence on the effects of socioeconomic status on both parental influence and aspirations and the negotiation of this influence by their children. In her study, Waters reported that middle class immigrant children were more likely to see strong differences between themselves and native-born

Black youth, primarily because of their perception of native youth as having values antithetical to those of their own immigrant parents. These immigrant parents usually stressed the importance of education, discipline, and a strong work ethic to their children and had other negative impressions of poor native-born Blacks. Correspondingly, poorer

West Indian youth were more likely to embrace the American identity and reject their parents’ negative perspectives of the Black American culture. As a result, such

35 contrasting perspectives have impacted immigrant schooling outcomes and how immigrant children perform relative to their native born counterparts. I expect that race will still be a significant factor for second-generation Caribbean immigrants, even though they may be less distinguishable as Caribbean than the first-generation, because culture, norms and values are still heavily inculcated into the second-generation; hence they will continue to achieve high levels of assimilation.

Marilyn Halter (1993) argues in ‘Between Race and Ethnicity,’ one of the biggest failures of all American assimilation theory is that race, if discussed at all, is treated either as a derivative of ethnicity or the terms are used interchangeably, as if they were the same (Gordon, 1964; Handlin, 1973; Higham, 1975; Kallen, 1924; Smith, 1991).

Halter (1993) further argues that race and ethnicity are two distinctive entities and the two should not be conflated in the assimilation literature. Classic assimilationist theory when applied to Blacks, as it has been of rural southern Blacks following the Great

Migration, depicts them as if they were no different from European peasants arriving in northern cities (Banfield, 1974; Glazer, 1971; Handlin, 1962; Lieberson, 1980; Kristol,

1966; Reid, 1939; Wirth, 1928).

Though it is usually difficult not to conflate race and ethnicity, my contribution to the Caribbean assimilation literature attempts to treat race and ethnicity as two separate entities. In doing so, I employ a sensitivity analysis whereby I examine the individual effects of race and ethnicity and national origin on the assimilation patterns of second- generation Caribbean immigrants. I expect than both race and nationality will impact

36 language proficiency, educational attainment, civic involvement and out-marriage among second-generation Caribbean immigrants.

THE MIGRATION OF CARIBBEAN IMMIGRANTS TO THE U.S.

This section gives a chronology of the ways in which Caribbean immigrants made their way into the United States from Spanish, English and French speaking countries. I provide an analysis of the colonial period in various islands prior to 1900s and I also discuss the push factors that fueled the movement of these immigrants into the United

States. I provide a background of the people who came to the U.S. from the British West

Indies, Cuba, Haiti and Dominican Republic. I also analyze the ways in which the United

States’ government shut the door on immigration and subsequently restricted the flow of

Caribbean immigrants into the U.S. I end the chronology by describing the new wave of

Caribbean immigrants and their reception and adaptation into the dominant culture. I subsequently provide insight on racial self-identification in the Caribbean and the ways in which self-identification can further impact assimilation into the host country.

The Arrival Of Afro-Caribbeans To The United States

At the beginning of the 1990s, the entire Caribbean region accounted for 32.5 million people; the Spanish speaking countries of Cuba, Dominican Republic and Puerto

Rico represented 61%. Twenty percent lived in predominantly French speaking islands,

17% lived on the islands where English was the official language and 2% of that figure represented those from Dutch speaking islands (Knight, 1990). These Caribbean islands varied in language, but also colonial histories, size, geography, demographics and

37 cultural heritage.

Many researchers describe the societies created on the Caribbean islands as

‘artificial or ‘manufactured’ societies; this is due primarily to the mixing of people from various races and ethnicities on the islands that many of them never really felt a sense of belonging or that they were actually from those islands (Lowenthal, 1972; Mintz, 1974 and Knight, 1990). The success and profitability of these societies was due to enslaved labor brought to the Caribbean over a period of 400 years; during this time an estimated

4.6 millions slaves were involuntarily brought to the islands (Curtin, 1969). Various

African cultures on the islands served to create a distinct Creole culture of its own among the islands and sugar plantations served to unite the islands linking them all together in unprecedented ways.

Afro-Caribbean people were brought to the United States from as early as the 17th century; the first were natives of Barbados who were taken by their British owners to

South Carolina. Most of the first Africans who came to the U.S. were men, women and children from the Caribbean (Reid, 1939). Following this involuntary period, the turn of the twentieth century marked the influx of a large number of British West Indians.

Between 1930 and 1965, the third wave of immigrants came, followed by the fourth wave, which is still taking place today (Halter, 1993). Slaves brought from the Caribbean to the North, New York specifically, continued to out-number Africans brought from the continent. Between 1715 and 1730, the ratio of Caribbean to African slaves was three to one. The largest number of captives came from Jamaica, followed by Africa, Barbados

38 and Antigua (Halter, 1993). Even in Charleston, South Carolina today, much of the

Caribbean influence and accents of the people were heavily influenced by the historic presence of Caribbean slaves. After the Civil War, the foreign Black Caribbean population increased five-fold with, the number increasing from four thousand to twenty thousand (Palmer, 1995; 1990).

Many distinguished Caribbean migrants paved the way in nineteenth-century Black

America in a number of ways. In the book ‘Immigration’ written by Halter (1993), West

Indians were said to play a prominent role politically, economically and intellectually in the leadership and development of many U.S. communities. These immigrants were craftsmen, scholars, teachers, preachers, and doctors. Many profound men and women who made significant contributions to the United States were descendants of the

Caribbean immigrants (Foner, 2001). Jan Earnst Matzeliger, inventor of the revolutionary shoe-making machine, Edward Wilmot Blyden, contributor to Black Nationalism, Robert

Brown Elliott, a U.S Congressman and Attorney General of South Carolina, W. E. B. Du

Bois, a legendary poet, songwriter, and activist; are all among some of the most distinguished sons and daughters of early Caribbean immigrants (Model, 1991; 1995).

The Arrival Of Cubans, Puerto Ricans And Dominicans

Cubans formed immigrant colonies in Key West, Tampa, New York City and New

Orleans by the end of the eighteenth century; a period which marked great political and economic turmoil for the immigrants’ countries of origin (Nijman, 1997). In the mid-

1880s, more than 20 percent of immigrants were Black or mulatto, many of whom

39 worked in the cigar-making industry, especially in Key West and Ybor City, Florida.

Puerto Rican migration began during the last third of the nineteenth century, at a time when hundreds of political exiles were living in New York City. Many of these exiles supported the island’s independence from Spain and in 1895 the Puerto Rican section of the Cuban Revolutionary Party was created, this party advocated the ‘Antillean

Confederation’ of the Cuban, Puerto Rican and Dominican Republics (Suchlicki, 1968).

The majority of the expatriates were White, affluent, and well-educated professionals who were physicians, teachers, and journalists. At the end of the Spanish-Cuban-

American War of 1898, when the United States occupied Puerto Rico, many returned home. Many of the immigrants who chose to stay in the U.S. experienced assimilation and adaption into the dominant culture very differently due to their already privileged backgrounds, high levels of human capital and high socio economic status.

Spanish-speaking immigrants came to the U.S. at a much later date than did the

Afro-Caribbeans, however their presence grew very quickly. In 2000, millions of immigrants who entered the U.S. were coming from Puerto Rico, Cuba and the

Dominican Republic. Not only has this large-scale migration transformed the sending countries family structure, religious practices, political ideology and business enterprises; this migration has also increased diversity and cultural complexity of countless regions of the United States.

Initially the Cubans who migrated to Miami were racially diverse with a significant number of them being Black and Mulatto; however the presence of Jim Crow laws and

40 severe discrimination in the South produced further migration to the North, especially to areas like New York City (Greenbaum and Poyo, 1985). Another contribution to the migration of Cubans to the North was the decline of the cigar industry; Cuban migration for workers in the U.S. sugar industry occurred early 1900s; most went to Tampa,

Florida, however due to discrimination Black Cubans migrated to the North. Black

Cubans were not the only ones subject to discrimination in the South, many White

Cubans faced similar hardships because they were not accepted into the dominant culture as purely White since their blood was considered to be racially and culturally mixed

(Grillo, 2000). As a result of Black Cuban migration, many of the Cubans who remained in Florida and who are represented in this research self-identify as Cuban Whites even though the literature suggests that they were not entirely accepted as White in the dominant culture.

Even though Cuban migration dates back to the nineteenth century (Greenbaum,

2002; Poyo, 1989), the majority of Cubans who live in the U.S. trace their origins to more recent times following the Cuban Revolution in 1959 (Pedraza, 1996). The first wave of immigrants who were granted refugee status after Fidel Castro’s ascent to power between

1959 and 1962 were known as the ‘Golden Exiles’- this first wave of immigrants was representative of Cuban elites (Garcia, 1996’Pedrazza, 1996). The second wave of Cuban immigrants took place over an eight- year period from 1965 to 1973. More than 260,000

Cubans arrived on daily flights from Cuba to Miami. This group included remaining elites as well as members of the Cuban middle class (Garcia 1996; Grenier and Perez

2003; Pedraza 1996; Portes and Stepick 1994). The third wave of immigrants represented a more diverse group of Cubans and is discussed in more detail below.

Continued Migration Of Cubans And Dominicans To The U.S.

Between April and September 1980, a large Cuban migration from Mariel harbor to

Key West took place. This was caused by the takeover of the Peruvian Embassy in

Havana by more than 10,800 Cubans who wanted to migrate (Larzelere, 1988). This resulted from the visits of more than 100,000 exiles to Cuba in 1979, which renewed social contacts with relatives and familiarized them with economic opportunities in the

United States. When the Cuban government opened the port of Mariel many were picked up by relatives living abroad; this began what became known as the "Freedom Flotilla" in the United States. During this exodus more than 125,000 Cubans arrived in Key West

(David, 1990). The socioeconomic profile of the Mariel exodus differed significantly from that of previous refugee waves, especially during the early 1960s. When the exiles arrived in Mariel aboard private boats and ships, Cuban officials forced them to take unrelated persons, some of who had been inmates at prisons or mental hospitals. The boatlift ended abruptly when Castro closed the harbor for further emigration (Portes and

Lief, 1989). The 1980 boatlift deepened the rifts between "old" and "new" Cubans in

Miami, where most of the latter settled. This ‘old’ and ‘new’ Cuban wave of Cuban migration is what accounts for the disparity in income levels among Cuban immigrants today (Larzelere, 1988).

Figure 2.3: Mariel Boatlift. Source: http://www.floridamemory.com/items/show/98692

Large-scale migration from the Dominican Republic began shortly after Trujillo's assassination in 1961, especially following the coup d'état against constitutional President

Juan Bosch (1963) and the civil war and U.S. military occupation of Santo Domingo

(Crasweller, 1966). As a result of these events, there were many socioeconomic forces which fueled the movement abroad of thousands of Dominicans over the past four decades (Duany, 2005). Economic crisis during the 1980s led to the increase of migrants from a variety of social classes. According to Duany (2005), vast amounts of unemployment and underemployment, high costs of living and the near collapse of the electricity, housing, water, health and education services, caused many to move abroad in search of better opportunities and living conditions. Urban working class, rather than the

43 middle class and peasantry were among the majority who migrated to the US.

PUSH FACTORS FROM THE CARIBBEAN

There were many push factors that contributed to the immigration of the Caribbean community to the United States. At the turn of the twentieth century, the British West

Indies was faced by grave economic hardship; during this same time period, the U.S. economy was expanding and high wages and the growth of employment opportunities played a major role in immigration from the West Indies (Palmer, 1990).

The collapse of the sugar economy experienced by the British West Indies drastically reduced the number of workers employed in the industry. The British West

Indies was unable to compete against cane sugar from Cuba or Brazil or even the sugar beets that were produced in Europe. Jamaican sugar dropped almost 80 percent between

1840 and 1900. The collapse of the sugar industry for this region marked the migration for those in search of jobs and better opportunity (Fogel, 2003).

Figure 2.4: Sugarcane workers in Montego Bay, Jamaica, 1900. Source: http://www.abdn.ac.uk/slavery/resource12c.htm

Fogel (2003) also argues that in an effort to sustain the sugar industry in Barbados, the ruling class instituted an exploitive system; workers were paid low wages and forced to work under very poor working conditions. This pattern of exploitation soon trickled over into other islands like Saint Kitts, Nevis, Montserrat, and Antigua. Both infant mortality and death rates skyrocketed for these islands; there was rampant malnutrition and starvation on these islands, which further resulted in the migration of population for those who were able to leave. Another major push factor for this region was its susceptibility to natural disasters, which were frequent and intense. Between 1880 and

1920 hurricanes, floods and droughts hit this region heavily. Additionally, the presence of colonial rule, racism and the stagnation of the Black middle class contributed greatly to

45 their movement to the United States (Holder, 1980). Black middle class workers in the

Caribbean realized that they were not fairly being remunerated for their skills; there was a lack of advancement for higher positions, especially for teachers who were under paid and received no pensions at the end of their careers. Additionally, many Whites and lighter skinned Blacks were favored for positions over the darker skinned Blacks

(Palmer, 1990). The structure and oppressiveness of colonial rule on the islands negatively affected not only on the workers and peasants but also on the aspiring Black middle class.

Those who immigrated to the U.S. were disproportionately literate and skilled, with a significant number being professionals or white-collar workers (Winston, 2002). The number of Afro-Caribbeans increased drastically during the first three decades of the twentieth century, and then began to fall off during the Depression. Despite the restrictive immigration laws, many Black immigrants passed through the U.S. ports between 1899-

1937 (Bryce-LaPorte, 1983). Many Caribbean immigrants came through Ellis Island and onto the shores of Manhattan. South Florida was also a primary destination; immigrants were coming from the Bahamas and Cuba. New York also represented a popular destination state for Caribbean immigrants, followed by Massachusetts. The peak years for migration was marked by 1913-1924 as many made their way to Brooklyn and

Harlem; Harlem being the largest West Indian City in the world by 1930 (Watkins-

Owens, 1996).

The first cohort of twentieth-century Caribbean immigrants to the United States

46 was more literate and skilled than their compatriots’ left behind; they were also more educated and skilled than the European immigrants who entered the country at the same time. Even more noticeable, this cohort of Caribbean immigrants was also more literate than the native-born White population in the United States. It was the arrival of these educated and skilled Caribbean immigrants who paved the way for future generations to come. By the 1930s, according to the book ‘Who's Who in Colored America,’ which covered the period from 1915 to 1932 many of the professionals coming from the British

West Indies and locating in New York City were doctors, dentists, and lawyers and entrepreneurs.

IMMIGRATION DURING THE 20TH CENTURY

By the second decade of the twentieth century there were deliberate attempts made to block the entry of members of the Black or African race into the U.S.- this was a reflection of racism in the legislative Acts in the U.S. Even though the rate of immigration declined due to World War I, Congress was still making it a point to reduce the number of newcomers to the country (Lodge, 1909). Advocates such as Booker T.

Washington reminded Americans of the indispensable labor that Afro-Caribbeans had performed in building the Panama Canal. According to Kasinitz (1992), even though laws were enacted to restrict immigration to the U.S., West Indians were still able to enter the

United States under Great Britain’s quota until the Depression.

The Immigration Act of 1924 dramatically reduced the number of Caribbean immigration to the United States. The act aimed at restricting non-White and Southern

47 and Eastern European immigration. An immigration quota system allowed only 2 percent of the foreign-born for each nationality and was detailed in the 1890 census. In addition to the Immigration Act, the Depression also led to more Caribbean people returning to their countries of origin due to economic hardship (Bogen, 1987). Despite all of this, people of foreign-born origin and their American-born offspring still continued to grow in the U.S. For them the United States represented newly found homes and a better way of life, and though the U.S. was facing a tough economic climate, the benefits of staying in the United States far out weighed returning to their countries of origin.

Caribbean Immigrants’ Migration Post World War II

Caribbean immigration picked back up after the U.S. entered World War II in 1941.

Many migrants were brought to work in agriculture and they took full advantage of the post war economic growth. Many immigrants working in the U.S. came from the

Bahamas, Jamaica, Barbados, Saint Vincent, Saint Lucia, and Dominica (Bryce La-Porte,

1983). Many Jamaican tobacco workers brought on contracts chose to flee to New York after their contracts were over, rather than return to their native countries; some but not all were able to transform their status and live legally in the U.S. During this same period many Puerto Ricans settled alongside Caribbean immigrants in the New York areas.

Puerto Ricans became the second largest minority in New York City, after African

Americans; the second largest Hispanic population in the United States, after Mexicans; and one of the most disadvantaged groups, together with American Indians and

Dominicans (Castro and Thomas, 2002). During the period of 1952 to 1965, permission

48 to enter the United States was given only to a handful of students and professionals and also to the relatives of earlier immigrants (Kasinitz, 1992).

The passage in 1952 of the Immigration and Nationality Act, also known as the

McCarran-Walter Act drastically reduced numbers of Caribbean farm workers to enter the U.S. The law had the desired effect of retarding the rate of Caribbean immigration to the U.S. Those who still came were overwhelmingly the close relatives of people already living in this country, rather than new immigrants as such. Fortunately, for those

Caribbean immigrants who still desired to come to the United States, a newly liberalized immigration law, commonly known as the Hart-Celler Act, was passed in September

1965 and went into full effect on July 1, 1968 (Reimers, 1965). This law launched a new wave of immigration from the Caribbean; the Hart-Celler Act favored those who sought better opportunities abroad. Between 1950 and 1960, the number of Caribbean immigrants grew including the massive flow of those leaving Cuba. For Jamaica alone the figures jumped more than eight-fold, six to ten thousand immigrants were arriving from Haiti annually and approximately half of these immigrants were settling in New

York City (Kasinitz, 1992). New York has and continues to be a major destination state for many who came from the Caribbean to settle in the U.S.

INTEGRATION OF CARIBBEAN IMMIGRANTS

The skin color of immigrants placed them in the same category as African

Americans and as a result this made them more susceptible to discrimination and violence. In the south, immigrants of Caribbean descent were singled out by the police as

49 troublemakers and they were targeted in the 1920s by the Ku Klux Klan; Hispanics did not fall shy of labels (Greenbaum, 2002). Up until 1980 Cuban exiles were considered to be hardworking, independent, law-abiding, and successful refugees from Communism.

However, following the Mariel exodus they received the reputation of undesirable, economically deprived, and criminally prone aliens (Skop, 2001). As the Clinton administration reversed the open door policy for Cubans and began the deportation of undocumented immigrants this began the end of special treatment for Cubans by the federal government. Following the Cold War period these new measures marked a profound shift in the U.S. public opinion regarding Cuban immigrants (Eckerson, 1966)

Similar to Puerto Ricans, Dominicans were one of the most stigmatized ethnic minorities in the United States. Many of the negative stereotypes given to African

Americans and Puerto Ricans were extended to Dominican Americans, in some cases primarily due to their darker skin colors (Grillo, 2000). According to the 2000 U.S. census data, when asked to classify their race, four out of five Cubans responded that they were White while, less than one-fourth of the Dominicans considered them White.

Less than 4 percent of the Cubans and nearly 9 percent of the Dominicans reported their race as Black. Moreover, a high proportion of people chose "some other race" or "two or more races".

RACIAL AND ETHNIC SELF-IDENTIFICATION

Self-identity is very important for immigrants who enter a new country for the first time. Labels given to the immigrant or perhaps even those adopted by them are crucial

50 when constructing a new identity and can lead to many consequences. Given the historical and emotional baggage the immigrants already arrive with, social construction of identity can be both challenging and confusing. Waters (2001) and Portes and

MacLeod (1996), all contend that the identity given to oneself is socially constructed based on what the person believes others see in them as or the identity others give to that person; similarly it is also possible to hold multiple self-identities at one point in time.

Ethnic identities were also highly situational for many Caribbean immigrants because the social construction of identity was contingent upon whom the immigrant was with, the circumstance the immigrant was under and who was doing the identifying or defining

(Waters, 2001).

According to the 1990 census, those immigrants arriving to the U.S. from Jamaica,

Barbados, Haiti and Grenada (traditionally Black societies) primarily identify as Black on the U.S. census. Both Trinidadians and Guyanese self-identified as Black in the U.S. however, they self-identified some other race in their countries of origin. The reason for this is that for many Caribbean islands there are many racial categories between Black and White. Caribbean people use a number of skin color categories that are also dependent on an individual’s social class. For example, someone of Indian and Black descent in the Caribbean self-identifies as ‘douglar’ a race that has been created in the

Caribbean region. Douglar is used to represent a hybrid identity; however, this categorization in the Caribbean as non-Black does not translate the same in the U.S.

Thus, when Dominicans, Trinidadians and Jamaicans of African ancestry arrive in the

United States, they often refuse the label “Black,” as it does not fit with their understanding of blackness (Bailey 2001; Itzigsohn and Dore- Cabral 2000).

Furthermore, the United States employs the one-drop rule which states that if a person has any trace of Black ancestry he or she is considered to be Black, this rule is not recognized in the Caribbean, consequently what it means to be Black in the U.S. does not have the same meaning attached to it in the Caribbean (Waters, 1999). Afro-Caribbean immigrants, therefore, face both a biological definition and a cultural association linked with blackness that may seem contradictory. Many Afro-Caribbeans also find their blackness being called in to question by African-Americans because the “one drop rule” defines them as Black, but African-Americans may reject this categorization. The differences in self-identifications by immigrants in their countries of origin and the

United States are indicative of how racial boundaries are blurred across regions. Factors such as skin color, hair textures, language, and social standing influence racial identification in the Caribbean (Rodriguez 2000), a process that varies considerably by country (Whitten and Torres 1998).

Race, as we know it is socially constructed phenomena and is likely to be fluid across societies. Accordingly, this volatility makes it very difficult to analyze the actual effects of race on assimilation in this research literature because very often, immigrants either fail to self-identify or when they actually do, their self-identification is based on what they believe others in the dominant culture see them as or what they perceive may grant them upward mobility into the dominant culture.

The study is unique compared to others because I examine the extent to which assimilation patterns are different between groups of Cubans with varying racial and ethnic self-identification. I further compare Cubans self-identifying as White and

Hispanic to Afro-Caribbeans and mixed Caribbeans. I also conduct a sensitivity analysis using national origin to test if there are any distinct differences in assimilation based on race/ethnicity and country of origin.

Racial Categories Proposed In The Study

As immigrants come to the United States, perceptions of race and ethnicity established in their countries of origin is often challenged by U.S. racial classification systems (Bailey 2001; Waters 1999). Previous self-classifications may not correspond with how they are classified in the destination country. Cubans’ immigration to the U.S. has been described as very successful due to their economic mobility relative to other immigrant groups (Nelson and Tienda, 1997). This Cuban success has been primarily defined by the economic advancement of the first two waves of Cuban immigrants who assumed the ‘exile identity’ (Greenbaum 2002; Mirabal 2003). The first two waves of

Cuban immigrants defined what it meant to be Cuban American. For this group, to be

Cuban American meant to be White and successful. This widespread presence of Cubans in the United States has also promoted the commonly held idea that Cubans are not Black

(Greenbaum 2002; Mirabal 2003). Greenbaum (2002) writes, “In the United States, Black

Cubans are invisible, to some extent unthinkable. The popular construction of Cuban-

American identity stands in deliberate opposition to Blackness”.

Cuba uses a “multiracial scheme,” portrayed by a color continuum, which in turn allocates rewards on the basis of color, idealizing Whiteness and despising Blackness

(Greenbaum 2002: 37). Skin tones have social relevance and consequences because darker skinned persons still face discrimination in Cuba (Fernandez 2001; Greenbaum

2002). This can be indicative of why so many Cubans are likely to self-identify as White as opposed to Hispanic because for this nationality whiteness is symbolic of economic upward mobility into the dominant culture.

The literature also shows that although Hispanic or Latino is used to describe someone of who is Spanish speaking of any race, historically, Hispanics have been racialized and singled out for differential treatment on the basis of physical characteristics

(Foley 1997; Menchaca 2002; Montejano 1987). Across the U.S. the terms Hispanic or

Latino are used to describe people who are of a brown complexion (Alcoff 2000;

Itzigsohn 2004; Rodriguez 2000). As Clara Rodriguez (2000) notes, the fact that Hispanic persons who appear phenotypically White are often referred to as “light-skinned” and not as White demonstrates the ways in which Latinos/Hispanics are continually racialized.

Another reason why many Cubans in the study are choosing to self-identify as White rather than Hispanic or Latino could possibly be to avoid being racialized and more importantly to gain upward mobility in the United States.

The research suggests that self-identification as White is an actual measure of integration, accordingly, I expect that there should be variation in assimilation between those Cubans who self-identify as White compared to those who self-identify as Hispanic

54 or ‘other’. For these reasons, I have divided the Cuban immigrants in the study into

“Cuban White”, “Cuban Hispanic” and Cuban Other”. I expect that those who choose to self-identify as Cuban White should be more assimilated compared to those who self- identify as Cuban Hispanic or Cuban Other. I further compare differences in assimilation not only among Cuban immigrants but also between Afro-Caribbean and mixed

Caribbeans based on four proposed proxies of assimilation.

CONTRIBUTION TO THE LITERATURE

In order to understand the assimilation patterns and the adaptations of Caribbean immigrants it is important to draw from each country’s unique histories and experiences of colonization. Migration has been the result of many Caribbean immigrants coping with economic and political upheavals of their home countries for many generations. The relationship between the sending and receiving countries is also a crucial element in teasing out the effects of the immigrants’ behaviors, attitudes and cultural adaptations.

The literature suggests that self-identification and race are very fluid across countries, thus one can question whether the examination of race is even beneficial to the research. I argue that race and ethnicity, though fluid and socially constructed has very rich meaning attached to it; moreover immigrants’ self-identification and self-perceptions can be very crucial to their experiences, adaptations and integration into the dominant culture.

The road for Caribbean immigrants has been a long and tumultuous one; the hardships faced by these groups are countless. The decision to leave one’s home country is primarily economic and political. Even when the decision to leave is made, entry into

55 another country is not always the easiest. For me, the bigger picture is what happens after the immigrants have entered the host country. I investigate the degree to which these immigrants are adapting to a new culture in the midst of cultural, social and emotional upheavals. Moreover, I examine the reasons why immigrants who come from the same region are assimilating differently into the same dominant culture. The aforementioned overview of the history of Caribbean immigrants provided some insight to this, but the proposed research gives a broader analysis of this and further investigated these differences.

CHAPTER THREE

REVIEW OF THE LITERATURE

Introduction

The following chapter provides an overview of the process of assimilation. I address the research question, “To what extent are race, class and gender indicative of assimilation among second-generation Caribbean immigrants?” To theoretically ground these answers I draw from two major perspectives, the classical linear assimilation model and the contemporary segmented assimilation model (Burgess, 1925, Park, 1930,

Gordon, 1964; Gans, 1992; Portes and Zhou, 1993; Zhou and Bankston, 1994; Zhou,

1997). I subsequently provide a critique of the assimilation models and my contribution to the existing literature. Next, I present a conceptual analysis of the individual impact of race, class and gender on assimilation. I further explain social class and assimilation by introducing both parental influences and family structure to examine the parent-child relationship on the adaptation process. I also propose four measures of assimilation quantified in the study, language proficiency, political participation, educational attainment and out-marriage/partnership. A conceptual analysis of these measures is provided relative to second-generation immigrants’ adaptation processes. The chapter ends by listing the hypotheses in the study relative to race, class and gender.

WHAT IS ASSIMILATION?

Assimilation continues to be a topic of continuing concern to many sociologists; it is described as a process by which groups with diverse beliefs and behavioral patterns become absorbed into another culture. The end product is usually the elimination of a group as a distinct cultural entity. Park (1914) describes assimilation as a process by which individuals spontaneously acquire one another’s language, characteristic attitudes, habits, and modes of behavior. This includes a progression where individuals and groups of individuals are incorporated into the dominant society. Park (1914) developed a model about the continuum of assimilation. He referred to this model as the race relations cycle and its components included contact, competition, accommodation and assimilation.

Critics such as Lipset (1950) and Lyman (1968) questioned both the sequence of Park’s cycle and its inevitability. Inherent in Park’s cycle is the notion that all immigrant groups move more or less systematically through the same basic cycle; contact with the host society, competition for scarce resources, accommodation with the members of the host society and finally assimilation into the host society. Both Lipset (1950) and Lyman

(1968) argue that it is not inevitable that each new immigrant group will ever reach accommodation let alone assimilation into the host society.

MODELS OF ASSIMILATION

Various models of assimilation have been developed by social scientists and each presents its unique network of explanatory variables (Dohrenwend and Smith, 1962; Roy,

1962; Taft, 1957; Uyeki, 1960; Weis, 1974). Gordon (1964; 1975) constructed what has

58 been said to be the most systematic paradigm for understanding and measuring the components of assimilation. According to Gordon, (1964:71) there are seven dimensions of assimilation: he refers to these as cultural, structural, marital, identificational, attitude receptional, behavior receptional and civic. Each of Gordon’s seven stages of assimilation is characterized into a sub-process or condition. Within the first stage of assimilation, which is the cultural stage, the sub process takes place by a group changing its behavioral patterns such as its religion to conform to those of the host society. The second stage, structural assimilation, includes a minority group’s large-scale entrance into cliques, clubs and institutions of the host society on a primary group level. The third stage, which is marital assimilation, refers to large-scale intermarriage between members of the host society and the immigrant group. The fourth stage, identificational assimilation, requires the development of “peoplehood” based exclusively on the host society. The fifth stage, attitude receptional assimilation suggests that the immigrant group has reached a point where it encounters little prejudiced attitudes. Under the sixth stage, behavioral receptional assimilation, there is the absence of discrimination. Lastly, civic assimilation is concerned with the absence of value and power conflict, this means that the minority group does not raise demands concerning the nature of the host society.

Since not all stages of assimilation aforementioned are easily measurable, sociologists and other social scientists that are concerned with the phenomenon usually choose those variables that are most easily measurable, such as cultural, structural, marital and identificational assimilation.

Traditional assimilation models emphasized a straight-line process where over time immigrants progressed from adaptation, through the adoption of mainstream society cultural mores, to complete incorporation into the host society. The model proposed by

Burgess (1925) and Park (1930) predicted a direct path along which successive immigrant generations traveled as they progressively acculturated until they became fully absorbed into and identified with the dominant society. These orthodox models have been placed under criticism and have been recognized as inadequate in their understanding and explanation of non-European assimilation, accordingly, they have been expanded in recent years to reflect the experiences of non-European contemporary immigrants. More contemporary studies (Gans, 1992; Portes and Zhou, 1993; Zhou and Bankston, 1994;

Zhou, 1997) all argue that far from being linear, assimilation is a multifaceted and complex process with economic, social and cultural ramifications that are closely intertwined.

The classical models of assimilation were developed at a time when conditions were quite different from those confronting settled immigrant groups today. Portes and

Zhou (1993) argue that descendants of European immigrants who confronted dilemmas of conflicting cultures were uniformly White. Even if their skin color was a darker hue than the natives, their skin color removed a major barrier to entry into the American mainstream. Thus the process of assimilation simply meant leaving the immigrant culture behind and embracing American ways. Portes and Zhou (1993) further argue that such an advantage obviously does not exist for Black, Asian and mestizo children of today’s

60 immigrants. Portes (1993), also points to the fact that the structure of economic opportunities has also changed. Fifty years ago the U.S. was the first industrial power in the world, this enabled the second-generation to move up gradually through better-paid occupations while remaining part of the working class. Sassen (1985) contends that such opportunities have increasingly disappeared in recent years due to national deindustrialization and global industrial restructuring. The result has been a widening gap between the minimally paid menial jobs that immigrants commonly accept and the high tech and professional occupations that require college degrees that native elites occupy.

This gradual disappearance of intermediate opportunities also bears directly on the race between first generation economic progress and second-generation expectations.

Adopting the outlooks and cultural ways of the native born does not represent, as in the past, the first step towards social and economic mobility and may in fact lead to the opposite. At the other end, immigrant groups who remain firmly ensconced in their respective ethnic communities may, by virtue of this fact, have a better chance for educational and economic mobility through use of the material and social capital that their communities make available (Coleman, 1988; Portes and Min Zhou, 1992; Wilson and Martin, 1982). Portes (1993) examines the experiences of Haitian immigrant groups and argues that the expected consequences of assimilation have not entirely reversed signs, but that the process has become segmented. Portes (1993) also examines what sector of American society a particular immigrant group assimilates. Instead of a

61 relatively uniform mainstream whose mores and prejudices dictate a common path of integration, today we observe several distinct forms of adaptation.

In response to the classical assimilation theory proposed by Gordon (1964), Gans

(1992) outlines several distinct trajectories that the new immigrants may follow, including downward as well as upward mobility among the possible outcomes. Further developing these ideas, Portes and Zhou (1993) propose the theory of “segmented assimilation.” According to these authors, one form of assimilation replicates the portrayal of growing acculturation and parallel integration into the White middle class; a second leads straight in the opposite direction to permanent poverty and assimilation into the underclass and a third associates rapid economic advancement with deliberate preservation of the immigrant community’s values and tight solidarity. Portes (1993) also argues that there are three features of the social contexts encountered by today’s newcomers that create vulnerability to downward assimilation. The first is color, the second is location, and the third is the absence of mobility ladders. Since the majority of immigrants today are non-White, it is by virtue of moving into a new social environment, marked by different values and prejudices that physical features become redefined and handicapped. Figures 3.1 and 3.2 illustrate transformations in legal immigration by region between the periods of 1820 and 2010.

Figure 3.1: Number of Legal Immigrants who came to the U.S. between 1820 and 2010. Source: http://teacher.scholastic.com/activities/immigration/pdfs/50_year/50_year_pie_chart.pdf

Figure 3.2: Number of Legal Immigrants who came to the U.S. from Various Regions between 1820 and 2010. Source: http://teacher.scholastic.com/activities/immigration/pdfs/by_region/region_chartmap.pdf

Portes and Zhou (1993) contend that the concentration of immigrant households in cities, and more so central cities, engenders a second vulnerability because new immigrant arrivals are now put in close contact with concentrations of native born minorities. Sullivan (1989) explains that location exposes second-generation children to the adversarial sub culture developed by marginalized native youths to cope with their own situation. This process of socialization may take place despite the economic advancement of first generation parents, since their children now have no objective reasons for embracing a counter cultural message. Many times if successful, the process can effectively block parental plans for intergenerational mobility.

A third source of vulnerability is related to changes in the host economy that have led to the evaporation of occupational ladders for intergenerational mobility. New immigrants are confined to occupations in the service and manufacturing industries that seldom offer channels for upward mobility. Due to economic restructuring, children of immigrant groups must cross a narrow bottleneck to occupations requiring advanced training in their careers if they are to keep pace with their U.S. acquired aspirations.

Portes and Zhou (1993) further contend that otherwise assimilation may not be into mainstream values and expectations but rather into impoverished groups who are confined to the bottom of the new economic hourglass.

Both classical and contemporary models of assimilation present conflicting ways in which immigrants have acculturated into the dominant culture. The contemporary

64 perspective of assimilation provides us with a critique of the classical model of assimilation. The following section in turn provides a critique of the segmented assimilation model.

CRITIQUE OF SEGMENTED ASSIMILATION

A major critique of segmented assimilation theory is that the experience of today’s immigrants and their offspring is not truly all that different from that of the 1890-

1920 wave of immigrants from Europe. For example, Alba and Nee (1997, 2003) argue that the offspring of earlier European immigrant groups often did not fully assimilate until the third or fourth generation. Perlmann and Waldinger (1997) are also skeptical of the idea that the racial distinctiveness of contemporary immigrants will be a long-term disadvantage. Because racial boundaries in the United States have proven to be fluid with regard to past “White” immigrants (Irish, Italians, and Jews, for example), they argue that contemporary Asian and Latin American immigrant groups may not be considered racially distinct in the long term. Critics have pointed out that the causal link between assimilation into the underclass and development of “oppositional cultures” among immigrant children is questionable.

Perlmann and Waldinger (1997, p.915) argue that second generation rebellion was not uncommon among earlier European groups, but that it did not ultimately hinder the upward mobility of these groups in later generations. Segmented assimilation theory has also been criticized for “essentializing central city Black culture in the image of the underclass” (Alba and Nee, 2003, p. 8). A variety of cultural models are found among

65 urban African Americans. It is thus naive to think that assimilation into native minority culture is necessarily downward assimilation into the underclass. In fact, Neckerman,

Carter, and Lee (1999) suggest that immigrants may well assimilate into the Black middle class, a possibility overlooked by proponents of segmented assimilation theory.

CONTRIBUTION TO THE LITERATURE

My study is unique compared to others in this empirical tradition in several ways.

First I will examine the assimilation of a distinctive group of second-generation immigrants in the Caribbean. I will focus on the same region however the population for the study is unique based on their languages, racial and ethnic composition, colonial histories and decisions to migrate. I further extend the research to include a comparative analysis using the respondents’ national origin as opposed to their racial and ethnic self- identification; this allows me to more thoroughly examine the impact of race and ethnicity on assimilation. I focus on race and ethnicity as an indicator of assimilation because much of the literature examines Latin American countries, however very little has been done on the English speaking traditionally Black societies. The research primarily compares Cuban immigrants to both English speaking and French speaking

Black Caribbean immigrants.

Second, I examine the impact of social class on assimilation by analyzing the parent-child relationship. I will explore the degree to which parents’ socioeconomic status influences the second-generation’s integration into the host society. The data also allows me to compare assimilation patterns over time and illustrate the ways in which

66 occurrences in time period one subsequently impact integration in later time periods. I also go beyond the traditional measures of social class typically used in the research and include other measures such as family’s economic situation, homeownership status, family structure and mother’s regular employment. Gender, which has been relatively absent in the assimilation literature, is another central component to my study. I will examine the distinct ways in which males and females integrate into the dominant culture.

Third, I will employ four measures of assimilation, characterized by the respondents’ language proficiency, political participation, educational attainment and out- marriage/partnership. These measures are widely used in the assimilation literature and are compelling indicators of subsequent assimilation into the dominant culture.

Similar to a number of studies in this genre, I will draw on the classical linear model to construct the four distinct measures of assimilation. I will also integrate the contemporary segmented assimilation model to indicate that these measures all represent upward assimilation into mainstream America. The main foundation of the study is not to examine the trajectory of assimilation as have been customary, but to innovatively explore the degree to which race, class and gender remain salient factors even amongst a relatively homogenous sub sample of second-generation Caribbean immigrants.

THE IMPACTS OF RACE, CLASS AND GENDER ON ASSIMILATION

Race/Ethnicity and Assimilation

Race and ethnicity are essential to my study in order to gain a more thorough understanding and analysis of the differences among Latin American Cuban immigrants,

Haitian, Jamaican and West Indian Black Caribbean immigrants and mixed Caribbean immigrants. Mixed Caribbean immigrants represent those who self identify as Hispanic, multi-racial or some other race from the islands of Jamaica, Haiti, Dominican Republic and the West Indies. Because the countries chosen for the study are traditionally Black

Caribbean societies, the population does not consist of a substantial number of White

Caribbeans; Whites are therefore represented in the research by those immigrants arriving from Cuba who self-identify as such. Black Caribbean immigrants as previously discussed in the last chapter fit a very unique profile that is distinct from African-

Americans; as such their adaptations and integration into the American culture have been unique. I contribute to the larger assimilation literature by comparing the differences in pattern of assimilation and integration among Cuban, Black Caribbean and mixed

Caribbean immigrants.

The new wave of immigrants has caused many researchers to re-examine and develop more insightful theories about the extent to which the experience of assimilation is now different from the earlier waves of European immigrants. Previous research findings on assimilation have all suggested that race and ethnicity are in fact predictors of assimilation. The focus of the study further investigates the degree to which race and

68 ethnicity continue to influence the assimilation patterns for a relatively uniform group of second-generation Caribbean immigrants.

As the population of non-native Blacks settling in the United States increased over the years, the concept of racial identification became an increasingly complicated issue. The first generation Black immigrants made it a point to emphasize their ethnic identities and national origins underplaying the more generic identification as Black. Due to prejudices against native Blacks, general aversion to an undistinguished Black identity as well as pride in their national identity, non-native Blacks in the U.S. resisted from identifying with American Blacks (Kasinitz, 1992; Arthur, 2000; Zephir, 2001). Mazrui

(1996) argue that new immigrants from Africa who possess a strong sense of their cultural heritage and African roots often use the term “American Africans” to describe themselves and their children. Although first generation Black immigrants from Africa and the Caribbean attempted to instill their aversion to an undifferentiated Black identity in their children, their offspring faced tremendous pressures to make ‘Black’ their principal identification (Woldemikael, 1989; Portes and Zhou, 1993; Waters, 1996).

According to Portes (1995) and Zhou (1997), although the loss of employment opportunities has potent effects, ethnic and racial characteristics also shape assimilatory patterns in the segmented model. Segmented assimilation postulates two cultural explanations for divergent assimilation paths. The first suggests that the stubborn cultural values of certain groups, particularly those of Asian origin, lead to resistance through enclave strategies, the formation of resilient families as well as an emphasis on success

69 among descendants; all of these characteristics represent features of upward assimilation.

A second explanation points to some other ethnic groups, particularly those of Latino and

Afro-Caribbean origin who appear to have weak cultural ties and may follow a downward assimilation path towards the norms of the marginalized populations of the inner cities (Portes and Rumbaut, 2001). This research examines four measures of upward assimilation and the ways in which second-generation Caribbean immigrants are depicting differences based on their race and nationality. I will further examine the extent to which those who self-identify as Black, White, Hispanic, or some other category are in fact achieving upward assimilation into the dominant culture.

Portes et al. (2005) further argue that children of Asian, Black, mulatto and mestizo immigrants cannot escape their ethnicity and race as defined by the mainstream, because the strong effects of discrimination throws a barrier in the path of occupational and social acceptance. Similarly Hernandez and Drake (1999) contend that the primary reason for a lack of economic improvement across generations among Mexican, Haitian and Dominican immigrants is that racial and ethnic stratification greatly limits their opportunities. The impact of discrimination against persons whose ethnicity or race is the same as that of the disadvantaged group in the U.S. is usually found among the children of immigrants and not so much among the first generation, who turn towards the seduction of ghetto cultures which become a primary reference for socialization (Portes and Rumbaut, 2001; Portes and Zhou, 1993; Rumbaut, 1994). Consequently segmented assimilation theory argues that, while cultural values might protect some groups from the

70 negative impact of American conditions, for many others ethnicity is a marker for discrimination by the majority White population and this in turn leads to negative outcomes for the immigrant. This study also examines the effects of discrimination on the assimilation patterns of second-generation Caribbean immigrants; I will investigate the extent to which discrimination is truly a salient effect on the adaptations of these immigrants. This study investigates the relationship of discrimination on political participation, English proficiency, educational attainment and out-marriages/partnership.

I expect that exposure to discrimination will lead to lower levels of language proficiency, political participation, out-marriage and educational attainment.

One critique of segmented assimilation and race is that while segmented assimilation claims that the non-White racial status of most current immigrants and the very different economy they face in twenty-first-century America puts them and their children at greater risk, Perlmann and Waldinger (1997), Waldinger and Perlmann (1998) and Gratton (2002) have all criticized segmented assimilation's characterization of the past. Furthermore, many studies find little evidence of second-generation decline or downward assimilation (Boyd, 2002; Hirschman, 2001; Farley and Alba, 2002; Kasinitz,

Mollenkopf and Waters, 2002, 2004; Smith, 2003; Waldinger and Feliciano, 2004;

Waldinger, 2007; Kasinitz et. al., 2008)

Parental Social Class and Assimilation

Many sociologists have argued that the characteristics of the immigrant parents trickle down to the children, affecting their general well being and adaptation process. In

71 this study, parents’ socioeconomic status is key in determining the assimilation patterns of second-generation children. Though parents’ socioeconomic status and the respondents’ own levels of education are typically used in the assimilation literature as measures of social class, the proposed research goes beyond these measures by including additional measures of family’s class, such as family’s economic situation, homeownership status, mother’s regular employment and family structure.

The social learning or role-model explanatory framework posits that parental influence is a major factor that is conceptually associated with the subsequent outcomes of their children’s learning and assimilation into the host society (Biblarz, Raftery, and

Bucur, 1997; Chen and Kaplan, 2001; Feng et al., 1999). This perspective further states that the desire of children to imitate their parents is crucial to understanding the process by which social indicators are transmitted across generations. Haveman, Wolfe and

Spaulding (1991) argue that parents with higher levels of education are expected to have a positive influence on the educational outcomes of their children because of the expectation that such parents will be seen as good role models by their children.

Similarly, the role model hypothesis predicts that children living with single parents who have lower levels of educational attainment will have less favorable schooling outcomes because of the absence of a positive model provided by a highly educated parent within the family and the negative influence of their own parents’ lower level of schooling.

Another crucial effect, which goes along with parental influence, is parental aspirations. Many studies have found evidence that parental aspirations do matter in

72 terms of schooling performance of immigrant children. According to Kao and Tienda

(1995), immigrant parents are more likely than their native born counterparts to promote educational achievement, through the imposition of strict rules on expected GPAs and homework assignments. Kao (2004) contends that while the parents of immigrant youths are generally less likely to talk about schooling, they are more likely to talk about college and have closer relationships with their children than their native born counterparts.

Fuligni (1997) argues that the strong emphasis on education shared by immigrant parents, their children and their children’s peers is an important contributor to the educational successes of East Asian, Filipino and European youths.

In the economics literature, Becker (1993) and Grawe and Mulligan (2002) frame the effects of parents’ educational attainment in terms of the human capital investment framework. This framework describes the positive relationship that exists between childrens’ schooling and the economic resources of their parents (Taubman, 1989;

Behrman and Knowles, 1999; Ermisch and Francesconi, 2001). The income effect is intuitive, since it shows that children in families with more resources (incomes and assets) are more likely to live in areas with better schools and have parents who can afford to pay for supplemental tutoring and other auxiliary services (Alderman, Orazem and Paterno, 2001; Gordon, Bridglall and Meroe, 2004). Parents with higher incomes also have greater ability to mold behavior of their children using pecuniary incentives compared with their less wealthy counterparts (Weinberg, 2001). Human capital investment models predict better schooling indicators among the children of some Black

73 immigrant groups that have been found to earn higher incomes compared with native- born Blacks (Butcher, 1994; Kalmijn, 1996; Massey et al., 2007). I expect that parents’ with higher socioeconomic status are more likely to rear children who are more proficient in English, more politically active and attain higher levels of education due to their higher levels of economic resources, higher levels of education and more prestigious jobs.

A major critique of parental influence on segmented assimilation is that it does not predict universal downward mobility anymore than classic assimilation predicts universal upward mobility. Alba and Nee (2001) and Portes and Zhou (1993) both argue that some members of the second-generation will do well compared to their parents, while others will not. Both theories posit that racial discrimination will make it much more difficult for those defined as non-White to achieve upward mobility in America's racially stratified economy. In the proposed research, I investigate the degree to which family’s social class (parent’s socioeconomic status, homeownership, family’s economic situation, family structure and mother’s regular employment) influences the process of assimilation. I also examine the effects of both discrimination and family’s social class in determining the level of assimilation into the dominant culture.

Family Structure and Assimilation

In order to study assimilation patterns of second-generation immigrants it is important to consider the family structure because characteristics of the family structure may subsequently influence the ways in which children of immigrants assimilate into the

74 dominant culture. In the research, family structure refers to immigrant children’s household guardians. The structure of the family is represented by the presence of both parents in the household versus some other type of living arrangement. For instance the respondent may be alternatively living in a single parent household, with step-family, friends or even grandparents. I use family structure as a measure of social class to illustrate the effects of the presence of both parents on the subsequent assimilation patterns of the second-generation immigrants.

Segmented assimilation theory responds to unexpected trends since 1960 in family structure, socioeconomic status, and educational achievement among descendants of certain immigrant groups (Waters, 1990; Gans, 1992; Skop, 2001). Similar to Portes and Zhou’s theory of assimilation, Waters 1994 and Fernandez-Kelly 1995 argue that second-generation youths confront a pluralistic fragmented environment that simultaneously offers a wealth of opportunities and major dangers to successful adaptation. When dealing with second-generation immigrants the question being posed is not whether these immigrants assimilate into American society but instead, to what segment of society do these immigrants assimilate? There are three major impediments to educational achievement and career success faced by today’s children of immigrants.

Portes and Zhou (1993); Rumbuat (1996); Portes and Rumbaut (2001) all argue that these challenges include the persistence of racial discrimination, the bifurcation of the

American labor market and its growing inequality and the consolidation of a marginalized population in the inner city. Based on a study of West Indian families and

75 teenagers in New York City, Waters (1996) contends that class mobility for immigrants and their children is no longer associated with increasing Americanization for all groups.

Different trajectories of socioeconomic incorporation and success and cultural integration describe the experience of different families.

Researchers further argue that these negative effects are salient in family structure such as the prevalence of marital disruption over time in the U.S. as well as in succeeding generations (Rumbaut, 1996; Gil and Vega, 1996). Rapid increases in marital instability among Dominicans, Puerto Ricans and Haitians mirror those among impoverished

African Americans and native Hispanics and constitute a form of assimilation towards

U.S. norms, albeit a negative one (Fernandez-Kelly and Schauffler, 1994; Landale and

Ogena, 1995; Portes, 1995; Portes and Zhou, 1994; Waters, 1996). Furthermore, higher rates of female-headed households among Mexican Americans in turn provides a model for Mexican immigrants and their descendants (Bean et al., 1980; Frisbie and Bean, 1995;

Landale and Oropesa (1995). The children and grandchildren of immigrants mimic these traits rather than those of the host society or of White natives (Portes, 1995).

According to Landale and Oropesa (1995) third-generation Latino children in

1990 were much more likely than first generation children to live in single parent families, and substantially more likely than non-Latino White children. Hernandez and

Drake (1999) located similar sharp increases in the risk of single parent families for third generation children. Brandon (2002) notes that Mexican, Black, and other Hispanic children increasingly live in single-mother families, which by all measures indicate

76 downward assimilation. The research examines the effects of family structure on assimilation and investigates the extent to which living in a household with both parents versus living in any other type of family structure has any significant effects on assimilation. I expect that families without marital disruptions will experience higher levels of assimilation than families that are not intact.

In a study of Mexican and Mexican American youths in a California high school,

Gibson (1997) found that minority students do better in school when they feel strongly anchored in the identities of their families, communities and peers and when they feel supported in pursuing a strategy of selective or addictive acculturation. Alternatively, those at greater risk of failure are those who feel disenfranchised from their culture.

Hirschman (2001) argues that familial and socio economic characteristics influence immigrant teenagers’ educational enrollment; poorly educated parents can therefore explain a significant share of some immigrant groups’ low enrollment rates due to the absence of parents and inner city residence. Strong effects of discrimination based on racial and ethnic differences, especially against Black persons, throws a barrier in the path of occupational mobility and social acceptance. Immigrant children’s identities, their aspirations, and their academic performance are affected accordingly (Fernandez-Kelly and Curan, 2001; Lopez, Stanton, Salazar, 2001).

Immigrant parents’ human capital and family composition do not exhaust the range of forces modeling types of acculturation and subsequent adaptation outcomes. The outside environment supplies the other key factor. When a favorable reception by the

77 government and society promotes the emergence of strong ethnic communities, the social capital grounded on ethnic networks provides a key resource in confronting obstacles to successful adaptation (Portes and Rambaut, 2001). Solidarity communities can be a valuable resource, since networks support parental guidance and parental aspirations for their children. Among immigrants of limited means, this function of social capital is vital

(Zhou and Bankston, 1996; Fernandez-Kelly and Konczal, 2005).

Alba, Farley, and Nee (Farley and Alba, 2002; Alba and Nee, 2003) all argue that parents typically begin at the very bottom of the occupational ladder makes upward, not downward mobility, the more likely outcome. Indeed, as shown by Farley and Alba

(2002), while second-generation persons originating in labor migrant groups do not possess the educational attainment of native-born Whites, their schooling performance represents significant advance as compared to the first generation. The same holds for occupational attainment, where sizeable growth in second-generation white-collar employment points to a major departure from the parental pattern.

There are many characteristics such as discrimination, parents’ education and family structure, which are all instrumental in giving a broader understanding of second- generation immigrant children modes of assimilation and which are included in the proposed research. These measures are all representative of the ways in which social class is significantly related to the assimilation of second-generation Caribbean immigrants.

Gender and Assimilation

Basic concepts, such as sex and gender too rarely enter the vocabulary or research design of immigration research. Globalization, immigration, transnationalism are significant sites for contemporary inquiries of gender. The growth in immigration research derived not from a social movement like feminism, but from the massive increase in literal, human movement across borders during the twentieth century.

Estimates show that as many as 150 million people live in nations other than where they were born. The growing literature on the topic has focused on economic activity of immigrant women in societies such as the United States, Canada, Australia, Latin

America, and Western Europe (Boyd, 1984; Evans, 1984; Gabaccia, 1992; Phizacklea,

1983; Stier, 1991; Stier and Tienda, 1992; Sullivan, 1984). These studies have arrived at three conclusions: First, immigrant women are characterized by higher rates of labor force participation than other women (Boyd, 1984; Castles and Kosack, 1973; Pedraza,

1991; Phizacklea, 1983; Simon, 1992; Tyree and Donato, 1986). Second, economically active immigrant women face greater hardships in the labor market when compared with immigrant men (Basavarajappa and Verma, 1990; Boyd, 1984; Evans, 1984; Phizacklea,

1983; Sullivan, 1984). Third, socioeconomic disadvantages of immigrant women are likely to differ by country of origin (Boyd, 1984; Phizacklea, 1983; Sullivan, 1984).

There has been considerable work undertaken on the matrifocal nature of the family in the Caribbean. The set of studies in the late 1980s of women in the Caribbean

(Massiah, 1986; Senior, 1991) provided a large body of material that profiles the

Caribbean woman as both worker and nurturer. For many Caribbean families the women usually assume head of the household and the breadwinner even when the male is present. Educational achievement research in Jamaica suggests that although boys are at an equivalent level to girls on entry to grade one in primary school, differences between them in academic achievement are beginning to be evident by grade three (Samms-

Vaughn, 2001), and become more marked by grade four based on the literacy test. The incidence of dropping out occurs as early as grade four at an average age of nine years old and a ratio of 8:5 boys to girls (Samms-Vaughn, 2001).

Gender differentials in exam results stirred dialogue among researchers as to why girls throughout the Caribbean seemed to be outperforming boys in many subjects; even as early as entry to grade one as well as relative to high school graduate and college graduates. Many researchers findings, such as Miller’s male marginalization theory

(which examines the feminization of education among other factors) and Figueroa’s

(1996) work on male privilege, point out that female socialization prepares girls better for the education system than boys. Parry’s (2000) examination of the “gendering” of academic subjects and Chevannes (1991) rejection of “male marginalization” suggest that many males, particularly those encumbered by poverty and poor initial education, have simply discarded education as a feasible route to desired goals, while women remain empowered by educational routes to greater independence. I expect that gender differences in childrearing and disparities in educational performance in Caribbean children will consequently lead to differences in language proficiency in children

80 immigrants. I expect that females will be more proficient in English compared to males due to their superior academic performance and stricter expectations placed on girls in childhood.

Elmelich and Lu (2004) argue that immigration seems to have an equalizing effect on the socioeconomic gender gap in the U.S. The reason being that immigration carries a greater risk for women than it does for men, this in turn explains the relatively low level of gender-poverty-gap among the Asian, central and southern Hispanic, and Mexican populations in the U.S. In a study conducted among immigrants in the Netherlands by

Veenman and Heij (2008), they introduce the gender gap convergence hypothesis where they argue that the immigration of women from lesser developed to developed countries actually allows them to improve their economic chances because there is more equal opportunities for paid labor and equal access to education in the host country. The reduction of the gender gap will be greater for women migrating from developing countries because the equalizing effects between male and female attainments will be more noticeable. The authors argue that the gender gap in education and income is reduced even further in subsequent generations of immigrants because migration from the developing world in fact allows women who were previously restricted due to their ascribed status to catch up with their male counterparts over time. The gender gap is therefore expected to converge even more for the second-generation onwards. I expect that according to the gender gap convergence hypothesis, second-generation Caribbean women will perform equally to Caribbean men; thus there will be no differences in

81 gender for the educational attainment assimilation model.

According to the Pew Research Center’s analysis of demographic data about new marriages in 2008, gender patterns in intermarriage vary widely. Some 22% of all Black male newlyweds in 2008 married outside their race, compared with just 9% of Black female newlyweds. Among Whites and Hispanics, by contrast, there are no gender differences in intermarriage rates. About 9% of both male and female White newlyweds in 2008 married a non-White spouse, and about a quarter of both male and female

Hispanic newlyweds in 2008 married someone who is not Hispanic. These trends indicate that males are more likely to out-marry compared to females. Additionally there seems to be much interracial marriages taking place.

Additionally, the changing of women’s roles, which are primarily due to their economic advancement in the labor market, has led to changes in family structures and partnerships. Less differentiated gender roles are now facilitating a reduction in the gender gap (Becker, 1981; Oppenheimer, 1988; Solsona, 1998; Luxán, Miret and

Treviño, 1999). Males and female now choose their mates based on alternative attributes that provide a less unbalanced relationship. According to these authors, a young female of lower socioeconomic status is no longer seeking out a male for financial security because that female is now advancing economically and financially in the labor market and is likely to choose a partner of equal status. Due to the transformation of women’s roles by their entry into the labor market, women are more likely to choose partners who are their equals. I expect that women will be more likely to out-marry/partner compared

82 to men. I further argue that gender is not neutral in out-marriage because both race and class interacts when choosing a spouse.

While research on gender and immigration has focused largely on the effect of immigration on women’s individual emancipation and empowerment (Hondagneu-

Sotelo, 1994; Hondagneu-Sotelo, 1992; Pessar, 1995), the proposed study is innovative because it explores the extent to which gender affects the assimilation patterns of second- generation Caribbean immigrants into the dominant culture. Furthermore I examine the degree to which males and females are integrated differently or similarly into the host society.

FOUR PROPOSED MEASURES OF ASSIMILATION

The following sections provide a conceptual overview of the four measures of assimilation employed in the research; language proficiency, civic involvement, out- marriage/partnership and educational attainment. I provide justification for the selection the measures and position these proxies of assimilation as a framework to the research.

1. Language Proficiency

Many researchers have cited a lack of English proficiency among first and second-generation as the primary reason for lower levels of academic achievement

(Rosenthal, Baker, Ginsung; 1983; Cosden et al., 1995). English proficiency is therefore noted in the literature as having very strong links to academic performance. Age of migration and generation status both influence English language proficiency (Stevens,

1999). Assimilation theorists argue that speaking English well is a key measure of

83 incorporation into the dominant culture. Immigrants who do not possess strong English speaking skills are more likely to be in poverty, more likely to rear children who perform poorly in school and additionally. These immigrants are also less likely to become U.S. citizens and by extension be involved in the political process.

With regard to gender, the literature shows that there is some difference between men and women and English proficiency. Women who immigrate to join their spouses as opposed to join the labor market are less likely to be proficient in English. Furthermore, this lack of entry into the labor force creates less monetary rewards for speaking English and in fact their costs of learning English may be even greater, since women who stay at home to take care of the home and raise children come into contact less with English speakers than if they were in the workforce. Accordingly, overtime women might show less improvement in English skills than men Carliner (2000). Aforementioned, one unique contribution to the literature is the examination of the immigrants’ language proficiency at an average age of fourteen years old. As mentioned in the previous section on gender and assimilation, I expect that gender roles, socialization and strict child rearing imposed on young girls in the Caribbean should lead to young girls being more proficient in English compared to young boys.

With regard to race and ethnicity, Veltman (1988) finds that school aged Hispanic immigrants are more likely to make the shift from Spanish to English speaking, however the speed of transition in language falls rapidly with age. According to Portes and

Schauffler (1994), 86 percent of Cuban-American children prefer speaking English to

Spanish. After using the 1900 and 1980 U.S. censuses to compare the English skills of immigrants, Jasso and Rosenzweig (1990) found that per capita income and literacy rate of the country of origin are also associated with better English skills.

Portes and Rumbaut (2006) state that country of origin also affects the acquisition of and proficiency in English after arrival. For example, those immigrants who intend to eventually return to their country of origin are less likely to invest and learn English than those (political refugees) who expect to remain in the U.S. for the rest of their lives. All of these reasons lead to variations in the average skill of English for immigrants after arrival into the host country. Learning English before entering the destination country is dependent on the linguistic distance of the source country’s language from English

(Chadwick and Miller, (1995). Jasso and Rosenzweig (1990) argue that per capita income of the source countries is positively correlated with proficiency in English.

The literature also suggests that race and ethnicity should have no effect on the immigrants’ language proficiency for later generations. According to Waters (2005), within this three-generation model of language assimilation the first generation makes some progress in language assimilation but remains dominant in their native tongue, the second generation is bilingual, and the third-generation speaks only English. Alba (2004) also argues that bilingualism is common between second-generation immigrants; this refers to those children who are growing up in immigrant households. These children are likely to speak an immigrant language at home, although almost all of them are proficient in English. I expect that because second-generation immigrants were either born or

85 brought to the U.S. at an early age, there might be no difference in language proficiency relative to race and ethnicity and nationality.

With regard to social class, higher levels of education are associated both with lower costs and greater benefits to learning English. Immigrants, who are not very proficient in English, often find jobs within ethnic enclaves that do not require that the immigrant be proficient in English. Alternatively, being hired in highly skilled jobs that may require immigrant workers to be in extensive contact with natives usually results in a high proficiency in English. Therefore, Kossoudji (1988) McManus (1990) and Carliner

(1996) argue that the wage premium for speaking English well rises with education.

Using data from the 1989 Census Bureau Current Population Survey Espenshade and Fu

(1997) find that an additional year of schooling in the United States has a larger effect on the English skills of the immigrant than one year of foreign schooling. Additionally, education, which is highly correlated with English proficiency, decreases the probability that new immigrants will settle in cities where there are high concentrations of fellow countrymen.

In addition to race, class and gender there are other additional independent variables that are related with language proficiency. Using data from the 1989 Census

Bureau Current Population Survey, Stevens (1994) finds that entry as a child and years since arrival also raise the English skills of immigrants. Stevens (1994) also contend that other modes of cultural integration such as U.S. citizenship, homeownership, marriage to

86 someone with a different mother tongue, especially English all have positive effects on the immigrants’ English skills.

The ability to learn the language (vocabulary, grammar, syntax and pronunciation) of the dominant culture declines with age; therefore age of entry into the host society, will also affect English skills (Dulay, Burt and Krashen, 1982; Spolsky,

1989; Johnson and Newport, 1991; Service and Craik, 1993). Immigrants who arrive in the destination country at a very young age are more inclined to invest in learning the language because they have a longer period by which they can recover the cost of their investments; this in turn increases the incentive to invest in learning the new language of the dominant culture. Accordingly, immigrants coming to the U.S. from non-English speaking countries at a very young age are more likely to learn more English than immigrants who arrive at a later age-- all things being equal.

Similar to the decision to out-marry; the decision to learn and invest in English is also heavily dependent on the size and cohesion of the ethnic enclave (Lazear, 1995).

Correspondingly, if an immigrant chooses to marry a native English speaker, this decision in fact improves his/her language skills. The proposed research examines the extent to which race, class and gender influence the language proficiency of second- generation Caribbean immigrant children.

2. Political Participation

Political participation is used in the research as the second measure of assimilation. I measure political participation by the respondents’ registration to vote.

The following section gives an analysis of the expected relationships between race, class, gender and other independent variables used in the study as they relate to the respondents’ political participation.

With regard to gender, both political scientists and sociologists have recognized significant variances in civic engagement between males and females. Verba et al. contends that men are more likely to engage in political activities, such as voting and contributing to a campaign, compared to women (Verba, Schlozman, and Brady, 1995).

In more contemporary research, the Center for American Women and Politics found that women of all ages had consistently higher voter turnout rates than men since 1984.

Employing the 2002 Civic and Political Health of the Nation Survey, (Keeta, Zuckin,

Andolina and Jenkins, 2002) all note that young women and men are similar in political participation, however, young men were seen to be much more attentive to the news and politics than young women, these differences were also observed among adult men and women. The authors state that even though there are some differences in rates of voluntarism and political knowledge and attentiveness between males and females, gender does not appear to be playing a significant role in shaping civic engagement among youth. According to contemporary research, I expect that there will be no difference in voting between males and females.

With regard to race and ethnicity, researchers find that when immigrants become politically involved within the dominant culture they have greater opportunity for the development of group based political power and there is greater representation for many

88 of these minority groups, who have been previously marginalized (Eisenstein, 1994,

Fraser, 1992 and Phillips, 1994). According to Western liberal enlightenment and political theory, a greater number of citizens participating in politics translate into a greater number of opinions and preferences being taken into consideration. More individual representation in politics and voices in democracy should lead to better representation and subsequent increases in equality and justice. As political activity increases among traditionally marginalized and politically underrepresented groups, this participation can act as a catalyst, creating public policies that take their interest into account, empowering those previously disenfranchised and advancing their interests.

Since both immigrants and racial and ethnic minority groups are politically involved at lower rates compared to Whites, these groups have been targeted through naturalization and voter registration drives in order to increase their political participation (Bobo, 2000;

Hochschild, 1996; Sniderman and Piazza, 1993).

Racial minorities are more engaged than their White counterparts when they live in high minority empowerment areas (Bobo and Gilliam, 1990), minorities are more likely to vote when there is a fellow minority on the ballot (Barreto, 2004, Gay, 2001), minority groups are also more responsive to racially framed issues (Bobo, 2000;

Hochschild, 1996; Sniderman and Piazza, 1993), minorities form social movements centered around their shared identity (Espiritu, 1992; McAdam, 1999; Padilla, 1985) and racial minorities are likely to sponsor their own political organizations (DeSipio, 1996;

Lien, 2001; Pinderhughes, 1992). Furthermore, both social class and geographical

89 location are dominant influences in voting behavior regardless of national origin Glazer

(1963).

The SES model has proven to be a resounding success in explaining political participation among non-White minority Americans. Once higher levels of education and family wealth among Whites have been controlled for, African American and Hispanic citizens have been found to participate at or near the same rates as their economically and educationally advantaged White counterparts (Bobo and Gilliam, 1990; Barreto, 2004,

Gay, 2001; Bobo, 2000; Hochschild, 1996; Sniderman and Piazza, 1993; Espiritu, 1992;

McAdam, 1999; Padilla, 1985; DeSipio, 1996; Lien, 2001; Pinderhughes, 1992).

With regard to social class, the socioeconomic status model postulates that people with more economic and social resources are more likely to be actively involved in politics and value and recognize the benefits of civic participation. According to

Wolfinger and Rosenstone (1980), the two most important predictors of participation are formal educational attainment and family income. Family income actually accounts for a larger share of the variation in citizen activity than formal education. For over fifty years, the American National Election Study has repeatedly shown that a higher level of education and more substantial economic resources increases participation in political involvement. I expect that class should influence political participation; accordingly, those with higher levels of education and higher socioeconomic status are more likely to be politically involved.

In addition to race, class and gender there are other factors that are related to political participation. In ‘A Portrait-Immigrant America’ written by Portes and Rumbaut

(1996) the authors postulate that politics and civic involvement have always been a difficult topic because many immigrants are torn between old loyalties and new realities.

Very often the politics of the first generation have been characterized by that immigrants’ preoccupation with the country of origin. Immigrants usually have difficulties giving up their loyalties and allegiances to their national origins.

There are four main characteristics of the sending countries from which immigrants arrive (Portes and Rumbaut, 1996). First, stateless nations are divided lands, which are contested by warring factions or occupied by foreign power. Second, hostile states are dictatorships that oppressed entire populations of people or singled out certain groups for persecution. The third characteristic of sending countries is consolidated and indifferent nation states; these types of states neither promoted nor acknowledged the migrant’s departure. The fourth are states that actually supported and facilitated the emigration of their nationals abroad as outposts serving their country’s interest. The authors argue that depending on the characteristics of the sending country, immigrants are more or less likely to be committed to political causes back home. These immigrants may either see themselves as representatives of their nation states abroad or they may turn away from their past and concentrate on building a new life in America. Though the sample population for the proposed research does not focus on first-generation

91 immigrants, the characteristics of the countries from which their parents came should be indicative of their own levels of political participation.

For mostly nineteenth century and early twentieth century immigrants, many of whom intended to return home, split loyalties resulted in them paying more attention, at least initially to the events in the sending countries than in the U.S. Portes and Rumbaut

(2006) argue that immigrants’ political participation differ based on their past political socialization, their commitment to return, and national situations left behind in the sending countries. These combined factors affect their participation and stance in

American domestic politics. Depending on their allegiance, some groups struggle for independent statehood of their countries while others were unaware that they even left countries behind.

Generally, immigrants who are politically inactive in one setting tend to translate this behavior into others. Involvement in transnational, political or civic activism essentially increases involvement in domestic politics. Portes and Rumbaut (1996) note that the passage of dual nationality and dual citizenship encourages immigrants to become U.S. citizens because immigrants subsequently lose the fear of giving up their original nationality and of being perceived as ‘defectors’ at home. The ability to hold on to their nationality and identities remove disincentives to acquire U.S. citizenship. As

U.S. citizenship and naturalization are acquired, immigrants can easily register to vote within the destination country. The authors further argue that even though immigrants are involved in transnational activities, their voices cannot be heard and they are not

92 represented in the political realm until they have naturalized. The proposed research examines the differences in voting patterns between those immigrants who are U.S. citizens and those who are naturalized citizens.

According to Portes and Rumbaut (1996), first generation immigrant groups opt for being ‘in the society’ but not ‘of the society’ avoiding naturalization at all costs.

Alternatively, subsequent immigrant groups opt to change their flag at the first opportunity granted. An important point to note is that higher numbers, greater concentrations and a higher rate of citizenship acquisition all contribute to the political strength of immigrant communities. The level of education, the length of residence in the country and the reversibility of migration affect the differences in the probability of naturalization among nationalities. Politics of the second and successive generations are said to pivot less around issues of class than those tied to a common ethnic origin.

The previous review of the literature suggests the sending country typically affects political participation for the first generation immigrants. I expect that nationality should affect political engagement due to immigrants’ split loyalties, ties to country of origin, intention to return home and political activity in the country of origin. I expect those racial and ethnic groups with weaker ties to the country of origin should be more politically active in the U.S. compared to those with stronger ties.

3. Out-Marriage/Partnership

Out-marriage/partnership is used as a third measure of assimilation in the study. I compare out-marriage/partnership by using three of the five racial and ethic categories

93 previously discussed (Cuban White, Cuban Hispanic and Caribbean Black). I eliminate those respondents in the Caribbean mixed and Cuban other categories because their racial identities were not clearly defined.

Researchers find that for gender, the roles of women have transformed over time and women are now more actively involved in the labor market. Additionally, in recent times, studies have indicated that more and more couples and women specifically are choosing to marry later in life. The rise in cohabitation actually blurs the meaning and interpretation of conventional statistical measures of union formation, marital and non- marital fertility and dissolution (Bumpass and Raley, 1995). Rising levels of cohabitation in fact slow the marital assimilation and reinforce group boundaries. Also as previously discussed, women are selecting their mates based on many other attributes due to their financial independence and movement into the labor force.

Race and ethnicity plays an integral part in the out-marriage and partnership of immigrants. Interracial marriages have increased in the U.S. since the 1970s and following the ruling of the Supreme Court in 1967 that deemed anti-miscegenation laws as unconstitutional. As a result of this, group boundaries have weakened and inter-social distance has declined. Labov and Jacobs (1998) argue that racial boundaries have been blurred over successive generations due to the growing population of mixed race children from intermarriage; this in turn has raised new questions about the conception of race.

Simmons and O’Connell (2003), state that interracial marriages accounted for only six percent of all marriages in 2000. Furthermore, marriage and social distance from Whites

94 varies widely across racial and ethnic groups (Lee and Edmonston, 2005 and Qian and

Litcher, 2000). As the immigrant population continues to grow, this in turn replenishes the marriage pool for their native born counterparts resulting in the reinforcement of cultural and racial identities and the formation of intergroup boundaries (Massey, 1995;

Qian and Litcher, 2001).

Gordon, 1964; Qian and Litcher, 2001; Sassler, 2005 all argue that intermarriage is viewed as the final step in the assimilation process. A very clear indication is that minority groups have adopted cultural patterns of the dominant culture (language, dress, customs) and that they have been absorbed both politically and economically into the mainstream society by the intermarriage with Whites. The economic assimilation and incorporation of many racial and ethnic minorities have taken place at a much slower rate and with less predictability than European White ethnics during the twentieth century.

For many racial and ethnic minority immigrants, prejudice and racial discrimination continue to limit opportunities and impede both economic success and assimilation into the dominant culture.

Group size is also an important factor when considering assimilation into the host society and intermarriage. Blau (1977) contends that as racial and ethnic minorities grow in population this decreases the likelihood of interaction and marriage with the majority group. This occurs because racial and ethnic minorities are more likely to interact and intermarry each other if their group is of a substantial size. Hence, as racial and ethnic

95 diversity continues to increase in America the speed of incorporation within society and intermarriage for that matter, is much more likely to decelerate.

Other factors such as residential concentration also affect opportunities for interaction with the mainstream (White) society. Many racial and ethnic minorities tend to be residentially segregated due to a lack of socioeconomic mobility. They are also more likely to live ethnic enclaves, and this in turn magnifies their social distance from

Whites compared to their native born counterparts (Massey, 1995; Portes and Bach,

1985). Qian and Litcher (2001) argue that social distance between the minority and majority society reduces the likelihood of interracial marriages because racial and ethnic minorities are less likely to cross the racial divide.

Waters (1999) contends that many Whites perceive minority immigrants in the same negative ways as their native born counterparts, and as a result these immigrants are less likely to enter romantic relationships or marriages with native born whites. As continuing waves of immigrants continue to grow there will be less intermarriage with

Whites and a higher level of intermarriage between immigrants. I expect that race matters relative to out-marriage/partnership; racial and ethnic minorities such as

Caribbean Blacks and Cuban Hispanics will be less likely to out marry compared to

Cuban Whites because the sample in the study is representative of a larger number of

Caribbean Blacks and Hispanics compared to Whites.

Alba and Nee (2003) note that both race and ethnicity are socially constructed and can be crossed, blurred or shifted over time and across generations. These authors define

96 the concept of boundary crossing as the movement from one group to another without any changes to the boundary. When blurring of these boundaries occur, there is an increasing number of mixed racial and ethnic populations and these categories can no longer be easily differentiated. These shifts in boundaries are said to have occurred when

‘outsiders’ become transformed into or accepted as ‘insiders.’

With regard to social class, Alba and Nee (2003) also state that it may not necessarily be the intention of racial and ethnic minority groups to intentionally assimilate; however in order to achieve successful adaption (the desire to achieve higher levels of education may put these racial and ethnic minorities in contact with the majority) certain decisions can in turn give rise to changes in behaviors that in turn lead to assimilation. Other factors that may change the dynamics of marriage markets, depress intermarriage rates with Whites, and consequently reduce the process of marital assimilation are the recent changes in the educational composition of some racial and ethnic groups. This has taken place due to the influx of poorly educated new immigrants into the dominant culture. Additionally, more highly educated immigrants are likely to cross the racial divide and enter into interracial marriages. I expect that class should have an impact on out-marriage and partnership; those belonging to higher socioeconomic groups should be more likely to out-marry/partner because they are in greater contact with those of similar socioeconomic backgrounds. These contacts are formed thorough membership in organizations, the workplace, school, cliques and other social clubs.

4. Educational Attainment

The educational assimilation model examines the academic achievement of the respondent in young adulthood. The study analyzes the effects of race, class and gender on the academic achievement of second-generation Caribbean immigrants.

Researchers find that for gender there should be no difference between males and females in educational attainment. As previously discussed, the gender gap convergence hypothesis (Veenman and Heij, 2008) states that the immigration of women from lesser developed to developed countries essentially allows immigrant women to improve their economic chances and catch up to their male counterparts educationally. Migration is also said to create an equalizing effect between males and females relative to educational attainment.

With regard to race and ethnicity, there is significant educational performance variation among Hispanic, Asian and Black youth. Kao (2004) examines whether this variation is as a result of ethnic background, immigrant status or the experience of being an immigrant of color. Kao and Tienda (1995) argue that both foreign born and native- born children of immigrant parents were relatively more successful academically than their same-race native-born counterparts who had native-born parents. Kao (2004) examined what is it about immigrant families that could account for their children’s superior performance and he also questioned whether the association between parenting styles and educational ambitions differ by ethnicity.

The story of immigrant academic achievement is often predicated on the selection of immigrants with specific cultural values and beliefs that enhance or detract from their achievement in the United States (Zhou, 1997; Goyette and Xie, 1999; Kim, 2002).

Mexican, Nicaraguan, Haitian and Cambodian youth have all been found to exhibit lower academic performance than U.S. born peers, while Vietnamese and Chinese origin youth have been cited as surpassing their third or higher generation peers (Portes and Rumbaut,

2001; Suarez-Orzoco, 2001; Kim, 2002).

Researchers find that for social class, parents’ academic aspirations and value for education directly relates to higher educational achievement. This was the case even without hands on involvement with homework assignments, and other school related activities. For many Caribbean families, the importance of education and value of educational attainment is key to the success of their children; immigrant parents are also strongly in favor of educational activities and are in constant contact with children’s educational progress and achievement. Education represents a key component of culture in childrearing for many immigrants and children are supported and encouraged in all of their endeavors by their parents (Roopnarine et al., 2006).

Researchers have identified a number of family factors in immigrant children’s educational attainment: socioeconomic status, parental language, length of residence in the US, parental expectations, family structure, sibship size, and parental support and involvement all affect educational attainment (Suarez-Orozco, 1989; Kao and Tienda,

1995; Rumbaut and Cornelius, 1995; Fuligini, 1997; Portes and Rumbaut, 2001; Glick and White, 2003; Kao, 2004).

Other researchers argued that school context is a source of inequality between immigrants groups or between immigrant and native students (Portes and MacLeod,

1996; Portes and Hao, 2004). More recent research on the schooling of immigrant children has turned to the influence of social context of the neighborhood (Sampson,

Squires and Zhou, 2000). More surprisingly in the literature is that some groups of immigrant youth outperform others even when these youth are from equally disadvantaged immigrant communities and attend the same disadvantaged schools (Zhou and Logan, 2003).

The literature suggests that family’s social class directly influences the outcomes of immigrant children. Parents belonging to lower social classes are more likely to position their children into sectors of the educational system that have access to fewer resources or that are to some degree problematic (Crosnoe, 2005). Parents of lower social classes are also more likely to rear immigrant children who are less prepared and less equipped to compete with those of higher socioeconomic backgrounds (Crosnoe, 2007).

Parents’ access to economic resources strongly influences the level of stress, isolation and environmental disadvantages for the child, a precursor for effective learning. A lack of economic resources is also associated with increased risk and a limitation to learning

(Magnuson et al., 2006). In addition to social class, limited English proficiency and parental educational attainment also affect children’s outcomes; this results because the

100 inability to speak English prohibits parents from helping their children with school assignments and homework (Aspiazu et al., 1998). I expect that class matters in educational achievement and subsequent assimilation, parents with higher socioeconomic status and greater economic resources are more likely to raise children who achieve higher levels of education compared to parents’ with lower socioeconomic status.

Education among immigrant children remains an important thread in the assimilation theory. The educational achievements of the second-generation are often seen as a precursor to assimilation. The Foundation for Child Development (2007) states that one out of every four children who are under the age of eight years old lives in a family where at least one of the parents is an immigrant. Paradoxically, approximately

93% of immigrant children are thus born in the U.S. and are naturalized citizens. Pre- immigration status is of considerable importance when looking at the types of mobility afforded by children of immigrants. Pre-immigration status represents social characteristics such as class, educational levels of parents, income and the sociocultural features of reception (Confield and Arzubiago, 2004; Portes and MacLeod, 1999; Portes and Rumbaut, 2001).

HYPOTHESES

The main research question investigates the degree to which race, class and gender impact the assimilation of second-generation Caribbean immigrants. Each of the four measures of assimilation proposed in the research examines different types of

101 integration into the dominant culture. Because the outcome variables represent four distinct proxies of assimilation, I have grouped the hypotheses for each dependent variable based on the three central components to the study-race, class and gender. I expect that race, class and gender should affect each outcome differently. Across the board I expect race will affect political participation, out-marriage, educational attainment and to a lesser extent language proficiency. I expect that class will be a salient factor for all four measures of assimilation. Gender will affect both language proficiency and out-marriage, however gender should not be significantly related to educational attainment and political participation due to convergence theories aformentioned. The following hypotheses are organized based on the four proxies of assimilation.

1. Language Proficiency

Hypothesis One: Race and Ethnicity

Second-generation immigrants are either born in the U.S. or brought to the country at an early age. According to the three-generational language model, the first- generation will speak the mother tongue, the second generation will be mostly bilingual and the third-generation will speak the language of the dominant culture. I hypothesize that there will no differences among the racial and ethnic groups proposed in the research because second-generation immigrants are more likely to be bilingual.

Hypothesis Two: Social Class

Parents who have higher levels of economic resources and higher levels of educational attainment are more likely to rear children who are proficient in English. I

102 hypothesize that in addition to parent’s socioeconomic status, family’s social class represented in the study by homeownership, mother’s regular employment, family’s economic situation and family structure should have positive influences on children’s language proficiency.

Hypothesis Three: Gender

The research shows that gender differences in childrearing and disparities in educational performance in Caribbean children will consequently lead to gender differences in language proficiency. As previously noted, researchers state that girls tend to have superior academic performance and stricter expectations placed on them in childhood compared to boys. I hypothesize that that girls will be more proficient in

English compared to boys.

2. Political Participation

Hypothesis One: Race and Ethnicity

National origin affects political engagement because some immigrants are faced with split loyalties, they have strong ties to their country of origin, and many have intentions to return home. I hypothesize that those racial and ethnic groups with weaker ties to the country of origin should be more politically active compared to those groups with stronger ties to their country of origin.

Hypothesis Two: Social Class

As previously noted in the literature, immigrants with higher levels of education and higher socioeconomic status are more likely to be politically involved. I hypothesize

103 that family’s social class (parent’s socioeconomic status, family’s social class represented in the study by homeownership, mother’s regular employment, family’s economic situation and family structure) and the respondents’ own level of education will all have positive effects on political participation.

Hypothesis Three: Gender

According to the literature, there has been some variation in voting patterns between men and women. However, more contemporary research finds that young women and men are similar in political participation, even though young men were seen to be much more attentive to the news and politics than young women. I hypothesize that there should be no differences in political participation in young adulthood among immigrants.

3. Out-Marriage/Partnership

Hypothesis One: Race and Ethnicity

As racial and ethnic minorities grow in population, it decreases the likelihood of interaction and marriage with the majority group. This occurs because racial and ethnic minorities are more likely to interact and intermarry each other if their group is of a substantial size. With regard to group size, which in used in the literature as predictor of out-marriage/partnership and subsequent assimilation, I hypothesize that racial and ethnic minorities such as Caribbean Blacks and Cuban Hispanics will be less likely to out- marry/partner compared to Cuban Whites.

104

Hypothesis Two: Social Class

Immigrants belonging to higher socioeconomic classes will be more likely to out- marry because they are in greater contact with those of similar socioeconomic status groups through membership in various cliques and organizations. I hypothesize that those respondents whose families are of a higher socioeconomic status should be more likely to out-marry/partnership compared to those whose families are of lower socioeconomic status. I also hypothesize that additional measures of family’s social class such as family structure, family’s economic situation, homeownership and mother’s regular employment will all positively influence out-marriage/partnership.

Hypothesis Three: Gender

Due to the transformation of women’s roles by their entry into the labor market, women are now choosing partners that are their equals. Women are also choosing to marry/partner based on their potential spouse’s social class. The literature suggests that gender is not neutral in out-marriage because women are now gaining financial independence and becoming less co-dependent on their mates. I hypothesize that women should be more likely to out-marry/partner compared to men.

4. Educational Attainment

Hypothesis One: Race and Ethnicity

More positive examples of Black achievement and less rigid social classification of racial groups in the Caribbean region have led to a stronger orientation toward socioeconomic achievement among Caribbean Blacks. This argument is more applicable

105 to the British Caribbean where Blacks form the majority than it is to the Hispanic

Caribbean where they generally constitute a minority. Success among Black Caribbeans can also be attributed to the fact that Afro-Caribbeans are economically driven to migrate compared to Cubans who are politically driven. As such, I hypothesize that Black

Caribbeans should be more likely to achieve higher levels of educational attainment, compared to Cubans and mixed Caribbeans.

Hypothesis Two: Social Class

The role model hypotheses states that parents with higher socioeconomic status and economic resources are more likely to raise children who achieve higher levels of education compared to parents’ with lower socioeconomic status. I hypothesize that parents’ socioeconomic status should positively influence the educational attainment of second-generation Caribbean immigrants. I also hypothesize that additional measures of family’s social class such as family structure, family’s economic situation, homeownership and mother’s regular employment will all positively influence educational attainment.

Hypothesis Three: Gender

The gender gap convergence hypothesis states that the immigration of women from lesser developed to developed countries essentially allows immigrant women to improve their economic chances. This in turn equalizes the effects of male and female educational attainment. I hypothesize second-generation Caribbean women will perform

106 equally to Caribbean men; thus there will be no differences in gender for the educational attainment assimilation model.

Hypotheses For Other Variables

I also include other social factors such as citizenship, homeownership, current economic situation, family structure, mother’s regular employment, discrimination and aspirations. The immigration literature sees these variables as also influencing assimilation and they are also connected to my key variables of focus. As previously discussed, homeownership, family’s economic situation, family structure and mother’s regular employment are used in the study alongside parent’s socioeconomic status as additional measures of family’s social class. Even though the literature cites parents’ socioeconomic status as the primary predictor of second-generation immigrants’ assimilation, my contribution to the literature goes beyond the existing body of work done in this field and further examines additional measures of social class that may potentially be related to assimilation. Citizenship status is also included in three of the four models to further demonstrate differences in behaviors between those immigrants that are U.S. born citizens and those immigrants that are naturalized citizens and non- citizens. I expect that immigrants patterns of assimilation should vary based on citizenship status.

There has been an ongoing debate in the assimilation literature which questions the extent to which class remains a significant factor as the new wave of immigrants continue to position themselves into the host society. Many contemporary researchers

107 argue that the new wave of immigrants is now met with considerable discrimination and as a result, this weakens the significance of family’s social class. I include discrimination in the study to analyze its influence on assimilation patterns and further examine the degree to which family’s social class continues to be salient while accounting for this factor. I argue that family’s social class still matters in assimilation, and continues to be significant even when discrimination is accounted for. Lastly, I include aspirations as a predictor of assimilation since the literature suggests that both parental and children educational aspirations have positive influences on integration into the dominant culture.

The proposed research analyzes the extent to which children’s aspirations influence the aforementioned measures of assimilation.

Portes et al. (2005) argue that children of Asian, Black, mulatto and mestizo immigrants cannot escape their ethnicity and race as defined by the mainstream, because the strong effects of discrimination throws a barrier in the path of occupational and social acceptance. I hypothesize that exposure to discrimination should lead to lower levels of language proficiency, political participation, out-marriage and educational attainment.

Portes and Rumbaut (1996) suggest that immigrants’ ties to their countries of origin are likely to affect civic engagement. The authors further state that citizenship status will influence political participation. I hypothesize that U.S born citizens should be more likely to attain higher levels of assimilation compared to naturalized citizens and non-citizens because they have lived in the dominant culture for a longer period and should therefore have weaker ties to their country of origin.

108

As noted previously, both parental and children’s academic aspirations are positively related to the immigrants’ subsequent educational outcomes. I hypothesize that immigrants with higher academic aspirations in wave one of the study should be more likely to attain higher levels of assimilation in subsequent years both in education and civic involvement.

109

CHAPTER FOUR

METHODOLOGY

Introduction

This research focuses on the ways in which race, class and gender affect the assimilation patterns of second-generation Caribbean immigrants into the dominant culture. The four measures of assimilation chosen for the study are political participation as an indicator of civic involvement, English language proficiency, educational achievement, and out-marriage/partnership. These measures are representative of the seven measures of assimilation Gordon (1964) outlined in his linear assimilation theory.

Drawing from Portes and Zhou (1996) assimilation theory, these measures are all indicators of upward assimilation into the dominant culture. I will implement a quantitative methodological approach, using logistic regression analysis to determine the extent to which the immigrants’ civic involvement, language proficiency, educational achievement and out-marriage/partnership are all influenced by the key variables of interest to me, namely race, nationality, class and gender.

110

This chapter provides the research questions that are central to the study, a description of the data and the selected sample of the population chosen for the study as well as the limitations to the study. An explanation, operationalization and a brief review of the literature on the four measures of assimilation are given. The chapter also includes a justification of the research technique used- logistic regression analysis and I provide the limitations to this quantitative technique as it relates to the study. A detailed account of the independent, dependent and control variables are also presented. I indicate the relevant recoding of the variables, the construction of new variables and the use of confirmatory factor analysis and alpha reliability to test for high levels of correlation among those variables that were indexed.

DESCRIPTION OF THE DATA

The Children of Immigrants Longitudinal Study was conducted in the cities of

Miami, Fort Lauderdale and San Diego between the years of 1996 and 2001. The survey collects information from second-generation children immigrants; the second generation refers to those children who are either born in the United States with at least one immigrant parent, or those born in another country but brought to the U.S. at a very early age. The study was conducted by Alejandro Portes from Princeton University and Ruben

Rumbaut from University of California, Irvine and funded by the Russell Sage

Foundation, the Andrew W. Mellon Foundation, the Spencer Foundation and the National

Science Foundation. The mode of data collection was both face-to-face interviews and written interviews.

111

The primary purpose of this research was designed to study the adaptation process of second-generation immigrants at three periods in time. The first wave of the study was conducted in junior high school, when the respondents were at an average age fourteen, the second wave of the study was just prior to high school graduation (or dropping out of school), at an average age of seventeen, and the third wave of the study was at the beginning of respondents’ work careers (or continuing schooling), at an average age twenty-four. The original survey was constructed in 1992, and gathered baseline information on the immigrants’ families, children’s demographic characteristics, language use, self-identities and educational attainment. Second-generation children immigrants attending public and private schools in the eighth and ninth grades in metropolitan areas of Miami, Fort Lauderdale and San Diego were originally interviewed. The original sample size consisted of 5,262 respondents representative of 77 different nationalities.

Three years later, corresponding to the time in which these students would graduate from high school, the first follow up interview was conducted with the second- generation immigrant children. The focus of this first follow up interview was to document the evolution of key adaptation outcomes over time. More specifically, the survey sought to establish the number of students who dropped out of high school, prior to their graduation dates. Additionally, the investigators were interested in finding out the ways in which adaptation such as language knowledge and preference, ethnic identity, self-esteem, and academic attainment transformed over the adolescent years. The first

112 follow up interview was successful in obtaining 81.5% of the original sample, which was equivalent to 4,288 of the original respondents. A parental interview was also conducted alongside the first follow up interview. A total of 2,442, parents were interviewed, this represented 46% of the original student sample. The purpose of interviewing the parents of the immigrants was to ascertain the parents’ and families’ direct characteristics as well as their aspirations and future plans for their children.

The final wave of the study was conducted between 2001 and 2003, a decade after the original survey. The purpose of this survey was to readily assess patterns of adaptation in early adulthood. At this point, the respondents averaged approximately 24 years of age. Since the respondents had already left school by this time, surveys were now conducted at the workplace as well as at their places of residence. Respondents were now scattered throughout 30 different states, and some surveys were even returned from military bases. For the third interview, mailed questionnaires were the principle source of completed data. Information from 3,163 respondents was retrieved, which represented

68.9 percent of the original sample and 84.3 percent of the first follow-up interview. Each sample wave retrieved approximately 85 percent of the preceding one.

There were a number of adaption outcomes measured during this third and final interview. At an average age of 24 years old for the respondents, the investigators were interested in documenting the young adults’ educational attainment, employment and occupational status, income, civil status and ethnicity of spouses/partners, political

113 attitudes and participation, ethnic and racial identities, delinquency and incarceration, attitudes and levels of identification with American society, and plans for the future.

SUB-SAMPLE POPULATION OF CARIBBEAN IMMIGRANTS

For purposes of the research, I selected five of the 77 nationalities represented in the original data set; this narrows my focus to second-generation Caribbean immigrants from the countries of Cuba, Haiti, Jamaica, Dominican Republic and the West Indies. My motivation for choosing these countries is primarily due to the countries’ distinctive racial and ethnic composition, diverse languages, and disparity in wealth. Moreover, as discussed earlier, this region has been understudied, and previous researchers in the field tend to focus primarily on the Latin American countries in the Caribbean as opposed to the English or French speaking islands.

My contribution to the literature is three-fold; first I examine the extent to which race, nationality, class and gender remain significant even among a geographically and culturally homogeneous group of Caribbean immigrants. Historically researchers have found that race, class and gender affect the ways in which immigrants position themselves into the dominant culture. As previously noted, there has been a lack of attention paid to Afro-Caribbean immigrants. My research is unique because I compare the assimilation patterns of Cuban immigrants to Afro-Caribbeans and mixed Caribbean immigrants arriving from five different countries. Second, I conduct a sensitivity analysis to determine the differences in assimilation using the respondents’ nationality. This comparative analysis allows me to determine respondents’ assimilation based on both

114 nationality and racial and ethnic self-identification. The third contribution of the research is the examination of the parent-child relationship over time. Because investigators interviewed both parents and children, I am able to analyze the extent to which family’s social class subsequently affects the assimilation patterns of their children. I rely mainly on the first survey where baseline information was collected from the parents and the children for my independent variables and one of my outcome variables (language proficiency). I also rely on the final survey, which was conducted when the respondents were at an average age of 24 years old for three of my outcome variables (political participation, out-marriage/partnership and educational attainment) and additional independent variables. Using data from both surveys allows me to illustrate the degree to which events occurring in previous time periods subsequently affect outcomes in successive time periods.

Even though the CILS dataset appears to be a longitudinal study, this is not entirely the case, since the same questions were not asked over the three different time periods. Longitudinal studies usually involve repeated observations of the same variables over time. The investigators do track the same respondents over time; however there were many new questions added to the surveys and very few questions were repeated in all three-time periods. This made it difficult to accurately observe changes over time. To counteract this issue, I chose to conduct my analysis using cross-sections of the data from each of the time periods. I was then able to select the necessary variables for my study that provided information on the adaptation of the respondents, at the average ages of

115 fourteen and twenty-four years. The following gives a brief overview of the four measures of assimilation used in the study.

LIMITATIONS OF THE DATA

I have identified a few limitations to the study based on the data set. When conducting quantitative research using secondary data, the researcher is usually somewhat limited in terms of what can be performed because the data was originally collected for a different purpose and this in turn limits the options available to the researcher. Consequently, the dataset has limited the study with regard to the site of data collection. Interviews were primarily conducted in Miami, Fort Lauderdale and San

Diego. The population sample chosen for the study was mainly representative of those second-generation Caribbean immigrants living in Florida. Due to the location of the interview, the sample size for each country chosen was not evenly represented; Cubans and Dominicans for example, far out numbered West Indians, Jamaicans and Haitians. A more even distribution of the population for each nationality would have contributed more richly to the study; this could have been obtained if respondents were interviewed from across the country as opposed to just two states. Even though the analysis appears to focus primarily on Cuban immigrants there is a robust representation of Afro-Caribbeans, this in turn adds to the depth of the study especially when results for these smaller groups are significantly different from Cubans. Another point to note is that since Cuban immigrants have been widely studied in the assimilation literature, using Cubans as a

116 reference group in the study provides a useful comparison and further illustrates how patterns of assimilation varies throughout the Caribbean among different groups.

The second limitation to the dataset is that this group of immigrants is representative of a higher socioeconomic status; as a result it was difficult for me to measure the impact of poverty and subsequent inequality between groups on the process of assimilation. A clear indication of this is that very few families received any type of public assistance. To counteract this shortcoming, I included parents’ socioeconomic status into the study as a measure of social class, however even these characteristics signify that these families are generally well off.

Another limitation but also an advantage to the study is the age of the respondents. While on one hand the young age cohort of the respondent provides insight into a particular segment of the population and the ways in which their assimilation patterns are unique, one limitation is that overtime the attitudes, behaviors and beliefs may alter which may paint a new picture. One important point to note is that these assimilation patterns are representative of second-generation immigrants at specific ages in their lives and that these patterns are likely to be changed over time based on experiences and changes in general lifestyles.

PROBABILITY MODELING

I have chosen to use logistic regression for my analysis because all four of my dependent variables are dichotomous indicating the likelihood of assimilation into the dominant culture. Additionally logistic regression modeling accounts for the independent

117 variables in the model, which are both categorical and discrete. The benefit of logistic models in comparison to other types of modeling is that it allows the researcher to make use of predictor variables that are either continuous or categorical (Lemeshow and

Hosmer, 2000). The use of this type of modeling is beneficial to the research because it predicted the probability that the process of assimilation was a success or not through the proxies that were proposed. I use logistic regression as opposed to its competitor probit modeling because many studies conducted in the assimilation literature commonly use logistic regression as a methodological technique.

Logistic models are utilized for numerous reasons. One of the main reasons why researchers utilize this type of modeling is because the dichotomous nature of the dependent variable violates the assumption of ordinary least squares. Another issue when using dichotomous data is that all normality tests are invalid because the error terms are not normally distributed. Unlike ordinary sum of square regression, logistic modeling does not restrict the researcher; both homogeneity of the variance and a normal distribution of the dependent data are not necessary (Menard, 2002). The underlying principles of logistic models are simply based on the nature of the log-curve and probabilities; hence the research must only assume that the logistic transformation is linear, the dependent variable is discrete, and there are no outliers included in the logarithmic curve. Many researchers have recommended the use of logistic regression, especially when normality assumptions are violated (Press and Wilson, 1978).

118

There are some limitations to logistic regression, which the research has to take into account. For instance, sample size can become very problematic when using logistic regression and in order to achieve stable and robust results the sample size must not be too small; small sample sizes can therefore lead to inaccurate estimates of the parameters.

Thus, there must be an adequate number of data points per predictor in order to produce meaningful results. Nonetheless, logistic regression continues to be widely used by researchers due to its practicality, applicability to real world situations and intuitive assumptions (Lemeshow and Hosmer, 2000).

I use odds ratio to interpret the regressions; odds ratio refers to the ratio of the odds divided by 1 minus the odds. Logit refers to the natural log of an odds ratio; often called log odds even though it really is a log odds ratio. The logit scale is linear and functions much like a z-score scale (Kohler and Kreuter, 2009). The logistic function shows that the change in probability is not constant (linear) with constant changes in X.

This means that the probability of a success (Y = 1) given the predictor variable (X) is a non-linear function, specifically a logistic function.

) eu Y i = u uABXBXL BX 1+ e =+11 + 2 2 ++KK

The above equation shows the logistic function where Y-hat is the estimated probability that the ith case is in a category and u is the regular linear regression equation

The logistic function is written as follows:

ebbX011+ πˆi = bbX 1+ e 011+

119

By algebraic manipulation, the logistic regression equation can be written in terms of an odds ratio for success:

⎡⎤PY(1|)= X ⎡⎤πˆ i ==+exp(bbX ) ⎢⎥(1PY ( 1 | X ))⎢⎥ (1ˆ ) 011i ⎣⎦−=i ⎣⎦−π

Exponentiated (logit) OR = odds ratio

Probability = odd ratio / (1 + odd ratio)

In order to convert logistic regression to log odds we take the natural log of both sides. It is also necessary to convert the logistic regression back to log odds in the study in order to calculate the z scores and subsequently test the validity of the hypotheses. The following equation can then be written in terms of logits (log-odds) for a single predictor:

⎡⎤⎡⎤PY(1|)= X πˆ ln==+ ln bbX ⎢⎥⎢⎥ˆ 011 ⎣⎦⎣⎦(1−PY (= 1 | X )) (1−π )

A BRIEF OVERVIEW OF THE FOUR ASSIMILATION MODELS USED IN

THE RESEARCH

As previously discussed, since assimilation is not readily observable or quantifiable, assimilation theorists have developed proxies of adaptation. Gordon (1964,

1975) has constructed what has been said to be the most systematic paradigm for understanding and measuring the components of assimilation. According to Gordon,

(1964:71) there are seven dimensions of assimilation; he refers to these as cultural, structural, marital, identificational, attitude receptional, behavior receptional and civic.

Each of Gordon’s seven stages of assimilation is characterized into a sub process or

120 condition. Chapter three gives an in depth review of the seven stages that include structural, marital, identificational, receptional, behavioral receptional and civic assimilation. Since not all stages of assimilation aforementioned are easily calculable, sociologists and other social scientists who are concerned with the phenomenon usually choose those variables which are most easily measurable such as cultural, structural, marital and identificational assimilation. For the proposed research, I have quantified for measures of assimilation: language proficiency, political participation, out- marriage/partnership and educational attainment. The following sections give a brief review of these measures and their operationalization. Chapter three also provides further analysis of these measures relative to the process assimilation.

Language Proficiency

In assimilation theory, language proficiency is a measure of how well the immigrant is able to speak English; it also looks at the individuals’ loss of the mother tongue. The three-generation model of language assimilation indicates that subsequent generations of immigrants eventually lose the ability to speak the mother tongue. Waters

(2005) postulates that within this three generation model of language assimilation the first generation makes some progress in language assimilation but remains dominant in their native tongue, the second generation is bilingual, and the third-generation speaks only

English. Alba (2004) also argues that bilingualism is common between second-generation immigrants; this refers to those children who are growing up in immigrant households.

These children are likely to speak an immigrant language at home, however almost all of

121 them are proficient in English. According to the U.S. 2000 census, among Hispanics,

92% speak English well or very well, even though 85% of them speak at least some of their mother tongue at home. Alba (2000) also argues that within the third generation or later, children speak only English at home, making it highly unlikely that they will be bilingual as adults. He contends that bilingualism is more common today than in the past, and especially for Latin Americans most children to some extent speak the mother tongue at home, but if they are born in the US; they speak English well or very well.

The proposed language proficiency assimilation model analyzes the impact of race, nationality, class and gender on the English proficiency of second-generation

Caribbean immigrants. I examine the degree to which these factors influence the trajectory of assimilation into the dominant culture. The respondent’s proficiency in

English is a determinant of the level of upward assimilation into the dominant culture, because it is an indication that the respondent is giving up his or her origins to become more like the dominant society.

Political Participation

Other than marriage, citizenship is one of the most significant factors in assimilation. Civic involvement has been used for a very long time as a measure of assimilation. An immigrants’ home country affects the acquisition of citizenship and willingness to vote in the United States (Bueker, 2005; Portes and Rumbaut, 2006).

Moreover, the country of origin affects the propensity of political incorporation, all other individual socioeconomic factors taken into account. Bueker (2005) tests five hypotheses

122 concerning voting and naturalization patterns. The first hypothesis is reversibility, which implies that difficulties associated with return migration would encourage higher rates of naturalization among certain immigrant groups. The second is translation; this states that political practices at the origin have an impact on political behavior at destination.

Thirdly mobilization argues that certain groups will exercise their political rights more fully in the wake of highly developed ethnic communities. The fourth and fifth hypotheses are assimilation and gender diversity.

Wolfinger and Rosenstone (1980) argue that the two most important predictors of participation are formal educational attainment and family income; a higher level of education and more substantial economic resources increases participation in political involvement. The political participation assimilation model in this study will test the impact of race, nationality, class and gender on the respondent’s political participation and subsequent assimilation, based on a number of variables that also affect civic involvement.

Out-Marriage/Partnership

Out-marriage is another very strong predictor of an immigrant’s assimilation into the dominant culture. Gordon (1964) argues that the increasing contact between groups brought about by structural assimilation would lead naturally to other forms of assimilation, particularly intermarriage. Widespread intermarriage, in turn, would gradually erase the social boundaries that had previously separated the immigrant group from the host society. In a way, then, structural assimilation was the lynchpin of the

123 assimilation process. Immigrant out-marriage with other native-born ethnic groups or with Whites represents a potentially useful but understudied indicator of social assimilation. Indeed, out-group marriage rates imply structural assimilation on other more commonly studied traits (e.g., ethnic group segregation or socioeconomic status)

(Anderson and Saenz, 1994; Fuand Heaton, 2000; Liang and Ito, 1999).

As such, intermarriage provides a measure of "social distance" between groups and of the strength of ethnic group boundaries or solidarity (Bogardus, 1968).

Intermarriage among different national-origin groups and generations helped foster a sense of "Whiteness" as a distinctive ethnic or cultural group, and it hastened the process of Americanization (Houtand Goldstein, 1994; Pagnini and Morgan, 1990). During the past century, out-group marriage among White ethnics tended to increase with length of residence in the United States and over successive generations (Alba, 1990; Gurakand

Fitzpatrick, 1982). Assimilation into American society accelerates with intermarriage and vice versa. Pagnini (1990) also argues that intermarriage reduces the ability of families to pass on to their children a consistent ethnic culture, and thus is an agent of assimilation.

The marital assimilation model will test the assimilation of the respondent based on the racial and ethnic self-identification of the spouse or partner. Respondents who are more likely to marry outside of their race/ethnicity are said to achieve a higher level of assimilation than those who do not. I test the impact of race, class and gender on out- marriage of second-generation Caribbean immigrants and their subsequent assimilation into the dominant culture.

124

Educational Attainment

In sociology the socioeconomic status of an individual is measured by income/wealth, occupation and education. Socioeconomic status as a measurement of assimilation indicates if the immigrant is able to catch up to the native born in terms of human capital characteristics. Many researchers argue that the parents of children immigrants tend to play a crucial role in the success stories of their children and that it is the parents who contribute significantly to the academic achievement or the failure of their immigrant offspring. Hirschman (2001) argues that familial and socio economic characteristics influence immigrant teenagers’ educational enrollment. Accordingly, poorly educated parents can therefore explain a significant share of some immigrant groups’ high enrollment rates, the absence of parents and inner city residence. Haveman,

Wolfe and Spaulding (1991) argue that parents with higher levels of education are hypothesized to have a positive influence on the educational outcomes of their children because of the expectation that such parents will be seen as good role models by their children. Respondents who achieve a high level of education are said to assimilate upwardly into middle class American society. I expect that the respondents’ educational attainment at age 24 will be influenced by the respondents’ gender, race/ethnicity, nationality and parents’ socioeconomic status.

125

MEASURES AND THEIR OPERATIONALIZATIONS

Dependents Variables

1. Language Proficiency

Language proficiency is used in the model both as an independent and a dependent variable. English language proficiency represents one of the four key measures of assimilation, and it is also used as a predictor of political participation, educational attainment and out-marriage/partnership. Language proficiency is measured in time- period one, when the respondents were at an average age of 14 years old. The language proficiency variables from the CILS sample fall into four categories (how well does the respondent speak English, read English, write English and understand English?). These responses were all coded on a scale of one to four (1=very little, 2=not well, 3=well,

4=very well). The variables were each transformed into binary code where 1=excellent

English and 2/4=non-excellent English proficiency. Based on confirmatory factor analysis (CFA) an additive index of language proficiency (alpha reliability= .82) was created. I used CFA because I believed theoretically, how well the respondent spoke, read, write and understood English represented several dimensions of the same concept.

A final recoding of the variable was done where respondents were divided into two categories where 1=respondents who were excellent on all four categories, and

0=non-excellent language proficiency. I chose to recode language proficiency into a dichotomous variable because most respondents were proficient in English. Recoding the variable to being proficient on all four levels of English compared to less than four levels

126 subsequently produced more meaningful results. Table 4.1, provides a descriptive summary of those respondents who are proficient in all four levels of assimilation compared to those who were proficient on less than four levels.

English Proficiency Frequency Total

Proficient on all four levels 1342 75.52 Proficient on less than four 435 24.48 Total 1777 100.00

Table 4.1: English Proficiency

2. Political Participation

Political participation is measured in wave three of the study. I employ variables from wave one, which are likely to influence subsequent political participation in wave three. Those respondents who are not U.S. citizens are eliminated from this particular assimilation model. I also compare the political participation of U.S. born citizens to naturalized citizens. The independent variables associated with political participation in the study are gender, race, parent’s socioeconomic status, respondents’ own education, family’s economic situation, homeownership status, discrimination, whether respondent lives with mother, mother’s regular employment, family structure, English proficiency, respondents’ aspirations and citizenship status.

The political participation variable measures if the respondent is registered to vote or not. The survey question asks, “Are you currently registered to vote?” Those

127 respondents who are not registered to vote are represented by the value zero and those who are currently registered to vote are represented by the value one. The variable is used in the model as a dependent variable because it represents one of the proxies of assimilation; thus indicating the respondents’ civic involvement within the dominant culture. Because only citizens can vote in the United States, those respondents who are not citizens were dropped from the model. Table 4.2 provides a descriptive summary for those respondents who are registered to vote compared to those who are not.

Registration To Vote Frequency Percent Yes 823 77.94 No 233 22.06 Total 1056 100.00

Table 4.2: Registration to Vote

3. Out-Marriage/Partnership

Out-marriage/partnership is surveyed using wave three of the study. The sample size for this model is considerably reduced compared to the other three assimilation models because of the respondents’ age. At an average age of twenty-four many respondents are unmarried/partnered, accordingly, those respondents who are single are eliminated from the study. The independent variables associated with out- marriage/partnership in the study are gender, race, parent’s socioeconomic status, respondents’ own education, family’s economic situation, homeownership status,

128 discrimination, whether respondent lives with mother, mother’s regular employment, family structure, English proficiency, respondents’ aspirations and citizenship status. I expect those immigrants who marry or partner outside of their own race to be more assimilated than those who marry or partner within their own race (Gordon, 1964; Qian and Litcher, 2001; Sassler, 2005).

This variable defines the ethnicity of the respondents’ spouse or partner and is used as a predictor of out-marriage/partnership The CILS survey question asks, “What is the ethnicity of your spouse or partner?” In order to create out-marriage/partnership categories, I first grouped the respondents’ spouse/partner based on the racial self- identification into five different categories: (1) White (those who self identified as

American White/ White American, Filipino-White, European nationality, European

American) (2) Hispanic/Latino (those who self identified as Hispanic, Hispanic

American, Latin American/Latino Americano, Mexican-American and other hyphenated

Latin American identity), (3) Black/West Indian(African-American, Black/Black

American, Haitian, West Indian), (4) partnered with other (partnered with any other race) and (5) not partnered (assumed to be not partnered if response not given in wave three)

These five race categories for the respondent’s spouse/partner were further collapsed into three categories: 1=Not-married/partnered, 2=in-marriage/partner and

3=out-marriage/partner. The respondent’s own race and ethnicity was used as an indicator of out/in partnership. In-marriage/partnership occurred when respondents in one group partnered with someone in the same group and out marriage/partnership was

129 represented by the mixing of groups. Table 4.3 presents the distribution of out- marriage/partnership and in-marriage/partnership for each racial and ethnic group in the study. Only three racial and ethnic categories are used in this model to determine out- marriage/partnership because ‘other’ and ‘mixed’ categories did not provide sufficient information for classification.

Not % In- % Out- % Race/Ethnicity married married married Total Cuban White 126 47 21 8 119 45 266 Cuban Hispanic/Latino 127 43 129 43 41 14 297 Cuban other 72 56 0 0 56 44 128 Caribbean Black 87 56 53 34 16 10 156 Caribbean Mixed 54 82 0 0 12 18 66 Total 466 - 203 - 244 - 91

Table 4.3: Respondents in the study that are single, out-married/partnered and in-married/partnered.

4. Educational Attainment

Educational attainment is measured in the study by the achievement of a college degree of higher compared to less than a college degree. As previously noted, I expect differences in educational attainment based on race/ethnicity, family’s social class and gender. The independent variables associated with educational attainment are gender, race/ethnicity, parent’s socioeconomic status, family’s economic situation, homeownership status, discrimination, whether respondent lives with mother, mother’s regular employment, family structure, English proficiency, respondents’ aspirations and citizenship status.

130

A new variable with five categories was created for the education variable. I indexed these variables into one categorical variable of education because theoretically, the receipt of a GED certificate, high school diploma, associates degree, bachelor’s degree and master’s degree are all measures of educational attainment; therefore I wanted to construct a new variable that would reflect this. The CILS measured educational attainment using six main variables. The first variable was the highest number of school years completed; this was measured using 10, 12, 14, 15, 16 and 18 years. The remaining education variables measured GED or high school equivalent degree, high school diploma, associates or junior college degree, bachelor’s degree and master’s degree.

In order to create the new education variable, I collapsed all of these responses into five categories where 1=less than high school, 2= high school diploma or GED, 3= associates degree, 4= bachelor’s degree and 5= masters degree. Those with 10 or 12 years of education and no high school diploma or GED were placed into category one. Those with a GED/high school certificate and high school diploma were put into category two.

The CILS did not separate those with GEDs from those with high school diplomas. Those with a junior college/associate degree or 14 or 15 years of education were put into category three. Those with a bachelor’s degree or 16 years of education were placed into category 4, and those with a master’s degree or 18 years of education were placed into category 5.

These five categories were then created into a binary variable where categories 1,

2 and 3 represented less than a college education, and categories 4 and 5 represented a

131 college degree or higher. I chose to create a dichotomous variable for educational attainment because it is more meaningful to the study to compare those respondents with a college degree of higher to those with less that a college degree in order to illustrate patterns on assimilation. Table 4.4 provides a descriptive summary of the respondents’ educational attainment prior to the recoding of he variables.

Educational Attainment Frequency Percent Less than high school 57 5.10 High school diploma/GED 483 43.20 Associates degree 267 23.88 Bachelors degree 279 24.96 Masters degree 32 2.86 Total 1118 100.00

Table 4.4: Educational Attainment

Independent Variables

The following gives an overview, description and justification of the independent variables used in the research. I describe relevant recoding and/or changes made to the variables, and preliminary testing done prior to the estimation of the models. Percentages and standard deviations for each dependent and independent variables are presented in appendix A. Correlation matrices for all variables used in each model are also presented in appendix B.

132

Gender

Gender remains an essential component of the assimilation literature, though much of the emphasis has primarily been on the disparities between men and women’s experiences in the labor force. My aim is to extend the literature beyond its current scope and further examine the extent to which second-generation Caribbean youth are integrating differently or similarly within the dominant culture. The proposed research investigates the role of gender in the immigrants’ language proficiency using wave one of the data. The model analyzes children immigrants at an average age of fourteen and investigates the degree to which Caribbean boys and girls display variations in English proficiency. I also assess the extent to which male and female Caribbean immigrants exemplify differences in political participation, educational attainment and decisions to out-marry/partner in young adulthood. I expect that there should be differences between

Caribbean immigrant youth both in middle school and young adulthood relative to the four aforementioned measures of assimilation.

The segmented assimilation literature is primarily concerned with the fate of

Black men assimilating into the underclass. In an ethnographic study done on Dominican immigrants and second-generation youth in New York; Nancy Lopez (2003) argues that gendered differences of a similar sort remains no less conspicuous today. Gender differences are said to be greater among immigrants than the native born, since for daughters, ‘becoming American’ is said to increase the earning potential as compared to their mothers. As previously discussed, convergence hypotheses have been used to

133 explain gender-neutral effects on assimilation. The movement of immigrants from developing nations to the developed world has reduced the gap between immigrant men and women because women are now given equal opportunities to succeed professionally and educationally. Additionally, the absence of a more rigid patriarchal system in the developed world also promotes achievement and success among immigrant women.

Table 4.5 presents the distribution for gender; males are used as the reference group in the study.

Respondent's Sex Frequency Percent Male 849 47.67 Female 932 52.33 Total 1781 100.00

Table 4.5: Distribution of Males and Females in the Sample.

Race/Ethnicity and Nationality

A review of the literature indicates that there is no uniform measure of race and ethnicity in assimilation and migration studies (Kao, 2004; Pong and Hao, 2007; Ellis and

White 2006; Waters and Lieberson, 1987). The proposed research examines the impact of race and ethnicity on civic involvement, educational achievement, out- marriage/partnership and language proficiency. The CILS defines race by the racial and ethnic self-identification of the respondent. The survey provides a list of racial categories and asks the respondents to self-identify. Though the dataset oversamples Cubans, it also

134 provides a substantial representation of English speaking countries. The sample used for the study includes respondents from Cuba, Dominican Republic, Haiti, Jamaica and the

West Indies.

I used the respondent’s racial and ethnic self-identification and nationality to construct a race-nationality variable. Since the Caribbean region is ethnically and racially diverse, using just the nationality variable proved to be insufficient for the study because it did not represent the distinct races of each country. As previously discussed, I counteract this problem by conducting a sensitivity analysis, using only the respondents’ nationality in the study. The countries included in the research all represent a diversity of people who range from Black, White, Latino/ Hispanic, Asian, and many multiracial groups. In order to capture these races I constructed five race-nationality categories.

Because the sample represents a substantial number of Cubans who self identify differently as discussed in chapter two, I first classify Cubans into three different groups.

Cuban Whites represent those Cubans who self-identify as White. Cuban Hispanic/Latino are those Cubans self-identifying as Hispanic, Latino, Latino-American. Cuban Other represent those Cubans who did not self identify as either Cuban White or Cuban

Latino/Hispanic.

Secondly, I create an Afro-Caribbean category because Haiti, Jamaica and the

British West Indies are predominantly Black societies, thus the majority of the respondents self-identify as Black. Accordingly, the Caribbean Black group represents those Haitians, Jamaicans, Dominicans and West Indians who self-identify as Black.

135

Table 4.6 shows the demographics for each country used in the study; the statistics are taken from the CIA fact book. I do not include the racial demographics for West Indians in this table because as mentioned previously, West Indians do not represent one country in the study; this category is made up of several former British colonies that are also predominantly Black.

Black Mixed Other White Jamaica 91.2 6.20 2.60 . Haiti 95.00 5.00 . . Dominicans 73.00 45.00 2.00 . Cuba 10.08 23.84 1.02 65.05

Table 4.6: Demographics for the countries used in the Study. Note: Data is taken from the CIA Factbook.

Next, I create a Caribbean mixed racial classification, which represents Haitians,

Jamaicans, Dominicans and West Indians, who self-identify either as multi-racial,

Hispanic or some other race. Since Haiti, Jamaica and the West Indies are predominantly

Black countries; there is very low representation of White and Hispanic immigrants.

Immigrants who do not self-identify as Black from these countries primarily self-identify as either multi-racial or other. Dominicans are also placed into the Caribbean mixed category because not only is this group of immigrants under-represented in the data, but also, most Dominicans either self-identify as Hispanic or multiracial in the study. The

136

Caribbean mixed category is thus used in the study to represent a combination of multiracial, Hispanic and other Caribbean immigrants from four of the five nationalities.

White Cubans were used as the reference group for the race-nationality category because there are no native born Whites in the CILS dataset and thus Cuban Whites are representative of the dominant group. Cuban Whites are also used as the reference category because the segmented assimilation theory indicates that those who are visible minorities will assimilate differently than those who are not. Table 4.8 presents a distribution of the respondents based on racial and ethnic self-identification; as discussed these figures were used to create the five racial and ethnic categories presented in Table

4.7.

Respondents' Race/Ethnicity Frequency Percent Cuban White 364 26.94 Cuban Hispanic 386 28.57 Cuban Other 197 14.58 Caribbean Black 251 18.58 Caribbean Mixed 153 11.32 Total 1351 100.00

Table 4.7: Racial and Ethnic Self-Identification of Respondents

137

Dominican Race/Ethnicity Cuban Republic Haiti Jamaica West Indies Total White 364 5 0 0 4 373 Black 6 5 103 87 56 257 Asian 0 1 0 6 4 11 Multi-racial 112 15 12 14 14 167 Cuban 25 0 0 0 0 25 Cuban-American 19 0 0 0 0 19 Haitian 0 0 10 0 0 10 Haitian-American 0 0 3 0 0 3 Jamaican 0 0 0 2 0 2 Other West-Indian 0 1 0 1 1 3 Latin American nation 8 2 0 0 0 10 Latino 7 4 0 0 0 11 Hispanic 372 38 0 1 1 412 Hispanic-American 7 1 0 0 0 8 Human race 10 0 2 0 1 13 Other 17 5 1 3 1 27 Total 947 77 131 114 82 1,351

Table 4.8: Data Used to Create the Five-Proposed Racial and Ethnic Categories

As previously mentioned, national origin is used in the research alternatively to race and ethnicity in order to determine the impacts of race, class and gender on assimilation for both classifications. The respondents’ nationality is represented in the dataset as the country identified by the respondent as his or her national origin. The assimilation literature suggests that nationality affects both educational attainment and language proficiency of immigrants (Chiswick and Miller, 1992, 1995; Dustmann, 1994).

138

Table 4.9 provides the distribution of the respondents for the five nationalities selected for the research; Cubans are used as the reference group in the study.

Respondent's National Origin Frequency Percent Cuba 1226 68.84 Dominican Republic 105 5.90 Haiti 178 9.99 Jamaica 156 8.76 West Indies 116 6.51 Total 1781 100.00

Table 4.9: The Distribution of Respondents from Five Different Caribbean Islands.

Social Class

Many classical sociologists have examined the degree to which the control and ownership of the means of production is directly related to social stratification. Marx

(1848) being critical of capitalism argued that the social class system kept the masses in their place and reinforced the power and dominance of the elite. Weber (1930) extended the work of Marx by including three primary resources, which he believed shape a person’s primary position: class, status and party. Similar to Marx, Weber believed that class includes the ownership of the means of production; however he went a step further to extend this theory by adding income, wealth and skill knowledge to the equation. The research draws upon the work of classical sociologists on stratification and inequality, primarily Max Weber’s premise on social class.

139

Weber (1930) argued that skill is an important commodity in the labor market and the development of a person’s skill through going to college served to enhance a person’s class position. Weber also used the term status group to refer to people who share the same perceived level of prestige. He refers to party as the capacity to organize and accomplish some particular goal. Weber uses class, status and party to illustrate that through class, power can be achieved. Connecting Weber’s view of social class to the assimilation literature, I argue that those immigrants with higher levels of class, status and power should achieve higher levels of assimilation and integrate more quickly into the dominant culture. I choose a classic measure of social class similar to that derived by

Weber (1930) to illustrate how social class is positively related to higher levels of language proficiency, political participation, educational attainment and the decision to out-marry/partner.

Kao (2004) defines family socioeconomic status by father’s education, mother’s education, father’s occupational prestige and mother’s occupational prestige. Glick

(2007) also uses a component of parents’ socioeconomic status when comparing academic performance of young immigrant children and the significance of race and national origin. He measures parents’ SES by family income and mother’s education; in the case of missing data for the mother, father’s education is used. Similar to race, there is no uniform way the assimilation literature measures socioeconomic status. Consistent with the assimilation literature I use parents’ socioeconomic status, as a measure of social class. I also go beyond the research to include four additional measures of family’s social

140 class. I expect other measures of family’s social class such as homeownership, family’s economic situation, family structure and mother’s regular employment should positively affect the assimilation patterns of immigrants. The respondents own education is also employed in political participation and out-marriage/partnership models to illustrate the degree to which the immigrants’ own education influences his/her integration into the dominant culture.

The parent-child relationship on assimilation allows me to enter socioeconomic status variables used by Portes and Rumbaut (2001), in their conceptual framework, however these variables were not employed in analytical modeling. Similar to Portes and

Rumbaut (2001) I will examine the ways in which parents’ socioeconomic status subsequently affects the second-generation immigrant’s integration into the host society.

I chose parents’ SES as the primary measure of social class for each of the four assimilation models because this has been customary in the assimilation literature. I subsequently introduce alternative measures of social class aforementioned to examine the overall significance of family’s social class.

Preliminary statistical tests using bivariate analyses and variance inflation factor indicates moderate degrees of collinearity among mother’s education, father’s education, mother’s occupational prestige and father’s occupational prestige. Results are presented in table 4.14. I also present the correlation matrix for the various measures of family’s social class in table 4.15. The means and standard deviation for each of these variables

141 are presented in Appendix A. The following sections list the various measures of social class employed in the research.

Parents’ Socioeconomic Status

As previously discussed, embedded into parents’ socioeconomic status is occupational prestige and parents’ education. Parents’ occupational prestige is measured in wave one of the Children of Immigrants Longitudinal Survey. The investigators employ the Treiman’s standard international prestige scale to produce scores for socioeconomic status. Sociologists have used this scale by identifying prestige rankings for more than 700 occupations. The Treiman scale remains one of the most widely used and influential measures of occupational prestige in the social sciences (Boyd, 1984; Castles and Kosack, 1973; Pedraza, 1991; Phizacklea, 1983; Simon, 1992; Tyree and Donato,

1986). Parents’ socioeconomic status is used in all four models as a primary measure of social class. Parents’ socio-economic status is used in all four assimilation models.

Family’s Economic Situation

I use the family’s economic status in wave one as a measure of social class to determine the respondents’ subsequent assimilation in wave three. This measure is employed in all four assimilation models. Family’s economic status is used in the research as an indicator of financial stability and general socioeconomic status of the family. On a five-point scale the survey asked the question, “What do you think your family’s current economic status is?” (1=wealthy, 2=upper middle class, 3=lower middle class, 4=working class, 5=poor.) I recoded this variable into a binary where (1 and

142

2=upper class or higher) and (0=less than upper class). Table 4.10 provides a description of the family’s economic situation representative of time period one.

Family's Economic Situation Frequency Percent Wealthy 59 3.35 Upper class 740 42.02 Lower middle class 594 33.73 Working class 351 19.93 Poor 17 0.97 Total 1761 100.00

Table 4.10: Family’s Economic Situation

Homeownership

Homeownership is used in the study as a measure of social class. I use family’s homeownership status in wave one to determine subsequent assimilation patterns in wave three for all four models. The literature suggests that homeownership has a positive effect on integration (Espenshade and Fu, 1997). To survey the question the respondents were asked, “Do your parents (or adult guardians) own or rent the house or apartment where you now live?” The responses were coded on a five-point scale (own, rent, lives with relatives, likes with friends/non relatives and other). For purposes of the research I create a binary variable where (1=own) (0=not own). Table 4.11 provides the distribution of parents’ homeownership status.

143

Homeownership Status Frequency Percent Own 1167 66.34 Do not own 592 33.66 Total 1759 100.00

Table 4.11: Homeownership Status

Family Structure

I use family structure in all four models of assimilation in the study to illustrate the respondents’ family unit. The CILS illustrates family structure in a number of ways: presence of both parents, single parent, step-families, living arrangements with friends or living arrangements with grandparents. Segmented assimilation theory responds to unexpected trends since 1960 in family structure, socioeconomic status, and educational achievement among descendants of certain immigrant groups as predictors of assimilation (Waters, 1990; Gans, 1992; Skop, 2001). The literature also states that higher rates of marital disruptions and female-headed households lead to downward assimilation (Fernandez-Kelly and Schauffler, 1994; Landale and Ogena, 1995; Portes,

1995; Portes and Zhou, 1994; Waters, 1996).

To survey the question the respondents were asked in wave one of the interviews,

“Which of the following best describes your current living situation?” This variable was recoded into a binary where 1=mother and father and 0=any other arrangement. Family structure was used as a measure of family’s social class in the study to demonstrate the extent to which disrupted family units or intact families subsequently influence patterns

144 of assimilation. Table 4.12 illustrates those percentages of families with and without martial disruptions.

Family Structure Frequency Percent Mother and Father 967 54.66 Otherwise 802 45.34 Total 1769 100.00

Table 4.12: Family Structure

Mother’s Regular Employment

Mother’s regular employment is included in the study as an indicator of financial stability and family’s social class. The question is surveyed in wave one of the study and used to predict assimilation outcomes in wave three of the study. To survey the regular employment of respondents’ mothers, the investigators ask, “Is your mother working in her current occupation?” The response categories are binary where 1= yes and 0 = no.

This variable is used as an indicator of social class because it represents the well being of the family through employment status. Mother’s regular employment is used in all four assimilation models. Table 4.13 provides a descriptive distribution of the respondent’s mother’s regular employment.

145

Mother Employed In Regular Occupation Frequency Percent Yes 1493 90.10 No 164 9.90 Total 1657 100.00

Table 4.13: Mother Currently Employed in Regular Occupation

Mother's Father's Father's Educational Educational Education Prestige Prestige Mother's educational prestige 0.402 Father's educational prestige 0.290 0.336 Mother's education 0.545 0.351 0.448

Table 4.14: Correlation Matrix for Parents’ Education and Job Prestige

Homeown Economic Family Mother’s ership situation structure reg. emp. Economic situation 0.198

Family structure 0.196 0.08

Mother’s reg. emp. 0.065 0.461 0.054

Parent SES 0.544 0.296 0.115 0.079

Table 4.15: Correlation Matrix for all Five Measures of Social Class

Highest Number of Years of Education

Another measure of social class employed in the study is the respondent’s own educational attainment. The assimilation literature states that one additional year of

146 schooling in the United States has a larger effect on the English skills and subsequent assimilation of the immigrant compared to one year of foreign schooling (Espenshade and Fu, 1997). The variable is entered into the political participation, and out- marriage/partnership models as a continuous variable. I do not use this variable in the language proficiency model because English proficiency attained in wave one cannot be predicted by education gained in wave three of the study. Also, this variable is excluded from the educational attainment model because it is used in the construction of the dependent variable for the educational attainment. The highest number of school years completed was measured using 10, 12, 14, 15, 16 and 18 number of years in school.

Table 4.16 provides a distribution of the respondents’ highest number of years of schooling.

Highest Number Of School Years Frequency Percent 10 49 4.40 12 169 15.18 14 355 31.90 15 223 20.04 16 222 19.95 18 95 8.54 Total 1113 100.00

Table 4.16: The Highest Number of School Years Completed.

147

Grade Point Average

Grade point average is used in the research as an indicator of the respondents’ education in wave one of the study for the language proficiency model. This variable represents the child’s education at an average age of fourteen years old because the respondent had not yet attained a tertiary education. The respondent’s grade point average is measured on a scale of 0-5 in the CILS dataset. The reason why this variable is only used in the language proficiency model is because I was then able to use formal education in the political participation and out-marriage/partnership models that were all surveyed in wave three of the study.

Other Independent Variables

Educational Aspirations

Aspirations are commonly used in the literature as measures of the adaptations of immigrant children. Generally, findings indicate that earlier aspirations tend to have some effect on assimilation patterns later in life (Portes, 1995). The expected education variable measures the respondent’s highest expected level of education. The survey asks,

“Realistically speaking, what is the highest level of education you think you will get?”

The respondent’s answer was indexed on a scale of one to five where (1=less than high school, 2=finish high school, 3= finish some college, 4= finish college, 5=finish graduate degree) for purposes of this research, I recoded this variable into three categories, where categories 1-3=less than college aspirations, category 4= college aspirations and

5=graduate aspirations—less than college aspirations is used as the reference group in the

148 study. Table 4.17 provides a descriptive summary of the respondents’ future aspirations at age 14.

Aspirations Frequency Percent Less than high school 7 0.39 Finish high school 113 6.37 Finish some college 168 9.48 Finish college 629 35.48 Finish graduate degree 856 48.28 Total 1773 100.00

Table 4.17: Respondent’s Education Expectation at Age 14.

Exposure to Discrimination

Portes and Zhou 1993; Rumbaut 1996; Portes and Rumbaut (2001) argue that the persistence of racial discrimination, the bifurcation of the American labor market and its growing inequality and the consolidation of a marginalized population in the inner city propels immigrants into the underclass and prevents upward assimilation. Discrimination is used in the study to determine the extent to which immigrants’ exposure to discrimination affects their subsequent assimilation into the dominant culture for all four models of assimilation. The question is surveyed in wave one of the study and used to predict successive patterns of assimilation in wave three. The survey examines discrimination by asking the respondents if they ever felt discriminated against because

149 of their race or ethnicity. The response is coded as a binary, where 0=no and 1=yes. Table

4.18 provides a description of discrimination.

Respondent Ever Felt Discriminated Against Frequency Percent Yes 2884 54.80 No 2338 44.40 Total 5262 100.00

Table 4.18: Respondent’s Exposure to Discrimination

Respondent Lives With Mother

The social learning or role model explanatory framework posits that parental influence is a major factor that is conceptually associated with the social indicators of their children (Biblarz, Raftery, and Bucur, 1997; Chen and Kaplan, 2001; Feng et al.,

1999). As mentioned above, stability in family structure and the presence of role models impact the patterns of assimilation. The investigators survey this question by asking,

“Does your mother live with you?” The responses are indexed into three categories

(1=yes, 2=no, 3=dead/unknown) for this study. However, the variable is recoded to (0=no and 1=yes). Category 3 was eliminated because none of the respondents reported their mother as dead or unknown. This variable was only used in the language proficiency and education assimilation models. Table 4.19 provides a descriptive analysis of the respondents’ living arrangement.

150

Respondent Lives With Mother Frequency Percent Yes 1640 94.04 No 104 5.96 Total 1744 100.00

Table 4.19: Respondent Currently Residing with Mother

Citizenship Status

The citizenship variable was constructed using length of stay in the U.S. and citizenship status in the CILS dataset. First, I recoded the citizenship of the respondent into four categories. The first category included those who were citizens and those who said they lived in the U.S. all of their life. The second group included those who were naturalized citizens. The third group included those who listed themselves as non-citizens under citizenship status and those who listed themselves as living in in the U.S. for more than 10 years under length of stay in the U.S. Finally the fourth category included those who were also non-citizens and lived in the U.S. for less than 10 years. In order to categorize the more difficult cases, I included those who lived in the U.S. all their lives as citizens (category 1) and those who said they held dual citizenship but did not live in the

U.S. all their lives in the U.S. as naturalized citizens (category 2).

Three categories were used in the study-- categories one and two represented U.S. born and naturalized respectively and categories three and four were collapsed to represent non-citizens. For the political participation model all non-citizens were dropped

151 from the model; however both non-citizens and naturalized citizens were compared against U.S. born citizens in both the education and out-marriage/partnership models. I do not include citizenship status in the English proficiency model because outcomes in wave one cannot be predicted by outcomes in wave three. Table 4.20 provides a descriptive summary of the respondents’ citizenship status; aforementioned, non-citizens are used as the reference category.

Citizenship Status Frequency Percent

U.S. citizens 1,440 77.08 Naturalized citizens 229 15.48 Non-citizens 110 7.44 Total 1479 100.00

Table 4.20: Respondents’ Citizenship Status

Length Of Stay In The US

Length of stay in the United States is used as a separate variable in the study as well as in the construction of the citizenship status variable. I used length of stay in the

English proficiency model because the question was surveyed in wave one of the study when the respondents were at an average age of fourteen, also at this time, citizenship status had not yet been surveyed. The literature suggests that length of stay in the U.S. is indicative of the immigrant’s language proficiency and subsequent assimilation into the

152 dominant culture (Stevens, 1994). Accordingly, immigrant children who were brought to the U.S. at a later age are less likely to be proficient in English than those who lived in the U.S. all of their lives. The CILS asked the question, “How long have you lived in the

U.S.?” Responses are provided on a five point scale where 1=all my life 2= more than 10 years, 3= 5 to 9 years and 4=less than 5 years. The variable was recoded to a binary where 1= all my life and 0= less than all of my life. Table 4.21 provides a descriptive summary of the respondents’ length of stay in the U.S.

Length Of Stay In The US Frequency Percent All my life 1080 60.67 Ten years or more 431 24.21 Five to nine years 261 14.66 Less than five years 8 0.45 Total 1780 100.00

Table 4.21: Length of Stay in the US

EXPECTED RELATIONSHIPS FOR INDEPENDENT VARIABLES

The assimilation literature has suggested many correlations between race, class and gender and integration into the dominant culture. The following section outlines a number of independent variables that are pertinent to the aforementioned four measures of assimilation. Due to the role model hypothesis and parental influence, I expect a positive relationship between social class and assimilation for all four measures. I

153 anticipate that Caribbean Blacks will be more upwardly mobile than both Cubans and

Caribbean non-Blacks. I expect females will be more likely to be proficient in English, however, less likely to out-marry. I also expect family stability and financial success to have positive outcomes on integration. Lastly, I anticipate that discrimination will negatively affect the process of second-generation Caribbean assimilation into the dominant culture. Table 4.22 presents the expected signs and relations for the variables in the study.

154

Out- Language Political Marriage/ Educational Proficiency Participation Partnership Attainment Gender Males - n/d - n/d Females + n/d + n/d Race/Ethnicity ? Cuban White n/d ? + - Cuban Hispanics n/d ? - - Cuban Other n/d ? n/a - Caribbean Black n/d ? - + Mixed Caribbean n/d n/a - Social Class Parents' socioeconomic status + + + + Current economic situation + + + + Homeownership + + + +

Mother regularly employed + + + + Family structure + + + + Grade Point Average + n/a n/a n/a Respondent's own education n/a + + n/a Other independent variables Discrimination - - - - Live with mother + + ? + English proficiency n/a + + + Less than college Aspirations n/a - - - Aspirations College Aspirations n/a + + + Graduate degree aspirations n/a + + + Citizenship status Citizens n/a + + + Naturalized citizens n/a - - - Non-citizen n/a - - -

Table 4.22: Expected Signs for all Four Measures of Assimilation. Notes: + = Positive relationship - =Negative relationship ? = Insufficient evidence to determine relationship n/a = Variable not used in the model n/d = No difference among variables

155

CHAPTER FIVE

RESULTS AND ANALYSIS

Introduction

This chapter presents a comparative analysis of the assimilation patterns of second-generation Caribbean immigrants. Though the research focuses on the impacts of race, class and gender; I further compare the effects of race and nationality on assimilation into the dominant culture. I examine the extent to which second-generation immigrants assimilate differently based on their nationality or alternatively based on their racial and ethnic self-identification. As mentioned in the previous chapter, I constructed five racial and ethnic categories: Cuban White, Cuban Hispanic, Cuban Other, Caribbean non-Black and Caribbean Black. I use country of origin as a reference category because this allowed the sample to be re-distributed from racial and ethnic groupings to the respondents’ nationality. This in turn allowed me to examine if there were any dissimilarities in assimilation between the two classifications. The results are presented in this chapter in two separate analyses.

My goal is to assess the degree to which the immigrants’ English proficiency, political participation, educational attainment and decision to out-marry/partner are

156 influenced by race, class and gender. The chapter provides an analysis of the results, indicating the hypotheses that are supported by the research findings and also highlighting the limitations of the study, in the event that the research findings suggest otherwise.

In addition using Cuban Whites as the reference category for race and ethnicity, I also compare the assimilation of Cuban Hispanics to Cuban other and Black Caribbean to non-Black Caribbeans. These results are presented in separate tables in the chapter.

Prior to running the analyses, I performed both two-way and three-way interaction tests on race, class and gender to determine if one influenced the other. Though three-way interactions proved to be insignificant for all four models; two-way interactions between gender and race and gender and class were in fact significant for the out- marriage/partnership model. I present the findings of the two-way interaction effects of gender, race and class later in the chapter.

RESEARCH QUESTIONS

The general research question analyzes the extent to which race, class and gender are related to assimilation of second-generation Caribbean immigrants. I employ four proxies of assimilation that are quantified in the study. Assimilation as a process is not readily observable or measurable, correspondingly, over the years researchers and assimilation theorists have used a number of measures to estimate and make testable the process of adaptation (Beuka 2005; Alba, 1990; Gurakan and Fitzpatrick, 1982;

Haveman, Wolfe and Spaulding 1991). Four specific measures are implemented as

157 proxies of assimilation; these measures are central to the argument and shape the following research questions.

(1) How do race and ethnicity of second-generation Caribbean immigrants impact the language proficiency, political participation, out-marriage/partnership, educational attainment and subsequent assimilation patterns?

(2) How does social class of second-generation Caribbean immigrants affect the language proficiency, political participation, out-marriage/partnership, educational attainment and subsequent assimilation patterns?

(3) How does gender of second-generation Caribbean immigrants impact the language proficiency, political participation, out-marriage/partnership, educational attainment and subsequent assimilation patterns?

LANGUAGE PROFICIENCY

The data used to measure English proficiency contains 1586 respondents. English proficiency is estimated using only wave one data from the CILS dataset because events in subsequent time periods could not be used to estimate outcomes in a previous time period. The variable is represented as a binary and coded 1 if the respondents are excellent on all four levels of English proficiency (ability to speak, read, write, understand) and 0 otherwise.

Logistic regression results and the calculations of race comparisons for each dependent variable are presented in the following analysis. Models one, two and three in each table represent the dependent variable regressed on gender race and class separately.

158

Model four represents the joint effects of race, class, and gender on the dependent variable. I employ parents’ socioeconomic status and the respondents’ own education as the main measures of social class in model 4 because these are the primary measures used in the literature to predict assimilation.

I subsequently introduce other measures of family’s social class (economic situation and homeownership) to model five, which shows the introduction of the first set of independent variables (current economic status, homeownership, grade point average, length of stay in the U.S., mother’s regular employment and whether the respondents’ live with his/her mother). Grade point average is only used in the language proficiency model, as an indication of class because this model reflects time-period one and subsequent educational attainment had not yet been achieved. Model six represents an extension of model five and the second set of independent variables (family structure,

English proficiency, college aspirations and citizenship status.) Citizenship status is not included in the language proficiency model because the question was surveyed in time- period three. For each of the four assimilation models, I introduce two separate sets of independent variables to illustrate the extent to which race, class and gender remain salient or are alternatively weakened with the introduction of new variables. The following section gives an analysis of the language proficiency assimilation model

First I introduce race, class and gender exclusively to the model to test the effects if any, on language proficiency. Model 1, Table 5.1 shows that the odds of females being proficient in English are 56% more than males; these results are in support of the

159 hypothesis three. Model 2, Table 5.1 shows that mixed Caribbeans are 64 percent more likely than Cuban Whites to be proficient in English at the 5 % significant level; these findings are not consistent with the race and ethnicity hypothesis for this model. In support of hypothesis two, model 3, table 5.1 shows that a one-unit increase in parents’ socioeconomic status increases the odds of being proficient in English by 43%.

Independently race, class and gender all have significant effects on English proficiency and all hypotheses are supported for this model except hypothesis one.

I subsequently examine the simultaneous effects of race, class and gender on

English proficiency. All hypotheses are now supported within the model. Model 4, table

5.1 shows that the odds of females being more proficient in English compared to males increases to 73% at the one percent significance level. Consistent with the literature, greater expectations are placed on young girls in the Caribbean to succeed compared to boys; this in turn leads to higher English proficiency for females. According to Brown and Miller (2006), “greater attention is given to girls to learn at school and develop social skills and values; it is more common for boys to be encouraged to learn “fending” skills and income generating skills at an early age in the Caribbean.” Brown and Miller (2006) further argue that young girls continue to outperform boys in the Caribbean sometimes even as early as entry to grade one.

Model 4, table 5.1 shows that the effect of race and ethnicity is no longer significant once class and gender are introduced. This is not unexpected because these immigrants belong to the second generation, thus they are less likely to speak the mother

160 tongue and more likely to speak the language of the dominant culture. As hypothesized, I expect there to be no difference between race and ethnicity relative to language proficiency because second-generation immigrants are already proficient in English. The linguistic assimilation literature notes that within the three-generation model, the first generation makes some progress in language assimilation but remains dominant in their native tongue, the second generation is bilingual, and the third-generation speaks only

English (Waters, 1995). Consistent with the literature we see that bilingualism is in fact common among second-generation immigrants, thus those children immigrants who are growing up in immigrant households are all likely to have high levels of English proficiency (Alba, 2004).

Model 4, table 5.1 shows that social class is also significantly related to language proficiency; a one unit increases in the parents’ socioeconomic status index increases the odds of the child’s English proficiency by 51% at the one percent significance level.

Concurrent with hypothesis two, parents with higher levels of education and greater economic resources are more likely to rear children who are more proficient in English

(Taubman, 1989; Behrman and Knowles, 1999; Ermisch and Francesconi, 2001;

Alderman, Orazem and Paterno 2001; Gordon, Bridglall and Meroe, 2004). The simultaneous introduction of race, class and gender is in support of all of three hypotheses for English proficiency. Both gender and class affect assimilation, however race and ethnicity has no effect.

161

Model 5, table 5.1 shows that by adding the first set of independent variables the odds of females being more proficient in English compared to males reduces to 57% at the one percent significance level. Race and Ethnicity have no significant effects on language proficiency. Class is significant at the five percent level, and the odd ratios are reduced to 29%. One important point to note is that the only significant measure of class for this model is parent’s socioeconomic class. Other measures of family’s social class such as the family’s economic situation, homeownership, family structure and mother’s regular employment do not predict language proficiency and subsequent assimilation.

Model 5, table 5.1 also shows that a one-unit increase in grade point average, which is used as a measure of the respondent’s education, increases language proficiency by 57% and length of stay in the U.S. positively affects language proficiency by 88%.

Even though the introduction of additional independent variables reduces the magnitude of class and gender, all hypotheses proposed are supported for the Language proficiency model.

In summary, the research concludes that the odds of being proficient in English, and subsequent assimilation into the dominant culture are significantly influenced by gender and social class. Model 2, table 5.1 shows that race and ethnicity matters initially, however the effects do not remain salient as additional independent variables are introduced. Models 4 and 5, table 5.2 also show that there are no significant differences between those who self-identify as Cuban Hispanics and Cuban other as well as those who self-identify as Caribbean Black and Caribbean mixed on English proficiency.

162

Model 5, table 5.1 shows that other social factors such as grade point average and length of stay in the U.S. also influence language proficiency and subsequent assimilation, however the immigrants exposure to discrimination has no significant effect for this model.

163

Language Proficiency Model 1 Model 2 Model 3 Model 4 Model 5 Gender Females 1.560 *** 1.732*** 1.571*** (1.73) (0.217) (0.203) Race/Ethnicity Cuban Hispanics 1.182 1.047 0.994 (0.173) (0.169) (0.166) Cuban Other 0.980 0.874 0.770 (0.177) (0.170) (0.156) Caribbean Black 1.071 0.974 0.953 (0.179) (0.184) (0.190) Caribbean Mixed 1.640* 1.454 1.420 (0.372) (0.357) (0.359) Family’s Social Class Parents' socioeconomic status 1.434*** 1.513*** 1.291* (0.119) (0.141) (1.520) Current economic situation 1.198 (0.159) Homeownership 0.862 (0.138) Mother regularly employed 0.941 (0.204) Family structure 0.924 (0.128) Grade point average 1.570*** (0.116) Other Independent Variables Length of stay in the US 1.875*** (0.247) Discrimination 0.851 (0.109) Live with mother 1.326 (0.365) Model Statistics LR chi2 (14) 39.82 0 107.72 0 PR> chi2 0.000 0.000 Log likelihood -807.853 -773.904 N 1526 1526

Table 5.1: Language Proficiency Logistic Regression. Notes: Odds Ratios are listed above and standard errors shown in parentheses. Statistical significance differences based on t-test or chi- square as appropriate: +< .1; *p <. 05; ** p <. 01; *** p <. 001.

164

Language Proficiency Model 4 Model 5

OR Chi2 OR Chi2 Cuban Hispanic to Cuban Other 0.860 1.040 1.290 1.290 Caribbean Black to Caribbean Mixed 0.661 2.120 0.671 1.930

Table 5.2: Language Proficiency Racial and Ethnic Comparisons. Notes: Statistical significance differences based on t-test or chi-square as appropriate: + < .1; *p <. 05; ** p <. 01; *** p <. 001.

POLITICAL PARTICIPATION

Political participation is measured using a sample size of 884 respondents because at an average age of twenty-four years old, a reduced number of the respondents indicated their civic involvement. These results are presented in tables 5.5. and 5.6. The variables used in this analysis were taken from wave one and wave three of the CILS dataset. I used logistic regression to estimate the effects of race, class, gender and other social factors on the voter registration and subsequent assimilation of the respondents.

The dependent variable is coded in the model as a binary variable where 1=registered to vote and 0=not registered to vote. All none U.S. citizens were dropped from the model and U.S. born citizens were used as a reference category for naturalized citizens

I first analyze the individual effects of race, class and gender on political participation. Consistent with hypothesis three, model 1, table 5.3 shows that gender has no independent effect on political participation. Model 2, table 5.3 shows that odds of

Cuban Hispanics being registered to vote are 52% more compared to Cuban Whites at the ten percent significance level; these findings support hypothesis one. Model 3, table 5.3

165 shows that parents’ SES has no effect on registration to vote, these results are not in support of hypothesis three which state that family’s social class should be positively related to the respondents’’ political participation.

I then evaluate the joint effects of race, class and gender on political participation.

Model 4, table 5.3 indicates that there is still no difference between male and female voter registration; these findings are consistent with hypothesis three which states that the gender gap is closing in voting globally.

As hypothesized, race is a significantly related to civic involvement but only at the ten percent level; the odds of Cuban Hispanics being registered to vote are 51% more compared to Cuban Whites and the odds of Caribbean Blacks being registered to vote are

74% greater than Cuban Whites.

Parents socioeconomic status, which is the primary measure of social class, has no effect on political participation, however the respondents’ own education is significant at the ten percent level; the odds of being registered to vote increase by 10% with every additional year of schooling. These findings are consistent with the literature, which states that higher levels of education lead to higher civic involvement due to increased knowledge, awareness and information about current issues and affairs within the host country (Junn, 1994; Wolfinger and Rosenstone, 1980).

Model 5, table 5.3 shows gender is not significantly related to political participation even when additional variables are introduced. However, race and ethnicity are significant at the ten percent level; the odds of Cuban Hispanics and Caribbean

166

Blacks being registered to vote compared to Cuban Whites are now increased to 48% and

76% respectively. With regard to social class, both parents’ SES and the respondents’ own level of education have no significant effects on political participation. Additionally, other measures of family’s social class such as family’s economic situation, homeownership family structure and mother’s regular employment do not significantly influence voter registration.

The second set of independent variables are introduced in model 6, table 5.3.

Results show that the odds of Hispanic Cubans being registered to vote increases from

51% to 62% and the odds of Black Caribbeans being registered to vote compared to

Cuban Whites increases from 74% to 103% both at the 5% significance level. As hypothesized immigrants are more likely to be politically engaged when they have weaker ties to their country of origins. Findings suggests that Hispanic Cubans and Black

Caribbeans are characteristic of those immigrant groups having weaker ties to their national origins and consequently being more likely to be politically active.

Model 6, table 5.3 also shows that naturalized citizens are 51% less likely compared to U.S. born citizens to vote. Consistent with the literature we see that those immigrants who are naturalized are less likely to be politically involved perhaps due to ties to their countries of origin. College aspirations have no significant effects on civic involvement, and even though education predicts civic involvement, aspirations to attain higher levels of education do no predict voter registration. Additionally, the immigrants’ exposure to discrimination has no effect on civic engagement.

167

In summary, results show that gender does not influence political participation, as previously discussed; these results can be further explained by convergence theories in the assimilation literature. As shown in model 4, table 5.3, the only measure of social class that is related to political participation is the respondents’ own level of education.

These results must be carefully interpreted because they are significant only at the ten percent level and the effects are subsequently weakened in models 5 and 6 with the introduction of additional variables. Additionally, other measures of family’s social class also have no effect on political participation. Consistent with the literature, race and ethnicity have significant effects on voter registration; both Cuban Latinos and Black

Caribbeans are more likely to be registered to vote compared to White Cubans; model 6, table 5.3 shows that these results are significant at the 5% levels. Models 4 and 5, table

5.4 also show that there is no significant difference between those who self-identify as

Cuban Hispanics and Cuban other as well as Caribbean Black and Caribbean mixed with regard to registration to vote.

168

Political Participation Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Gender Females 1.112 1.090 1.106 1.160 (0.192) (0.205) (0.209) (0.224) Race/ Ethnicity Cuban Hispanics 1.523+ 1.513+ 1.489+ 1.621* (0.334) (0.356) (0.352) (0.390) Cuban Other 1.274 1.256 1.277 1.223 (0.352) (0.360) (0.370) (0.357) Caribbean Black 1.542 1.745+ 1.768+ 2.030* (0.449) (0.573) (0.589) (0.688) Caribbean Non-Black 1.394 1.306 1.336 1.411 (0.456) (0.436) (0.451) (0.483) Family’s Social Class Parents' socioeconomic 1.143 1.073 0.950 0.903 status (0.145) (0.156) (0.166) (0.160) Respondent's own education 1.101+ 1.091 1.091

(0.058) (0.058) (0.061) Current economic situation 0.924 0.869 (0.181) (0.173) Homeownership 1.450 1.378 (0.354) (0.341) Mother regularly employed 1.000 1.014 (0.326) (0.334) Family structure 1.192 1.362 (0.236) (0.277) Other Independent Variables Discrimination 1.093 1.152 (0.206) (0.221) Live with mother 1.066 0.959 (0.488) (0.445) English proficiency 1.285 (0.296) College Aspirations 1.553 (0.466) Graduate degree aspirations 1.234

(0.358)

Citizenship status Naturalized citizens 0.490*** Model Statistics LR chi2(14) 10.16 14.21 29.30 PR> chi2 0.18 0.36 0.03 log likelihood -386.17 -384.14 -376.59 N 884 884 884 Table 5.3: Political Participation Logistic Regression. Notes: Odds Ratios are listed above and standard errors shown in parentheses. Statistical significance differences based on t-test or chi-square as appropriate: +< .1; *p <. 05; ** p <. 01; *** p <. 001.

169

Political Participation Model 4 Model 5 Model 6

OR Chi2 OR Chi2 OR Chi2

Cuban Hispanic to Cuban Other 1.204 0.35 1.166 0.23 1.325 0.76

Caribbean Black to Caribbean Mixed 0.876 0.47 1.323 0.43 1.438 0.71

Table 5.4: Political Participation Racial and Ethnic Comparisons. Notes: Statistical significance differences based on t-test or chi-square as appropriate: +< .1; *p <. 05; ** p <. 01; *** p <. 001.

OUT-MARRIAGE/PARTNERSHIP

Out-marriage and partnership is measured in the research by whether or not the respondent identifies his/her spouse into a racial and ethnic group different from his/her own racial/ethnic self-identification. The models for out-marriage/partnership are shown in table 5.5. Only three of the five racial categories were used in this model (Cuban

White, Cuban Hispanic and Black Caribbean). Those immigrants who self-identified as

Cuban other or Caribbean mixed were dropped from the model because I was unable to specify out-marriage/partnership from the ‘other’ classifications. I use Cuban Whites as the reference category for this assimilation model. Racial and ethnic comparisons were not made for this model since only three categories were used. The model is representative of 335 respondents. Two-way interaction effects were additionally employed to test the relationship of gender on class and gender on race relative to out- marriage. Findings show that there is no significant relationship between race and class

170 with regard to out-marriage/partnership. The following gives an analysis and presentation of the results.

I first look at the individual effects of race, class and gender on out- marriage/partnership. Models 1 and 3, table 5.5 show that gender and class have no individual effects on out-marriage/partnership. Model 2, table 5.5 shows that the odds of both Cuban Hispanics and Caribbean Blacks out-marrying/partnering are significantly less compared to White Cubans. The initial models are not in support of hypotheses two and three, but hypothesis one is supported.

I further examine the simultaneous effects of race, class and gender on out- marriage/partnership. Model 4, table 5.5 indicates that gender has no significant impacts on out-marriage/partnership. However, the odds of Cuban Hispanics and Black

Caribbeans out-marrying compared to White Cubans are 97% and 98% less respectively.

As expected, the size of the racial and ethnic groups is seen to significantly influence out marriage; those who self identify as minorities are less likely to out-marry because racial and ethnic minorities are more likely to marry each other if their group is of a substantial size. With regard to class, parents’ socioeconomic status and the respondents’ own level of education have no effect on out-marriage/partnership. Without the introduction of additional independent variables, we see race and ethnicity are the only determinants of out-marriage/partnership. These finding are only in support of hypothesis one.

Models 5 and 6, table 5.5 show the first and second set of independent variables that are introduced; results indicate that race and ethnicity are the only salient factors on

171 out-marriage/partnership. By introducing the second set of variables, the odds of Cuban

Hispanics and Caribbean Blacks out-marrying compared to Cuban Whites are still statistically significant, however the magnitude of the effects are reduced to 47% and

43% respectively. None of the additional measures of family’s social class have statistically significant effects on out-marriage/partnership. Additional independent variables such as discrimination and citizenship status were not related to out- marriage/partnership, however the immigrants’ academic aspirations significantly influences the decision to out-marry/partner at the 10% significance level.

I successively perform two-way interactions to examine the effects of gender in relation to race and class and to further test the validity of my hypotheses. Model 7, table

5.5 shows that gender significantly interacts with both race and class for out- marriage/partnership. The odds of Cuban White females out-marrying are now 235% greater compared to Cuban White males given that parents’ SES equals to zero. These results support hypothesis three which states that due to the transformation of women’s roles and the entry of women into the labor force, the marriage pool for this group is increasing and women now have greater options in choosing a spouse than previously.

Another explanation for these findings could be that females tend to out-number men in the study, thus their mate selection is smaller within the immigrant group.

Model 7, table 5.5 also shows that the odds of Hispanic Cuban males out- marrying are 88% lower compared to Cuban White males. Black Caribbean males are

90% less likely than White Cuban males to out-marry. According to Blau (1977), group

172 size matters and larger immigrant groups are more likely to intermarry due to a higher selection of mates within the group. Paradoxically, even though the number of immigrants self-identifying as Afro-Caribbean in the study are smaller compared to

Cuban Whites, findings suggests that Cuban Whites are still more likely to out- marry/partner. One possible explanation for this is that Cubans have filtered themselves into two distinct categories- Cuban White and Cuban Hispanic. Thus readers ought to consider if Cuban Whites marrying Cuban Hispanics can truly be considered out- marriage.

With regard to social class, a one-unit increase in parents’ SES increases the odds of out-marriage/partnership for men by 161% and decreases out marriage for women by

1%. Other measures of family’s social class do not significantly change the model.

Consistent with hypothesis two we see that higher levels of economic resources place immigrants in contact with other cultures through social groups and cliques and consequently promote intermarriage.

The two-way interaction effects of gender and race also indicate that Cuban

Hispanic women are 97% less likely than Cuban White women to out marry. One plausible explanation could be that that Cuban Hispanic women out-number Cuban

White women thus they are more likely to inter-marry/partner. The study also shows that the effect of being Black Caribbean rather than a White Cuban is not significantly different by gender. As noted before, two-way interaction effects of gender and class further illustrate that women with higher socioeconomic status are 1% less likely to out-

173 marry/partner compared to those with lower socioeconomic status. These effects are not consistent with hypothesis one because we expected all women should be more likely to out-marry/partner; nonetheless 1% is still a very small margin when analyzing the likelihood of out-marriage/partnership for women with higher SES.

Additional independent variables such as discrimination and citizenship status have no effects out-marriage/partnership, however immigrants’ academic aspirations negatively affect immigrants’ decision to out-marry/partner at the 5% significance level.

In summary the results of the two-way interaction out-marriage/partnership model indicate that Caribbean Black men are the least likely group to out-marry and Cuban

White men are the most likely group to out-marry. Cuban Hispanic men are also significantly less likely to out marry compared to Cuban White; paradoxically Cuban

White females are the ones who are more frequently out-marrying. This can possibly be explained by the size of the Cuban White female group. As hypothesized, parents’ socioeconomic status is positively related to out-marriage/partnership however the difference does no vary by racial and ethnic groups—as it only holds for men.

Consistent with all hypotheses for this model, findings suggest that race, class and gender are all salient factors in predicting out-marriage/partnership for Caribbean immigrants. However, the results of this model should be carefully interpreted because the population size is small and more notably the respondents are young. At an average age of 24 the decision to out-marry/partner may not be truly representative of the entire immigrant group and the attitudes and behaviors of these young adults are likely to

174 change over time. Additionally, the self-identification of Cubans as Cuban White and

Cuban Hispanic raises the question the extent to which Cubans truly out- marrying/partnering.

175

Out-marriage/partnership Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Gender Females 0.905 1.027 1.034 1.176 3.35* (0.196) (0.299) (0.304) (0.362) (1.75) Race/ Ethnicity Cuban Hispanics 0.056*** .0628*** .0625*** .0534*** 0.120*** (0.017) (0.019) (0.019) (0.018) (0.059) Caribbean Black 0.053*** .060*** .0629*** .0579*** .0973** (0.019) (0.024) (.0255) (0.024) (0.076) Family’s Social Class Parents' socioeconomic status 1.122 1.120 1.290 1.390 2.611* (.169) (0.246) (0.357) (0.392) (1.06) Respondent's own education 0.954 0.963 0.984 0.988 (0.073) (0.074) (0.078) (0.082) Current economic situation 0.970 1.026 1.044 (0.279) (0.305) (0.314) Homeownership 0.727 0.757 0.763 (0.273) (0.288) (0.303) Mother regularly employed 0.803 0.824 (0.384) (0.402) Family structure 1.276 1.270 (0.386) (0.39) Other Independent Variables Discrimination 1.083 1.111 (0.319) (0.337) Live with mother 0.545 0.654 (0.354) (0.428) English proficiency 0.593 0.550 (0.220) (0.210) College aspirations 0.574 0.548 (0.261) (0.254) Graduate degree 0.434+ 0.370* aspirations (.196) (.173) Continued

Table 5.5: Out-marriage/partnership Logistic Regression. Notes: Odds Ratios are listed above and standard errors shown in parentheses. Statistical significance differences based on t-test or chi- square as appropriate: +< .1; *p <. 05; ** p <. 01; *** p <. 001

176

Table 5.5 continued.

Out-marriage/partnership Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Model 7 Citizenship status Naturalized citizens 1.495 1.789 (0.516) (0.656) 1.080 1.184 Non-citizen (.528) (.592) Interaction terms Gender-Cuban Hispanic- 0.241* interact (.158) Gender-Caribbean-Black- interact 0.461 (0.43) Gender-Parent SES-interact 0.377* (0.181) Model Statistics LR chi2 (14) 120.47 122.14 131.020 140.45 PR> chi2 0.000 0.000 0.000 0.000 Log likelihood 171.308 170.47 166.03 161.31 N 335 335 335 335

EDUCATIONAL ATTAINMENT

Educational attainment is employed in the research as a fourth measure of assimilation; the analysis is based on 1,005 respondents. The dependent variable is binary, where 1= attainment of a college degree of higher and 0= attainment of less than a college degree. The following section presents the hypotheses and findings of the model.

First I introduce race, class and gender exclusively into the model to test their relationships to educational attainment. Model 1, table 5.6 shows that gender is not significantly related to educational attainment, these results support hypothesis three.

Model 2, table 5.6 shows that race has no effect on the respondents’ educational attainment; these results do not support hypothesis one. Model 3, table 5.6 shows that a

177 one unit increase in parents’ socioeconomic status increase the respondents’ attainment of a college degree of higher by 102% at the one percent significant level—these findings are in support of hypothesis two.

I then look at the joint effects of race, class and gender on educational attainment.

Model 4, table 5.6 indicates that gender has no immediate effect on the respondent’s educational attainment. These findings support hypothesis three; consistent with the gender convergence hypothesis (Veenman and Heij, 2008), one can argue that immigration equalizes the effects for both education and income between men and women. The migration of immigrants from developing to developed nations allows women immigrants’ equal opportunities to education and employment and subsequently closes the gender gap particularly for the second and later generations of immigrants.

Model 4, table 5.7 indicates that the odds of attaining a college degree or more are

84% higher for Caribbean Blacks compared to mixed Caribbean at the 10% significance level. These findings support hypotheses one which states that Black Caribbeans are economically driven to migrate, thus they are expected to achieve higher levels of education (Chiswick, 1979, Palmer, 1974). Additionally, Caribbean Blacks have a greater propensity to be motivated and succeed due to less rigid classifications of racial groups in the Caribbean and their exposure to Black achievement (Wilson, 1987). Another important point to note is that the mixed Caribbean group in the study is made up of predominantly Dominican immigrants. Grillo (2000) suggests that traditionally,

Dominican immigrants have been negatively stereotyped and have also been one of the

178 least likely immigrant groups to achieve upward mobility in the U.S. which in turn their lower levels of academic achievement.

In support of hypothesis two, model 4, table 5.6 suggest that class is also a significantly related to educational attainment; a one-unit increase in parents’ socioeconomic status increases the odds of obtaining a college degree or higher by 218% at the one percent significant level. Consistent with the role model hypothesis parents’ socioeconomics status significantly influences the educational attainment of second- generation Caribbean immigrants (Taubman, 1989; Behrman and Knowles, 1999;

Ermisch and Francesconi, 2001; Alderman, Orazem and Paterno 2001; Gordon, Bridglall and Meroe, 2004).

I then introduce the first set of independent variables, model 5, table 5.6 shows that gender has no effect on educational attainment. Model 5, table 5.7 shows that race remains statistically significant; the odds of Caribbean Blacks obtaining a college degree compared to Caribbean mixed increases to 92% at the ten percent significant level.

Model 5, table 5.6 shows that class is also still significant; an increase in parents’ socioeconomic status increases the odds of obtaining a college degree or higher by 208%.

Of all the social class measures employed in the study, the only significant predictors of educational attainment are parents’ SES and family structure. Findings suggest that the odds of obtaining a college degree or higher increase by 71% if respondents come from families with both parents present.

179

Model 6, table 5.6 shows the results of educational attainment when all independent variables are taken into account; the odds of Caribbeans Blacks obtaining a college degree compared to Cuban Whites have now increased to 55%. The introduction of additional variables has further substantiated the effects of race on educational attainment because these findings have now emerged in the model as significant at the

10% level. Also, model 6, table 5.7 shows that the odds of Caribbean Blacks obtaining a college degree compared to Caribbean mixed immigrants increases from 90% to 104%.

Model 6, table 5.6 shows that class remains a salient predictor or educational attainment; a one-unit increase in parents’ socioeconomic status increases the odds of obtaining a college degree or higher by 181% with all other variables taken into account. This model also illustrates that family structure, aspirations and citizenship status all have significant effects on educational attainment.

Consistent with hypothesis one, findings suggest that Black Caribbeans are more likely to hold college degrees or higher compared to both Mixed Caribbeans and White

Cubans. As hypothesized, parents’ socioeconomic status is significantly related to the educational attainment of the child, this finding supports the reproduction of inequality noted in hypothesis two, where families with greater economic resources are more likely to rear children who receive higher levels of education and the opposite is true.

Consistent with hypothesis three, gender does not have an effect on educational achievement; convergence theories are used to explain the closing gap in education among immigrants in young adulthood. In summary, findings indicate that race and class

180 significantly predicts educational attainment and subsequent assimilation into the dominant culture and gender remains neutral.

Educational Attainment Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Gender Females 0.964 1.188 1.225 1.173 (0.129) (0.181) (0.189) (1.870) Race/ Ethnicity Cuban Hispanics 0.997 0.954 0.93 0 0.939 (0.169) (0.179) (0.176) (0.181) Cuban Other 1.200 1.164 1.181 1.146 (0.256) (0.271) (0.278) (0.275) Caribbean Black 1.227 1.353 1.410 1.555+ (0.25) (0.307) (0.332) (0.383) Caribbean Non-Black 0.790 0.736 0.731 0.759 (0.225) (0.206) (0.209) (0.211)

Family’s Social Class Parents' socioeconomic status 3.026 *** 3.183 *** 3.080 *** 2.813 *** (0.341) (0.397) (0.448) (0.422) Current economic situation 1.099 1.013 (0.169) (0.161) Homeownership 0.98 0.897 (0.211) (0.199) Mother regularly employed 0.999 1.000 (0.276) (0.285) Family structure 1.710* 1.718** (0.282) (0.291) Other Independent Variables Discrimination 0.1005 0.942 (0.153) (0.147 Live with mother 1.251 1.268 (0.596) (0.611 English proficiency 1.192 (0.240) College Aspirations 1.994* (0.648) Graduate degree aspirations 3.257*** (1.020) Continued

Table 5.6: Educational Attainment Logistic Regression. Notes: Odds Ratios are listed above and Standard errors shown in parentheses. Statistical significance differences based on t-test or chi- square as appropriate: +< .1; *p <. 05; ** p <. 01; *** p <. 001

181

Table 5.6 continued.

Educational Attainment Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Citizenship status Naturalized citizens 1.094 (0.211) Non -citizen 0.225*** (0.095) Model Statistics LR chi2 (14) 104.32 117.79 158.96 PR> chi2 0.000 0.000 0.000 Log likelihood -551.69 -544.962 -524.381 N 1005 1005 1005

Educational Attainment Model 4 Model 5 Model 6

OR Chi2 OR Chi2 OR Chi2

Cuban Hispanic to Cuban Other 0.811 0.65 0.787 0.91 0.819 0.61

Caribbean Black to Caribbean Mixed 1.838+ 3.61 1.92+ 4.11 2.04* 4.54

Table 5.7: Educational Attainment Racial and Ethnic Comparisons. Notes: Statistical significance differences based on t-test or chi-square as appropriate: +< .1; *p <. 05; ** p <. 01; *** p <. 001.

GENERAL FINDINGS FOR OTHER INDEPENDENT VARIABLES

The following section provides the results for additional independent variables used in all four models. Though discrimination, aspirations and citizenship status are not central components of the study, they were directly related to the four measures of assimilation. As previously discussed, I hypothesized that these three factors should be

182 related to language proficiency, political participation, out-marriage/partnership and educational attainment.

Across all four measures of assimilation, discrimination has no effect on second- generation Caribbean immigrants’ integration into the dominant culture. Contrary to the findings of Portes et al., (2006), exposure to discrimination is not seen in the research to have any significant effects on language proficiency, political participation, out-marriage and educational attainment; these findings are therefore not supported by my hypotheses.

Portes and Zhou (1996) argued that parental influence no longer mattered in assimilation; they further contended that parental influence was previously a significant factor in determining assimilation, however with newer waves off immigrants, discrimination has now trumped parents’ socioeconomic status. My research findings suggest otherwise; in fact for this segment of second-generation Caribbean immigrants’ parental socioeconomic status is salient across all four measures of assimilation and paradoxically discrimination has no effect on the immigrants’ integration into the dominant culture.

College aspirations were employed in the political participation, out- marriage/partnership and educational attainment assimilation models. I did not use aspirations in the language proficiency model because I wanted to examine the effects of prior aspirations on subsequent patterns of assimilation. Accordingly, both aspirations and language proficiency were surveyed in wave one of the study when the respondents’ were at an average age of fourteen. Findings suggest that the respondents’ aspirations for college are not significant in the decision to out-marry/partner or political involvement,

183 however aspirations are salient for educational attainment at the one percent level.

Results indicate that second-generation Caribbean immigrants with higher levels of educational aspirations in early teenage years are more likely to attain higher levels of academic achievement in young adulthood. Consistent with the assimilation literature higher aspirations among immigrants usually lead to greater academic achievement.

Similarly, citizenship status was employed in the political participation, out- marriage/partnership and educational attainment models of assimilation. Results indicate that citizenship status is statistically significant for both political participation and educational attainment, however the decision to out-marry/partner did not vary by citizenship status. Non-citizens are seen to be less likely to attain a college degree or higher compared to citizens and naturalized citizens are less likely to be registered to vote compared to U.S. born citizens. These findings are consistent with the literature which notes that naturalized citizens have a certain level of apprehension when it comes to severing ties with their countries of origin and becoming fully integrated into the dominant culture, consequently they are less likely to be fully assimilated into the dominant culture (Portes and Rumbaut, 2006).

SENSITIVITY ANALYSIS USING NATIONALITY

The following sections present the findings of the sensitivity analysis employed in the research to determine if changing the respondents’ racial and ethnic categories to nationality explains any distinct differences in the study. As previously discussed, respondents’ national origin is used in the study as a comparison to the five racial and

184 ethnic categories (Cuban White, Cuban Hispanic, Cuban Other, Caribbean Black and

Caribbean non-Black). Similar to the preceding analysis, I examine the impact of nationality, class and gender on language proficiency, political participation and educational attainment. Five countries in the research- Cuba, Dominican Republic, Haiti,

Jamaica and the West Indies represent the respondents’ national origin. I am unable to analyze the effects of nationality for out-marriage/partnership because the dataset does not provide the country of origin for the respondents’ spouse or partner. Accordingly, out-marriage is only examined on the basis of the five racial and ethnic groups. The following sections analyze the results of models 2 and 6 in subsequent tables for each of the four measures of assimilation used.

Language Proficiency

I first look at individual effects of the respondent’s national origin on language proficiency. Model 2, table 5.8 shows that the odds of Jamaicans being proficient in

English are 46% higher than Cubans. These results are not unexpected since English is the native language of Jamaicans whereas Cubans and Dominicans are Spanish speaking and Haitians and French speaking.

I then examine the effects of all independent variables on language proficiency.

Model 5, table 5.8 shows that at an average age of fourteen, girls are 58% more proficient in English compared to boys at the one percent significance level.

Similar to Model 5, table 5.1 that shows no difference in racial and ethnic self- identification relative to language proficiency, the sensitivity analysis indicates that there

185 is also no difference among nationalities with regard to language proficiency. Model 5.9 also shows that there are no significant differences between Dominicans and Haitians and

Jamaicans and other West Indians relative to language proficiency. Additionally, class is a significant indicator of English proficiency; both the respondents’ own education

(shown by grade point average) and parents’ socioeconomic status are significantly related to language proficiency at the one percent level.

Both analyses of language proficiency (Models 5, tables 5.1 and 5.8) show that

Caribbean girls are more likely to out-perform Caribbean boys in English. Parents’ socioeconomic status and the respondents’ grade point average are also significantly related to language proficiency. Additionally, there are no statistically significant differences using both racial and ethnic self-identification and national origin with regard to English proficiency when all independent variables are taken into account.

The sensitivity analysis was therefore advantageous to the study in explaining the lack of significant differences between the respondents’ racial and ethnic self- identification and nationality when predicting language proficiency. Both measures produced analogous results and ultimately supported the all expected hypotheses.

186

Language Proficiency Model 1 Model 2 Model 3 Model 4 Model 5 Gender Females 1.560*** 1.747*** 1.589*** -0.173 -0.219 -0.205 Nationality Dominicans 0.834 0.889 1.071 -0.188 -0.224 -0.281 Haitians 0.872 1.110 1.191 -0.157 -0.237 -0.269 Jamaicans 1.461+ 1.280 1.470 -0.316 -0.312 -0.385 West Indians 1.500 1.088 1.051 -0.377 -0.282 -0.282 Family’s Social Class Parents' socioeconomic status 1.439*** 1.501*** 1.288* -0.117 -0.140 -0.153 Current economic situation 1.186 -0.158 Homeownership 0.853 -0.138 Mother regularly employed 0.916 -0.200 Family structure 0.948 -1.310 Grade point average 1.554*** -0.114 Other Independent Variables Length of stay in the US 1.933*** -0.257 Discrimination 0.879 -0.115 Live with mother 1.338 -0.368 Model Statistics LR chi2 (14) 37.870 105.790 PR> chi2 0.000 0.000 Log likelihood -808.830 -774.867 N 1526 1526

Table 5.8: Language Proficiency-Nationality Logistic Regression. Notes: Odds Ratios are listed above and standard errors shown in parentheses. Statistical significance differences based on t-test or chi-square as appropriate: +< .1; *p <. 05; ** p <. 01; *** p <. 001.

187

Language Proficiency Model4 Model5 OR Chi2 OR Chi2 Dominican to Haitian 0.8 0.51 0.899 0.11 Jamaican to West Indians 1.172 0.22 1.4 0.92

Table 5.9: Language Proficiency-Nationality comparisons. Notes: Statistical significance differences based on t-test or chi-square as appropriate: +< .1; *p <. 05; ** p <. 01; *** p <. 001.

Political Participation

For political participation I find the following relationships-- model 2, table 5.10 shows that by exclusively introducing nationality, the odds of Dominicans being registered to vote are 44% less than Cubans at the 10% level. These findings are consistent with hypothesis one which states that immigrants with stronger ties to their countries of origin will be less likely to vote in the dominant culture.

I further investigate changes in political participation taking all independent variables into account. Model 6, table 5.10, indicates that gender is not significantly related to voter registration. Consistent with hypothesis three, both analyses show that gender has no effect on registration to vote.

Model 6, table 5.10 also shows that nationality is significantly related to registration to vote at the 10% level. Paralleling the two analyses (racial and ethnic self- identification and nationality) model 6, table 5.3 shows that by using racial and ethnic self-identification, Cuban Latinos and Black Caribbean are more likely to be registered to vote compared to Cuban Whites; these results can be explained due to these immigrants

188 having weaker ties to their countries of origin. Alternatively, the sensitivity analysis indicates that Dominicans are representative of those immigrants who have the strongest ties to their countries of origin and are subsequently less likely to vote compared to both

Cubans and Haitians.

With regard to social class, I find that respondents’ own level of education is significantly related to political participation at the ten percent level without the introduction of additional independent variables. However these effects are subsequently weakened as additional variables are added. Models 6, tables 5.3 and 5.10, which include the addition of all independent variables for both analyses, demonstrate that family’s social class is not significantly related to political participation.

Similar to the findings in model 6, table 5.3, naturalized citizens are significantly less likely than U.S. born citizens to vote. Correspondingly, discrimination and other measures of family’s social class do not predict political participation.

Even though the sensitivity analysis shows that nationality is significantly related to political participation, these results are only significant at the ten percent level. The use of racial and ethnic self-identification produced more meaningful results for the study because these findings were significant at the five percent level. Nevertheless, the sensitivity analysis was insightful to the study because results emerged for Dominican immigrants that were previously concealed by using racial and ethnic self-identification in determining political participation. The analysis was able to produce meaningful results for Dominicans in comparison to both Cubans and Haitians. The sensitivity

189 analysis was beneficial to the study because it created a multi-dimensional view of

Caribbean immigrants relative to civic involvement.

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Gender Females 1.112 1.100 1.115 1.17 (0.192) (0.208) (0.211) (0.228) Nationality Dominicans 0.560+ 0.531+ 0.550+ 0.530+ (0.189) (0.070) (0.194) (0.191)

Haitians 1.542 1.339 1.313 1.452 (0.599) (0.461) (0.528) (0.591) Jamaicans 1.177 1.362 1.435 1.837 (0.401) (0.433) (0.576) (0.763) West Indians 1.530 1.874 1.931 1.762 (0.634) (0.196) (0.947) (0.869) Family’s Social Class Parents' socioeconomic status 1.143 1.060 0.941 0.899 (0.145) (0.155) (0.165) (0.159) Respondent's own education 1.099+ 1.088 1.088 (0.058) (0.059) (0.061) Current economic situation 0.918 0.862 (0.180) (0.173) Homeownership 1.449 1.372 (0.354) (0.340) Mother regularly employed 0.984 0.976 (0.321) (0.323) Family structure 1.205 1.398 (0.238) (0.285)

Continued Table 5.10: Political Participation-Nationality Logistic Regression. Notes: Odds Ratios are listed above and standard errors shown in parentheses. Statistical significance differences based on t-test or chi-square as appropriate: +< .1; *p <. 05; ** p <. 01; *** p <. 001.

190

Table 5.10 continued.

Model 1 Model 2 Model 3 Model 4 Model 5 Model 6

Other Independent Variables Discrimination 1.075 1.138 (0.204) (0.220) Live with mother 1.085 0.994 (0.496) (0.460) English proficiency 1.248 (0.288) College Aspirations 1.461 (0.440) Graduate degree aspirations 1.157 (0.338) Citizenship status Naturalized citizens 0.483** (0.106) Model Statistics LR chi2 (14) 11.91 16.02 30.46 PR> chi2 0.1037 0.247 0.0232 Log likelihood -385.292 383.23 -376.016 N 884 884 884

Political Participation Model 4 Model 5 Model 6 OR Chi2 OR Chi2 OR Chi2

Dominican to Haitian 0.396+ 3.35 0.418+ 2.91 0.365+ 3.72

Jamaican to West Indians 0.726 0.28 0.743 0.24 1.042 0

Table 5.11: Political Participation- Nationality Comparisons. Notes: Statistical significance differences based on t-test or chi-square as appropriate: +< .1; *p <. 05; ** p <. 01; *** p <. 001.

191

Educational Attainment

First, I analyze the individual relationship between nationality and educational attainment. Model 2, table 5.12 shows that there is no significant relationship between national origin and the respondents’ educational attainment; these findings are not consistent with hypothesis one and the results of the previous analysis (see model 2, table

5.6.).

Next, I investigate the relationship between all independent variables and the respondents’ national origin. I find that for gender (model 6, table 5.12) there is no effect on educational attainment; these findings support hypothesis three. Consistent with the gender convergence hypothesis, gender has no effect on education for both analyses.

With regard to nationality, model 6, table 5.12 shows that there is no difference among nationalities relative to educational attainment; these results do not support hypothesis one. Table 5.13 also shows that there are no significant differences between

Dominicans and Haitians and Jamaicans and West Indians relative to educational attainment.

The sensitivity analysis indicates that when Caribbean Blacks are divided based on nationality as opposed to grouped together by race; there are no significant differences in educational attainment. The redistribution of Caribbean immigrants by racial and ethnic self-identification is seen as a more useful method and produces more meaningful results in the study. One possible explanation for this is that because the sample distribution is skewed towards Cubans, there may be some difficulty in obtaining

192 significant results when making comparisons to substantially smaller groups of Black

Caribbeans. Accordingly, a more even distribution of the sample using the five racial and ethnic categories produces significant findings.

With regard to social class, model 6, table 5.12 shows that family’s social class is positively related to the respondents educational attainment; these results are significant at the one percent level and support hypothesis two. Both analyses also show that family structure, parents’ socioeconomic status, aspirations and citizenship status are all significantly related to educational attainment.

Overall findings of the model suggest that a comparison of race/ethnicity to nationality is beneficial to the study because it reveals variation in education by race, which is not captured using just nationality. Additionally, findings of the sensitivity analysis indicate that by using a more diverse subset of Cubans produces more meaningful results when making comparisons to Afro-Caribbean immigrants.

In summary, results of both analyses are consistent. In both analyses I find that gender is not significantly related to educational attainment. Additionally, family’s social class is significantly related to the respondents’ educational attainment. Although the use of national origin does not show significant differences in educational attainment, the sensitivity analysis demonstrated why it was useful to the study to use the respondents’ racial and ethnic self-identification as an alternative determinant of educational attainment.

193

Educational Attainment Model 1 Model 2 Model 3 Model 4 Model 5 Model 6 Gender Females 0.964 1.2 04 1.24 0 1.196 (0.129) (0.184) (0.192) (0.191)

Nationality Dominicans 0.562 0.629 0.638 0.652

(0.201) (0.241) (0.246) (0.275)

Haitians 0.657 0.841 0.87 0.966 (0.173) (0.247) (0.262) (0.304) Jamaicans 1.348 1.129 1.202 1.330 (0.313) (0.289) (0.320) (0.375) West Indians 1.099 1.066 1.030 1.089 (0.307) (0.324) (0.319) (0.349) Family’s Social Class Parents' socioeconomic status 3.026*** 3.123 *** 3.013 *** 2.759 *** (0.340) (0.390) (0.441) (0.418) Current economic situation 1.089 0.996 (0.168) (0.158) Homeownership 0.992 0.909 (0.215) (0.202) Mother regularly employed 1.012 1.013 (0.280) (0.289) Family structure 1.670** 1.681 (0.276) (0.286) Other Independent Variables Discrimination 0.980 0.925 (0.151) (0.146) Live with mother 1.295 1.307 (0.618) (0.629) English proficiency 1.158 (0.233) College Aspirations 2.086* (0.678) Graduate degree aspirations 3.410*** (1.066) Continued Table 5.12: Educational Attainment-Nationality Logistic Regression. Notes: Odds ratios are listed above and the standard errors shown in parentheses. Statistical significance differences based on t-test or chi- square as appropriate: +< .1; *p <. 05; ** p <. 01; *** p <. 001

194

Tables 5.12 continued.

Educational Attainment Model 1 Model 2 Model 3 Model 4 Model 5 Model 6

Citizenship status Naturalized citizens 1.076 (0.210) 0.238** Non-citizens (0.101) Model Statistics LR chi2 (14) 102.13 114.79 155.56 PR> chi2 0.000 0.000 0.000 Log likelihood -552.795 546.461 526.083 N 1005 1005 1005

Educational Attainment Model 4 Model 5 Model 6 OR Chi2 OR Chi2 OR Chi2 Dominican to Haitian 0.747 0.39 0.732 0.44 0.674 0.65 Jamaican to West Indians 1.059 0.02 1.165 0.17 1.217 0.24

Table 5.13: Educational Attainment- Nationality Comparisons. Notes: Statistical significance differences based on t-test or chi-square as appropriate: +< .1; *p <. 05; ** p <. 01; *** p <. 001.

BENEFITS OF USING A SENSITIVITY ANALYSIS

Comparisons of race and ethnicity to national origin allowed me to capture a broader more comprehensive picture of assimilation in the study. The results remained consistent across both models and there were new findings that emerged which painted an even clearer picture of assimilation among Caribbean immigrants. The sensitivity analysis was useful in confirming the findings of the original model as well as elaborating

195 on new results based on national origin. Additionally, there were some findings that were captured using race that I was unable to reveal with just the use of nationality. Employing a comparative approach proved to be beneficial to the analysis because it allowed for the distinction among Caribbean countries. Whereas in the preceding analysis, comparisons could only be made between those who self identified as Whites, Blacks and Hispanics and Other, nationality also allowed for the contrast among Jamaicans, Haitians,

Dominicans and West Indians and Cubans.

Since self-identification as White serves as an indication of greater assimilation and integration into the dominant culture, I felt it would be a valuable contribution to literature to demonstrate how different self-identified groups emigrating from the same country assimilate differently; hence the creation of sub-categories for Cuba. Both race, ethnicity and country of origin all provide meaningful dimensions to the study because both analyses reveal differences in assimilation among a very small segment of immigrants. Results suggest that both race and ethnicity are salient factors when predicting the assimilation of second-generation Caribbean immigrants. The next section summarizes these results more completely.

RACE, CLASS, GENDER AND SECOND-GENERATION CARIBBEAN

IMMIGRANTS

Table 5.14 summarizes my findings for the effects of race, class and gender on the four measures of assimilation. General findings indicate that race, class and gender all influence the assimilation of second-generation Caribbean immigrants. The use of four

196 measures of assimilation allowed me to examine the individual effects of race/ethnicity, class and gender on each of the measures and to further assess which of the three had greater effects on each of the models.

My findings imply that although race, class and gender all affect integration into the dominant culture, family’s social class has the greatest effects on assimilation because it is statistically significant across all four models at the 1% and 5% levels. Of the six measures of family’s social class employed in the study, parents’ socioeconomic status is seen to be the most consistent and significant predictor of assimilation for all four models.

Race has the second largest effect on assimilation. The effects of race are consistent both in the baseline models and with the introduction of additional independent variables for three of the four measures of assimilation; political participation, out- marriage/partnership and educational attainment, and at the 5% and 1% significance levels respectively. For language proficiency, race is exclusively salient within the model, however the effects are subsequently diminished by the introduction of gender and class and other additional independent variables.

Lastly, gender is significantly related to assimilation for the language proficiency and out-marriage models at the 1% and 5% significance level, respectively. Convergence theory is used in the study to further explain why gender remains neutral for both the political participation and educational attainment models. The following chapter concludes the research findings and also provides suggestions for future research.

197

Out- Language Political Marriage/Part- Educational

Proficiency Participation nership Attainment Gender Females *** * Race/Ethnicity Cuban Hispanics * *** Cuban Other Caribbean Black * *** + Mixed Caribbean Hispanic Cuban to Cuban Other Afro-Caribbean to Mixed Caribbean * Nationality Dominicans + Haitians Jamaicans West Indians Dominicans to Haitians + Jamaicans to West Indians Social Class Parents' socioeconomic status * * *** Current economic situation Homeownership Mother regularly employed Family structure *** Grade Point Average *** Respondent's own education + Interaction Terms Cuban-Hispanic women * Gender-Afro-Caribbean women Parent SES-women *

Table 5.14: Overall Effects of Race, Class and Gender on Four Measures of Assimilation. Notes: Statistical significance differences for the final models based on t-test or chi-square as appropriate: + p < .1; * p <. 05; **p <. 01; *** p <. 001.

198

CHAPTER SIX

CONCLUSIONS

Migration is one of the processes through which minority groups are formed.

Sociologists believe that the shifting of political boundaries also facilitate the integration of minority groups with the dominant culture. Assimilation theorists argue that when minority groups come into contact with the dominant culture there are consequences such as prejudice and discrimination that reduce the immigrants’ mobility and hinder assimilation within the host society.

Classical assimilation theorists argue that complete assimilation is inevitable and when this occurs there is the absence of any form of prejudice or discrimination towards to the minority group. Contemporary assimilation theorists have challenged this view, arguing that there are permanent barriers that prevent immigrants from achieving complete assimilation such as the color of their skin, institutionalized discrimination, segmented labor markets and the concentration of immigrants in the inner city. They further contend that these barriers are also instrumental in propelling the new wave of immigrants into the underclass. Both perspectives on assimilation are used to inform this

199 research; however this study focuses on the intersectionality of race, class and gender as it relates to the assimilation of second-generation Caribbean immigrants.

This chapter provides a summary of my findings, which are informed by both the classical and contemporary models of assimilation, contributions to the research, the general hypotheses, limitations of the study and recommendations for future research. In general my findings are that race, class and gender are all significantly related to the assimilation of second-generation Caribbean immigrants into the dominant culture, however family’s social class demonstrates greater effects than both race and gender.

Race is significantly related to three of the four models of assimilation. The study shows that independently, race is significantly related to language proficiency even though its effects are later diminished. Race also significantly influences political participation; as previously noted, Cuban Hispanics, Afro-Caribbeans and Dominicans all demonstrate distinctive patterns of civic engagement. With regard to out- marriage/partnership, both Cuban Hispanics and Afro-Caribbeans are less likely to out- marry/partner compared to Cuban Whites. Furthermore, results demonstrate that Afro-

Caribbeans are more likely to attain college degrees or higher compared to Cuban Whites and mixed Caribbeans.

Family’s social class is also a significant determinant of assimilation. Of all the measures of social class proposed in the study, parents’ socioeconomic status remains the most salient determinant of assimilation for language proficiency, out- marriage/partnership and particularly educational attainment. Other measures of social

200 class such as family structure and the respondents’ own level of education is also significantly related to assimilation. The respondents’ own level of education is related to political participation, although its effects are subsequently weakened with the addition of new variables.

Gender is also significantly related to assimilation in the research. Both language proficiency and out-marriage/partnership are considerably influenced by gender.

Additionally for immigrant women, the research shows that theories of convergence have created an equalizing effect with their male counterparts in educational attainment. With regard to political participation, my research shows that there are no differences in voting patterns between men and women both for immigrants and non-immigrants.

CONTRIBUTIONS TO THE ASSIMILATION LITERATURE

I use both the classical and contemporary theories of assimilation to theoretically guide the research, however there are a number of shortcomings that limit each perspective from providing an accurate and comprehensive analysis of assimilation. One limitation of contemporary research on assimilation is that the literature has primarily focused on Latin American countries; these studies have conducted comparative analyses primarily with Asians and African immigrant groups. Another common caveat in the literature has always been the lack of attention to Afro-Caribbean immigrants and the ways in which their integration in the United States has differed from other immigrant groups. Accordingly, this dissertation is unique because it investigates the degree to which Cuban immigrants display distinctive patterns of assimilation compared to

201 predominantly Afro-Caribbean immigrants as well as those immigrants of mixed heritage.

I chose to focus on these groups because Cubans characterized in the literature as being upwardly mobile and achieving high levels of assimilation. Correspondingly, racial and ethnic studies have also described Afro-Caribbeans as economically driven to migrate and similarly attaining high levels of upward mobility in the U.S. Since both groups are relatively homogeneous based on mobility, I wanted to further examine the degree to which these immigrants are representative of distinct patterns of integration based on the four aforementioned measures of assimilation. In addition to examining

Caribbean immigrants based on their racial and ethnic self-identification, I extended my research by conducting a sensitivity analysis, which demonstrated differences in patterns of assimilation based on nationality.

The Children of Immigrants Longitudinal Study has an uneven distribution of

Caribbean immigrants and bias towards Cuban immigrants. Since the distribution of

Cubans was substantially greater compared to the other nationalities, I was able to gain a better racial and ethnic representation of the sample by creating five racial and ethnic categories based on the immigrants’ self-identification. Not only did this reclassification of immigrants produce unique results, but I was also able to demonstrate differences in assimilation patterns among Latin American, Afro-Caribbean and mixed Caribbean immigrants. The use of racial and ethnic groups, however, conceals the effects of nationality. By conducting a sensitivity analysis and separating out the effects of

202 nationality I could compare differences in assimilation among Caribbean immigrants based not only on racial and ethnic self-identification but also nationality.

This study shows that race, class and gender all act independently for three of the four models of assimilation. With regard to the out-marriage/partnership model, two-way interaction effects revealed that gender interacts with class and race when predicting assimilation. There were no interaction effects for language proficiency, political participation and educational attainment. Consequently, overall findings of the study suggest that race, class and gender act independently when predicting Caribbean immigrants language proficiency, political participation and educational attainment.

Findings suggest that class is the most consistent and significant predictor of assimilation. Class trumps both race and gender as Caribbean immigrants position themselves into the dominant culture. Wilson (1980) argues that economic class has become more important than race in determining occupational mobility as signaled by the growth of the Black middle class and the concurrent representation of a disproportionately large Black underclass, a marginal position Blacks share with poor

Whites. Parallel to the findings of Wilson (1980), the results of this study indicate that regardless of the immigrants’ racial and ethnic self-identity, class continues to significantly influence all four measures of assimilation. The continuing significance of class is further demonstrated by the out-performance of Cuban White immigrants by

Afro-Caribbeans both in educational attainment and political participation. The salience of class further exemplifies the reproduction of inequality and the ability of the first

203 generation immigrants to transmit both economic resources as well as inherent disadvantage to the second-generation.

The declining significance of race is further exemplified in the research by weak effects of discrimination to significantly predict subsequent assimilation of Caribbean immigrants. These findings challenge the view of Portes and Rumbaut (2006) who argue that class is essentially becoming less important because the new wave of immigrants are confronted with discrimination due to the color of their skin. Additionally, segmented labor markets into highly skilled jobs and menial jobs prevent these immigrants from gaining upward mobility. The authors further argue that these experiences were not encountered by earlier waves of European immigrants. Contrary to these views, this study shows that discrimination is not significantly related to any of the four measures of assimilation; these findings contribute to sociological inquiry and also merit deeper investigation.

Although the assimilation literature typically characterizes Cubans as upwardly mobile, this study shows that this is not a distinguishing feature for this group because

Afro-Caribbeans are essentially out-performing Cubans particularly in educational attainment. One plausible explanation for this is that Afro-Caribbeans are economically motivated to migrate compared to Cubans who politically driven. The research also shows that Afro-Caribbeans are also gaining higher levels of upward mobility compared to Cubans in political participation. Both educational attainment and political participation are two notable measures of assimilation employed in the literature.

204

Gender is a significant correlate of assimilation. Accordingly, the transformation of women’s roles and their entry into the labor market are used to explain significant differences in assimilation in the research (Becker, 1981; Oppenheimer, 1988; Solsona,

1998). Moreover, values, norms and discipline that are inculcated in young girls and a strict childrearing culture geared towards females in the Caribbean help explain why immigrant girls are significantly divergent from boys in language proficiency (Figueroa,

1996).

Convergence theories are also used in the study to possibly explain that for many immigrant women, migration from developing to developed countries closes the gap over subsequent generations and creates an equalizing effect with their male counterparts within the dominant culture. The broader finding is not that gender has no effect on assimilation, but more significantly, one plausible explanation for those models where there are no differences between men and women, could be that the gender gap is essentially closing. As noted in previous studies, this equalizing effect could result in women gaining equal mobility to men over subsequent generations.

HOW DO MODELS OF ASSIMILATION INFORM THE RESEARCH?

The ability of racial and ethnic groups to adapt and assimilate within the dominant culture is beneficial both on the micro and macro level because the advantages are multi- dimensional. Both countries and individuals benefit economically, politically, socially and culturally. The blending of cultures is by no means a seamless and uniformed process; findings suggest that there is the continuous reproduction of inequality, a lack of

205 mobility for many immigrant groups and the permanent struggle for acceptance and recognition for many immigrants who belong to lower social classes (Biblarz, Raftery, and Bucur, 1997; Chen and Kaplan, 2001; Feng et al., 1999).

Proponents of the classical model of assimilation (Park, 1930; Burgess, 1925;

Gordon, 1964) argue that over time immigrant groups completely abandon their native customs and eventually adopt the norms and values of the dominant culture, ultimately achieving complete assimilation. Conversely, segmented assimilation theorists (Portes and Zhou, 1993; Gans, 1992; Zhou and Bankston, 1994) contend that the new wave of immigrants will be faced with a segmented pathway due to their exposure to discrimination, which results from the color of their skin and the restructuring of the economy which in turn presents these immigrants with a segmented labor market. The segmented view further states that unlike earlier waves of European immigrants, the new wave will be less likely to achieve upward mobility and alternatively assimilate downwardly into the underclass.

My findings suggest that contrary to the classical view, immigrants do not always obtain complete assimilation. First, consistent with previous findings, second-generation immigrants are bilingual and are not entirely abandoning their native language.

Furthermore, immigrants have not been studied long enough to provide sufficient evidence, which demonstrates that all immigrants will eventually relinquish their mother tongue and adopt the language of the dominant culture. Second, not all immigrants are assimilating through civic involvement. The study shows that many Caribbean

206 immigrants are maintaining ties with their countries of origin and are thus exhibiting lower rates of political participation. Third, there is the disinclination of some Caribbean immigrants to out-marry/partner. The research shows that immigrants belonging to larger ethnic enclaves are less likely to out-marry/partner. Fourth, not all immigrants are gaining upward mobility through the attainment of a college degree or higher. Diverging from the classical view of assimilation, findings of the research demonstrate that bilingualism, a lack of civic invlovement, intermarriage and the inability of some immigrants to attain a college degree or higher are indicative of the extent to which many

Caribbean immigrants are not gaining complete integration into the dominant culture.

This study also challenges the segmented view of assimilation in which the new wave of immigrants are characterized by the tendency to assimilate downwardly based on exposure to discrimination due to their skin color. As previously noted, findings indicate that discrimination is not significantly related to measures of assimilation. Accordingly, the research shows that the experiences of Caribbean immigrants are not marked by discrimination and that second-generation Caribbean immigrants are in fact gaining upward mobility regardless of their skin color. Consequently, for second-generation

Caribbean immigrants, integration into the dominant culture does not ultimately follow a downward pathway.

207

GENERAL FINDINGS

Hypotheses were constructed based on race, class and gender for each model of assimilation. I also proposed general hypotheses for the study based on additional independent variables such as discrimination, citizenship status and aspirations.

Language Proficiency

For the language proficiency model, I hypothesized that there would be no differences among immigrants based on race and ethnicity due to the three-generational language model. Though race and ethnicity proved to be significant independently of class and gender prior to the introduction of other independent variables, these effects were subsequently diminished. In support of hypothesis one, results showed that race and ethnicity are not significantly related to language proficiency.

I expected that family’s social class would lead to higher levels of English proficiency due to parents’ greater levels of economic resources and human capital.

Results indicated that parents’ socioeconomic status was the only significant indicator of family’s social class for language proficiency. Though other measures of family’s social class were not significant, the primary measure used in the assimilation literature remained salient for this model and supported hypothesis two.

I also hypothesized that at age fourteen, girls should be more proficient in English due to higher expectations placed on girls and stricter child rearing in the Caribbean.

Consistent with hypothesis three, the study showed that gender is related to language proficiency and Caribbean girls are in fact more likely to outperform boys in English

208 during middle school. Overall, results of the language proficiency assimilation model indicate support for all hypotheses.

Political Participation

I hypothesized that immigrants with weaker ties to their countries of origin should be more politically active in the dominant culture compared to those with stronger ties to their national origin. As hypothesized, findings indicated that Cuban Hispanics and Black

Caribbeans were representative of those groups having weaker ties to their countries of origin compared to Cuban Whites. Additionally the sensitivity analysis indicated that

Dominicans were also characteristic of those immigrants with stronger ties to their national origin compared to Haitian and all Cubans.

I expected that higher levels of human capital and economic resources should lead to more political awareness and increased voter registration. Though, parents’ socioeconomic status was not related to political participation findings suggest that the respondents own level of education had significant relationships with political participation prior to the introduction of additional independent variables. Other measures of family’s social class such as economic situation, family structure, homeownership and mother’s regular employment had no significant impacts on the immigrants’ political participation. In general, there is sufficient evidence in support of hypothesized class relationships.

I expected that due to the gender convergence theory in voting; there should be no significant difference between women and men’s voting patterns. Consistent with the

209 gender convergence hypothesis, findings showed that gender has no significant relationship with voter registration, further indicating that over time the gender gap is closing between men and women. Overall findings suggest that all hypotheses for this model were also supported.

Out-Marriage/Partnership

For the out-marriage/partnership model, I hypothesized that immigrants belonging to larger ethnic enclaves should be less likely to out-marry/partner due to less contact with the dominant culture and a broader marriage pool within their enclave. In support the hypothesis, my findings showed that both Cuban Hispanics men and Caribbean

Blacks men were less likely to out-marry/partner. This potentially occurs because they belonged to substantially larger immigrant groups compared to Cuban White men.

I also expected that higher levels of economic resources and human capital should subsequently lead to out-marriage/partnership due to the immigrants establishing friendship ties and being in greater social contact with the dominant culture. Consistent with this hypothesis findings showed that parents’ socioeconomic status is related to out- marriage /partnership even though other measures of family’s social class were not significant in the process of assimilation.

I hypothesized that women should be more likely to out-marry/partner compared to males due to the transformation of gender roles and a higher percentage of women entering the labor force. This hypothesis was also supported in the study because findings

210 suggested that women were out-marrying/partnering at substantially higher rates compared to men. In general, all hypotheses were supported for this model.

Educational Attainment

With regard to the educational attainment model, I hypothesized that Black

Caribbeans should be most likely to attain a college degree or higher because these groups are economically driven to migrate as opposed to politically driven. Results indicate that racial and ethnic self-identification was a more sophisticated measure of predicting educational attainment compared to nationality because findings showed that national origin was not significantly related to educational attainment. Thus, the study produced sufficient evidence to support hypothesis my hypothesis about race and ethnicity.

I also hypothesized that families with higher levels of economic resources and education should rear children who are more likely to attain a college degree or higher.

The presence of positive role models and adequate financial resources should subsequently propel the second-generation into middle class America. Consistent with this hypothesis, both parents’ socioeconomic status and family structure were significant predictors of educational attainment even though other measures of family’s social class were not salient.

With regard to gender, I expected that there should be no significant differences between males and females levels of educational attainment in young adulthood due to the gender convergence hypothesis. This hypothesis was supported, as educational

211 attainment for this population remained gender neutral. Overall, results indicated that all hypotheses were supported for the educational attainment assimilation model.

As mentioned before, in addition to the race, class and gender hypotheses for each model, I proposed relationships for additional independent variables used in the research.

First, I hypothesized that exposure to discrimination should negatively affect the immigrants’ language proficiency, political participation, out-marriage-/partnership and educational attainment. Findings indicated that discrimination had no effect on any of the four measures of assimilation proposed in the study, providing no support for this hypothesis.

Secondly, I hypothesized that U.S. born citizens should be more assimilated compared to naturalized citizens for the political participation model and that U.S. born citizens as well as naturalized citizens should also be more integrated compared non- citizens for the out-marriage/partnership and educational attainment models. The study showed that this hypothesis was only supported for political participation and educational attainment models however, there were no significant differences in citizenship status for the out-marriage/partnership model.

Lastly, I hypothesized that higher aspirations should lead to higher levels of political participation, out-marriage/partnership and educational attainment; this hypothesis was only supported for the educational attainment model.

Generally, most of the hypotheses proposed in the study were supported, however the research produced some new findings, which challenged conventional views of the

212 assimilation literature. These findings are discussed in the concluding sections of the chapter. Some of these new findings were essentially as a result of the unique sample and produced meaningful results from which future bodies of work ought to build. The following sections discuss limitations of the study, findings and suggestions for future research.

LIMITATIONS OF THE STUDY

While providing a more detailed look at Caribbean migration than most other studies, this dissertation has limitations that are broader to the study and others were specific to each model of assimilation. As previously discussed the site of the interviews, age of the respondents, socioeconomic status of the families and the racial and ethnic self-identification of the immigrants create some limitations.

Because the respondents were predominantly from Miami and Fort Lauderdale, the sample used in the study were relatively similar based on residential location. Also, many of the respondents attended the same high schools thus the socioeconomic status of the families were somewhat uniformed. This is shown in the data as very few families in the sample are dependent on public assistance, many of the families identified as middle class and family incomes are also relatively high. In turn, this made it challenging to compare lower income immigrant groups to higher income groups simply because most families were upwardly mobile. Consequently all measures proposed in the study were representative of upward assimilation and I was unable to provide comparisons for downward assimilation. This limitation has also been an advantage to the study because I

213 was able to provide analysis and meaningful results for a unique, segment of the population.

The age of the respondents also restricted the study for some of the assimilation models. One advantage of using young adults was that the investigators were able to track changes over time and incorporate a parent-child relationship. Additionally during young adulthood, the respondents are at the formative years in their lives and the study is able to provide a snapshot of assimilation in young adulthood for this unique cohort.

Conversely, at an average age of twenty-four many respondents had not yet completed college, many had not chosen a partner or spouse and were not politically involved.

Accordingly, the sample size for each model measured in young adulthood was reduced.

The analysis of young adults meant that the attitudes, behaviors and experiences for this cohort of immigrants were not fully stabilized and were subject to change over time.

As previously noted, another limitation was the overrepresentation of Cubans compared to Afro-Caribbeans and mixed Caribbean immigrant groups. This bias towards

Cuban immigrants created some difficulty in effectively measuring race and ethnicity and obtaining significant differences among the five nationalities. The overrepresentation of

Cubans was also a benefit to the study because this allowed me to divide these respondents based on their racial and ethnic self-identification; this in turn provided interesting relationships among Cuban Hispanics, Cuban Whites and Cuban mixed immigrants.

214

The respondents’ racial and ethnic self-identification was both a benefit and limitation to the study. The use of racial and ethnic self identification was advantageous because it allowed me to create unique categories based on the respondents’ own perception of their race and further contrast these groups. However, the disadvantage of this which is likely to be common across many other studies (Kao, 2004; Pong and Hao,

2007; Ellis and White, 2006; Waters and Lieberson, 1987) that employ race and ethnicity as a central component is that whenever respondents self-identify their race it is somewhat difficult to produce meaningful conclusions because not only is race fluid across countries, but also, many immigrants racially self-identify based on what they believe would grant them upward mobility into the dominant culture. As a result, the researcher has to question the extent to which those who self-identify as White in the study are truly White, or perhaps they may perceive whiteness differently from the dominant culture. This blurring of racial and ethnic self-identity made it challenging to arrive at well-defined conclusions on why certain racial groups depict distinct patterns of assimilation. As noted, I was able to abate this limitation to some degree by conducting a sensitivity analysis using nationality. The sensitivity analysis confirmed the findings of the models that used racial and ethnic self-identification and further expounded on the assimilation process by contrasting Cubans with Haitians, Dominicans, Jamaicans and other West Indians.

With regard to the language proficiency assimilation model, one limitation was that most respondents were generally proficient in the ability to speak, read, write and

215 understand English, therefore creating a skewed distribution of the data. In order to create more meaningful results, I analyzed language proficiency as a binary variable and further classified the respondents based on two levels of English proficiency.

For the political participation, registration to vote was the only measure of civic involvement employed in the study due to restrictions of the Children of Immigrants’

Longitudinal Survey. The addition of alternative measures of participation would have been useful in extending the research, and facilitating the examination of those immigrants who maintained political involvement through alternative activities besides voter registration.

With regard to the out-marriage/partnership model, the respondents’ racial and ethnic self-identification as ‘other’ limited the number of categories that could be used to determine out-marriage/partnership, therefore, I was only able to employ those categories with specific self-identities (Cuban White, Cuban Hispanic and Black Caribbean).

Although there were limitations to the study, the study produced results that not only supported the proposed hypotheses but also challenged conventional views in the assimilation literature. The following section provides recommendations for future research.

RECOMMENDATIONS FOR FUTURE RESEARCH

Research findings indicate that Afro-Caribbeans are upwardly mobile in the dominant culture and are correspondingly out-performing Cuban immigrants. To further extend the body of work, a unique contribution to the literature would be a cross-national

216 approach between Afro-Caribbean and African immigrants. The analysis ought to investigate the extent to which assimilation patterns differ among predominantly Black immigrants based on regional and cultural differences. Although previous studies have paralleled Caribbean immigrants to African Americans, the assimilation literature lacks adequate examination of Afro-Caribbean and African immigrants in young adulthood. A cross-national approach between the two groups would also allow for the investigation of downward assimilation due to variations in socioeconomic status. Since many cross- national studies have focused on Asian immigrants and their tendency to gain upward mobility into the dominant culture, further analysis ought to explore the extent to which immigrant groups of the same race but different ethnicity show similar or distinct patterns of assimilation.

As previously noted, gender has been a common inadequacy in the assimilation literature, consequently, future research ought to continue to build upon the relationship between men and women’s assimilation, testing the relevance and applicability of convergence hypotheses across countries. Furthermore, the inclusion of additional variables such as religion, political orientation, friendship ties, attitudes towards

American culture and the immigrants’ involvement transnational activities can further explain the process of assimilation and subsequently create a more complete investigation of immigrants’ adaptations.

Aforementioned, one novel finding of the research is the inability of discrimination to significantly predict the assimilation patterns of Caribbean immigrants.

217

To further build on understanding the reason for this finding, future research ought to examine additional measures of discrimination for this population and also analyze the effects of discrimination cross-nationally, to illustrate to degree to which these findings are consistent across additional immigrant groups.

Moreover, the continuing analysis of race, class and gender will be needed for future studies because these are all longstanding sociological concepts, which are all needed for a complete understanding of assimilation. There has been a longstanding history of theory and debates about immigrants and patterns of assimilation.

As sociologists, we understand that the complexity of immigrants is changing.

Due to distinct assimilation trajectories into the host society, immigrants can no longer be categorized as an impoverished, underprivileged and disadvantaged population who automatically integrate into the underclass. This and many other studies (Kao and Tienda,

1995; Fulgini (1997) show that new immigrants are well educated, highly proficient in

English and politically active—these immigrants are also gaining mobility and blending with the dominant culture in unprecedented ways.

Government policies and laws are correspondingly being enacted and amended taking these transformations into account. The Obama administration continues to advocate for undocumented immigrants, particularly those under the age of 30. In June

2012 the administration announced that law-abiding young undocumented immigrants who came to the U.S. as children will no longer be deported, and instead they would be

218 granted work permits. These provisions would serve to accelerate to the rates of assimilation as well as facilitate the integration of cultures.

The road for immigrants continues to be long and tumultuous, however as the face of immigrants continue to change and their belief in American meritocracy is recognized by those who have the power and ability to enact laws and legislature, as well as those who are willing to embrace and accept these newcomers—only then can the pathway to complete assimilation for this new wave of immigrants truly begin.

219

REFERENCES

Alba, Richard and Nee, Victor. Rethinking Assimilation Theory for a New Era of Immigration in Charles Hirschman, Phillip Kasinitz and Josh DeWind (eds.) The Handbook of International Migration: The American Experience. New York: Russell Sage Foundation, 1999. Alba, Richard and Nee, Victor. Remaking the American Mainstream: Assimilation and Contemporary Immigration. Cambridge, Mass.: Harvard University Press, 2003. Alderman Harold, Orazem Peter, Paterno Elizabeth M. School Quality, School Cost, and the Public/Private School Choices of Low Income Households in Pakistan Journal of Human Resources; 36:304–26, 2001.

Altonji, Joseph, G., and David Card. The Effects of Immigration on the Labor Market Outcomes of Less-Skilled Natives in J.M. Abowd and R.B. Freeman (eds.), Immigration, Trade, and the Labor Market Chicago: University of Chicago Press, 1991.

Aspiazu, Gary. G., Bauer, Scott C., and Spillett, Mary. Improving academic performance of Hispanic Youth: A Community Education Model. Bilingual Research Journal, 22, 127–147, 1998.

Banfield, Edward. The Unheavenly City. Boston, MA: Little Brown, 1974.

Bankston, Carl L. and Min Zhou. Effects of Minority-Language Literacy on the Academic Achievement of Vietnamese Youths in New Orleans. Sociology of Education 68(1): 1-17, 1995. Bankston, Carl L., and Min Zhou. The Social Adjustment of Vietnamese American Adolescents: Evidence for a Segmented-Assimilation Approach. Social Science Quarterly 78(2): 509-523, 1997. Bankston, Carl L. and Min Zhou. Being Well vs. Doing Well: Self-Esteem and School Performance among Immigrant and Nonimmigrant Racial and Ethnic Groups. International Migration Review 36(2): 389-415, 2002.

220

Baretto, Matt A. Ethnic Cues: The Role of Shared Ethnicity in Latino Vote Choice. Paper presented at the annual meeting of the American Political Science Association, Chicago, Illinois, September, 2004. Bean, Frank D. Frisbie, W. Parker. and Eberstein, Isaac. W. Recent Changes in marital instability among Mexican Americans: Has the Trend Been Converging with that of Blacks? Social Forces, 58, pp. 1205-20, 1980.

Becker, Gary. Nobel Lecture: The Economic Way of Looking at Behavior. Journal of Political Economy. Pp: 101:385–409, 1993.

Becker, Gary. A Treatise on the Family. London: Harvard University Press, 1981.

Behrman, Jere. R, Knowles, James. C. Household Income and Child Schooling in Vietnam. World Bank Economic Review. Pp.13:211–56, 1999.

Bennett, Neil G., Ann Klimas Blanc, and David E. Bloom. Commitment and the Modern Union: Assessing the Link between Premarital Cohabitation and Subsequent Marital Stability. American Sociological Review 53:127–38, 1988.

Biblarz, Timothy. J. Adrian. E. Raftery, and Alexander, Bucur. Family Structure and Social Mobility. Social Forces, 75:1319-41, 1997.

Biblarz Timothy, J, Raftery Adrian, E. Family Structure, Educational Attainment, and Socioeconomic Success: Rethinking the Pathology of Matriarchy. American Journal of Sociology. pp;105:321–65, 1999.

Blackwell, Debra L. and Daniel T. Lichter. Mate Selection among Married and Cohabiting Couples. Journal of Family Issues 21:275–302, 2000.

Blackwell, Debra L. and Daniel T. Lichter. Homogamy among Dating, Cohabiting, and Married Couples. The Sociological Quarterly 45:719–37, 2004.

Blau, Peter M. Inequality and Heterogeneity: A Primitive Theory of Social Structure. New York: Free Press, 1977.

Bobo, Lawrence. Race and Beliefs About Affirmative Action: Assessing the Effects of Interests, Group Threat, Ideology, and Racism. in Racialized Politics. eds. David Sears, Jim Sidanius, and Lawrence Bobo. Chicago: University of Chicago Press, 2000.

Bobo, Lawrence and Frank Gilliam. Race, Sociopolitical Participation and Black

221

Empowerment. American Political Science Review 84(2): 379-393, 1990.

Bogen, Elizabeth. Immigration in New York New York : Praeger, 1 xiv, 268 p., [12],1987.

Borjas, George. The earnings of male Hispanic immigrants in the United States. Industrial and Labor Relation Review, 35: 343-53, 1982.

Borjas, George. International Differences in the Labor Market Performance of Immigrants. W.E. Upjohn Institute for Employment Research, Michigan, 1988.

Borjas, George. Friends or Strangers: The Impact of Immigrants on the US Economy. Basic Books, New York, 1990.

Boyd, M.onica. At a Disadvantage: The Occupational Attainments of Foreign-Born Women in Canada.” International Migration Review 18(4):1091–1119, 1984.

Boyd, Monica. Educational attainments of immigrant offspring: success or segmented assimilation? International Migration Review. pp;36(4):1037–60, 2002.

Brandon, Peter. David. The living arrangements of children in immigrant families in the United States. International Migration Review, V36, 416−436, 2002.

Bumpass, Larry L. and R. Kelly Raley. Redefining Single-Parent Families: Cohabitation and Changing Family Reality. Demography 32:97–109, 1995.

Butcher Kristin F. Black Immigrants in the USA: Comparison With Native Blacks and Other Immigrants. Industrial and Labor Relations Review. 47:265–84, 1994.

Bryce-Laporte, Roy, and Delores Mortimer. Caribbean Immigrants to the United States. 2nd ed. Washington, D.C.: Smithsonian Institution, 1983.

Carliner Geoffrey. The language ability of US Immigrants: Assimilation and Cohort Effects. NBER Working Paper W5222, 1995.

Carliner Geoffrey. The Endogeneity Between Language and Earnings: International Analyses,” Journal of Labor Economics 13:246-88, 1995.

Carliner Geoffrey. The Wages and Language Skills of US Immigrants. NBER Working Paper W5763. Chiswick. Washington, DC: The AEI Press, pp. 229-96, 1996.

222

Carliner Geoffrey. 1998. Language Skill Definition: A Study of Legalized Aliens. International Migration Review, 32:877-900, 1998.

Castells, Manuel. End of Millennium. Oxford: Blackwell Publishers pp. 128-149, 2000. Castro, Max, and Thomas D. Boswell. The Dominican Diaspora Revisited: Dominicans and Dominican-Americans in a New Century. The North-South Agenda, Paper No. 53. Miami: North-South Center, University of Miami, 2002.

Chen, Zeng-Yin. and Howard B. Kaplan. Intergenerational Transmission of Constructive Parenting. Journal of Marriage and the Family 63: 17-31, 2001. Chevannes, Barry.What We Sow and What We Reap: Problems in the cultivation of male identity in Jamaica. Grace, Kennedy Foundation Lecture Series, 1999.

Chiswick, Barry R. Sons of Immigrants: Are They at an Earnings Disadvantage? American Economic Review 67:376-80, 1977.

Chiswick, Barry, and DebBurman, Noyna. Pre-school Enrollment: An Analysis by Immigrant Generation.” Social Science Research, 35, 60–87, 2006.

Chiswick, Barry R. The Economic Progress of Immigrants: Some Apparently Universal Patterns. Contemporary Economic Problems. Ed. W. Fellner. Washington, DC: The American Enterprise Institute for Public Policy Research. Pp. 357-399, 1979.

Clasen, Donna. R., and Brown, Bradford B. The Multidimensionality of Peer Pressure in Adolescence.” Journal of Youth and Adolescence, 14, 451-468, 1985. Coleman, James. Social Capital in the Creation of Human Capital. American Journal of Sociology, vol. 94. No. (Supplement), pp. S95-121, 1988. Cornfield, Daniel. B., and Arzubiaga, Angela. Immigrants and Education in the U.S. interior: Integrating and Segmenting Tendencies in Nashville, Tennessee. Peabody Journal of Education, 79, 157–179, 2004.

Crane, Jonathan. The Epidemic Theory of Ghettos and Neighborhood Effects on Dropping Out and Teenage Childbearing. American Journal of Sociology, Vol. 96(5): 1226-1259, 1991. Crassweller, Robert D. Trujillo: The Life and Times of a Caribbean Dictator. MacMillan, New York ,1966.

223

Crosnoe, Robert. Double Disadvantage or Signs of Resilience? The Elementary School Contexts of Children from Mexican Immigrant Families. American Educational Research Journal, 42, 269–303, 2005.

Crosnoe, Robert. Early Child Care and the School Readiness of Children from Mexican Immigrant Families. International Migration Review, 41, 152–181, 2007.

Curtin, Philip D. The Atlantic slave trade: A census. Madison: University of Wisconsin Press, 1969.

DeSipio, Louis. Counting on the Latino Vote Latinos as the New Electorate. Charlottesville, Virginia: University Press of Virginia, 1996.

Dohrenwend, Bruce P. and Robert J. Smith. Toward a Theory of Acculturation. Southwestern Journal of Anthropology 18:30-9, 1962.

Dornbusch, Sanford. M., Ritter, Phillip. L., Leiderman, P. H., Roberts, Donald. F., and Fraleigh, Michael. J. The Relation of Parenting Style to Adolescent School Performance. Child Development, 58 , 1244 -1257, 1987. Duany, Jorge. Dominican Migration to Puerto Rico: a Transnational Perspective Centro Journal, vol. XVII, núm. 001, Spring, pp. 242-269. The City University of New York, 2005.

Dulay, Heidi. C., and Burt, Marina.K. Goofing: an Indicator of Children’s Second Language Learning Strategies. LL, 22, 235-251, 1972.

Dustmann Christian, Van Soest Arthur. Language Fluency and Earnings: Estimation with Misclassified Language Indicators. Review of Economics and Statistics 83(4): 663–674, 2001.

Earls, Felton and Maya Carlson. Promoting Human Capability as an Alternative to Early Crime Prevention. Pp. 141-68 in Integrating Crime Prevention Strategies: Propensity and Opportunity, edited by R. V. Clarke, 1996. Eckerson, Helen F. Immigration and National Origins. Annals of the American Academy of Political and Social Science. The New Immigration 367: 4–14, 1996.

Eisenstein, Zillah. The Color of Gender: Re-imaging Democracy. Los Angeles: University of California Press, 1994.

224

Ermisch J, Francesconi M. Family Matters: Impacts of Family Background on Educational Attainments Economica 68:137–56, 2001.

Espenshade, Thomas .J. and Haishan. Fu. An Analysis of English-Language Proficiency among U.S. Immigrants. American Sociological Review 62:288-305, 1997.

Espiritu, Yen Le. Asian American Pan-ethnicity Bridging Institutions and Identities. Philadelphia: Temple University Press, 1992.

Farley, Reynolds and Richard Alba. The New Second-Generation in the United States. International Migration Review 36(3): 669-701, 2002.

Feliciano, Cynthia. Educational Selectivity in U.S. Immigration: How Do Immigrants Compare to Those Left Behind? Demography 42:131–152.], 2005.

Feng D, Giarrusso R, Bengston Vern. L, Frye N. Intergenerational Transmission of Marital Quality and Instability. Journal of Marriage and the Family. 61:451–63, 1999.

Fernandez-Kelly, M. P., and Schauffler, Divided fates: Immigrant Children in a Restructured U.S. Economy. International Migration Review, 28 pp. 662-6, 1994. Fernandez-Kelly, M. Patricia Social and cultural capital in the urban ghetto: Implication for Economic Sociology of Immigration: Essays in Networks, Ethnicity and Entrepreneurship. New York: Russell Sage Foundation, pp. 213-47.89, 1995.

Fernandez-Kelly, Patricia and Lisa Konczal. Murdering the Alphabet: Identity and Entrepreneurship among Second Generation Cubans, West Indians, and Central Americans. Ethnic and Racial Studies 28 (November): 1153-1181, 2005.

Fong, Eric and Rima Wilkes. Racial and Ethnic Residential Patterns in Canada. Sociological Forum 18: 577-602, 2003.

Foner, Nancy. Race and Color: Jamaican Migrants in London and New York City. International Migration Review 19:708-27, 1985. Fraser, Nancy. Re-thinking the public Sphere: A contribution to the critique of actually existing democracy. In C. Calhoun (Ed.), Habermas and the public sphere, 1992. Frey, William. The New White Flight. American Demographics 16(4): 40-48, 1994.

Frey, William H. and Liaw, Kao.Lee. The Impact of Recent Immigration on Population Redistribution within the United States. In The Immigration Debate: Studies of

225

Economic, Demographic, and Fiscal effects of Immigration, Smith, J.P., Edmonston B. (eds). National Academy Press: Washington DC; 388-448, 1998.

Frey, William .H. and Liaw, Kao-Lee. Internal Migration and Foreign-Born Latinos and Asians: Are They Assimilating Geographically? In Migration and Restructuring in The United States, eds. K. Pandit and S. Davies Withers. Maryland: Rowman & Littlfield Publishers, Inc, 1999.

Frisbie, W. Parker. and Bean, Frank. D. The Latino Family in Comparative Perspective: Trends and Conditions. In C. K. Jacabson (ed.), American Families: Issues in Race and Ethnicity (pp. 29-71) New York: Garland, 1995.

Fuligini Andrew J. The Academic Achievement of Adolescents in Immigrant Families: The Roles of Family Background, Attitudes, and Behavior. Child Development. 68:351– 63, 1997.

Gabaccia, Donna. Seeking Common Ground: Multidisciplinary Studies of Immigrant Women in the United States.” Westport, Connecticut: Greenwood Press, 1992.

Gans, Herbert. Second Generation decline: Scenarios for the economic and ethnic futures of the post-1965 American Immigrants. Ethnic and Racial Studies. Vol. 15. Pp 173-92, 1992. Gibson, Daniel. M., and Jefferson, Renee. N. The Effect of Perceived Parental Involvement and the use of Growth-Fostering Relationships on Self-Concept in Adolescents Participating in Gear Up. Adolescence 41(161), 111-125, 2006.

Glazer, Nathan and Daniel Moynihan, Beyond the Melting Pot (Cambridge, MA:MIT and Harvard University Press, 1963.

Glazer, Nathan. Blacks and Ethnic Groups: The Difference and the Political Difference It Makes. Social Problems 18, 1971.

Glick, Jennifer. E., and Michael. J. White The Academic Trajectories of Immigrant Youths. Demography 40(4): 759–784, 2003. Gordon, Milton. 1964. Assimilation in American Life. New York: Oxford University Press, 1964. Gordon Edmund. W, Bridglall Beatrice L, Meroe Aundra S. Supplementary Education: The Hidden Curriculum of High Academic Achievement. Lanham, MD: Rowman and

226

Littlefield, 2004.

Goss, Jon and Bruce, Lindquist. Conceptualizing International Migration: A Structuration Perspective. International Migration Review (29) 2 317-351, 1995. Gratton Brian. Race, the Children of Immigrants, and Social Science Theory. Journal of American Ethnic History. 21(4):74–84, 2002.

Grawe Nathan D, Mulligan Casey B. Economic Interpretations of Intergenerational Correlations Journal of Economic Perspectives 16(3):45–58, 2002.

Greenbaum, Susan. More Than Black: Afro-Cubans in Tampa. Gainesville, FL” University Press of Florida, 2002. 15-25, 2002.

Grillo, Evelio. Black Cuban, Black American Houston, Texas: Arte Publico Press, 2000.

Halter, Marilyn. Between Race and Ethnicity: Cape Verdean Americans Immigrants 1860-1965 Urbana, IL: University of Illinois Press, 1993.

Handlin, Oscar. The Newcomers: Negroes and Puerto Ricans in a Changing Metropolis” Garden City, NY: Doubleday, 1962.

Handlin, Oscar. The Uprooted. Boston, MA: Little, Brown, 1973.

Harris, Kathleen Mulllan. The Health Status and Risk Behaviors of Adolescents in Immigrant Families. pp. 286-347 in Donald J. Hernandez (ed.), Children of Immigrants: Health adjustment, and Public Assistance. Washington, DC: National Academy Press, 1999. Haveman R, Wolfe B, Spaulding J. Childhood Events and Circumstances Influencing High School Completion. Demography. 28:133–57, 1991.

Hernadez, Donald J. Children of Immigrants: Health, Adjustment, and Public Assistance. Washington, DC: National Academy Press, 1999.

Hernandez, Donald. J., and Drake, K. Socioeconomic and demographic risk factors among children in immigrant and native born families: 1910, 1960, and 1990. In D. J. Hernandez (Ed.), Children of Immigrants: Health Adjustment and public assistant (pp. 19-53) Washington, DC: National Academy Press, 1999. Higham, John. Send These to Me: Jews and Other Immigrants in Urban America. New

227

York: Athenaeum, 1975.

Hirschman, Charles. The Educational Enrollment of Immigrant Youth: A Test of the Segmented Assimilation Hypothesis.” Demography. Vol. 38 (August), pp. 317-336, 2001. Holder, Calvin. The Rise of the West Indian Politician in New York City, 1900-1952. Afro-Americans in New York Life and History. 1980.

Holder, Calvin. The Causes and Composition of West Indian Immigration to New York City, 1900-1952. Afro-Americans in New York Life and History 11, no. 2 January 1987.

Holder, Calvin. Making Ends Meet: West Indian Economic Adjustment in New York City, 1900-1952. Wadabagei: A Journal of the Caribbean and its Diaspora 1, no. 1 Winter/Spring 1998.

Hondagneu-Sotelo, Pierrette., and Cynthia. Cranford. Gender and Migration. In Handbook of the Sociology of Gender. Ed. J. S. Chafetz. New York: Kluwer. Pp. 105– 126, 1999.

Hochschild, Jennifer. Facing Up to the American Dream. Princeton: Princeton University Press, 1996.

Jasso, Guillermina. and Mark.R. Rosenzweig. The New Chosen People: Immigrants in the United States. New York: Russell Sage Foundation, 1990.

Jencks, Christopher, and Susan. E. Mayer. Residential Segregation, Job Proximity, and Black Job Opportunities Inner City Poverty in the United States. Ed. E. Laurence, J. Lynn, and G. H. M. Michael. Washington, DC: National Academy Press, 1990. Jensen, Lief. The Demographic Diversity of Immigrants and their Children. In Rueben G. Rumbaut and Alejandro Portes (eds), Ethnicities: Children of Immigrants in America, Berkley, CA: University of California Press and Russell Sage Foundation pp. 21-56, 2001.

Johnson, Jacquelin. and Newport, Elissa. Critical period effects in second language learning: The influence of maturational state on the acquisition of English as a second language. Cognitive Psychology, 21, 60-99, 1989.

Joyner, Kara and Grace Kao. Interracial Relationships and the Transition to Adulthood.

228

American Sociological Review 70:563–81, 2005.

Junn, Jane. Participation in a Liberal Democracy: The Political Participation of Immigrants and Ethnic Minorities in the United States. American Behavioral Scientist, 42, 1417-38, 1999.

Kallen, Horace. Culture and Democracy in the United States: Studies in the Group Psychology of the American Peoples. New York: Boni and Liveright, 1924.

Kalmijn, Matthijs. The Socioeconomic Assimilation of Caribbean American Blacks Social Forces, 74(3): 911-930, 1996.

Kao Grace, Tienda Marta. Optimism and Achievement: The Educational Performance of Immigrant Youth. Social Science Quarterly. 76:1–19, 1995.

Kao, Grace. Psychological Well-Being and Educational Achievement Among Immigrant Youth. Pp. 410-477 in Donald J. Hernandez (ed.), Children of Immigrants: Health, Adjustment, and Public Assistance. Washington, DC: National Academy Press, 1999.

Kao Grace. Parental Influences on the Educational Outcomes of Immigrant Youths. International Migration Review. 2004;38:427–49, 2004.

Kao, Grace and Rutherford, Lindsay. T. Does Social Capital Still Matter; Immigrant Minority Disadvantage in School-Specific Social Capital and its Effects on Academic Achievement. Sociological Perspectives, 50, 27–52, 2007.

Kasinitz, Philip. Caribbean New York: Black Immigrants and the Politics of Race. Ithaca, N.Y.: Cornell University Press, 1992.

Kasinitz, Philip, Mollenkopf John H, Waters Mary C, Holdaway Jennifer. Inheriting the City: The Children of Immigrants Come of Age. New York: Russell Sage Foundation; 2008.

Kasinitz, Philip, Mollenkopf John, Waters Mary C. Becoming New Yorkers: Ethnographies of the New Second Generation. New York: Russell Sage Foundation; 2004.

Kasinitz, Philip, Mollenkopf John, Waters Mary C. Becoming American/becoming New Yorkers: Immigrant Incorporation in a Majority Minority City. International Migration Review. 36(4):1020–36, 2002.

229

Knight, Franklin W. The Caribbean: The Genesis of a Fragmented Nationalism. 2nd ed. New York: Oxford University Press 10-18, 1990.

Koser, Khalid. and Salt, John. The Geography of Highly Skilled International Migration. International Journal of Population Geography, 3: 285–303, 1997. Koser, Khalid. Social Networks and the Asylum Cycle: The Case of Iranians in the Netherlands. International Migration Review, Vol.31 (3):591-611, 1997. Kossoudji Sheme A. English Language Ability and the Labor Market Opportunities of Hispanic and East-Asian Immigrant Men. Journal of Labor Economics 6(2): 205–228, 1988. Kristol, Irving. The Negro Today Is Like the Immigrant of Yesterday. New York Times Magazine, 11 September pp. 50-124, 1966.

Labov, Teresa G. and Jerry A. Jacobs. Preserving Multiple Ancestry: Intermarriage and Mixed Births in Hawaii. Journal of Comparative Family Studies 29:481–502, 1998.

Landale, Nancy. S., and Ogena, Nimfa. B. Migration and union dissolution among Puerto Rican women. International Migration Review, 29, pp. 671-92, 1995. Landale, Nancy. S. and Oropesa, R. Salvador. Immigrant Children and the children of immigrants: Inter and Intra-ethnic group differences in the United States. Population Research Group Research Paper 95-02, 1995. Larzelere, Alex. The 1980 Cuban Boatlift. Washington, DC: National Defense University Press, 1988.

Lee, Sharon M. and Barry Edmonston. New Marriages, New Families: U.S. Racial and Hispanic Intermarriage. Vol. 60. Washington, DC: Population Reference Bureau, 2005.

Lichter, Daniel T. and Zhenchao Qian. Marriage and Family in a Multiracial Society. New York and Washington, DC: Russell Sage Foundation and Population Reference Bureau, 2004.

Lichter, Daniel T., Zhenchao Qian, and Leanna M. Mellott. Marriage or Dissolution? Union Transitions among Poor Cohabiting Women. Demography 43:223–40, 2006.

Lieberson, Stanley. A Piece of the Pie: Blacks and White Immigrants since 1880. Berkeley, CA: University of California Press, 1980.

230

Lien, Pei-te. The Making of Asian America Through Political Participation. Philadelphia: Temple University Press, 2001.

Lien, Pei-te, M. Margaret Conway and Janelle Wong. The Politics of Asian Americans Diversity and Community. New York: Routledge, 2004.

Light, Ivan. Ethnic Economies. In Handbook of Economic Sociology. Ed. N. Smelser and R. Swedberg. New York: Russell Sage Foundation, 1994.

Lipset, Seymour M. Changing social status and prejudice : the race relations theories of pioneering American Sociologist. Commentary 9:475-9, 1950.

Lopez, David. E., and Stanton-Salazar, Ricardo. D. Mexican Americans: A Second- Generation at Risk. In R. G. Rumbaut & A. Portes (Eds.), Ethnicities, 2001.

Lowenthal, David, West Indian Societies. London Oxford Press, 1972.

Luxán, M., Miret, P. and Treviño, R. Is the male-provider model still in place? Partnership formation in contemporary Spain in Gender inequalities in Southern Europe. edited by González, M.J., Jurado, T. y Naldini, M.. Women and work welfare in the 1990s, Vol.4, 1999.

Lyman, Stanford M. The Race Relations Cycle of Robert E Park.” Pacific Sociological Review 2:16-22. Attending Selective Colleges and Universities in the United States.” American Journal of Education 113:243-71, 1950.

Magnuson, Katherine., Lahaie, Claudia., and Waldfogel, Jane. Preschool and School Readiness of Children of Immigrants. Social Science Quarterly, 87, 1241–1262, 2006.

Massey, Douglass, and Nancy. Denton. The Dimensions of Residential Segregation. Social Forces 67:281–315, 1988.

Massey, Douglas. S., Social Structure, Household Strategies and the Cumulative Causation of Migration. Population Index, Vol. 56, No. 1, pp. 3-26, 1990. Massey, Douglas. S., Juaquin Arango, Graeme Hugo, Ali Kouaouci, Adfela Pellegrino, J. Edward Taylor. Theories of International Migration: A Review and Appraisal. Population and Development Review, Vol. 19, No. 3, pp. 431-466, 1993.

231

Massey, Douglas. S., Juaquin Arango, Graeme Hugo, Ali Kouaouci, Adela Pellegrino, J. Edward Taylor. An Evaluation of International Migration Theory: The North American Case. Population and Development Review, Vol. 20, No. 4, pp. 699-751, 1994. Massey, Douglas S. The New Immigration and Ethnicity in the United States. Population and Development Review 21:631–52, 1995. Massey, Douglas. S., Margarita. Mooney, Kimberly. Torres, and Camille. Charles. Black Immigrants and Black Natives. American Journal of Education 113 February, by The University of Chicago, 2007. Massiah, Joycelyn. Women in the Caribbean Project, Institute of Social and Economic Research (ISER), UWI: Cave Hill, 1986.

Mazrui, Ali. A. The African Diaspora and Globalization. Paper presented at the conference on African Diaspora. Suny Binghamon. Bloomington: Indiana University Press, 1996.

McKenney, Nampeo. R., and Cresce, Arthur. R. Measurement of Ethnicity in the United States: Experiences of the U.S. Census Bureau. In Challenges of Measuring an Ethnic World: Science, politics an d reality. Statistics Canada and U.S. Bureau of the Census. Washington, D.C, 1992.

McLaughlin, James., Liljestrom, Ana., Lim, Jae. Hoon and Meyers, Dawn. A Community Study about Latino Immigrants and Education. Education and Urban Society, 34, 212– 232, 2002.

McManus Walter, S. Labor Market Costs of Language Disparity: an Interpretation of Hispanic Earnings Differences. The American Economic Review 75: 818–827, 1985.

McManus Walter, Gould William, Welch Finis. Earnings of Hispanic Men: the Role of English Language Proficiency. Journal of Labor Economics 1(2): 101–130, 1983.

Miller, Errol. The Marginalisation of the Black Jamaican Male: Insights from the Development of the Teaching Profession. Kingston: Institute of Social and Economic Research, UWI; and (1991) Men at Risk. Kingston: Jamaica Publishing House, 1986.

Mintz, Sydney, W. Caribbean Transformations. Chicago: Aldine Pub. Co, 1974.

Miyares, Ines, M.; Skop, Emily, H.; Skop, Emily H. The Magnetism of Miami: Segmented Paths in Cuban Migration. Geographical Review (American Geographical

232

Society) 87 (4): 504–519, 1997.

Model, Suzanne. Caribbean Immigrants: A Black Success Story? International Migration Review 25, no. 2, 1991.

Model, Suzanne. West Indian Prosperity: Fact or Fiction? Social Problems 42, no. 4 November, 1995.

Morenoff, Jeffrey and Robert J. Sampson. Violent Crime and the Spatial Dynamics of Neighborhood Transition: Chicago, 1970- 1990. Social Forces 76:31-64, 1997. Nagasawa, Richard, Zhenchao, Qian, and Paul Wong. Theory of Segmented Assimilation and the Adoption of Marijuana Use and Delinquent Behavior by Asian Pacific Youth. The Sociological Quarterly 42(3): 351-372, 2001. Newbold, K.Bruce. Evolutionary Immigrant Settlement Patterns: Concepts and Evidence. In Migration and Restructuring in The United States, eds. K. Pandit and S. Davies Withers. Maryland: Rowman & Littlefield Publishers, Inc, 1999.

Nijman, Jan. Globalization to a Latin Beat: The Miami Growth Machine. Annals of the American Academy of Political and Social Science 551 (1): 164–177, 1997.

Okazaki, Sumie., and Sue, Stanley. Cultural Considerations in Psychological Assessment of Asian Americans. In J. N. Butcher (Ed.), Clinical personality assessment: Practical approaches (pp. 107-119). New \brk:Oxford Press, 1995. Oppenheimer, Valerie. A Theory of Marriage Timing. The American Journal of Sociology 94.: 563-591, 1988.

Padilla, Felix. Latino Ethnic Consciousness: The Case of Mexican Americans and Puerto Ricans in Chicago. Notre Dame, Indiana: University of Notre Dame Press, 1985.

Palmer, Ransford. A Decade of West Indian Immigration to the United States, 1962- 1972: An Economic Analysis. Social and Economic Studies, 23:571-587, 1974.

Palmer, Ransford, In Search of a Better Life: Perspectives on Migration from the Caribbean. New York: Praeger Publishers, 1990.

Palmer, Ransford. Pilgrims from the Sun: West Indian Migration to America. New York: Twayne Publishers, 1995.

233

Park, Robert. E. Racial Assimilation in Secondary Groups. American Sociological Society. 8: 66-72, 1914.

Park, Robert E. Assimilation. Social Pages 281-83 in Encyclopedia of the Social Sciences. Ed. E.R.A. Seligman and A. Johnson. New York: Macmillan Company, 1930.

Park, Robert E., and Ernest W. Burgess. Introduction to the Science of Sociology. Chicago: University of Chicago Press., 1969.

Pedraza, Silvia. Women and Migration: The Social Consequences of Gender. Annual Review of Sociology 17:303–325, 1991.

Pérez Carreón, G., Drake, Corey., and Barton, Angela. C. The Importance of Presence: Immigrant Parents’ School Engagement Experiences. American Educational Research Journal, 42, 465–498, 2005.

Perlmann, Joel and Roger Waldinger. Second Generation Decline? Children of Immigrants, Past and Present – A Reconsideration. International Migration Review 31(4): 893-922, 1997. Pessar, Patricia. A Visa For a Dream: Dominicans in the United States Boston, Mass. Allyn and Bacon, 1995. Pessar, Patricia. The Role of Gender, Households, and Social Networks in the Migration Process: A Review and Appraisal. In: The Handbook of International Migration: The American Experience. Charles Hirschman, Philip Kasinitz, Josh DeWind (eds.) New York: Russell Sage Foundation, 1999. Pessar, Patricia. Engendering Migration Studies: The Case of New Immigrants in the United States." American Behavioral Scientist 42: 577-600, 1999.

Pessar, Patricia. Gender and Family. In: The New Americans: A Guide to Immigration Since 1965. Mary C. Waters and Reed Ueda with Helen B. Marrow (eds.); Cambridge, Mass.: Harvard University Press, 2007. Phillips, Anne. Democracy and Difference. University Park: Pennsylvania State University Press, 1994. Phizacklea, Annie. One Way Ticket: Migration and Female Labor. London: Routledge, 1983.

Pinderhughes, Dianne. The Role of African American Political Organizations in the

234

Mobilization of Voters. in From Exclusion to Inclusion, The Long Struggle for African American Political Power. eds. Ralph Gomes and Linda Faye Williams. Westport, CT: Greenwood Press, 1992.

Portes, Alejandro; Jensen, Leif The Enclave and Entrants: Patterns of Ethnic Enterprise in Miami Before and After Mariel. American Sociological Review 54 (6): 929–949 1989.

Portes Alejandro and Zhou, Min. The new second generation: Segmented assimilation and its variants among Post-1965 Immigrant youth. The Annals of the American Academy of Political and Social Sciences, vol. 530 pp. 74-96, 1993. Portes, Alejandro, and Richard Schauffler. Language and the Second Generation: Bilingualism Yesterday and Today. International Migration Review 28(4):640-6l, 1994. Portes, Alejandro and Jozsef Borocz. Contemporary Immigration: Theoretical Perspectives on its Determinants and Modes of Incorporation. In: Theories of Migration. Robin Cohen (ed.). Cheltenham, UK ; Brookfield, Vt., US : E. Elgar, 1996. Portes, Alejandro and Truelove, Cynthia. Making Sense of Diversity: Recent Research on Hispanic Minorities in the United States.” New York: McGraw-Hill, 1996. Portes, Alejandro. and MacLeod, Dag. Educational Progress of Children of Immigrants: The Role of Class, Ethnicity, and School Context”, Sociology of Education, 69(4): 255- 75, 1996.

Portes, Alejandro and Richard Schauffler. Language and the Second Generation: Bilingualism Yesterday and Today. In The New Second Generation, edited by Aljandro Portes. New York: Russell Sage Foundation, 1996. Portes, Alejandro and Rumbaut, Rueben G. Immigrant America: A Portrait. Berkley CA: University of California Press, 1996. Portes, Alejandro Social Capital: Its origins and applications. In modern sociology.” Annual Review of Sociology, vol. 24 Pp. 1-24, 1998. Portes, A. and MacLeod, Dag. Educating the Second Generation: Determinants of Academic Achievement among Children of Immigrants in the United States, Journal of Ethnic and Migration Studies, 25(3): 373-96, 1999.

Portes, Alejandro. “Legacies: The Story of the Immigrant Second-Generation. Berkley, CA: University of California Press and Russell Sage Foundation, 2001.

235

Portes, Alejandro and Josh DeWind. A Cross-Atlantic Dialogue: The Progress of Research and Theory in the Study of International Migration. International Migration Review 38(3): 828-851, 2004. Portes Alejandro, Fernandez-Kelly Patricia, Haller William. Segmented Assimilation on the Ground: the New Second-Generation in Early Adulthood.” Ethnic and Racial Studies. 28(6):1000–40, 2005.

Portes Alejandro, Rumbaut Ruben G. Legacies: The Story of the Immigrant Second Generation.” Berkeley, CA: University of California Press, 2001.

Poyo, Gerald E. The Anarchist Challenge to the Cuban Independence Movement, 1885- 90. Cuban Studies/Estudios Cubanos, 1985.

Qian, Zhenchao.. Breaking the Racial Barriers: Variations in Interracial Marriage between 1980 and 1990. Demography 34:478–500, 1997.

Qian, Zhenchao and José A. Cobas. Latinos’ Mate Selection: Variations by National Origin, Race, and Nativity. Social Science Research 33:225–47, 2004.

Qian, Zhenchao and Daniel T. Lichter. Measuring Marital Assimilation: Intermarriage Among Natives and Immigrants. Social Science Research 30:289–312, 2001.

Reid, Ira de A., The Negro Immigrant: His Background, Characteristics and Social Adjustment.” 1899-1937 (New York: Columbia University Press, 1939.

Reimers, David. Still the Golden Door: The Third World Comes to America. Columbia University Press, 1985.

Robert, William Fogel, Slavery in the New World”. Without Consent or Contract: The Rise and Fall of American Slavery. p 21-23, 1994.

Rong, Xue Lan and Frank Brown. The Effects of Immigrant Generation and Ethnicity of Educational Attainment among Young African and Caribbean Blacks in the United States.” Harvard Educational Review 71(3): 536-565, 2001. Romano, Renee C. Race Mixing: Black-White Marriage in Postwar America. Cambridge, MA: Harvard University Press, 2003. Roopnarine, J. and Evans, M. Family Structural Organisation, Mother-Child and Father- Child Relationships and Psychological Outcomes in English-speaking African Caribbean

236 and Indo-Caribbean Families.” Prepared for participants in the meeting of researchersFort Young Hotel, Dominica, May 25th to 28th 2006. Prepared by Janet Brown and Sian Williams, 2005.

Roopnarine, J. L., Krishnakumar, A., Metindogan, A., and Evans, M. Links Between Parenting Styles, Parent-Child Academic Interaction, Parent-School Interaction, and Early Academic Skills and Social Behaviors in Young Children of English-Speaking Caribbean Immigrants. Early Childhood Research Quarterly, 21, 238–252, 2006.

Root, Maria P. Love’s Revolution: Interracial Marriage. Philadelphia, PA Temple University Press, 2001.

Roy, Prodipto. The Measurement of Assimilation: the Spokane Indians. American Journal of Sociology 67:541-51, 1962.

Rumbaut, Rueben. G. The Crucible within: Ethnic Identity, Self Esteem and Segmented Assimilation among Children of Immigrants. International Migration Review 28 (4):748- 94, 1994.

Rumbaut, Ruben. Ages, Life Stages, and Generational Cohorts: Decomposing the Immigrant First and Second-Generations in the United States. International Migration Review, 38, 1160–1205, 2004.

Salvo J, Ortiz R. The Newest New Yorkers: An Analysis of Immigration into New York City during the 1980s. June. DCP #92-16, 1992.

Sampson, Robert J., Stephen Raudenbush, and Felton Earls. Neighborhoods and Violent Crime: A Multilevel Study of Collective Efficacy. Science 277:918-24, 1997. Samms-Vaughan, M. The Profiles Project, Report No.1, A Profile of the Status of Jamaican Preschool Children and their Learning Environments.” Department of Child Health with Caribbean Child Development Centre, UWI: Mona, 2001.

Samms-Vaughan, M. A Comprehensive Analysis of Jamaican Children’s Exposure to Violence at 11-12 Years. Published by PAHO, 2005.

Sassen, Saskia New York City’s informal economy. In A Portes , M. Castells and L. A. Benton (eds), The Informal Economy: Studies in Advanced and Less Developed Countries, Baltimore MD: The John Hopkins University Press, pp. 60-77, 1989.

237

Sassler, Sharon. Gender and Ethnic Differences in Marital Assimilation in the Early Twentieth Century. International Migration Review 39:608–36, 2005.

Schaller, A., Rocha, L. O., & Barshinger, D. (2007). Maternal Attitudes and Parent Education: How Immigrant Mothers Support their Child’s Education Despite Their Own Low Levels of Education. Early Childhood Education Journal, 34, 351–356, 2007.

Schuman, Howard, Charlotte Steeh, Lawrence Bobo, and Maria Krysan. Racial Attitudes in America: Trends and Interpretations. Cambridge, MA: Harvard University Press, 1997.

Semyonov, M., Lerenthal, T. Country of origin, gender, and the attainment of socioeconomic status: a study of stratification in the Jewish population of Israel. Research in Social Stratification and Mobility 10, 327–345, 1991.

Senior, Olive. Working Miracles: Women’s Lives in the English-Speaking Caribbean. ISER, UWI: Cave Hill. London: James Currey. Bloomington: Indiana University Press, 1991.

Service, E. and Craik, F.I.M. Differences Between Young and Older Adults in Learning a Foreign Vocabulary. Journal of Memory and Language, 33, 59-74, 1993.

Simmons, Tavia and Martin O’Connell. Married-Couple and Unmarried-Partner Households: 2000. Washington, DC: U.S. Census Bureau, 2003.

Sjaastad, A. Larry. The Costs and Returns of Human Migration. The Journal of Political Economy, Vol. 70, No. 5, Part 2: Investment in Human Beings (Oct., 1962), pp. 80-93, 1962. Skop, Emily H. and Miyares, Ines M. The Magnetism of Miami: Segmented Paths in Cuban Migration. Geographical Review (American Geographical Society) 87 (4): 504– 51, 1997.

Skop, Emily H. Race and Place in the Adaptation of Mariel Exiles. International Migration Review 35 (2): 449–471, 2001.

Smith, Elsie J., Ethnic Identity Development: Toward the Development of a Theory Within the Context of Majority/ Minority Status. Journal of Counseling and Development, 70 pp. 181-88, 1991.

Smith, James. Assimilation Across the Latino Generations. American Economic Review.

238

2003; 93(2):315–19, 2003.

Smith, James and Barry. Edmonston. The New Americans: Economic, Demographic, and Fiscal Effects of Immigration. Washington: National Academy Press, 1997.

Sniderman, Paul and Thomas Piazza. The Scar of Race. Cambridge, MA: Harvard University Press, 1993.

Sohn, Soomin, and Wang, X. Christine. Immigrant Parents’ Involvement in American Schools: Perspectives from Korean Mothers. Early Childhood Education Journal, 34, 125–132, 2006.

Solsona, M. 1998. The Second Demographic Transition from a gender perspective. Innovation 11: 211-225, 1998.

Soong, Roland. Racial Classifications in Latin America Zona Latina. Retrieved 2007-12- 22, 1999.

Sowell, Thomas. Ethnic America. New York: Basic Books, 1981

Spolsky, Bernard. Maori Bilingual Education and Language Revitalization. J. Multilingual and Multicultural Development, 10, 2, 89-106, 1989.

Spolsky, Bernard. Review of Key Issues in Bilingualism and Bilingual Education'. Applied Linguistics 10 (4), 449-451, 1989.

St-Hilaire, Aonghas. The Social Adaptation of Children of Mexican Immigrants: Educational Aspirations Beyond Junior High School. Social Science Quarterly 83(4): 1026-1043, 2002. Stevens, Gillian. The Social and Demographic Context of Language Use in the United States. American Sociological Review 57(2):171-85, 1992. Stevens, Gillian. Age at Immigration and Second Language Proficiency among Foreign- Born Adults." Language in Society 28(4):555-78, 1999. Stier, Haya. Immigrant women go to work: Analysis of immigrant wives labor supply for 6 Asian groups. Social Science Quarterly, 72(1), 67-82, 1991.

Suarez-Orozco, Carola and Marcelo M. Suarez-Orozco. Children of Immigration. Cambridge, MA: Harvard University Press, 2001.

239

Suchlicki, Jaime. The Cuban Revolution: A Documentary Bibliography, 1952-1968. Miami: University of Miami, 1968.

Sudhanshu Handa. Female Schooling Achievement in Jamaica: A Market and non- Market Analysis. Department of Economics, UWI (Mona) Parry, O. (2000) Student Choices in Kingston High Schools, 1996.

Sullivan, Mercer L. Getting Paid: Youth Crime and Work in the Inner City. Ithaca, NY: Cornell University Press, 1989.

Taft, Ronald. A Psychological Model of the Study of Assimilation. Human Relations 10:141-56, 1957.

Taubman P. The Role of Parental Income on Educational Attainment. American Economic Review. 79:57–61, 1989.

Tienda, Marta. and Karen. Booth. Gender, Migration, and Social Change." International Sociology, 6: 51-72, 1991.

Uyeki, Eugene. S. Correlates of ethnic Identification. American Journal of Sociology 65:468-74, 1960.

Veltman, Calvin. Language Shit in the United States. Mouton. . 1988. "Modeling the Language Shift Process of Hispanic Immigrants. International Migration Review 22(4): 545-62, 1983.

Waldinger, Roger and Cynthia Feliciano. Will the New Second Generation Experience ‘Downward Assimilation? Segmented Assimilation Re-Assessed.” Ethnic and Racial Studies 27(3): 376-402, 2004. Waldinger, Roger. Did Manufacturing Matter? The Experience of Yesterday's Second Generation: A Reassessment. International Migration Review. 41(1):3–39, 2007.

Waldinger, Roger, Perlman Joel. Second Generations: Past, Present, Future. Journal of Ethnic and Migration Studies. 24(1):5–24, 1998.

Warner, W.L., and Srole, Leo. The Social Systems of American Ethnic Groups. New Haven, CT: Yale University Press in A. Portes (ed.), The New Second Generation, New York: Russel Sage, pp.197-220, 1945. Waters, Mary C. Ethnic Options: Choosing Identities in America. Berkeley: University of

240

California Press. Cloth and paper, 1990.

Waters, Mary C. The Role of Lineage in Identity Formation Among Black Americans. Qualitative Sociology vol. 14, no. 1, Spring, 1991.

Waters, Mary C. Ethnic and Racial Identities of Second-Generation Black Immigrants in New York City. International Migration Review, vol. 28 pp. 795-820, 1994.

Waters, Mary C 1996. West Indian Family Resources and Adolescent Outcomes, paper presented at the meetings of the American Association for the Advancement of Science, Baltimore, February, 1996.

Waters, Mary C. Black Identities: West Indian Immigrants Dream and American Realities. Cambridge: Harvard University Press, 1999.

Watkins-Owens, Irma. Blood Relations: Caribbean Immigrants and the Harlem Community1900-1930. Bloomington: Indiana University Press, 1996.

Wilson, Kenneth L., and W. Allen Martin. Ethnic Enclaves: A Comparison of the Cuban and Black Economies in Miami. American Journal of Sociology 88(1):135-160, 1982.

Wilson, William J. The Truly Disadvantaged. University of Chicago Press, 1987.

Wilson, William J. The Declining Significance of Race University of Chicago Press, 1980.

Wolfinger, Raymond. E., and Rosenstone, Steven. J. Who Votes CT: Yale University Press, 1980.

Wirth, Louis, The Ghetto” Chicago: University of Chicago Press, 1928.

Woldemikael, Teckle.M. Becoming Black American: Haitians and Americans Institutions in Evanston, Illonois. New York: AMS Press, 1989.

Wright, Richard., and Mark, Ellis. Race, Region, and the Territorial Politics of Immigration in the United States. International Journal of Population Geography 6(3):197–211, 2000.

Zephir, Flore. Trends in Ethnic Identification among second generation Haitian Immigrants in New York City. Westport, Conn.: Bergin & Garvey, 2001.

241

Zhou, Min and Bankston III, Carl N. 1996. Social Capital and Adaptation of the Second Generation: The Case of Vietnamese Youth in New Orleans, in A. Portes (ed.), The New Second Generation, New York: Russel Sage, pp.197-220, 1996.

Zhou, Min. 1997. Segmented Assimilation: Issues, Controversies, and Recent research on the New Second Generation. International Migration Review 31, (4): 975-1008, 1997. Zhou, Min. Growing up American: The Challenge Confronting Immigrant Children and Children of Immigrants. Annual Review of Sociology 23: 63-95, 1997. Zhou, Min. Growing up American: how Vietnamese children adapt to life in the United States Carl L. Bankston, III New York : Russell Sage Foundation, 1998.

242

APPENDIX A: DESCRIPTIVE STATISTICS

Standard Mean/Percent Deviation Dependent Variables Language Proficiency 75.52 0.430 Political Participation 77.94 0.415 Out-Marriage/Partnership 26.72 0.442 Educational Attainment 27.81 0.448 Independent Variables Gender Males 47.67 0.499 Females 52.33 0.499 Race Cuban Whites 26.94 1.350 Cuban Hispanics 28.57 0.412 Cuban Other 14.58 0.314 Caribbean Black 18.58 0.348 Caribbean non-Black 11.32 0.280 Nationality Cubans 68.84 0.688 Dominicans 5.90 0.235 Haitians 9.99 0.300 Jamaicans 8.76 0.282 West Indians 6.51 0.247 Social Class Parents' socioeconomic status 0.09 0.693 Homeownership 66.30 0.473 Current economic situation 45.40 0.498 Mother employed regularly 90.10 0.299 Grade point average 2.29 0.905 Respondents' highest number of school 14.46 1.861 Family Structure 54.66 0.498 Other Independent Variables Length of stay in the US 60.67 0.489 Respondent' lives with mother 9.40 0.237 Continued

Table A1: Mean and Standard Deviation for all Variables Used in the Study.

243

Table A1 continued.

Discrimination 15.34 0.499 Educational Aspirations High school Aspirations 16.24 0.369 College Aspirations 35.47 0.479 Graduate Aspirations 48.28 0.499 Citizenship status Citizens 77.08 0.600 Non-Citizens 15.48 0.262 Naturalized citizens 7.44 0.362

244

APPENDIX B: CORRELATION MATRICES FOR ASSIMILATION MODELS

245

engpro~b Gender2 Gender1 nation~1 nation~2 nation~3 nation~4 nation~5 parent~s GPA usstay disc livewm~r

engprof_b 1.0000 Gender2 0.1037 1.0000 Gender1 -0.1037 -1.0000 1.0000 nationality1 -0.0216 -0.1271 0.1271 1.0000 nationality2 -0.0169 0.0438 -0.0438 -0.3820 1.0000 nationality3 -0.0002 0.0435 -0.0435 -0.4906 -0.0824 1.0000 nationality4 0.0367 0.0602 -0.0602 -0.4544 -0.0763 -0.0980 1.0000 nationality5 0.0156 0.0752 -0.0752 -0.4015 -0.0674 -0.0866 -0.0802 1.0000 parentses 0.0993 -0.1282 0.1282 0.0817 -0.0590 -0.1297 0.0489 0.0042 1.0000 GPA 0.1866 0.1175 -0.1175 -0.0090 -0.0501 -0.0031 0.0415 0.0223 0.1923 1.0000 usstay 0.1337 -0.0684 0.0684 0.2196 -0.0819 -0.1036 -0.1753 -0.0109 0.1742 0.0335 1.0000 disc -0.0321 -0.0330 0.0330 0.2471 -0.0380 -0.1187 -0.1688 -0.0934 -0.0111 0.0019 0.0621 1.0000 livewmother 0.0394 0.0252 -0.0252 -0.0136 -0.0029 -0.0044 -0.0046 0.0384 0.0234 0.0786 -0.0030 -0.0060 1.0000 econsit 0.0613 -0.0257 0.0257 0.0091 -0.0479 -0.0658 0.0670 0.0325 0.2899 0.0121 0.1060 -0.0007 -0.0080 homeown 0.0441 -0.0968 0.0968 0.0348 -0.0546 0.0151 0.0001 -0.0303 0.5487 0.1421 0.1460 0.0028 0.0346 motheremp 0.0077 -0.0267 0.0267 0.0505 -0.0149 -0.1282 0.0629 0.0024 0.0796 0.0649 0.0278 -0.0118 0.0589 famstruct 0.0108 -0.0273 0.0273 0.1248 -0.0345 -0.0623 -0.0818 -0.0336 0.1154 0.1587 -0.0509 0.0259 0.2676

econsit homeown mother~p famstr~t

econsit 1.0000 homeown 0.1875 1.0000 motheremp 0.0544 0.0717 1.0000 famstruct 0.0775 0.1945 0.0479 1.0000

. . B 1: Correlation Matrix for English Proficiency Using Nationality

engpro~b Gender2 rncat2 rncat3 rncat4 rncat5 racenat parent~s GPA usstay disc livewm~r econsit

engprof_b 1.0000 Gender2 0.1124 1.0000 rncat2 0.0077 0.0072 1.0000 rncat3 -0.0430 -0.1270 -0.2617 1.0000 rncat4 -0.0078 0.1048 -0.2922 -0.1962 1.0000 rncat5 0.0456 0.0601 -0.2260 -0.1518 -0.1694 1.0000 racenat 0.0164 0.0811 -0.2673 0.1290 0.4884 0.6441 1.0000 parentses 0.1092 -0.1224 0.0294 0.0267 -0.0647 -0.0567 -0.0845 1.0000 GPA 0.1682 0.1012 -0.0224 0.0110 0.0096 0.0073 0.0134 0.1828 1.0000 usstay 0.1316 -0.0736 0.0297 0.0755 -0.1866 -0.0906 -0.1941 0.1983 0.0237 1.0000 disc -0.0432 -0.0525 0.0454 0.0185 -0.2295 -0.0994 -0.2634 -0.0049 -0.0316 0.0608 1.0000 livewmother 0.0595 0.0180 -0.0401 -0.0317 0.0199 0.0354 0.0203 0.0097 0.0468 -0.0164 0.0024 1.0000 econsit 0.0555 -0.0578 0.0221 -0.0076 0.0259 -0.0330 -0.0058 0.2909 0.0186 0.0984 0.0036 0.0039 1.0000 homeown 0.0582 -0.0926 0.0075 0.0062 -0.0097 -0.0444 -0.0443 0.5429 0.1339 0.1726 -0.0178 0.0125 0.1811 motheremp 0.0186 -0.0013 0.0255 0.0580 -0.0334 -0.0620 -0.0479 0.0569 0.0494 0.0428 0.0007 0.0601 0.0695 famstruct 0.0005 -0.0483 0.0458 0.0135 -0.1034 -0.0370 -0.1003 0.0935 0.1282 -0.0529 0.0033 0.2579 0.0797

homeown mother~p famstr~t

homeown 1.0000 motheremp 0.0559 1.0000 famstruct 0.1631 0.0634 1.0000

Figure B 2: Correlation Matrix for English Proficiency Using Race

246

educre~i Gender2 nation~2 nation~3 nation~4 nation~5 parent~s disc livewm~r econsit homeown mother~p famstr~t

educredbi 1.0000 Gender2 -0.0215 1.0000 nationality2 -0.0472 0.0719 1.0000 nationality3 -0.0495 0.0556 -0.0698 1.0000 nationality4 0.0348 0.0570 -0.0689 -0.0900 1.0000 nationality5 0.0156 0.1176 -0.0583 -0.0761 -0.0751 1.0000 parentses 0.3054 -0.1705 -0.0561 -0.1112 0.0600 0.0197 1.0000 disc -0.0012 -0.0620 -0.0515 -0.1267 -0.1712 -0.0888 0.0141 1.0000 livewmother 0.0519 0.0091 0.0446 0.0195 -0.0012 0.0486 0.0323 -0.0481 1.0000 econsit 0.1181 -0.0735 -0.0327 -0.0495 0.0588 0.0189 0.3041 -0.0127 0.0178 1.0000 homeown 0.1655 -0.1174 -0.0734 0.0302 0.0098 -0.0072 0.5409 -0.0157 0.0513 0.1835 1.0000 motheremp 0.0305 0.0183 -0.0227 -0.0816 0.0553 0.0202 0.0665 -0.0153 0.0521 0.0674 0.1070 1.0000 famstruct 0.1274 -0.0598 -0.0024 -0.0331 -0.0766 0.0106 0.0918 -0.0166 0.2426 0.0737 0.1734 0.0513 1.0000 csnat 0.0149 0.0582 -0.0368 0.0369 0.1408 -0.0740 -0.0975 0.0322 -0.0236 -0.0820 -0.0801 0.0008 0.1080 csnon -0.1459 0.0868 -0.0566 0.1697 0.0847 0.0092 -0.1579 -0.0756 0.0417 -0.0884 -0.1384 -0.0481 -0.0226 engprof_b 0.0604 0.0740 0.0023 0.0162 0.0672 0.0210 0.0700 -0.0449 0.0241 0.0499 -0.0190 -0.0000 -0.0398 EduExpect8~2 -0.0845 -0.0870 -0.0235 -0.0057 -0.0005 0.0306 -0.0557 -0.0107 -0.0302 -0.0301 -0.0278 -0.0161 -0.0467 EduExpect8~3 0.1813 0.0649 -0.0549 0.0321 -0.0036 -0.0322 0.1651 0.0439 0.0332 0.1095 0.0954 0.0121 0.0504

csnat csnon engpro~b EduExp~2 EduExp~3

csnat 1.0000 csnon -0.1573 1.0000 engprof_b -0.0720 -0.0261 1.0000 EduExpect8~2 -0.0374 0.0088 -0.0220 1.0000 EduExpect8~3 0.0143 -0.0330 0.1077 -0.7666 1.0000

. Figure B 3: Correlation Matrix for Educational Attainment Using Nationality

educre~i Gender2 rncat2 rncat3 rncat4 rncat5 parent~s disc livewm~r econsit homeown mother~p famstr~t

educredbi 1.0000 Gender2 -0.0215 1.0000 rncat2 -0.0094 -0.0227 1.0000 rncat3 0.0163 -0.0993 -0.2295 1.0000 rncat4 0.0414 0.1252 -0.2365 -0.1531 1.0000 rncat5 -0.0481 0.0795 -0.1924 -0.1245 -0.1284 1.0000 parentses 0.3054 -0.1705 0.0274 0.0068 -0.0230 -0.0298 1.0000 disc -0.0012 -0.0620 0.0562 0.0149 -0.1821 -0.0923 0.0141 1.0000 livewmother 0.0519 0.0091 -0.0314 -0.0214 0.0298 0.0440 0.0323 -0.0481 1.0000 econsit 0.1181 -0.0735 0.0423 -0.0128 0.0094 -0.0202 0.3041 -0.0127 0.0178 1.0000 homeown 0.1655 -0.1174 0.0468 -0.0121 0.0266 -0.0481 0.5409 -0.0157 0.0513 0.1835 1.0000 motheremp 0.0305 0.0183 0.0123 0.0482 0.0129 -0.0654 0.0665 -0.0153 0.0521 0.0674 0.1070 1.0000 famstruct 0.1274 -0.0598 0.0595 -0.0102 -0.0594 -0.0016 0.0918 -0.0166 0.2426 0.0737 0.1734 0.0513 1.0000 csnat 0.0149 0.0582 0.0629 -0.0675 0.0757 0.0077 -0.0975 0.0322 -0.0236 -0.0820 -0.0801 0.0008 0.1080 csnon -0.1459 0.0868 -0.0519 -0.0274 0.1191 -0.0060 -0.1579 -0.0756 0.0417 -0.0884 -0.1384 -0.0481 -0.0226 engprof_b 0.0604 0.0740 -0.0362 -0.0074 -0.0043 0.0595 0.0700 -0.0449 0.0241 0.0499 -0.0190 -0.0000 -0.0398 EduExpect8~2 -0.0845 -0.0870 -0.0001 -0.0094 0.0129 -0.0180 -0.0557 -0.0107 -0.0302 -0.0301 -0.0278 -0.0161 -0.0467 EduExpect8~3 0.1813 0.0649 0.0069 0.0360 0.0078 -0.0503 0.1651 0.0439 0.0332 0.1095 0.0954 0.0121 0.0504

csnat csnon engpro~b EduExp~2 EduExp~3

csnat 1.0000 csnon -0.1573 1.0000 engprof_b -0.0720 -0.0261 1.0000 EduExpect8~2 -0.0374 0.0088 -0.0220 1.0000 EduExpect8~3 0.0143 -0.0330 0.1077 -0.7666 1.0000

Figure B 4: Correlation Matrix for Educational Attainment Using Race

247

marcat2 Gender2 rncat2 rncat3 rncat4 rncat5 parent~s V407A disc livewm~r econsit homeown mother~p

marcat2 1.0000 Gender2 0.0948 1.0000 rncat2 -0.2029 -0.0287 1.0000 rncat3 0.1639 -0.1004 -0.2829 1.0000 rncat4 -0.1691 0.1430 -0.3075 -0.1853 1.0000 rncat5 -0.0628 0.0530 -0.1911 -0.1152 -0.1252 1.0000 parentses -0.0233 -0.1750 0.0110 0.0068 -0.0429 -0.0100 1.0000 V407A -0.0508 -0.0206 0.0391 -0.0487 0.0078 -0.0271 0.2950 1.0000 disc 0.0719 -0.0701 0.0370 -0.0179 -0.2220 -0.0857 -0.0043 -0.0085 1.0000 livewmother -0.0676 0.0052 -0.0297 -0.0367 0.0305 0.0266 0.0437 0.0698 -0.0452 1.0000 econsit -0.0501 -0.1001 0.0470 -0.0084 0.0095 -0.0035 0.3049 0.0778 -0.0247 0.0268 1.0000 homeown -0.0433 -0.1211 0.0332 -0.0331 0.0104 -0.0458 0.5437 0.1937 -0.0303 0.0151 0.1874 1.0000 motheremp -0.0113 0.0217 0.0034 0.0396 0.0086 -0.0666 0.0595 0.0468 -0.0339 0.0625 0.0568 0.0750 1.0000 famstruct -0.0117 -0.0799 0.0563 -0.0225 -0.0767 -0.0269 0.0876 0.1175 -0.0393 0.2422 0.0804 0.1382 0.0597 csnat 0.0077 0.0683 0.0539 -0.0792 0.0706 -0.0065 -0.1223 0.0214 0.0341 -0.0170 -0.0642 -0.1127 -0.0117 csnon -0.0051 0.0934 -0.0365 -0.0188 0.1586 -0.0121 -0.1468 -0.1559 -0.0579 0.0305 -0.0739 -0.1324 -0.0231 engprof_b -0.0146 0.0651 -0.0389 0.0064 -0.0055 0.0801 0.0638 0.0648 -0.0791 0.0575 0.0283 -0.0134 0.0332 EduExpect8~2 -0.0186 -0.0906 -0.0078 -0.0089 0.0106 0.0056 -0.0552 -0.0829 -0.0016 -0.0184 -0.0324 -0.0355 -0.0069 EduExpect8~3 -0.0108 0.0606 0.0004 0.0292 0.0036 -0.0724 0.1471 0.2462 0.0198 0.0136 0.1024 0.0957 0.0134

famstr~t csnat csnon engpro~b EduExp~2 EduExp~3

famstruct 1.0000 csnat 0.1085 1.0000 csnon -0.0233 -0.1529 1.0000 engprof_b -0.0471 -0.0620 -0.0343 1.0000 EduExpect8~2 -0.0282 -0.0394 -0.0049 -0.0035 1.0000 EduExpect8~3 0.0259 0.0283 -0.0077 0.0822 -0.7826 1.0000

B 5: Correlation Matrix for Out-Marriage/Partnership Using Race

V442 Gender2 nation~2 nation~3 nation~4 nation~5 parent~s V407A disc livewm~r econsit homeown mother~p

V442 1.0000 Gender2 -0.0419 1.0000 nationality2 -0.0455 0.0695 1.0000 nationality3 -0.0528 0.0480 -0.0673 1.0000 nationality4 -0.0109 0.0621 -0.0682 -0.0882 1.0000 nationality5 0.0444 0.1183 -0.0571 -0.0738 -0.0749 1.0000 parentses 0.0951 -0.1683 -0.0443 -0.1139 0.0613 0.0247 1.0000 V407A 0.1084 -0.0195 -0.0757 0.0102 0.0716 -0.0258 0.3072 1.0000 disc 0.0246 -0.0517 -0.0451 -0.1125 -0.1735 -0.0946 0.0072 -0.0184 1.0000 livewmother 0.0061 0.0142 0.0446 0.0165 -0.0026 0.0489 0.0344 0.0773 -0.0439 1.0000 econsit 0.0419 -0.0755 -0.0274 -0.0800 0.0676 0.0180 0.3203 0.0990 -0.0147 0.0140 1.0000 homeown 0.1181 -0.1183 -0.0602 0.0284 0.0241 -0.0211 0.5351 0.2051 -0.0194 0.0465 0.1931 1.0000 motheremp 0.0418 0.0207 -0.0135 -0.0843 0.0535 0.0315 0.0645 0.0403 -0.0122 0.0562 0.0699 0.1115 1.0000 famstruct 0.0445 -0.0549 -0.0099 -0.0389 -0.0719 0.0226 0.0868 0.1228 -0.0371 0.2482 0.0780 0.1815 0.0488 csnat -0.0185 0.0663 -0.0438 0.0519 0.1518 -0.0719 -0.1119 0.0244 0.0349 -0.0251 -0.0811 -0.0947 -0.0050 engprof_b 0.0271 0.0707 0.0236 0.0238 0.0663 0.0070 0.0627 0.0587 -0.0585 0.0338 0.0461 -0.0178 -0.0010 EduExpect8~2 0.0231 -0.0837 -0.0296 0.0108 -0.0029 0.0102 -0.0484 -0.0746 -0.0269 -0.0257 -0.0402 -0.0227 -0.0022 EduExpect8~3 0.0270 0.0581 -0.0514 0.0165 0.0003 -0.0190 0.1630 0.2474 0.0440 0.0317 0.1092 0.1007 0.0053

famstr~t csnat engpro~b EduExp~2 EduExp~3

famstruct 1.0000 csnat 0.1111 1.0000 engprof_b -0.0440 -0.0735 1.0000 EduExpect8~2 -0.0458 -0.0380 -0.0181 1.0000 EduExpect8~3 0.0425 0.0163 0.1014 -0.7707 1.0000

B 6: Correlation Matrix for Political Participation Using Nationality

248

V442 Gender2 rncat2 rncat3 rncat4 rncat5 parent~s V407A disc livewm~r econsit homeown mother~p

V442 1.0000 Gender2 -0.0419 1.0000 rncat2 0.0520 -0.0271 1.0000 rncat3 0.0178 -0.1022 -0.2294 1.0000 rncat4 -0.0236 0.1234 -0.2369 -0.1522 1.0000 rncat5 0.0176 0.0843 -0.1909 -0.1226 -0.1267 1.0000 parentses 0.0951 -0.1683 0.0257 0.0097 -0.0301 -0.0159 1.0000 V407A 0.1084 -0.0195 0.0506 -0.0117 0.0413 -0.0312 0.3072 1.0000 disc 0.0246 -0.0517 0.0513 0.0093 -0.1687 -0.0874 0.0072 -0.0184 1.0000 livewmother 0.0061 0.0142 -0.0227 -0.0251 0.0286 0.0432 0.0344 0.0773 -0.0439 1.0000 econsit 0.0419 -0.0755 0.0429 -0.0166 0.0007 -0.0208 0.3203 0.0990 -0.0147 0.0140 1.0000 homeown 0.1181 -0.1183 0.0464 -0.0036 0.0164 -0.0293 0.5351 0.2051 -0.0194 0.0465 0.1931 1.0000 motheremp 0.0418 0.0207 0.0115 0.0542 0.0166 -0.0525 0.0645 0.0403 -0.0122 0.0562 0.0699 0.1115 1.0000 famstruct 0.0445 -0.0549 0.0678 -0.0211 -0.0542 0.0054 0.0868 0.1228 -0.0371 0.2482 0.0780 0.1815 0.0488 csnat -0.0185 0.0663 0.0573 -0.0704 0.0877 0.0084 -0.1119 0.0244 0.0349 -0.0251 -0.0811 -0.0947 -0.0050 engprof_b 0.0271 0.0707 -0.0466 -0.0071 -0.0044 0.0711 0.0627 0.0587 -0.0585 0.0338 0.0461 -0.0178 -0.0010 EduExpect8~2 0.0231 -0.0837 -0.0024 -0.0015 0.0222 -0.0287 -0.0484 -0.0746 -0.0269 -0.0257 -0.0402 -0.0227 -0.0022 EduExpect8~3 0.0270 0.0581 0.0073 0.0339 0.0038 -0.0461 0.1630 0.2474 0.0440 0.0317 0.1092 0.1007 0.0053

famstr~t csnat engpro~b EduExp~2 EduExp~3

famstruct 1.0000 csnat 0.1111 1.0000 engprof_b -0.0440 -0.0735 1.0000 EduExpect8~2 -0.0458 -0.0380 -0.0181 1.0000 EduExpect8~3 0.0425 0.0163 0.1014 -0.7707 1.0000

B 7: Correlation Matrix for Political Participation Using Race

249