University of Montana ScholarWorks at University of Montana
Graduate Student Theses, Dissertations, & Professional Papers Graduate School
1983
Theoretical and methodological exploration of portrayals of Native Americans in children's fiction
Karl Ormond Edwards The University of Montana
Follow this and additional works at: https://scholarworks.umt.edu/etd Let us know how access to this document benefits ou.y
Recommended Citation Edwards, Karl Ormond, "Theoretical and methodological exploration of portrayals of Native Americans in children's fiction" (1983). Graduate Student Theses, Dissertations, & Professional Papers. 3867. https://scholarworks.umt.edu/etd/3867
This Thesis is brought to you for free and open access by the Graduate School at ScholarWorks at University of Montana. It has been accepted for inclusion in Graduate Student Theses, Dissertations, & Professional Papers by an authorized administrator of ScholarWorks at University of Montana. For more information, please contact [email protected]. COPYRIGHT ACT OF 1976
THIS IS AN UNPUBLISHED MANUSCRIPT IN WHICH COPYRIGHT SUB SISTS. ANY FURTHER REPRINTING OF ITS CONTENTS MUST BE APPROVED BY THE AUTHOR,
MANSFIELD LIBRARY UNIVERSITY OF MONTANA DATE : 19 83
A THEORETICAL AND METHODOLOGICAL EXPLORATION
OF PORTRAYALS OF NATIVE AMERICANS
IN CHILDREN'S FICTION
By
Karl Ormond Edwards
B.S., University of Utah, 1973 B.A., University of Utah, 197-4
Presented in partial fulfillment of the requirements for the degree of
Master of Interdisciplinary Studies
UNIVERSITY OF MONTANA
1983
Approved by:
Dean, Graduate Scho*
!- 3~ to Date UMI Number: EP35744
All rights reserved
INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted.
In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. UMI Dissertation Publishing
UMI EP35744 Published by ProQuest LLC (2012). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code
uest'
ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106-1346 Edwards, Karl Ormond, M.I.S., August 1983 Interdisciplinary
Studies
A Theoretical and Methodological Exploration of Portrayals of Native Americans in Children's Fiction (282 pp.)
Director: Dr. Henrietta V. Whiteman
This is an exploratory study into the theoretical and methodo logical issues concerning portrayals of Native Americans in children's fiction. The purpose of this study is to analyze the portrayals of Native Americans with the goal of determining whether such portrayals in children's fiction are descriptive or evaluative. This study develops a theoretical foundation for the analysis of portrayals of Native Americans in children's fiction and selectively applies known techniques and methods to measure such portrayals. Through this process methodological issues concerning content analysis are evaluated and their data findings compared. Both quantitative and qualitative methods are used to measure and analyze the portrayals of Native Americans and Non-Native Americans in children's fiction. Quantitatively, Pratt's (1971; 1972) Evaluative Coefficient Analysis method is used to measure the use of evaluative words to portray subjects in such books. Semi-quantitatively, Garcia's (1978; 1979) thematic analysis is used to measure the presence and frequency of themes within the portrayal of Native Americans. Qualitatively, the categories of story plot, character and group portrayals, and impressions and impressionistic techniques are employed to measure the portrayals of Native Americans and Non-Native Americans. The sample, composed from a ten percent random sample from a previously identified population, consists of six books published between 1968 and 1982. This study concludes that portrayals of Native Americans in children's fiction tend to be more evaluative, negative, and judgmental than for Non-Native Americans. It is concluded that through stylistic methods, reinforced by perpetuated themes, portrayals of Native Americans remain static over time. More importantly, however, is the conclusion that the portrayals of Native Americans now are more subtle, more latent, and harder to manifestly perceive. Such conclusions imply the need for increased theoretical and methodological studies, as well additional in-depth analyses of recent books portraying Native Americans. To my loving wife, Cindy, and to our beautiful son, Kristoffer, this thesis is dedicated.
iii PREFACE
Since my youth I have held an insatiable desire to learn as much as I could about the Indians of North America. Over the years
I have read, and often reread, as many books about Indians as I could find. Always I read these books with a sympathetic heart and wondered how other people could treat Indians as they did throughout history. In college the opportunity to historically study Indians broadened my interests and knowledge, and I soon realized that
Indians are people that only live in the minds and attitudes of people. The term Indian brings many and varied ideas to mind, foremost of which is the notion of a people characterized by socially standardized images—socially from the Non-Indian point of view.
Who really are the Indians? In coming to grips with this question,
I soon realized that I held some very basic and commonly shared attitudes—that Indians are Indians. Indeed, the study of Indians became synonymous with the study of the history of Indian-white political, military, religious, economic, and to some degree, educational relations.
In the early 1970s there emerged for me a new and exciting group of people—Native Americans. Native Americans implied Indians, but the name also meant much more, most important being that Native
Americans are people that are alive today. Furthermore, Native
iv Americans are seen as having cultural and social histories that encompass far more than just Indian-white relations, and the name
Native Americans implies plurality whereas Indians purports unity.
This means the study of Native Americans is an interdisciplinary
adventure of many diverse and unique groups of people. Yet I had long had a sense that specific groups I had studied did not fit
into the image of Indians; these people were Apaches, Nez Perce,
and other distinct nations of people. These people had unique cultures and, as such, stood out from Indians. They are, however,
Native Americans in the sense that Germans are Europeans.
Gradually, then, an awakening came about that my earlier
attitudes about Native Americans rested upon faulty awareness and learning. As my understanding progressed I realized the tremendous role of the media in developing my attitudes about Native Americans.
I am luckier than many, however, because I continually read, and such reading created anomolies in the information and attitudes I developed about Native Americans. Additionally, I had been taught in history and English, as well as elsewhere, to read and learn critically—to question everything. Others are less fortunate and read far less. There seem to be, however, some fundamental differ ences between my perceptions concerning the treatment of Native
Americans in books and others' perceptions of the books. More
v importantly, I have not developed a repertoire of erroneous, incomplete, and unjust attitudes about Native Americans. This study, then, is the analysis of a sample of books to determine one small aspect of how books affect attitudes about Native
Americans.
vi ACKNOWLEDGEMENTS
First and foremost I would like to especially thank my wife, Cindy, for all the support and assistance she has given me, particularly in developing and preparing the universe, population, and sample., in typing and retyping, and in performing the tedious work of statistical calculations. A special thanks is given to my chair, Dr. Henrietta Whiteman, for her constant support, invaluable criticisms, and enthusiasm, and for her long hours of assistance. A very special thanks is also given to
Dr. Rod Brod for his enduring patience and determination to get me through this study, and for his continual presence and support over the years when others could not be there to help me. I would also like to thank Dr. Carling Malouf for suggesting my program of study and for continually encouraging me in my studies. Lastly, a note of thanks to Dr. Joseph Epes Brown for interesting and encouraging me in my studies of Native Americans.
vii CONTENTS
Page
List of Tables x
List of Figures xii
Chapter
1. INTRODUCTION 1
Theoretical Framework and Rationale 4
Related Methodological Research 29
Research Problem 39
2. METHODOLOGY 46
Research Population and Samples 46
Measurement Procedures 54
3. RESULTS 60
Qualitative Findings 62
Quantitative Results for Individual Books 69
Comparison of Books 121
Comparison of Quantitative Results with Qualitative Results 156
4. CONCLUSIONS 160
Implications 172
Endnotes 174
References Cited 178
viii Page
Appendixes
A. DISSERTATIONS AND THESES ON NATIVE AMERICANS IN LITERATURE 220
B. REFERENCES ON NATIVE AMERICANS RELEVANT TO THE STUDY OF NATIVE AMERICANS IN LITERATURE 234
C. PRINCIPLES/SCOPE OF PROPAGANDA 254
D. REFERENCES ON CONTENT ANALYSIS 260
E. STUDY INSTRUMENTS 267
F. ECO WORD LIST 273
G. DATA SUMMARY 277
H. LIST OF OBSERVED THEMES 281
ix TABLES
Table Page
1 . Examples for Calculating "Evaluative Assertion Analysis" 37
2. Study Universe, Populations, and Samples 50
3. Akavak, an Eskimo Journey (Book #1): Observed Word Frequencies 71
4. Akavak, an Eskimo Journey (Book #1): Page and Word Frequencies, Coefficient of Evaluation, and Evaluativeness Score 73
5. Akavak, an Eskimo Journey (Book #1): Positive, Negative, and Neutral Word Rankings 77
6. Sing Down the Moon (Book #2): Observed Word Frequencies 79
7- Sing Down the Moon (Book #2): Page and Word Frequencies, Coefficient of Evaluation, and Evaluativeness Score 83
8. Sing Down the Moon (Book #2): Positive, Negative, and Neutral Word Rankings 84
9. The Spirit is Willing (Book #3): Observed Word Frequencies 88
10. The Spirit is Willing (Book #3): Page and Word Frequencies, Coefficient of Evaluation, and Evaluativeness Score 90
11. The Spirit is Willing (Book #3): Positive, Negative, and Neutral Word Rankings 93
12. Squaw Man's Son (Book #4): Observed Word Frequencies 96
13. Squaw Man's Son (Book #4): Page and Word Frequencies, Coefficient of Evaluation, and Evaluativeness Score 99
x Table Page
14. Squaw Man's Son (Book #4): Positive, Negative, and Neutral Word Rankings 103
15. Sarah Bishop (Book #5): Observed Word Frequencies 106
16. Sarah Bishop (Book #5): Page and Word Frequencies, Coefficient of Evaluation, and Evaluativeness Score 108
17. Sarah Bishop (Book #5): Positive, Negative, and Neutral Word Rankings 111
18. Where the Buffaloes Begin (Book #6): Observed Word Frequencies 115
19. Where the Buffaloes Begin (Book #6): Page and Word Frequencies, Coefficient of Evaluation, and Evaluativeness Score 119
20. Where the Buffaloes Begin (Book #6): Positive, Negative, and Neutral Word Frequencies 120
21 . Observed Frequencies by Book and Ethnicity 124
22. Observed Frequencies by Book and We-They 128
23. Observed Frequencies by Book with Ethnic Subjects. . 131
24. Observed Frequenices by Book with We-They Subjects . 135
25. Observed Frequencies by Ethnicity 137
26. Observed Frequencies by We-They 140
27. Mean Scores Over Time 145
28. Coefficient of Evaluation Scores 278
29. Evaluativeness Scores 279
30. Summary of Data by Book, Ethnicity, and We-They. . . 280
xi FIGURES
Figure Page
1. Evaluations: (+)(0)(-) 147
2. Coefficient of Evaluation Over Times 151
3. Evaluativeness Scores Over Time 153
xii Chapter 1
INTRODUCTION
The purpose of this study is to analyze the portrayals of
Native Americans in children's fictional literature. The goal is to determine whether portrayals of Native Americans in children's books are descriptive or evaluative. Descriptive portrayals are those terms which discuss, or are used by, the subject of interest and are simply neutral, such as "small," "winter," "running,"
"Nez Perce," and so forth. Conversely, evaluative portrayals are those in which the subject either utilizes, or is discussed in, terms that are biased in their contextual presentation or which make a judgment about the subject, such as "smelly," "quick," "savage,"
"conscientious," and so forth. Other studies, which are discussed below, have led to indifferent or conflicting conclusions concerning evaluative or judgmental portrayals of Native Americans. Such studies, however, have not included any analyses of the descriptive content of such portrayals. This exploratory study, then, evaluates the results and compares the findings concerning portrayals of
Native Americans using both quantitative and qualitative measure ments. The study also presents a theoretical foundation for the analysis of portrayals of Native Americans.
Literature is one of the most important means of reaching
1 2
people through the mass media. It is, however, along with other
printed media, the only vehicle of communication that has no
restrictions as to particular times and/or places. Furthermore,
through literature one can: (1) develop and carefully edit a subject as desired; (2) present multiple ideas to anyone who can read; and (3) present a format that allows the reader to receive
the message at his/her leisure and even to reread the message.
Many books, additionally, stand the test of time—i.e., each successive generation can read and thereby be affected by the same literature—which facilitates the perpetuation of ideas and attitudes about Native Americans. Just the same, many new books on Native Americans successively appear that may be more valid or
persuasive and thereby influence changes in ideas and attitudes.
In other instances the popularity of a book may expose people with no previous knowledge and/or attitudes about Native Americans to the subject, and thus, create new ideas and attitudes for them.
A reader of literature, then, is affected by the reading experience in various ways and in varying degrees. In discussing the effects of reading.it is useful to distinguish between (1) the
preconditions of publication, (2) the conditions of distribution of the literature, (3) the supply, content, and style of the literature or the publication itself, and (A) the readers' pre dispositions (Waples, et. al., 1940:7-21, 136-137). Although a 3 complete study of the effects of literature would analyze each of these factors, monetary and temporal limitations often restrict studies instead to discussing individual or partial factors. The factor of interest to this study, and most other studies, is that of the publication itself. Furthermore, most studies focus upon the partial factor of the contents of the publication; i.e., subject classification, sophistication or maturity, locale, action or situation, social relationships, validity, and characters
(Waples, et. al., 1940:140-141). That is, researchers seek to describe the characteristics of the content or what the contents contain. Such studies (1) describe trends in communications content,
(2) relate known characteristics of sources to the messages they produce, and (3) audit or compare communication content against prescribed standards. Other studies, however, focus upon the partial factor of style; i.e., difficulty, sophistication or maturity, literary and aesthetic qualities, semantics, and modes of presentation (Waples, et. al., 1940:140). This study, then, seeks to describe characteristics of style in the content or how the messages are presented.
Very few people ever actually interact with or encounter people from cultures and societies identified most often as
1 "American Indians" or "Native Americans." Most people, however, can express opinions, attitudes, facts, or arguments about Native 4
Americans. Much of this ability to categorize Native Americans, therefore, results from learning about them through various forms of media and face-to-face communications. It is of interest to this researcher to analyze how literature categorizes or portrays
Native Americans in an effort to partially explain attitudes held by Non-Native Americans with little or no experience with Native
Americans. Similarly, it is of interest to see how such portrayals may reinforce or alter previously held attitudes about Native
Americans. Native Americans were selected, then, because over time their portrayals have been perceived as being static, and also because of the frequency with which they are encountered in various types and forms of literature.
Theoretical Framework and Rationale
Before focusing on the topic of this study it is necessary to conceptualize the theoretical framework and, thereby, the sub sequent rationale for conducting the present research. This discussion formulates an explicit paradigmatic continuity that generally is lacking in other studies (Rosengren, 1980; Wilhoit and de Bock, 1980). Such implicit treatment, or total avoidance of the researcher's theoretical perspective, has created and perpetuated a number of misconceptions and questionable results, and has allowed researchers to measure and analyze "characterizations," "themes," 5
and other categories of portrayals of Native Americans (and other subjects) under the erroneous rubric of "stereotyping." While some
of the more recent studies avoid explicit use of "stereotyping"
(Garcia, 1978, 1980), they have merely become less explicit about
their theoretical framework and rationale. No known study, then,
explicitly explains the theoretical connection for any type of
measurement concerning portrayals of Native Americans in literature.
Many fields of study (e.g., education, political science,
English, sociology, psychology) focus upon forms of literature to
evaluate their quality and effectiveness. The field of social
psychology, however, along with its inherent interdisciplinary
nature, provides the greatest possibilities. It is noted that
interpersonal communications, a newly created interdisciplinary field of study focusing upon communications, is conceived here to be an allied/subdiscipline of social psychology. Social psychology,
then, is the field of study from which interpersonal communications ultimately derives the majority of its assumptions and concepts, if not theoretical perspectives. Conversely, a fundamental problem with social psychology, as a field of study for discussing the various
theories, assumptions, concepts, and so forth, applicable to this study, is the absence of any explicit scientific paradigms
(Kuhn, 1970). This is not to say that social psychology has no
theoretical perspectives, but rather that there are no generally accepted theoretical paradigms in social psychology (Kuhn, 1970;
Rosenberg and R. Turner, 1981; Shaw and Costanzo, 1970; Slawski,
1981). Indeed, the field of social psychology often is viewed by
its approaches, its contents and problems (substantive areas), or
its conceptual interests (Lindzey, 1968; Newcomb, et. al., 1965;
Sampson, 1964; Sherif and Sherif, 1956). Others have synthesized
theories used in social psychology into theoretical orientations
(Rosenberg and R. Turner, 1981; Shaw and Costanzo, 1970) and per spectives (Slawski, 1981). Shaw and Costanzo's orientations
contain too many overlapping assumptions and exemplars for a
useful paradigm, while Rosenberg and Turner's present sociological
perspectives and orientations are, for the most part, clarifications
of theories, approaches, and concepts, and are of no paradigmatic
usefulness. Slawski's perspectives are influenced more by their
methods than by their images of the subject matter and exemplars
(Kuhn, 1970).
These various ways of organizing studies in the field of social psychology do not provide any encompassing paradigm(s)
through which a researcher can establish the exemplars, methods, ways of viewing the subject matter, and theories relevant to a given study. It is not the purpose of this study, however, to
propose such a paradigm; indeed, Slawski states that "to attempt such a task in detail ... is a lifetime's work" (1981:100). 7
George Ritzer's presentation of sociological paradigms, on the
other hand, provides a very tenuous paradigm for social psychology
in his Social Definition Paradigm (1980:83-140). The applicability
of the definition paradigm to social psychology is demonstrated
most effectively through their shared theoretical orientations,
their shared assumptions, and their similar conceptualizations
of the subject matter.
One of the primary theoretical orientations of the Social
Definition Paradigm is that of symbolic interactionism (Ritzer,
1981:96-108; J. Turner, 1982:305-388). Rosenberg and R. Turner,
and Slawski explicitly state the superior utility of the symbolic
interactionism orientation or perspective:
Symbolic interactionism is generally regarded as the most distinctive theoretical contribution to SSP ["sociological social psychology"] (Rosenberg and R. Turner, 1981:xvi).
Symbolic interactionism is a metatheoretical frame of refer ence, in part a kind of philosophy which can easily be used to begin to integrate or incorporate all other social psycho logical theories within its scope (Slawski, 1981:100).
While Stryker in Rosenberg and R. Turner posits:
Of the theoretical orientations underlying work in social psychology, it is symbolic interactionism that has had its major development among sociologists. . . . the appeal of symbolic interactionism to sociologists reflects the funda mental compatibility of this social-psychological perspective with the structural concerns of sociology proper. A theme of this chapter is fit, . . . between more general sociological theory and symbolic interactionism as social psychological theory (1981:3). 8
Similarly, the subject matter for the Social Definition
Paradigm is not just social facts or simple stimulus-response
behavior, but rather the way individuals define their social
situations and the effect of such upon ensuing actions or behavior
(Ritzer, 1980:27, 89-91). Broadly defined, the field of social
psychology is interested in such interaction between individuals
and their environment, which thereby deduces the applicability of
the definition paradigm to social psychology as the Social Psycho
logical Defintion Paradigm.
This proposed Social Psychological Definition Paradigm
primarily serves the functions of (1) establishing the way the
subject matter is conceived, (2) expositing some basic assumptions,
and (3) providing an orientation from which theoretical models,
substantive theories, and guiding theoretical concepts can be
derived. (The commonality of exemplars is beyond the scope of
this discussion). Generally speaking, then, the image of the
subject matter is conceived of as being similar in several con- 2 structs. The most important idea is that individuals are active
creators of their own social reality. A second important construct
is that social reality is not a static set of coercive social facts,
or even further, that reality is only that which is created by man
at the moment. Theoretical interests within this paradigm focus
upon the "minding" process (Ritzer, 1980:90) or what takes place 9 within the mind of the individual (e.g., attitudes):
Something occurs in a man's mind between the time a stimulus is applied and the response is emitted and it is this creative activity that is the concern-of the social [psychological] definitionist. To the social factist, external social con straints determine behavior, while to the behaviorist there is no room for creative responses since contingencies of reinforcement determine behavior (Ritzer, 1980:90).
A third construct is the conception of a social process or a tendency to focus upon the process of social interaction rather than static structure or behavior. As a result of these conceptu alizations of man's free will to define the social world, Ritzer further posits a fourth construct in the nature of an ideological theme that "sociologists of both liberal and radical persuasion tend to gravitate toward social definitionism" (1980:90). Lastly, unlike the pure Social Definition Paradigm, the subject matter is seen as being responsive to external stimuli as well as internal definition; that there is, rather, an interaction between the social definitionist's and the social factist's paradigms within the Social
Psychological Definition Paradigm. This last construct more closely affiliates the image of the subject matter with the broader definition of social psychology. .It is the interaction of the individual who to the best of his ability defines the social world and its environ ment, which in turn attempts to control the individual through norms, values, and social control agencies that are often provided by reference groups. 10
This image of the subject matter, therefore, allows one to
exposit theoretical assumptions, models, theories, and concepts
relevant to a proposed study. Theoretical models, then, will be
taken from general communications models, while communications,
propaganda, attitudes-behavior, and awareness/perception theories;
and stereotyping, prejudice/discrimination, and minority-majority
relations concepts will be incorporated to define the theoretical
framework of this study.
Theories, models, and/or concepts of human communications,
which implicitly underlie many studies of literature, are seen as
a process explicitly stated by the paradigm's image of the subject
matter. Very little work has been done in the area of general com
munications theory, however, as most researchers prefer to utilize
general models and/or theories and concepts borrowed from one or
more of the multi-disciplines incorporated into communications.
This absence of high level or grand theories is partially the result of the broad spectrum of topics encompassed by the field of communi
cations, in addition to the dilemma posed by being both an art and
a science, and because.of the general lack of interest in proposing
a comprehensive grand theory. For the most part, therefore, scien
tific studies in the field of communications rely upon models and/or
taxonomies of the communications process to guide their research
(Cronkhite, 1976:49-74; Hooper, 1976:33-54; Mortensen, 1972:29-68). 11
Thereby, such models should ideally provide the freedom to incorporate middle and low range theories from which to derive specific problems.
A review of previous research concerning portrayals of groups or individuals in various forms of literature, however, demonstrates their unanimous failure to state their guiding theories or the communications model that they incorporated. Although theoretical models are generally missing from research on portrayals in liter ature, the method of measurement—content analysis—is usually explained. By such explicit utilization of content analysis methodology, researchers implicitly (if not unintentionally) indicate the use of the communications process model which underlies content analysis methodology, as promulgated by Harold D. Lasswell:
Who Says What In Which Channel To Whom With What Effect? (DeFleur, 1970:123; Lasswell, 1948:37-51).
This model has also been expressed mathematically by Shannon and
Weaver (Borden, 1971:9; Cronkhite, 1976:57; Hooper, 1976:39-42;
Mortensen, 1972:36-40; Shannon and Weaver, 1962:98):
Information Source Transmitter Receiver Destination CUD Message Signal A Signal Message Transmitted Received
Noise Source 12
Lasswell's communications process model has additionally been expanded into a process or communications act for achieving
"isomorphism of meaning" (DeFleur, 1970:90-96):
MASS MEDIUM DEVICE
SOURCE TRANSMITTER -• CHANNEL -• RECEIVER
NOISE S
DESTINATION RECEIVER CHANNEL TRANSMITTER SOURCE
FEEDBACK DEVICE
This model of the communications process—implicit to most studies through the use of the content analysis method—is assumed by the founders of content analysis (i.e., Lasswell, et. al., 1949), and explicitly stated in texts by Berelson (1952) and Holsti (1962) discussing the methods of content analysis. As this model is appropriate to this study's paradigmatic perspectives, it will be used as the general underlying model to guide this research.
Some very important theoretical conceptualizations and middle range theories are encompassed within the communications models (both the one used here and others): (1) awareness; 13
(2) attitudes and behaviors (including stereotyping, prejudice/
discrimination, and minority-majority relations); (3) contemporary
theories of mass communications; and (A) theories of persuasion
and propaganda (which includes the effects of literature). It
can be seen that if one cannot perceive communication, one cannot
be aware of communication nor can they respond to the communication.
This is inherent in the definition of communication:
Human communication has occurred when a human being responds to a symbol; . . . [and] ... a symbol, as used in this definition, is a sign which stands for something else only because it is used to stand for something else, whereas a signal is a sign of something else because of a natural physical relationship (Cronkhite, 1976:20, 44; emphases in the original).
Gary Cronkhite, in his text Communication and Awareness, poignantly
stresses the importance of awareness and sees the level of awareness
for most people, when they communicate, as being quite low because
it has become reflexive, almost like breathing (1976:4-5).3 The
concept of awareness, as used here, includes becoming (1) aware of
communication, (2) aware of people, (3) aware of messages, and
(4) aware of contexts. While all four processes are important, the
third—which will be discussed below—is of most interest to this
study, hence the first and fourth functions will be held as theoret
ical assumptions. The awareness of people, however, is of secondary
interest in that we are interested in how people become aware of
other people (i.e., Native Americans) through literature. 14
Psychologists and social psychologists have long been interested in social and racial awareness (Burrow, 1924-25; Clark and Clark,
1940; Diamont, 1969; Goodman, 1952; Katz, 1978; Landbreth and
Johnson, 1953; Meltzer, 1941; Renninger and Williams, 1966;
Stabler and Johnson, 1971; Stevenson and Stewart, 1958; Vaughan,
1963; Williams and Roberson, 1967; Yancey and Singh, 1975). These studies are important in that they show that awareness, especially of racial groups, begins at an early age and is most tractable in younger school age children.
Perception, a part of becoming aware, interacts with individual attitudes in that the way one perceives something is influenced (e.g., what to look for, what to see) by one's attitudes about it, while perceptions similarly influence (e.g., develop, change, reinforce) attitudes about it. Theoretical concepts of attitudes, therefore, are integrated with the concept of awareness.
The concept of attitudes, one of longstanding importance in the field of social psychology (Allport, 1968:59-60; Lambarth, 1980:
182-183; Oskamp, 1977:4-19) is a pivotal concept in the theoretical orientation of this study. The nature of attitudes has been widely discussed (Allport, 1968; Fishbein, 1967, 1971; McGuire, 1969;
Oskamp, 1977; Rokeach, 1968; Rosenberg, et. al., 1960; Thomas, 1971) and is generally viewed in one of two ways: (1) as having three components—cognitive, affective, and behavioral; and (2) as 15
dimensions labeled beliefs, attitudes, and behavioral intentions
(Oskamp, 1977:19). Furthermore:
Beliefs, opinions, values, and habits are concepts which are related to the concept of attitude, but are not synonymous with it. Whereas an attitude is a broad evaluative orientation toward an object, a belief or an opinion is narrower and more cognitive in nature. Public opinion refers to the shared opinions of large groups of people (Oskamp, 1977=19).
Equally elusive is any agreed upon definition (Oskamp, 1977:7-17),
but generally attitude can be defined as "a readiness to respond
in a favorable or unfavorable manner to a particular class of objects" (Oskamp, 1977:19). Associated with the concept of
attitude is the concept of behaviors—the observable and measurable actions—which are the manifestations of social interaction; that
is, ultimately we are interested in the relationships between communications (e.g., literature), attitudes (e.g., prejudices or
beliefs about Native Americans), and their manifestation (e.g.,
discrimination) in overt behavior (DeFleur and Westie, 1958). In
the context of the foregoing definitions and for the purpose of this study, attitudes will be viewed as being both associated with overt
behavior and composed of cognitive, affective, and behavioral or conative components.
Prejudice, and its associated behavior of discrimination, is the primary attitude of interest to this study, and is also implicit to most studies of minorities in communications.
Discrimination and prejudice are also components of minority-majority 16 relations, but they revolve around racial attitudes or prejudice.
Within this complex field, researchers are interested in (1 ) the development (and awareness) of prejudice or racial attitudes,
(2) the relationship between prejudice and discrimination or racism, (3) the modification of both prejudice and discrimination, and (4) the effects of discrimination and prejudice. Scholars have tried to synthesize this broad area of study (Barth and Noel, 1972;
Blalock, 1967; Bogardas, 1930; Brown, 1934; Frazier, 1957; Glick,
1955; Harding, et. al., 1968; Horton, 1966; Lieberson, 1961; Noel,
1968; Palmer, 1956; Rothman, 1977; Williams, 1966), but as of yet no acceptable theoretical perspective(s) exists. At best, several approaches to the studies of prejudice and minority-majority relations exist:
. . . The "ethnic group" approach. Here the focus is on a single racial, religious, linguistic, or nationality group, usually one that is a clearly identifiable minority in some modern national state. An attempt is made to describe the group's historical development, cultural traditions, distinctive values, relations with neighboring groups, and changing political and economic fortunes. . . . The "social interaction" approach. Here the focus is not a single ethnic group, but on the interrelationships of two or more such groups, frequently on the relationships of all the major ethnic groups in a particular geographical area. Very commonly the analysis is in terms of the major forms of social interaction distinguished by Park and Burgess (1921): compe tition, conflict, accomodation, and assimilation. . . . The "social problems" approach. Here the focus is on "the behavior of discrimination, the attitude of prejudice, and the consequences for the minority and majority groups of both of these (Rose, 1964:571)" (Harding, et. al., 1968:1). 17
The "social problems" perspective with direct emphasis upon the attitude of prejudice is the guiding approach used in this study.
Harding, et. al. distinguish between "ethnic" or "inter- group" attitudes and prejudice, which is appropriate to this study:
An ethnic attitude is an attitude which a person has toward some or all members of an ethnic group, provided that the atti tude is influenced to some degree by knowledge (or presumed knowledge) of the other individual's group membership. Ethnic attitudes are frequently referred to as "intergroup attitudes," although they include the attitudes of individuals toward the groups of which they themselves are members (1968:3; emphases added).
On the other hand, they state that researchers desire to evaluate— not just describe—ethnic attitudes, and "thus arises the concept of prejudice" (Harding, et. al., 1968:4; emphasis in the original):
Consequently, it seems most useful to us to define prejudice as a failure of rationality £r a failure of justice or a failure of human-heartedness in an individual's attitude toward members of another ethnic group (Harding, et. al., 1968:6; emphases in the original).
This definition clearly incorporates the three components of an attitude—the cognitive, the affective, and the conative—which also are discussed explicitly by Harding, et. al. for the more specific ethnic or intergroup attitudes, as originally discussed by Smith (1947), Kramer (1949), and Chein (1951). The cognitive component consists of perceptions, beliefs, and expectations (or the individual's picture of the minority group) that one holds about others. Beliefs and expectations are often conceived of as varying along a number of dimensions. They include: (1) simple 18
(or undifferentiated) versus complex (or differentiated); (2) cen tral in consciousness versus peripheral; (3) believed tentatively versus believed with assurance; (4) inadequately grounded versus grounded on appropriate evidence; (5) inaccurate versus accurate; and (6) tenacious versus readily modified (Harding, et. al., 1968:4;
Kramer, 1949:393-394, 395-398). It is within the cognitive component that the concept of stereotype, coined by Walter Lippmann (1922:
79-156; 1927:435-440), is found to derive from:
A belief that is simple, inadequately grounded, at least par tially inaccurate, and held with considerable assurance by many people . . . (Harding, et. al., 1968:4).
The affective component, on the other hand, includes the friendliness or unfriendliness to the minority group and the various feelings
(e.g., admiration, sympathy, hate, identification, contempt, fear, envy, alienation) or emotions evoked in an individual by the actual or symbolic stimulus of the minority group (Harding, et. al., 1968:4;
Kramer, 1949:394, 398-400). Lastly, the conative component encompasses beliefs, perceptions, and expectations about "what should be done" and action orientations toward a minority group; it is the tendency, desire, and/or disposition to act in certain ways (Harding, et. al.,
1968:4; Kramer, 1949:393, 400-404).
Despite the absence of theoretical guidance, there exist numerous studies in the area of prejudice/discrimination (Allport,
1954; Bell, 1979; Bettelheim and Janowitz, 1964; Clark, 1963; 19
Cole, 1975; Council for Interracial Books for Children, n.d.;
Daniels and Kitano, 1970; Hartley, 1946; Kluckhohn, 1960; Knowles and Prewitt, 1969; Kramer, 1949; Kutner, et. al., 1952; Linn, 1965;
Morland and Suthers, 1980; Park, 1927; Pate, 1981; Prewitt, n.d.;
Raab and Libset, 1968; Rose, 1964; Royce, 1927; Schuman and Harding,
1964; Swayne, n.d.; TeSelle, 1973; Thomas, 1903-04, 1904, 1927;
U.S., 1977; Vander Zanden, 1983; Van Til, 1975; Young, 1925, 1927a,
1927b). Similarly, there are also a prolific number of relevant studies concerning minority-majority relations (Baker, 1974; Bennett,
1971; Berelson and Salter, 1946; Berg, 1975; Bernard, 1925-26;
Bettelheim and Janowitz, 1949; Bogardas, 1925a, 1925b; Carlson,
1971; Dinnerstein, 1979; Ehrlich and Feldman, 1977; Faris, 1925;
Finney, 1926; Frazier, 1957; Gist, 1932; Gregory, 1971; Handlin, 1957;
Haydon, 1978; Iglitzin, 1972; Kinton, 1974; Mack and Duster, 1973;
Miller, .1976; Nash and Weiss, 1970; Noar, n.d.; Park, 1924a, 1924b;
Radke, et. al., 1949; Stabler, et. al., 1970; Stanton, 1960; Sullivan and Goodell, 1975; Takaki, 1979; Wasserman, 1976; Williams and Edwards,
1969; Young, 1927a, 1927b, 1927c; Zeligs, 1936, 1938). These areas overlap and are often one and the same as denoted by the above referenced studies which could have just as easily been grouped together. The cognitive component and the concept of stereotypes, or more appropriately, the mismeasurement of stereotypes, however, are more relevant to this study. 20
The broad definition of a stereotype—a picture in one's mind—is used too often by professionals and laymen alike. The infamous Katz and Braly (1933) study is perhaps the greatest contribution to this erroneous measurement and use of the concept of stereotypes; or at least it..has dominated the field of study concerning the cognitive component of ethnic group attitudes and prejudice. The Katz and Braly method of study had college students select from a list of eighty-four "traits" those that they considered most characteristic of various ethnic groups (1933:282-284). Based on these samples Katz and Braly then constructed an index of
"definiteness of the ten racial stereotypes" (1933:288) by counting the fewest number of traits required to include fifty percent of the five hundred choices possible by all subjects, which they found to range from 4.6 for Negroes to 15.9 for Turks (1933:287). On the basis of the degree of consensus by their subjects, Katz and Braly described their research as a study of "racial stereotypes." Other studies (Gilbert, 1951; Karlins, et. al., 1969) replicated Katz and
Braly and substantiated their findings (Vander Zanden, 1983:20).
However, the theoretical and empirical basis for such quantum inferences must be questioned because:
Actually they (Katz and Braly) made no attempt to ascertain whether or not any particular set of judgements by an individual student did or did not result from stereotyped thinking of the sort described by Lippmann (1922). In effect, Katz and Braly were using the term "stereotype" simply as a pejorative desig nation for "group concept" (Harding, et. al., 1968:7-8; emphasis added). 21
Therefore, many of the numerous studies that have based their research upon the Katz and Braly position and methodology are in fact invalid because they are not measuring stereotypes (Bain, 1939;
Blake and Dennis, 1943; Brody, 1973; Child and Doob, 1943; Edwards,
1940; Gardner, et. al., 1975; Gordon, 1962; Gregory, 1941; Hamil,
1970; Johnson and Greenbaum, 1980; Katz and Braly, 1935; LaViolett and Silvert, 1951; Martin and Halverson, 1981; Meenes, 1943; Meltzer,
1939; Razran, 1950; Saenger and Flowerman, 1954; Sherif, 1935).
Despite the prolific number of erroneous studies of stereotypes, correct uses of the concept are found in a few earlier studies.
(Lippmann, 1922, 1927; Lipsky, 1927; Rice, 1926a, 1926b; Weeks,
1927). A few recent works also promulgate acceptable alternatives or correct uses of the stereotype concept (Bogardas, 1950; Brigham,
1973; Ehrlich and Rinehart, 1965; Funk, 1976; Gonzalez and Williams,
1981; Hamilton, 1981; Hesselbart, 1975, 1978; Smith, 1980). Regardless of the term used—stereotypes or prejudice—the data collected are not measurements of these concepts, but rather descriptions of ethnic groups. This, then, raises considerable questions about the validity of the bulk of the studies purporting to analyze the stereotypes of
Native Americans and other ethnic groups (see Appendixes A and B).
Neither this researcher nor Harding, et. al. could find any study that attempted "to determine the actual basis of the judgements recorded or to the extent to which they involved a failure of rationality" (1968:8). This position, thereby, makes explicit that while the concepts of stereotypes and prejudice are of ultimate
theoretical interest, very limited techniques for the measurements
of such currently exists. This issue will be discussed further in conjunction with the effects of reading and persuasion/propaganda
theory.
Melvin L. DeFleur (1970) presents one of the few attempts at organizing communications concepts and ideas into theories,
particularly theories of mass communications. The Social Categories
Theory (DeFleur, 1970:122-124), expressly derived from the communi cations model discussed previously, is the most relevant because it:
. . . assumes that there are broad collectivities, aggregates, or social categories in urban-industrual societies whose behavior in the face of a given set of stimuli is more or less uniform. . . . knowledge of several very simple variables—age, sex, and educational attainment—provides a resonably accurate guide to the type of communication content a given individual will or will not select from available media (DeFleur, 1970: 122-123).
This theory is more in line with the Social Facts Paradigm
(Ritzer, 1980:35-82), but also overlaps with the Social Behavior
Paradigm (Ritzer, 1980:141-186). Its basic assumption is that people with similar characteristics also have similar folkways, and therefore, they will select similar communications content.
This idea is then extended to assume that the content will cause the members of a particular category to respond in similar manner
(basic Stimulus-Response (S-R) assumption of the Social Behavior 23
Paradigm) (DeFleur, 1970:123). This theory, however, does not exclude the notion of the "minding" process—essential to the
Social Psychological Definition Paradigm—that is seemingly implicit to the adopted communications model. In fact, the assumption of the theory as presented by DeFleur merely focuses upon the content component of the model. If, however, one further develops the assumptions of the whole model it becomes apparent that there is indeed an integration of a "minding" process explicitly presented in the source, the receiver, and the encoding/decoding processes.
DeFleur also discusses an extension of The Social Categories
Theory that he labels the Social Relationships Theory. This approach, however, is conceived of here as being more similar to the former than different and, therefore, will be assumed to be included in the first theory. The main contribution of the Social Relationships Theory, derived from research on the "effects" of mass communications
(Lazarsfeld, Berelson, and Gaudet, 1940), is its emphasis upon:
the important role of group ties, as a complex of intervening variables between media and audience influence, . . . [and] . . . the recognition that informal social relationships play a signifi- . cant role in modifying.the manner in which a given individual will act upon a message which comes to his attention via the mass media (DeFleur, 1970:125, 127; emphasis in the original).
This theory is clearly an enlargement of the focus from the content
(in The Social Categories Theory) to other factors that emphasize the "minding" process and the individual's control over the communications process. 24
Related to these first two perspectives is what DeFleur calls the Cultural Norms Theory (1970:129-139), which he considers the basis for the criticisms of the harmful effects of the media:
Essentially, the cultural norms theory postulates that the mass media, through selective presentations and emphasis of certain themes, creates impressions among their audiences that common cultural norms concerning the emphasized topics are structured or defined in some specific way. Since individual behavior is usually guided by cultural norms (or the actor's impressions of what the norms are) with respect to a given topic or situation, the media would then serve indirectly to influence conduct. Stated in social psychological terms, the media are said to provide a "definition of the situation" which the actor believes to be real. This definition provides guides for action which appear to be approved and supported by society. Therefore, conduct is indirectly shaped by exposure to communications (1970:129-130; emphasis in the original).
The Cultural Norms Theory succinctly summarizes the implicit theory behind most studies of the treatment of Native Americans and minor ities in various forms of media. This essentially social factist position can be altered to a social psychological definitionist one by including the idea that individuals have a degree of freedom through their "minding" process that allows the media to (1 ) rein force existing patterns, (2) create new shared convictions, or
(3) change existing norms and reference groups (DeFleur, 1970:130).
Such a process in embraced by persuasion and propaganda theory.
Persuasion and propaganda theories, which are often conceived of as simple "extensions and utilizations of the contemporary theories of mass communications" (DeFleur, 1970:140; emphases added), are 25 related to these other theories of mass communications. DeFleur presents propaganda theory as resulting from the Mechanistic S-R
Theory (1970:112-117) and studies into the effects of mass communi cations during World War I (i.e., especially Lasswell, 1927).
Modern theories—now often called persuasion theories—have been developed by the systematic replacement and/or modification of the basic assumptions of S-R theory and propaganda. Theories of persuasion have a history extending back to Aristotle's discussions of rhetoric, and continue to be a primary theoretical orientation in communications theory (Abelson, 1959; Andersen, 1971; Applebaum and Anatol, 1974; Beisecker and Parson, 1972; Bettinghaus, 1968;
Brembeck and Howell, 1952; Bryson, 1948; Cronkhite, 1969; Ehninger,
1974; Fotheringham, 1966; Gordon, 1969, 1971; Miller, 1946; Reardon,
1981; Roloff and Miller, 1980; Rosnow and Robinson, 1967). Theoreti cally, two basic models are generally followed. The first is the psychodynamic model which is formulated upon the Individual Differences
Theory, and is essentially based upon a psychological S-R model.
The second model, appropriate to this study, is what DeFleur calls the sociocultural model, which is based upon both the Social Relation ships and the Cultural Norms theories and the "assumption that mass communicated messages can be used to provide an individual with new and seemingly group-supported interpretations of some phenomenon toward which the individual is acting" (1970:145). Schematically, 26 this model of persuasion is (DeFleur, 1970:146)
(Persuasive Defines (or Forming or alter Achieves [message _ redefines) ing definitions change in sociocultural for socially direction processes of approved behavior of overt group(s) for group members behavior
The relationship between propaganda and persuasion is a dubious one. It currently appears that propaganda is seen as counterfeit abuses of persuasion (Cronkhite, 1976; Ehninger, 1974;
Mathesson, 1977), or as a very negative aspect of persuasion; that is, a concept no longer accepted. It seems, however, that the many studies in the area of propaganda (Albig, 1939; Barghoorn, 1964;
Berelson and de Garzia, 1947; Berelson and Janowitz, 1966; Bruner,
1941; Doving, 1959; Ellul, 1965; Flowerman, 1947; George, 1959a;
Lasswell, 1927; Lasswell, et. al., 1935; Merton and Lazarsfeld, 1957;
Mosher, et. al., 1941; Murphy and Likert, 1938; Root, 1962; Smith, et. al., 1946; Whilton and Larson, 1964) and rumor (Allport and
Postman, 1946, 1948; DeFleur, 1962; Rosnow and Fine, 1976;
Shibutani, 1966) suggest that propaganda theory may instead include the concept of persuasion. The superior utility of propaganda is explicitly seen in Leonard Doob's monumental essay wich presents the theory of Propaganda: Its Psychology and Techniques (1935).
While presenting a propaganda theory, Doob also provides the most definitive definition of the concept of propaganda: 27
Intentional propaganda is a systematic attempt by an interested individual (or individuals) to control the attitudes of groups of individuals through the use of suggestion and, consequently, to control their actions; unintentional propaganda is the control of the attitudes and, consequently, the actions of groups of individuals through the use of suggestion (1935:89; emphases in the original).
Doob also establishes what he calls the "principles of propaganda," which similarly can be labeled scope statements (see Appendix C).
In discussing his theory of propaganda, Doob provides the basic criteria of a good theory (Cohen, 1980:191): (1) a relatively explicit theory; (2) explicit definitions; (3) a logical skeleton of theory; (4) clear scope statements as to the domain of applica bility; (5) the development of testable hypotheses; and (6) the formulation of abstract problems. Specifically for this study,
Doob's theory provides that:
Books . . . are without a doubt, one of the most important media for unintentional propagandists. For partially through books or through books interpreted by teachers does the younger generation obtain involuntarily attitudes toward the distinctive features of its cultural pattern (1935:393).
Doob's contention that propaganda is totally persuasive through the media, then, is a theoretical underpinning that should be used in..the analysis of literature and other communications, but one that seldom has been used (for an exception see Wright, 1939).
The final consideration in this theoretical rationale is that of the effects of communication, and literature in particular.
There have been numerous studies concerning the effects of literature 28 and efforts to alter attitudes (Beggs, 1956; Berreraan, 1940;
Dawson, 1959; Fischer and Torney, 1976; Georgeoff, 1968; Gezi and
Johnson, 1970; Jackson, 1944; Litcher and Johnson, 1969; Manley,
1944; Menefee, 1936; Morris, 1981; Poussaint, 1970; Ruth, 1969;
Smith, 1948; Weiss, 1968; Westphal, 1977; Yee and Fruth, 1971).
However, Waples, Berelson, and Bradshaw's study on What Reading
Does to People (1940), which is discussed above, is the most comprehensive and seminal as well:as neglected study.
It was stated above that the process of becoming aware of messages would be again discussed, as would the concepts of stereo types and prejudice. The concepts of awareness and prejudice are partially interwoven into the outline for the study of the effects of literature proposed by Waples, Berelson, and Bradshaw (1940).
Awareness is explicitly subject to the preconditions and distribution of materials and implicitly a part of the publications themselves and the reader's predispositions, while prejudice is explicitly a part of all four phases. That is, racial awareness (or awareness of availability of literature) is in part subject to the preconditions and distribution, but may or may not effect the content or the reader's predispositions. Conversely, attitudes of prejudice may alter the preconditions and distribution of materials, as well as the supply, style, and content of the publication. Similarly, the reader's attitudes of prejudice are paramount preconditions to the reading of materials, so that the effects will vary greatly. Related Methodological Research
The methodology selected for this study is primarily a quantitative descriptive form. Since it is a multi-disciplinary study, methodological considerations are drawn from various fields
(Bailey, 1978; Barchs, 1959; Berger, 1982; Cicourel, 1964; Condie and Christiansen, 1977; Conkling, 1975; Dixon, 1977; Emmert and
Brooks, 1970; Festinger and Katz, 1953; Galvin and Book, 1972;
Goode and Hatt, 1952; Gottschalk, Kluckhohn, and Angell, 1947;
Hansen and Parsons, 1968; Harrah, 1963; Kaplan, 1964; Lindzey,
1954; Lindzey and Aronson, 1968; Orenstein and Phillips, 1978;
Schwartz and Jacobs, 1979; Siegel, 1956; Smith, R.G., 1978;
Webb, et. al., 1966). Content analysis, the selected method for this study, has been previously discussed by the researcher
(Edwards, 1981), while other seminal methodological discussions include George (1959b), Kaplan (1943), Krippendorff (1980),
Lasswell, et. al. (1949), and Rosengren (1981). Many researchers emphatically perceive content analysis to be the primary or only method for the analysis of communications (Berelson, 1952;
Gerbner, et. al., 1969; Holsti, 1962, 1969; Krippendorff, 1980;
Stone, et. al., 1966) while others advocate both content analysis and other methods and techniques (Berger, 1982; Emmert and Brooks,
1970; Kaplan, 1943), and some, more recently, put the method in disrepute (Golding and Murdock, 1980). Regardless, content analys 30 has been a method of study since at least the eighteenth century
(Dovring, 1954) and the method or subject of numerous studies
(see Appendix D) on the various vehicles of communications
(i.e., newspapers, radio, motion pictures, literature, etc.) and their messages.
Content analysis, therefore, has been used in various forms to analyze literature and the effects of literature on people, especially children. Several general studies on literature and textbooks have adequately synthesized the research to date (see especially Bachtold, et. al., 1977; Banfield, 1975; Bekkedal, 1973;
Billington, 1966; Chambers, 1968; Hester, 1975). More specifically, methods of content analysis have been used to analyze (1) blacks
(Banks, 1969; Broderick, 1973; Nelson, 1926; Shuey, et. al., 1953;
Walker, 1971); (2) Mexican-Americans (Blatt, 1968; Gaines, 1971); and (3) racial/ethnic groups in general (Cornelius, 1971; Dale, 1970;
Elkin, 1965; Fisher, 1977; Gast, 1965, 1967, 1970; Kane, 1970a,
1970b; Miller, 1978; Monti, 1979; Pierce, 1977; Singh, 1975; U.S.,
1980; Weigel, et. al., 1980). Other studies have utilized content analysis to focus upon, and identify values, beliefs, and attitudes
(Albrecht, 1956; American Educator, 1983; Asman, 1983; de Charms and Moeller, 1962; Holsti, 1962; Moline, 1983; Reddick, 1934).
More recently other forms of bias and sexism have become the focus of content analysis studies and literature analysis (Caute, 1980; 31
Frasher and Walker, 1972; Kolbe and LaVoie, 1981; Queen, 1975;
Schlachter, 1977; Schnell and Sweeney, 1975; Scott, 1980; Sex
Discrimination Assistance Center, n.d.; United States Commission on Civil Rights., 1977.)- It should be noted, additionally, that literature—and in particular fiction—has been extensively critiqued and analyzed by more qualitative methodology (Brown, 1965; Eastman,
1965; Hazard, 1937; Leisy, 1929; Lubbock, 1957; Quinn, 1951;
Spiller, 1957; Walcutt, 1966).
Instruments for evaluating literature have been the subject of numerous essays. Qualitative techniques and methods include simple guidelines and suggested criteria (Byler, 1973; CIBC, n.d.;
Costo and Henry, 1970; EPIE, 1981; Gilliland, 1980; McGraw-Hill,
1975; Peltola, 1979a, 1979b; Rollins, 1941, 1967; Woodbury, 1979,
1980a, 1980b). Semi-quantitative methods include checklists and evaluation forms (EPIE, 1975; Garcia and Garcia, 1980; Haley, et. al.,
1978; Khatri, 1980; Rosenberg, 1973; SSEC, 1975). Quantitative methods encompass sophisticated techniques like values analysis
(Lester, 1982; White, 1947), and bias or evaluative assertions/loaded words analysis (McDiarmid and Pratt, 1971; Merrill, 1965; Pratt,
1969, 1971, 1972; Rosi, 1964). However, it is found that most researchers formulate their own instruments (Berelson and Salter,
1946; DeFleur, 1962; Gast, 1965, 1967, 1970; Holsti, 1969).
Occasionally, as was the case with the Gast method (1965), other 32 researchers will copy previous methods (Barron, 1981; Broderick,
1973; Cata, 1977; Gast, 1967, 1970).
Several problems are inherent to many of these instruments, particularly those formulated by one researcher and then used by others. Theoretically, as mentioned above, many of the studies infer
"bias," "prejudice," or "stereotypes" from what are essentially
"traits" or "characterizations" categories, which has been a major problem with many theses and dissertations analyzing Native Americans in literature (see Appendix A for a list of studies). A second problem, especially in earlier studies is the lack of formal documentation of research methods and instruments. Literally hundreds of studies, of which a part is listed in Appendixes A and B, have been done on Native Americans in various forms of media, but very few have included examples of their instruments and techniques. A third problem, specific to the quantitative studies, is that of validity and reliability. This problem is often compounded by the absence of examples of their instruments. Most importantly, however, is the problem that very few studies deal with validity and reliability at all (for exceptions, see Barron, 1981; Cata, 1977), which are often major handicaps to the method, as best exemplified by studies of
"stereotypes" of Native Americans.
It should be noted at this juncture that other methods have been used to look at the images of Native Americans such as historical 33 analysis (Berkhofer, 1978; Byler, 1973; Churchill, et. al., 1978;
Costo and Henry, 1970; Ewers, 1972; Fiedler, 1968; Henry, 1973;
Hoover, 1976; Hoy, 1979), while many others have used-literary criticism (Aitson, 1980; Bataille and Silet, 1980; Chapman, 1975;
Deloria, 1980a, 1980b; Fairchild, 1928; Jones, 1958; Keiser, 1933;
Lucas, 1980; Medicine, 1979; Sanders, 1973; Tenkate, 1970; Troy,
1972), which here is defined as inclusive of qualitative content analysis. Braroe and Braroe (1977) utilized social anthropological observation techniques in looking at images and identities of Northern
Plains Native Americans, while Edwards (1980) utilized observation and library research techniques to analyze the library cataloging system, as a form of institutional discrimination, to discover images presented by the cataloging system. Lastly, in a very important study of racial attitudes and self-identity in Native American children, Beuf (1977) utilized twelve methods—including content analysis—to ascertain images of Native Americans. These other methods, however, are not as applicable to a quantitative analysis of the portrayals of Native Americans in literature.
Over the past eighty years or more, many types of literature have been analyzed and critiqued for their treatment of Native
Americans. However, very little research has focused upon the partial factor of publication style or aspects of how Native Americans are portrayed. More specifically, they do not determine how evaluative 34 assertions or value judgments are employed by various sources—i.e., authors and/or publishing companies—of such literature. Generally, previous researchers have dwelt on the publication's content or what is being said which, as such, implies that what is included and/or omitted from the contents will account for the major factors of the literature's influence upon the reader. Furthermore, the methodologies of such studies almost invariably have been qualitative or "semi-quantitative" and, for the most part, non-systematic; that is, the researchers' methods and instruments are often not explicitly provided, which means they are not easily replicated and are of questionable validity.
While many studies of the treatment of minority groups
(including Native Americans) were made prior to World War II
(Bissell, 1923; Fairchild, 1928; Johnson, 1911; Keiser, 1933; Lew,
1923; Malone, 1927; Reddick, 1934), it was only after the war and the development of more systematic methodologies in content analysis that such treatments emphasized quantitative measurement. Berelson and Salter's 1946 study of the treatment of minorities in magazine fiction was one of the.first quantitative studies in this area, and has since become the hallmark for most research of this type. The focus of their study was minority characterizations as measured by eight broad categories, which the authors only partially defined in operational terms. Unfortunately, their instrument(s) for measuring 35 categories were not published anywhere and only implicitly presented in their report. A further problem with this study is that in discussing the results the authors analyzed only several of the categories in their study. To their credit, however, in doing their analysis they utilized demographic statistical data as an independent standard criteria with which to compare their results.
As a first effort at analyzing what the contents from one form of literature (popular magazine fiction) says about minorities,
Berelson and Salter's study is invaluable. Subsequently, many studies were spawned to analyze what books—especially children's fiction and textbooks—were saying about minorities, including countless quali tative and quantitative studies on presentations of Native Americans.
Whereas, most post-war investigations followed the path of
Berelson and Salter (and many pre-war studies) in analyzing what constitutes the treatment of minorities in literature, a few researchers have pursued assertion analysis as delineated by Janis in 1949:
Assertion analysis will probably be found to be the most productive type of content analysis, inasmuch as the "thematic content".corresponds most nearly to the overall signification of communication. . . . The assertions found in a communication are the primary content indicators of the intentions and motives of the communicator. Similarly, the effects which a communication produces on an audience are primarily due to the assertions content (p. 67).
One is, ultimately, interested in the potential effects on the audience, yet, as stated above, most researchers and literature critics assume that it is what is said and not how it is said, as
proposed by Janis, that primarily accounts for the effects of the communication on the audience/reader. It was, nonetheless, not
until 1956 that a usable instrument and methods were published by
Osgood, Saporta, and Nunnally. Their instrument, "Evaluative
Assertion Analysis," measures evaluative assertions through the connection of an evaluative (or common meaning) term/complement of variable direction and intensity to a subject (or attitude object) by means of a connector (or verb) of varying direction
and intensity, i.e., subject-verb-complement (Osgood, et. al.,
1956). For example, in the phrase "Indians were called savages,"
"Indian" is the subject or attitude object, "were called" is the
verb or connector, and "savage" is the complement or evaluative
term. The evaluative terms and connectors are weighted on a scale of +3 to -3 by an analyst, i.e., "were called" is scored as +2 and
"savage" is scored as -3. The analyst then multiplies these scores for each subject, and the sum of the products is divided by the absolute sum of the connector weights to provide a score indicative of the evaluation of the subject by the source, as shown in Table 1.
This method has been partially implemented by researchers studying similar problems (Lopatka, 1965; North, et. al., 1963; Osgood, et. al.
1957; Osgood, et. al., 1975; Snider and Osgood, 1969), but apparently
has been completely adopted to analyze minorities (in textbooks) only by McDiarmid and Pratt (1971). 37
TABLE 1
Examples for Calculating "Evaluative Assertion Analysis"
Sum of all C x CMT (or AO) Formula: AQ Sum of all absolute values of C
CMT2 Assertion C (or AO) Product
1. Indians were called = savages +2 X -3 -6 2. Indians were called
very desperate +2 X -1 - -2 3. Indians were said to have hurt the settler -1 X +1 .5 = -1.5 4. Indians were hostile +3 X -2 :: -6 5. Indians made savage = raids +3 X -3 -9 6. Indians raided settlers -2 X +1 .5 = -3 Sum of values I;13| -27.5
Overall evaluation score for "Indians" in one textbook = ~ "2-12
1 —C = Connectors 2—CMT = Common Meaning Terms (AO = Attitude Object)
In the course of using the Osgood, Saporta, and Nunnally method to analyze the treatment of various minority groups in
Ontario social studies textbooks, McDiarmid and Pratt (1971) encountered a "number of defects in the instrument" (Pratt, 1971:4). 38
As a solution to these defects, Pratt (1969) developed a similar instrument, the Evaluative Assertion Rating System (EARS), which
"attempted to provide an 'empirical' basis for the weights assigned to assertion components, and to take into account two dimensions of evaluative assertions implicitly ignored by Evaluative Assertion
Analysis" (1971:4). The two dimensions were the tense of the verb and the number of subjects. After constructing the EARS instrument,
Pratt applied it to evaluate the treatment of Arabs, French
Canadians, Indians, and Negroes in sixty-nine history textbooks
(1969), and again in the study discussed above (McDiarmid and Pratt,
1971). On the basis of these studies Pratt conducted a secondary analysis to determine the potential for creating an instrument usable by teachers and administrators, which resulted in the creation of the Evaluative Coefficient (ECO) Analysis method and instrument (1971). As part of the ECO method, Pratt developed a standard word list, which was published in toto in a comprehensive user's manual (1972).
In summary, the study of the treatment of Native Americans in literature has emphasized the partial factor of publication contents or what is being said, and for the most part these investigations have looked at "characterizations" rather than
"stereotypes" of Native Americans and/or other groups. Additionally, the focus of these studies has generally been children's fiction and 39 textbooks—especially during the last fifteen years—and the method ologies have been qualitative and/or very questionable. It should be noted, however, that recent studies have demonstrated greater reliability by investigating thematic frequency (Banks, 1969; Garcia,
1978, 1980; Shaughnessy, 1978). A substantial review of previous research indicates that the ECO method—which demonstrates the greatest usability to date—has been utilized primarily to study treatments of groups in textbooks. More importantly, the method has been essentially ignored by researchers. The only studies located that even partially adapted the method were Kirkness (1977)— also a Canadian study of textbooks—and, to a very limited extent,
Garcia (1978, 1979) where ECO was used in conjunction with thematic frequency. Other Canadian reports were found that have referenced the method (LaRoque, 1975; Manitoba Indian Brotherhood, 1974), but neither actually used it.
Research Problem "
The present study analyzes the value judgments about Native
Americans made by sources, over time, in children's fiction. The need for such a study is clearly demonstrated by the above discussion.
The present study is significantly related to the problems of racial awareness and understanding communications theory and it begins to narrow the research gap of how Native Americans are portrayed in children's fiction. Furthermore, it provides for the productive expansion of, and exploration with, known techniques; which will, hopefully, expose this useful method to other researchers as well.
This research project is extremely beneficial in that it opens up new horizons in understanding how communications messages (on Native
Americans) make evaluative assertions and judgments that may effect the attitudes—and ultimately behaviors—of bias and prejudice
(favorable or unfavorable), and discrimination.
Specific questions, therefore, that this research answers are
1. Are portrayals of Native Americans and other subjects evaluative? Or are they neutral?
2. If evaluative, are the judgments made by those sources favorable or unfavorable toward Native Americans?
3. How do judgments of Native Americans compare to those of Non-Native Americans within the source? Between sources?
4. What is the frequency of evaluative terms to the total space devoted to Native Americans? Non-Native Americans?
5. How frequently are specific terms or concepts applied to Native Americans? Non-Native Americans?
6. Have sources become less evaluative over time?
7. Are standard themes used?
The outcome of this study—or the answers to these questions—will provide a descriptive analysis of how Native Americans are portrayed by various sources in children's fictional literature. It is impor tant, however, to keep in mind several important restraints or assumptions involved in such an analysis. 41
First, the evaluative judgments (or ECO) analysis of how
Native Americans are portrayed constitutes only a subpart of the study of literature style. The style of literature, however, is but a partial factor of the publication itself, which is one of four factors involved in evaluating the effects of the reading experience:
It is entirely proper to discuss character of production, scope of distribution, slanting of the content, or predispo sitions of the readers [separately], provided that we [explain that we] are talking about factors. The impropriety consists in talking about one of several factors at work as though that one factor, if clearly established, were certain to produce a corresponding and equivalent effect. Our survey of the liter ature has shown not merely laymen and the popular writer but even the experienced students in the field to be careless about the distinction. They discuss factors, partial factors, as though they were effects. Among the most serious offenders are those who study the reading of students. We have yet to find any comprehensive study of the effects of students' reading. But we have found many studies of isolated factors and of partial effects upon which far-reaching educational recommend ations have been based (Waples, et. al., 1940:12; emphases in the original).
The assumption, then, is that identified evaluative assertions cannot be taken to be the only causal factor in the literature's effect upon the reader.
Second, it is assumed that fictional literature is, and probably will continue to be, read by children and young adults for a variety of reasons. In this connection, the third assumption is that such fictional literature is made available through either public or school libraries, particularly in this time of increasing book costs. Both of these assumptions are underpinned by our cultural emphasis upon reading.
A fourth general assumption, related to the first, is "that the expression of value judgements in written communications has some effect on the reader" (Pratt, 1971:2). Research has shown that students' attitudes toward social issues and minorities can be effected by reading materials (Beuf, 1977; Fisher, 1968; Haugh,
1952; Litcher and Johnson, 1969; Shirley, 1966; Trager and Yarrow,
1952), and that written communications affect the ideas and beliefs of readers (Okanes, 1956; Payne, 1964; Tannenbaum, 1953; Watts,
1967; Weiss, 1961). This is to say that the literature—specifically fiction—to be analyzed has effects upon the readers in ways which can be analyzed.
A fifth assumption is that the role of literature as a factor in the resulting effects is revealed by its manifest content and by the collection of symbols of which it consists. Related to this is the assumption that appropriate methods of analysis can identify the stimuli provided in given publications (Waples, et. al.,
1940:147).
A final assumption is that while the primary function of reading fiction may be manifestly for distraction or respite, other less than obvious social effects may result by conveying new meanings and stimulating emotions about the various subjects, regardless of the intent of either the source or the reader. A3
This analysis of previous research indicates an absence of theoretically grounded and methodologically sound research in the study of literature and its effects. Given a virtual absence of quality research on Native Americans in literature, the assumptions as outlined above are utilized to orient this study. In addition,
Garcia (1978, 1980) and Pratt (1969, 1971, 1972) serve as empirical exemplars or sources, and Pratts' ECO Analysis method provides the foundation for this study's methodology. The theoretical orientation for this study, as traced above, begins with the proposed Social
Psychological Definition Paradigm. More substantively, this study derives direction from Lasswell's expanded communications process/act model (see page 11). This model encompasses middle range theories concerning awareness, attitudes, mass communications, and persuasion/ propaganda. As applied to this research, then, the theoretical orientation provides: (1) that individuals are capable of thinking and interacting with their environment; (2) that communications is a process or act between two or more people; (3) that through a
"minding" process awareness of others is both controlled by and controls the communications process; (4) that the communications process, and more specifically literature, affects attitudes— particularly attitudes of prejudice or bias about Native Americans;
(5) that communications in literature portray given subjects, but in and of themselves do not create stereotypes; (6) that mass media 44 communications, as explained by the Cultural Norms Theory, reinforce existing patterns, create new shared convictions, and change existing norms and reference groups; (7) that extremely evaluative communi cation becomes more than persuasive and is unintentional propaganda, which if perpetuated over time becomes intentional propaganda; and
(8) that the communications act through literature affects the reader's attitudes in varying degrees.
Combining this paradigmatic orientation and the trends from the most recent studies (Bachtold, et. al., 1977; Banfield,
1975; Barron, 1981; Bekkedal, 1973; Broderick, 1973; Cata, 1977;
Garcia, 1978, 1980; Hester, 1975) the following are the expected results of this study:
1. That books published in the 1960s will be more evaluative than books published in the 1970s and 1980s.
2. That the evaluativeness of the books, based upon the sample average score, will be less than ten percent (<10%).
3. That the percentage of descriptive words for all subjects, groups, and books will be greater than ninety percent (>90%).
4. That books published after 1975 will have improved because of outside demands and, therefore, will have a descriptive percentage greater than ninety-five percent (>95%).
5. That books published after 1975, similarly, will have an even lower Evaluativeness Score of less than five percent (<5%).
6. That evaluative judgments will be favorable (i.e., >75%) toward Native Americans.
7. That judgments of Native Americans will be less favorable than those of others. 8. That the percentage of evaluative terms to the total space devoted to Native Americans will be greater than that for Non-Native Americans.
9. That the percentage of negative evaluative judgments about Native Americans will be greater than five percent (>5%), while negative evaluative terms for Non-Native Americans will be less than five percent (<5%).
10. That similar themes concerning Native Americans, found by Garcia in textbooks, will be found in the fiction books. Chapter 2
METHODOLOGY
Research Population and Samples
A major obstacle of content analysis studies is establishing the population's boundaries. In determining the population boundaries for this study, a review of previous sampling procedures and a pre- investigation study were conducted. Generally speaking, the majority of previous studies utilized some type of recommended readings list for their populations. Such lists often came from Portraits, Book- finder, and Wilson's Catalogs, while more recently Barron (1981) used bibliographies as well, and Cata (1977) utilized local sources and available books. Based upon the assumption that the primary way in which individuals come into contact with fictional books on
Native Americans is through their public and/or school libraries, an investigation also was made of their acquisition processes.
Accordingly, interviews with public and school librarians were conducted, along with reading various guides (e.g., Woodbury, 1979,
1980a, 1980b). The public library acquisition process consists of reading book reviews in The Library Journal, School Library Journal,
Booklist, The Hornbook Magazine, Publishers Weekly, Wilson's Catalogs,
CIBC Bulletin, Center of Children's Books Bulletin, and various announcements from publishing companies. Each librarian then
46 kl
indicates the books he or she feels should be ordered. Based upon
the recommendations of the librarians and rereading the reviews,
the librarian in charge of acquisitions then orders the suggested
books as allowed by current finances. Replacement of books is based
upon demand, while removal generally is based on the book's physical
condition, although complaints also are also considered. The
recommended process for school libraries is similar to that of the
public library, although the guides used for selecting the books
are different. Rather than using The Library Journal and others,
school libraries rely on Wilson's Catalogs first, then upon Book-
finder, Adventuring with Books, Books for You: A Booklist for
Senior High Students, Your Reading: Booklist for Junior High
Students, School Library Journal, and finally, Booklist.
Since the major guides used were found to be Wilson's
Catalogs, The Library Journal, Bookfinder, Booklist, Publishers
Weekly, School Library Journal, The Hornbook Magazine, and various
lists published by the National Council for Teachers of English
(NCTE), a preinvestigation study was made of these sources. During
the first phase of this research it was determined that books on
Native Americans are primarily found under the 300 (Social Sciences),
572/573 (Anthropology), and 900 (History) classifications of the
Dewey Decimal Classification System. Therefore, the researcher
decided to narrow the population to these three forms of nonfiction 48
literature. Subsequent research indicated that all the books listed
in the other major guides also were included in the more comprehen
sive Wilson's Catalogs, and this guide, therefore, was used as the
source for the research population. The Wilson's Catalogs used
were the Children's Catalog (1956, 1961, 1966, 1971, 1976, 1981,
1982), the Fiction Catalog (1960, 1971, 1975, 1980, 1981), the
Junior High School Library Catalog (1965, 1970, 1975, 1980, 1981),
the Public Library Catalog (1968, 1973, 1978, 1979, 1980, 1981),
and the Standard Catalog for High School Libraries (1957, 1962, 1967,
1972, 1977, 1978, 1979, 1980, 1981). The researcher and an associate
read all listings, looking for particular key items, to identify
books from this universe for the population. The primary indicator
was the work "Indian," while secondary identifiers included:
(1) specific Native American national names (i.e., Cheyenne, Apache,
Eskimo) for all groups north of Mexico; (2) specific Native American
individuals (i.e., Chief Joseph, Geronimo); and (3) specific histori
cal events/places known to have involved Native Americans (the only one, ultimately, being "Custer's Last Stand"). Utilizing these general requirements, all the titles in the various Wilson's Catalogs
published between 1960 and 1982 were placed into population listings.
After all titles had been recorded, the population listings were culled further by the researcher's associate who selected only those
books with an original copyright of 1960-1982. These books then 49 formed the research population. For sampling purposes the popu lation was left stratified by reading level (i.e., Preschool or
Easy, Children's, Junior High, Senior High, Adult), literature type (i.e., fiction, nonfiction), and Dewey Decimal Classification form (i.e., 300, 572/573900). Table 2 presents the research population by level, type, form, and respective subpopulation sizes, with a total ^population size of N^=1013 books.
It originally had been proposed (Edwards, 1983) to do a stratified random sample of the population developed in the pre- investigation study, as discussed above. A ten percent sampling was made based upon the researcher's interest to achieve a proportional error of .10; i.e., P + 2 (SE) or P +.10. In order s s to achieve this, a sample size of Ns=100 was required. Furthermore, because of the extremely small subpopulation sizes for anthropological
(i.e., 572/573) forms of literature, it was decided to purposively sample all the books in this literature form. Additionally, because of the conceived importance of preschool materials in the development of racial awareness and attitudes, this level also was purposively oversampled. The total sample size, therefore, was to be 112 books (see Table 2); however, the proposed procedures
(Edwards, 1983) did not work well with such a sample.
The first problem encountered was that of book size.
Pratt (1971 ) indicated that sampling within a textbook would Table 2 Study Universe, Populations, and Samples
Random Purposive Level Type Form Size Sample Sample
Preschool (Easy) Fiction/ Nonfiction NA 19
Children's Nonfiction Social Sciences 74 7 Anthropology 1 0 History 150 15 Fiction 62 6
Junior High School Nonfiction Social Sciences 34 3 Anthropology 0 0 History 141 14 Fiction 63
Senior High School Nonfiction Social Sciences 33 3 Anthropology 2 0 History 126 13 Fiction 29 3
Adult (Public Library) Nonfiction Social Sciences 27 3 Anthropology 1 0 History 113 11 (Fiction) Fiction 138 14
Totals P = 1013 S = 100 S2 = 12 51 provide equivalent results even at fifty percent sampling. Therefore, it was determined that a random sampling of those pages upon which the subjects were listed would be feasible. The next step was to determine the average number of words per page and per book, and the average number of pages per book for each class of books. Thirty percent of the sample books were selected randomly for this analysis.
The total number of pages for each book was first recorded and five randomly selected pages were then counted to obtain the total number of words per page. Next, the mean average number of words per page was calculated. Then, by multiplying this number by the total number of pages, the total average number of words for the book was obtained.
These figures for each book were then summed up for all the books in the class to obtain the class mean. From these figures it was decided that a ten percent sampling of pages should provide a theoreti cally sound and methodologically usable amount of data.
A second problem, which became apparent after initial attempts to analyze several books to validate the above sampling procedures, was endeavoring to determine how many subjects in a given book to evaluate. This was particularly true for nonfiction books that discussed numerous distinctive Native American nations and groups.
A third, and more crucial problem, was the amount of time required to sample one book at even ten percent, and the overwhelming amount of new data to process. At this point it was determined, 52 therefore, that the research population would have to be redefined into a more workable size.
Theoretically, younger children are the group most susceptible to influence by reading books as demonstrated by studies of racial awareness and the effects of reading discussed in chapter one. It also may be assumed that children would prefer to read more enter taining books, which are found most frequently in fiction. Theoretical support for this decision is found in Waples, et. al. (1941), where the authors state that recreational reading has greater influence because the reader already is interested in the material. Secondly, there has been very little research on nonfiction. It would be more useful at this point, therefore, to be able to compare stylistic portrayals of Native Americans in fiction with previous studies, rather than providing a strictly descriptive analysis of how non- fiction is presented.
The research population for this study, therefore, was redefined as those books listed in the fiction category in Wilson's
Children's Catalogs. The population for this level and category, as previously determined, is sixty-two books (N=62). Furthermore, the same random sample for this group was utilized, providing a more workable sample of six books (see Table 2; n=6).
The sampling procedure within each book was conducted through a random quota sampling of the identified subjects or strata resulting 53
in ten random samples per subject. As each subject did not appear
on each page randomly selected, the page served merely as a starting
point. The pages were read until the subject was located. From that
line on, data on the subject was measured by the procedures discussed
below for the number of lines equivalent to one page, which produced,
therefore, the equivalent of ten full page measurements for each
subject. The sampling procedure was justified upon the need to
obtain enough data to allow for statistical analysis, while also
limiting the amount of data to a workable level. At the children's
level the average book was found to contain a mean of 230 words per
page, which at ten pages per subject produced an average of twenty-
three hundred words per subject for analysis of descriptive and
evaluative terms. This, then, works out to be an approximately
seven percent (.0712) sampling for each subject, based upon average
figures. Within the research sample, two books were close to the
average size of book for that category, while two were larger and
two were smaller than average, indicating a fairly representative sampling based upon the size of the books.
The method of. measurement used, as stated in chapter one, was
the Evaluative Coefficient (ECO) Analysis developed by David Pratt, with some minor procedural changes to adapt it to fiction from
textbook types of literature. These procedures, originally delineated
by Pratt in "Value Judgements in Textbooks" (1971) and more 54 comprehensively explained in his "users" manual, How to Find and
Measure Bias in Textbooks (1972), were used to measure subjects in children's fiction.
Measurement Procedures
The following steps are followed in measuring how sources of children's fiction make evaluative and/or descriptive judgments about subjects treated in such books.
1. The first step, discussed in detail above, is to define the sources to be analyzed, which is accomplished by first defining a functional population from which a random sample is then drawn.
Sources are defined here as the author(s) and/or publisher(s) of each book selected to be analyzed.
2. The second step operationally defines the subjects of interest. In selecting subjects certain criteria are imposed.
First, during the initial reading each subject in the book is tracked by recording the number of each page on which that subject appears.
Secondly, each subject should appear on at least twenty percent (20%) of the total pages in the book. In determining subjects, minor characters and groups are collapsed into related groups (e.g.,
"General Canby" is counted as part of the subject "soldiers").
Lastly, the following broad definitions are used in forming subject categories that are to be measured using ECO Analysis: 55
a. "Indians" or "Native Americans" or "Amerindians" or "Eskimos": any group of people or images (e.g., animals, spacemen) identitied simply by one of these nomenclatures, or by illusion (e.g., "feathered creature") and/or description (e.g., dressed in buckskin and feathers) as being a member of such a groups.
b. " Indians/Eskimos/tribe": any group of people (not images) referred to by their proper (e.g., Dine) or ascribed nomenclature (e.g., Navajo, Apache).
c. "Non-Indian/Native American/Eskimo" or "(Name of specific group)": any group of people not of Native American/Indian/ Eskimo ancestry (e.g., "whites," "blacks," "soldiers," "settlers"). This category will be subdivided as required— that is, where specific groups of Non-Native Americans are the subject of more than ten evaluative statements, then separate scores for them are kept.
d. "Characters": any person (e.g., Geronimo, Custer) or singular image (e.g., general, chief, the boy, Yogi Bear) regardless of group affiliation that is referred to in the contents by the source. Again, separate score sheets for each "character" are made.
3. The next step is to prepare an adequate supply of score sheets or instruments (see Appendix E) and to:secure a copy of the ECO Word List (see Appendix F). Pratt's instrument was modified for this study to include columns for recording descriptive terms and the word type (e.g., verb, adjective).
4. The fourth step is to select and train analysts (Pratt,
1972:16-17). However, financial and time restraints precluded this step to self-training on the part of this researcher.
5. The fifth step in the measurement of portrayals in fiction is to read the individual books in their entirety. In reading the 56
book each subject, when first encountered, is listed by page number.
Each succeeding page mentioning the subject is recorded, which both
facilitates the random sampling and also provides an additional
measurement of frequency. Furthermore, a list of themes employed
by Garcia (1978, 1980) is used as an instrument for recording the
frequency and appearance of themes in the book, which also are
recorded by page number. This data then is transcribed onto the
ECO Analysis Administrative Data Instrument (see Appendix E).
6. The sixth step is to write a short qualitative analysis
and story description of the book.
7. The next step is to designate subject categories. This
is done with two primary guidelines: (1) representative sampling;
and (2) Native American/Non-Native American distinctions. Once this
is done, a series of ten randomly selected page numbers for each subject is generated. Each page, or the first succeeding page
mentioning the subject, then is analyzed. Each time a word is used
by or in reference to the subject—evaluative or descriptive—it is
recorded on the score sheet. Articles and prepositions are eliminated from the analysis. Individual score sheets are maintained for each subject. As this process proved to be very time consuming, an
improved technique was developed for this step. After choosing the
random pages they then are xeroxed. Utilizing different colored pens for each subject, each word is scored on the xeroxed page of the text, 57 thus combining this step with step number ten below. While most words that make value judgments or descriptions are usually adjectives, such words may also be adverbs, verbs, nouns, or pronouns:
X admired bravery X ruthlessly took the money X assisted in the crime X seemed to be insane X became more modest X's personality was stable X committed an offence X's prejudice was obvious X did a dishonorable action X talked disloyally X had courage X treatened Y X had very civilized ways X told a lie X lived a good life X tried to be charitable X may have been foolish X was known to be very wise X murdered Y Y disliked X's dishonesty
(Pratt, 1972:19; emphases in the original).
8. In measuring or recording words, the following guidelines are incorporated:
a. In the situation where the term is negatively applied to the subject, i.e., "Jim was not intelligent," the recorded term is the reversed term, "unintelligent."
b. In recording the passages the researcher has to be alert for instances of irony.
c. Quotations are treated as usual, unless the quotation is explicitly rejected by the source. This is because quotations may be used by the source to convey an attitude as much by the quotation selected as by the statements made by the source itself.
9. This step simply consists of recording, by type of word, each listed word as a verb, adjective, adverb, noun, or pronoun.
10. The next step, after listing all the terms in reference to the subjects of interest in the source, is to score each term as 58 favorable (+), unfavorable (-), or neutral (0) in the "direction" column of the score sheet. Words are scored according to the ECO
Word List. For words not listed, a synonym is used and/or the following criteria/guideline:
One way of deciding whether or not a word is evaluative is to apply the "congruency" test; that is, to ask whether the word would be most appropriately applied to "saints" or "heros," or alternately to "sinners" or "villains." Let us suppose an analyst is uncertain whether the word "contemptible" is evalu ative: use of this test would suggest that one is more likely to speak of a "contemptible sinner" than of a "contemptible saint." Similarly, a word like "cooperative" is more likely to be applied to a saint or hero than a sinner or villain. Hence "contemptible" and "cooperative" are evaluative terms. But words like "tall" or "powerful" could be applied equally appropriately to saints or sinners, heros or villains, and hence are descriptive or neutral (Pratt, 1972:18).
However, Pratt dictates that "the Word List should be followed except where to do so would clearly misrepresent the author's intent," and that "the cardinal rule in this kind of analysis is, [to] never violate the original meaning" (1972:20, 21; emphases in the original).^
11. Once the measurements are recorded and scored, the data is then analyzed statistically:
a. The first statistical analysis performed is the Coefficient of Evaluation (Pratt, 1972:22), which requires a minimum of ten (10) evaluative terms per subject. The formula used is:
100 (F) F + U
where F = favorable terms and U = unfavorable terms. The Coefficient of Evaluation will always be between zero (totally unfavorable) and one hundred (totally favorable) with fifty representing the point of ambivalence. It should be noted that Pratt also refers to fifty as "neutral," which 59
is a misnomer as "neutral" evaluations would really be descriptive rather than evaluative. It is for this reason that this study records descriptive, as well as evaluative, terms. The proportion of the evaluative terms used by a source to refer to a subject favorably and expressed as a percentage, then, is called the Coefficient of Evaluation.
b. The second analytical technique is to determine the Evalu ativeness Score, which is simply the proportion between the number of evaluative terms referring to the subject in the source to the total number of words devoted to the subject in the source and presented as a percentage. This analytical measure is indicative of the frequency with which value judgments are expressed about the subject relative to the amount of space devoted to the subject. The formula used is:
100 (Total Evaluative Words) Total Words
c. A third type of analysis consists of comparing various variables—e.g., "Indian" to "Non-Indian"—in contingency tables for variance within a source, and between sources, utilizing chi-square (X ).
d. A fourth type of analysis is to group the sources by publi cation dates to determine trends in evaluations of subjects in the sources over time.
e. A fifth analysis is a description of themes and their frequencies.
f. A sixth analysis is to regroup the books for within and between book analyses to evaluate for trends over time.
g. The final analysis consists of a comparative analysis of the quantitative and qualitative data.
Based upon these results, inferences are made about what kind of attitudes, impressions, values, and judgments are made by sources as a partial explanation of how Native Americans are portrayed in children's fiction. Chapter 3
RESULTS
The qualitative research findings will be presented first.
Next, the quantitative results for each book will be presented.
These within book analyses will be followed by a comparison of data between books, across books, and over several time periods:
(1) Time 1 = 1968-1970; (2) Time 2 = 1974-1978; and (3) Time 3 =
1980-1981. The expected results, again, of this study are:
1. That books published in the 1960s will be more evaluative than books published in the 1970s and 1980s.
2. That the evaluativeness of books, based upon the sample average score, will be less than ten percent (<10%).
3. That the percentage of descriptive words for all subjects, groups, and books will be greater than ninety percent (>90%).
4. That books published after 1975 will have improved because of outside demands, and therefore, will have a descriptive percentage greater than ninety-five percent (>95%).
5. That books published after 1975, similarly, will have an even lower Evaluative Score of less than five percent (<5%).
6. That evaluative judgments will be favorable (i.e., >75%) toward Native Americans.
7- That judgments of Native Americans will be less favorable than those of others.
8. That the percentage of evaluative terms to total space devoted to Native Americans will be greater than that for Non-Native Americans.
9. That the percentage of negative evaluative judgments about Native Americans will be greater than five percent (>5%),
60 61
while negative evaluations for Non-Native Americans will be less than five percent (<5%).
10. That similar themes concerning Native Americans, found by Garcia in textbooks, will be found in the fiction books.
The last part of this chapter will discuss parallelisms between the qualitative findings and the quantitative results.
Tables 5 through 10, discussed below in conjunction with each book, summarize the quantitative data collected on each subject.
Each cell in the contingency tables presents the observed frequency along with the percentages in parentheses. Additionally, each table provides, below the contingency table, the ratios of positive evalu ative terms to negative evaluative terms for each subject and for 2 each book as a whole. Lastly, the calculated chi-square (X ), the degrees of freedom (df), and the level of statistical significance
(e.g., p<.001) for the chi-square are given on each table. The
2 analytical statistic of chi-square (X ) tests to see if the terms used are independent of the subjects. This is done to more closely determine if the differences across the cells exist because the terms are related to the subjects, or, if the differences exist simply because of chance alone or because of sampling error. For this study, then, all chi-square tests of significance will begin with the following null hypothesis, alternative hypothesis, and level of significance or alpha error:
1. Hq: Evaluative and descriptive terms are not related to the subjects. 62
H : Evaluative and descriptive terms are related to the subjects.
2. Accepted level of alpha error'.oC- .01.
Tables 11 through 16, discussed in the second part of this chapter, are presented in the same format as Tables 5 through 10 and provide statistical results for comparisons between groups and books.
Summary data are provided in Appendix G. Table 28 in
Appendix G summarizes for each subject by book, for each group, and for the total book, the Coefficient of Evaluation scores; which are, once again, the proportion of evaluative terms used by a source to refer to a subject favorably and that is expressed as a percentage.
A high score would be a positive portrayal and a low score would be a negative portrayal. Table 29 in Appendix G delineates the second analytical measure or the Evaluativeness Score, a percentage which measures the frequency that value judgments are expressed about a subject relative to the amount of space or total descriptive and evaluative words.
Qualitative Findings
Procedurally, the qualitative measurement of the books in the sample is quite simple. (It is remembered from chapter two that while reading the book, frequency counts by page number were kept for both subjects and themes.) After reading the book the qualitative analysis was made first, and then the semi-quantitative analyses were 63 made utilizing the ECO Analysis Administrative Data Instrument.
In measuring portrayals qualitatively, then, the three very crude, basic, and broad categories are (1) the story plot, (2) the character and group portrayals, and (3) the impressions of the book.
It is clear, therefore, that the qualitative methods have greater scope than the quantitative, which measure portrayals only. Further more, the data is less structured and usually recorded in roughly two handwritten pages, which is about one-tenth the amount of space and/or data accumulated by the quantitative ECO Analysis Score Sheet
Instrument. Each book will be discussed by category first.
The story plot for each book is as follows:
1. Akavak, an Eskimo Journey. The time setting is not explicitly stated by the source, but through pictures and descriptions it appears to take place many years ago. The story is about Eskimos, and specifically a young man, Akavak, and his grandfather. The grandfather wishes to travel to see his brother because he feels that he will die soon. As the whole family cannot travel at that time, Akavak must take his grandfather. The majority of the story, then, is about their trip, which is filled with numerous environmental and spiritual obstacles. They finally get within sight of the brother's "village" and sled down the mountain, but on the way the grandfather dies. The book ends with the brother singing a song.
2. Sing Down the Moon. The time setting is the 1860s and the story is about the Navajos, other Native Americans, and Non-Native Americans. The story is narrated through the perspective of a young woman, "Bright Morning,'' who along with a friend, is kidnapped by two Spanish men and sold to an American woman. The two girls eventually escape with the help of a Nez Perce girl. Then the American soldiers come and burn their "village," after which the Navajos surrender and are removed to the Bosque Redondo. Bright Morning marries Tall Boy and together they run away to live secretly in the mountains near their home. They have a child and are 64
very happy until soldiers come nearby, then they leave again. The story ends with them returning to live in their old "village," supposedly happily ever after.
3. The Spirit is Willing. The time setting is the 1880s in Arizona and the story is about two Non-Native American girls, Portia and Carrie. Geronimo has "escaped" so Portia cannot leave town and she must spend the summer there. Carrie and Portia decide to hold a seance during which they believe they contact a spirit. They decide that it is the spirit of an "Indian mummy." They attempt to see the mummy, which is on display in a saloon, but are caught. Portia pretends to faint and to be under the spell of the mummy's spirit, which creates an uproar. As a result of all the publicity Portia's mother decides to leave town, regardless of the risk of Geronimo. It is then disclosed that the whole thing is a ruse, but the book ends when Carrie decides to continue the seances because she has information that others want to know and the predictions are harmless anyway.
4. Squaw Man's Son. The time setting is the 1880s in Oregon and California, and the story is about the Modocs, although it is allegedly about a "halfbreed," Billy. Billy's father sends his mother "back to her people" so the father can become respectable. After a period of time Billy's father remarries, but his step-mother makes him live at the livery. Eventually Billy runs away to live with his mother and the Modocs. The Modocs are told to move onto a reservation, but they want to stay where they are, and war breaks out. The was is long and violent, but eventually the Modocs surrender. While in prison, Billy's father comes to get him. At "home" things still do not work out so his father sends Billy to live with his father's brother. The book ends with Billy leaving the day Captain Jack is hung.
5. Sarah Bishop. The time setting is the 1780s during the Revolutionary War, and the story is about a young girl, Sarah Bishop. Sarah's father, a British sympathizer, is tarred and feathered and dies. Sarah tries to find her brother, who is in the American army and finally learns that he has died while a British prisoner. Sarah is blamed for starting a fire so she runs away. She finds a cave near a lake and decides to live there, going into a nearby town every once in a while for various reasons. She makes friends with the Longknife family, who shows her how to do things and helps her out every now and then. She starts 65
going to see a particular young man in town, but is soon accused of being a witch. Through her friend, Morton, she is acquitted. The book ends with her going back to her cave, but promising to come back to town soon.
6. Where the Buffaloes Begin. The time setting is not explicitly stated, but it is clearly set in the nineteenth century or earlier. The story is about a boy named "Little Wolf" who awakes one morning and decides to ride off to find out where the buffaloes begin. He rides for a long time and eventually comes to a misty lake from which he hears strange sounds. After some time he sees the buffaloes coming out of the lake, and soon they are running and Little Wolf is riding among them. After a long time, as they are nearing his village, Little Wolf spots the Assiniboines preparing to attack his village, so he directs the buffaloes to stampede the Assiniboines. The story ends with the buffaloes disappearing into the horizon after stampeding the Assiniboines and "saving" the village.
Qualitative analysis for character and group portrayals are as follows for each book:
1. Akavak, an Eskimo Journey. Eskimos in general are portrayed as simple, if not poor, people who believe in a variety of spirits and beings. They are inaccurately portrayed as residing in permanent villages. Spirits and other beings are negatively portrayed, and often put down or mocked by the source. Akavak appears to be positively portrayed, almost as a superman-type person. Grandfather seems to be ambivalently portrayed, having tremendous knowledge and experience, but also being "strangely" mystic or spiritual.
2. Sing Down the Moon. Native Americans are positively portrayed This is based primarily on the descriptions rather than the depth of characterizations. All portrayals are shallow and evaluative, but those for Native Americans are less so than those for Non-Native Americans. Individual Native American characters run the whole range of characterizations from positive to negative to descriptive portrayals.
3. The Spirit is Willing. Native Americans are generally mocked and very negatively portrayed, while Non-Native Americans are very positively portrayed. The main characters have very full and colorful characterizations, while the mummy 66
is ambivalently portrayed—at times positive, other times negative—and is imbued with a romantic "Noble Indian" characterization. Geronimo is the scapegoat and is portrayed as the general bad guy, while Apache Sam is ambivalently portrayed. Mexicans and Chinese are both negatively portrayed.
4. Squaw Man's Son. Native Americans are ambivalently portrayed in that Modocs are favorably presented but other Native Americans in the book and specific characters are very negatively portrayed. Non-Native Americans are neutrally portrayed, although many are portrayed as snooty and bigoted. Captain Jack is positively portrayed, as are Billy, his mother, and his step-family members. Other Modoc characters are ambivalently or negatively portrayed. Overall, portrayals are quite shallow or poorly developed, and characters and groups are more like props in a fictional ized documentary.
5. Sarah Bishop. Native Americans are neutrally to positively portrayed, while colonial Americans are often negatively or ambivalently portrayed, and the British positively to ambivalently portrayed. On the whole it is a very neutral book. It is probably the best in this study, although characterization of the Longknifes, while generally good, contains some demeaning and standard positive images.
6. Where the Buffaloes Begin. The character Little Wolf is ambivalently portrayed, while the group Assiniboines is incredibly negatively portrayed. Even the buffaloes are at times negatively portrayed in this highly evaluative book.
The final category is an analysis of the impression of the books. This includes the analyst's impressions and/or reactions to the books:
1. Akavak, an Eskimo Journey. This book is perceived as being basically a good book, although certain elements (e.g., the subject "Spirits") are very negatively portrayed and there are inaccuracies. The book leaves the reader with a feeling of futility in the Eskimo way of life, but pride in their endurance as well. 67
2. Sing Down the Moon. This book gives the overt impression of being very favorable toward Native Americans and very negative toward Non-Native Americans. After some time, however, the covertly negative impressions surface, so that the reader actually receives a very negative presen tation of Native Americans. This is particularly true of the source's portrayal of Navajo culture, which contains inaccuracies and omissions.
3. The Spirit is Willing. This book leaves the reader with the feeling and attitude that Native Americans are either lawless and/or romantic people. It is made clear that good and bad Anglos are better than Native Americans, Chinese, and Mexicans, although the reader is led to believe that Carrie is supposed to be more egalitarian thinking. This is to say that the main character is portrayed to be above reproach, but negative evaluations exist through other characters. The book contains defamation of characters, oversimplifications, questionable acuracy, omissions, and numerous other methods of creating impressions that negatively evaluate Native Americans.
A. Squaw Man's Son. This book is, from title to concluding hysterical note, replete with inaccuracies, innuendos, omissions, defamations, dehumanizations, and overgeneral- izations to name but a few of the propagandistic methods used in portraying people and historical events. The book attempts to tell the story from the perspective of a "half-breed youngster," obviously seeking realism and accuracy. The source, however, equates evaluative language, attitudes, and actions of the day to realism. Additionally, the source attempts to portray the Modocs as the "good guys," but simply ends up portraying most of the Non-Native Americans more negatively than some of the Modocs.
5. Sarah Bishop. The best book in this study, although Native Americans still seem to be negatively or demeaningly portrayed. The Longknifes are described neutrally, but their role and actions are clearly subservient to those of Sarah. The portrayal, then, is manifestly descriptive, but somehow—covertly and/or latently—negatively evaluative. As such, most would classify the evaluative portrayal of Native Americans as neutral to positive, but with the portrayals overall being descriptive. 68
6. Where the Buffaloes Begin. The contents of this book are overshadowed by the accompanying drawings, which are manifestly positive portrayals. Latently, however, they are very biased; i.e., (1) they never show people smiling or happy, (2) the people are either blurred or portrayed in an unkept manner, and (3) scenery is often portrayed gloomily. The text, however, is manifestly negative, especially toward the Assiniboines. The book contains loaded words, oversimplifications, and omissions in its portrayals.
The qualitative measurements generally yield, then,
negative critiques of the books, although several of the books
contain a greater range of portrayals from which some are
positive. An exception is the book Sarah Bishop (#5) by Scott
O'Dell, which is found to be more positively critiqued. !ln analyzing
the first category—the story plot—a major finding is that all the
books are historically set in the nineteenth century or earlier.
As such, this portrays Native Americans as being people of the
past. (It will be recalled that the universe and population were selected upon the criteria of Native Americans and did not include
Non-Native Americans, which means that inferences cannot be so
easily drawn concerning them; i.e., it cannot be inferred that
Non-Native Americans are usually portrayed as people of the past.)
A second major finding is that Native Americans are either (1) in conflict with either Native Americans or Non-Native Americans, or
are (2) helpers for Non-Native Americans—Native Americans never
help other Native Americans. A third finding, from the character
and group portrayals, is that Native Americans are generally 69 portrayed more negatively than Non-Native Americans. A last finding is that all the books utilize impressionistic techniques that more evaluatively portray Native Americans.
Quantitative Results for Individual Books
Book Number One (#1): Akavak, an Eskimo Journey (New York: Harcourt, 1969).
The first book analyzed is Akavak, an Eskimo Journey by
James Houston. The book is eighty pages long including the cover page, and is listed as being fourth to sixth grade reading level by Wilson's Children's Catalog (1971:416; 1976:532-533). There are two main Native American character subjects—"Akavak" and
"Grandfather"—and two minor group characterizations—"Eskimos" and "Spirits." As the subject "Eskimos" is not listed on a sufficient enough number of pages it is not measured. The subject
"Spirits" includes the following groups/individuals: (1 ) Night spirits; (2) Spirits of ancestors; (3) Mountain spirits; (4) Wind giants; and (5) Dwarf people. While this subject appears on less than ten pages, which was set as the sampling size, because of the smaller number of pages in the book the subject still appears in ten percent of the book and is, therefore, included in the study.
All three subjects in the book are Native American, but the source has provided a very clear we-they dichotomy. Resulting from the fact that the source establishes these entities as antagonists, the subject "Spirits" is treated as a "they" subject while the subjects "Akavak" and "Grandfather" are considered as "we" subjects for comparisons, although the dichotomy may be a false one for
the Eskimos themselves. 2 Table 3 indicates the chi-square (X ) test of significance, which, again, tests to see if the terms used are independent of the subjects or if differences exist simply because of chance or sampling error alone. The null hypothesis (H ) is set to be that evaluative
and descriptive terms are not related to the subjects. Conversely,
if statistically significant differences are found, then the
alternative hypothesis, (H ) that evaluative and descriptive terms a. are related to the subjects and not because of chance or sampling
error, is accepted. Similarly, the level of significance or the
accepted level of alpha error is oc = .01. The statistic chi-square 2 (X ), as found for the book Akavak, an Eskimo Journey, is calculated 2 X = 148.2 with four degrees of freedom (df = 4). Using the above null hypothesis and alpha level with the chi-square table, it is 2 2 found that X , =13.2. Since the critical X value is less .01 df=4 2 than the calculated X value, the null hypothesis is rejected.
This means that the evaluative and descriptive term distributional differences across the cells are statistically significant for the subjects in this book as reported in Table 3. Evaluative and descriptive term frequencies, then, are directly related to or dependent upon the subjects and who the subjects are. Furthermore, Table 3
Akavak, an Eskimo Journey (Book #1): Observed Word Frequencies
Term Direction Subjects Totals
Akavak Grandfather Spirits (#1) (#2) (#3)
+ 19 31 4 54 (3.32%) (4.70%) (4.04%) (4.06%)
0 458 546 * 39 1043 (80.07%) (82.73%) (39.39%) (78.36%)
_ 95 83 * 56 234 (16.61%) (12.58%) (56.57%) (17.58%)
Totals 572 660 99 1331 (100.00%) (100.01%) (100.00%) (100.00%)
Ratios .2000 .3735 .0714 .2308 +:-
X2 z 117.7 13.2 p<.001 df = A X.01 df=*» =
*—indicates statistical significance for individual cell distribution 72
Table 3 indicates that the descriptive and negative cell distributions for the subject "Spirits" (#3) are both individually statistically significant. This means that the observed term frequencies in each cell can, by themselves, account for the significance in the distribution of terms. This, then, makes these two frequencies uniquely significant in comparison to others in this book.
Since the chi-square analysis is statistically significant, it is of interest to further analyze the results. First of all, the ratios of positive to negative terms in Table 3 indicate and/or substantiate the imbalance of portrayals in the negative direction.
A ratio of 1.0000 would indicate a balanced or ambivalent portrayal, while a larger score would indicate a positive portrayal and a smaller score would indicate a negative portrayal. In this case, the ratios indicate that for the subject "Akavak" (#1), there are twenty positive terms used for every one hundred negative words used. For the subject "Spirits" (#3), however, there are only seven positive words used for every one hundred negative terms, which can be inferred -to be significant, also, because of the fact that the corresponding negative frequency cell is individually statistically significant.
The subjects selected for further analysis are listed below in Table 4 along with: (1) the frequency of pages mentioning the 73 subject; (2) the total frequency of descriptive and evaluative terms; (3) the Coefficient of Evaluation, or percentage of evaluative terms used by the source to refer to the subject favorably, for which a low score is negative, an average score is ambivalent, and a high score is positive in terms of the subject's portrayal; and (4) the Evaluativeness Score, or the percentage of value judgments or evaluative terms expressed about a subject to total words about the subject.
Table 4 Akavak, an Eskimo Journey (Book #1): .Page and Word Frequencies, Coefficient of Evaluation, and Evaluativeness Score
Coefficient a Evaluativgness Subject Page Word of Evaluation Score Number Subject Frequency Frequency % %
1 Akavak 56 (1)* 572 (2)* 17 20 2 Grand father 51 (2) 660 (1) 27 17 3 Spirits 8 (3) 99 (3) 7 61 " 100 (F) a—Coefficient of Evaluation = — , where F = total F + U favorable terms and U = total unfavorable terms 100 (Total Evaluative Words) b Evaluativeness Score = Tofcal Words *—indicates ranking
The page and total word frequencies are indicative of the major characters and groups in the book—that is, it would be lb expected that the major characters would be ranked highest on
both page (or scope) and word (or depth) frequencies. There are, however, some interesting discrepancies between the two measures for various subjects. In looking at the subject portrayals by
page frequency rankings for the character subjects it is seen that "Akavak" (#1) has slightly more scope, or page frequency, than the second subject, "Grandfather" (#2), giving the illusion of being the main character. Conversely, in looking at the depth of the portrayals indicated by the total number of evaluative and descriptive words it is observed that -the subject "Grandfather"
(#2) is latently a more major subject in the book. The most interesting descrepancy, however, is not the differences in rankings but rather the ratio of words to pages for the subjects. Subject number three—"Spirits"—is interesting particularly in that the manifest visibility of the subject is quite low in comparison to the other two subjects, but is portrayed by a high frequency of evaluative and descriptiveterms which combine to create a highly developed portrayal of the subject "Spirits." The depth of the portrayal is also found to be very evaluative. In looking at the
Evaluativeness Scores it is found that this source, more so than any other in this study, makes judgmental portrayals of the subject
"Spirits" nearly two-thirds of the time. Furthermore, it is observed that the Coefficient of Evaluation score indicates that 75
the subject "Spirits" is significantly negatively evaluated. That
is to say, the score is less than would be expected for a normal sample, which should range between twenty-five and seventy-five
percent. Conversely, if the score had been between twenty-five
and fifty percent the score would have been negatively evaluated or ambivalent in the negative direction. While the range for
ambivalence is left ambiguous, it is generally conceived of as
including scores between thirty-three and sixty-six percent.
Scores on either side of the normal range (25%-75%) are considered,
therefore, substantively significant. Scores greater than fifty
percent, however, indicate positive evaluative portrayals and scores less than fifty percent indicate negative portrayals. Thus, a score of thirty-eight percent is considered to be an ambivalent score in the negative direction, while a score of sixty percent is
considered to be ambivalent in the positive direction. Deriving from the low frequency of being mentioned, the unusually high rate of words to portray the subject, and the negatively skewed
evaluation, the "they" subject "Spirits" is given a significantly latent and negatively.evalative portrayal by the source.
In looking at the distribution of scores in Table 3 for
Akavak, an Eskimo Journey it is seen that the "we" subjects of
"Akavak" and "Grandfather" have much greater percentages of descriptive words than evaluative terms, but the "they" subject 76
"Spirits" has a larger proportion of negative evaluative terms
than both descriptive and positive words combined (i.e.,
56.37% > 39.39% + 4.04%). Furthermore, it is observed that the
percentage of descriptive terms for each subject is less than
predicted: "Akavak" (#1), 80.07% < 90%; "Grandfather" (#2),
82.73% < 90%; and "Spirits" (#3), 39-39% < 90%. Additionally, the
percentage of descriptive terms for the book as a whole is only
78.36%, which is much less than expected. Since only slightly more
than one-third of the portrayals of subject number three—"Spirits"—
and a little more than three-fourths of all portrayals in the book
are descriptive, the use of evaluative terms in this book is much
greater than that found in any of the other books in this study.
Such a result is substantively significant for this study, and suggests further analysis in this direction. In looking at the types of terms, then, the subjects are found to be ranked as provided in
Table 5.
The rankings in Table 5 indicate that the subject "Grandfather"
(#2) is portrayed the most descriptively, but that all three subjects
are clearly portrayed evaluatively. Furthermore, the rankings also verify other trends noted above (i.e., the subject "Grandfather" also is ranked the highest for positive evaluative judgments and the subject "Spirits" (#3) is clearly the most negatively biased
portrayal). It is interesting, however, that the subject "Akavak" 77
(#1) ranks the lowest in positive evaluations, which serves to
reinforce the very negative portrayal of this subject.
Table 5 Akavak, an Eskimo Journey (Book #1): Positive, Negative, and Neutral Word Rankings
Rank Positive Rank Negative Rank Neutral
1 (2)* Grandfather 1 (3) Spirits 1 (2) Grandfather (4.70%) (56.37%) (82.73%)
2 (3) Spirits 2 (1) Akavak 2 (1) Akavak (4.04%) (16.61%) (80.07%)
3 (1) Akavak 3 (2) Grandfather 3 (3) Spirits (3.32%) (12.58%) (39.39%)
*—indicates the subject number
Analysis of the results, then, demonstrates that the book
Akavak, an Eskimo Journey significantly portrays Eskimos and
their way of life negatively. Furthermore, the source specifically
portrays the "they" subject (#3) in a statistically significantly negative and judgmental style. Lastly, it is found that the "they" subject (#3) and the book as a whole are significantly more evaluative
than any of the others in this study.
Book Number Two (#2): Sing Down the Moon (Boston: Houghton Miflin Company, 1970).
The second book analyzed is Sing Down the Moon by Scott O'Dell. 78
The book is 137 pages long, and is listed as being fifth to seventh grade reading level by Wilson's Children's Catalog (1976:555;
7 1981:511-512). Although a number of individuals and groups are
presented in the story, only six subjects are measured. These include two Native American character subjects, three Native American group subjects, and one Non-Native American group subject. The subject
"Non-Indians" actually includes one unnamed character and two groups— the Spanish and the soldiers—while the subject "Indians" comprises all Native Americans other than Navajos and specific Navajo characters, who are presented either as separate subjects or as the group "Navajos."
It is seen from Table 6, using the same null hypothesis (HQ)
that evaluative and descriptive terms are not related to the subject
.and the sane level of significance or accepted level of alpha error
(oC- .01), that calculated X2 = 29.8 with df = 10. The table 2 2 X 23.2, which is less than the calculated X . Thus, the .01 df=10 null" hypothesis is rejected, meaning that the observed evaluative and descriptive term distributional differences across the cells are statistically significant for the subjects in this book, Sing Down the Moon, as reported in Table 6. Evaluative and descriptive term frequencies, then, are directly related to or dependent upon the subjects (H ) and who the subjects are. a. Since chi-square analysis is statistically significant, it is of interest to further analyze the results. First of all, Table 6 Table 6
Sing Down the Moon (Book #2): Observed Word Frequencies
Terra Direction Subjects Totals
Bright Morning Family Navajos Tall Boy Indians Non-Indians (#1 ) (#2) (#3) (#4) (#5) (#6)
+ 18 2 12 11 9 8 60 (3.01%) (0.73%) (1.92%) (1.46%) (4.23%) (2.24%) (2.13%)
0 558 258 590 692 197 317 2612 (93.16%) (94.16%) (94.40%) (92.14%) (92.49%) (88.80%) (92.66%)
_ 23 14 23 48 7 32 147 (3.84%) (5.11%) (3.68%) (6.39%) (3.29%) (8.96%) (5.21%)
Totals 599 274 625 751 213 357 2819 (100.01%) (100.00%) (100.00%) (99.99%) (100.01%) (100.00%) (100.00%)
Ratios .7826 .1429 .5217 .2292 1 .2857 .2500 .4082
2 2 X = 29.8 df ;10 X q1 df=10 = 23.2 p<.001 80 lists the ratios of positive to negative words, which provides some interesting results. The ratios range from a high of 1.2857 for the subject "Indians" (#5) to a low of .1429 for the subject "Family"
(#2). This means that for the subject "Indians" there are 128 negative words used for every one hundred negative words used, which indicates a substantively significantly positive portrayal of the subject "Indians." Furthermore, it is observed that the character subject "Tall Boy," which has a ratio of .2292, and the subject
"Non-Indians," which has a ratio of .2500, both have very low positive to negative ratios, indicating, again, significantly negative portrayals for these two subjects. Subjects further analyzed for this book are listed in Table 7 along with page and word frequencies and statistical scores for the Coefficient of Evaluation, or the percentage of evaluative terms used by the source to refer to a subject favorably, and the Evaluativeness Score, or the percentage of value judgments or evaluative terms expressed about a subject to total words about the subject.
In looking at the page and word frequencies several interesting results and trends are observed. First of all, the narrator
(Subject #1) of this first person story, while having the highest page frequency, ranks only third in word frequency. This means that although the subject appears to be the main character based upon the scope or number of pages upon which the subject is mentioned, 81 the depth or word frequency indicates that the subject is less important than manifestly portrayed. Latently, then, the subject
"Bright Morning (#1) is not the major character; that is, this subject is not the one best developed in terms of descriptive and evaluative characterizations. Conversely, the subject "Tall Boy" (#4) is ranked fourth in page frequency and, having almost half the number of the third ranked subject, would normally be considered a minor subject in the story. According to the subject's word frequency, however, the character subject "Tall Boy" is ranked number one for depth of portrayal. This means that manifestly the character subject
"Tall Boy" appears to be a rather unimportant one, yet in analyzing the word frequency it is inferred that the portrayal is latently greater than that for all other character or group subjects in the book Sing Down the Moon. In looking at the Coefficient of Evaluation scores some other interesting results stand out. Most striking is the score for the subject "Indians" (#5) that is found to be fifty-six percent, which is the only Coefficient of Evaluation score in the positive direction for this book, although the score is more an ambivalent one (i.e., <75%). That is, given a normal distribution, it is expected that most scores would be between twenty-five and seventy-five percent. This means that scores less than or greater than these extremes are significant as delineated by the proposed results. Three of the subjects' scores are significantly negative: 82
(1) "Non-Indians," 20% < 25%; (2) "Tall Boy," 19% < 25%; and
(3) "Family," 12% < 25%. The other two subjects—"Bright Morning" and "Navajos"—both are portrayed ambivalently, but in the negative direction. Additionally, two of the observed Evaluativeness Scores are of interest. First, the subject "Non-Indians" (#6) is found to be the most evaluated subject in the book with an Evaluativeness
Score of eleven percent, which means that eleven percent of the subject
"Non-Indians" portrayal is evaluative. The subject "Non-Indians" portrayal, therefore, is both significantly negative and broadly evaluative, particularly in comparison to other subjects. The other
Evaluativeness Score of interest is the one for the subject "Tall Boy"
(#4), which is, along with the subject "Indians" (#5), the second most evaluative portrayal in the book. When this finding is combined with the other results on the subject—"Tall Boy"—it can be inferred ; that the portrayal is latently and broadly judgmental, and significantly negative. Lastly, but perhaps more importantly, it is observed from the Coefficient of Evaluation scores that the subject "Family" (#2), which is a minor group subject in the story, is unexplicably the most negatively judged subject in the book Sing Down the Moon.
Because of the limited role of the subject, this negative evaluation of the subject "Family" is inferred to be an unconscious portrayal by the source. Similarly, however, the portrayal also is inferred to be latently judgmental, regardless of the subject's scope and 83 depth of portrayal, because it subtly or covertly maligns Navajo families through an unusually negative portrayal. The source, thereby, negatively evaluates Navajo culture through latently judgmental portrayals of the subjects "Family" (#2) and "Tall Boy" (#4), who symbolize Navajo men in this story.
Table 7 Sing Down the Moon (Book #2): Page and Word Frequencies, Coefficient of Evaluation, and Evaluativeness Score
Coefficient Evaluativeness Q Subject Page Word of Evaluationa Score Number Subject Frequency Frequency % %
1 Bright Morning 117 (1 )* 599 (3)* 44 7 2 Family 48 (6) 274 (5) 12 6 3 Navajo 99 (2) 625 (2) 34 6 4 Tall Boy 57 (4) 751 (1 ) 19 8 5 Indians 55 (5) 213 (6) 56 8 6 Non- Indians 92 (3) 357 (4) 20 11 100 (F) a—Coefficient of Evaluation = where F = total favorable F + U terms and U = total unfavorable terms 100 (Total Evaluative Words! b—Evaluativeness Score = Total Words *—indicates ranking
Table 6 also presents the distribution of scores for each subject in Sing Down the Moon. The first observation is that neutral words overwhelmingly account for the largest percentage of the 84
portrayals for all but one subject. The percentage of descriptive
terms for the subject "Non-Indians" (#6), however, is lower than expected (88.80% < 90%). Further analysis of the types of terms used for each subject would be interesting. Table 8, therefore, ranks each subject by frequency of type of term.
Table 8 Sing Down the Moon (Book #2): Positive, Negative, and Neutral Word Rankings
Rank Positive Rank Negative Rank Neutral
1 (5)* Indians 1 (6) Non-Indians 1 (3) Navajos (4.23%) " ' " " 18.96%)"~ "' (94.40%)
2 (2) Bright Morning 2 (4) Tall Boy 2 (2) Family (3.01%) (6.39%) (94.16%)
3 (6) Non-Indians 3 (2) Family 3 (1) Bright Morning (2.24%) (5.11%) (93.16%)
4 (3) Navajos 4 (1) Bright Morning 4 (5) Indians - (1.92%) (3.84%) (92.49%)
5 (4) Tall Boy 5 (3) Navajos 5 (4) Tall Boy (1.46%) (3.68%) (92.14%)
6 (2) Family 6 (5) Indians 6 (6) Non-Indians (0.73%) (3.29%) (88.80%)
*—indicates the subject number
In looking at the rankings in Table 8, several trends observed above are again found along with some other interesting results. 85
First of all, it is found that Non-Native Americans have the least descriptive portrayals of all the subjects and, therefore, all Native
American subjects are more descriptively portrayed than Non-Native
Americans. This result is substantively significant because it is
the only case in this study where a source does so in a book.
Second, it is observed that the subject "Indians" (#5) is ranked first in positive and last in negative portrayals. This reinforces
the fact that the subject also has the highest positive to negative
terms ratio. Third, it is seen that the subject "Tall Boy" (#4) is ranked fifth in positive and neutral evaluations, but is ranked second in negative judgments, which corroborates trends found by the
Coefficient of Evaluation and the Evaluativeness Scores to negatively
portray this subject. Interestingly, it is noted that the subject
"Navajos" (#3) is the most descriptively portrayed subject, but is ranked lower in positive and higher in negative evaluative portrayals
than the subject "Indians" (#5). It is inferred from these rankings,
therefore, that specific Native American groups, i.e., "Navajos",
are more descriptively portrayed than the ambiguous subject "Indians," and that the evaluative portrayal for "Navajos" is more negative than for "Indians." This result is also found by the Coefficient of
Evaluation, which is fifty-six percent for the subject "Indians" and thirty-four percent for the subject "Navajos." Lastly, while the overall portrayal of the subject "Family" (#2) is found to be 86
descriptive, the part of the portrayal that is evaluative is decidedly
biased in the negative direction. Substantiating the Coefficient of
Evaluation result, the subject "Family" is ranked lowest in positive
portrayals with less than one percent. That is, the source essentially
has nothing positive to say about the Navajo family, which is the
reverse of what would be expected based upon the knowledge of the
importance of the family in Navajo life. This result becomes even
more significant when it is observed that, conversely, the subject
"Family" ranks high in negative judgments with over five percent.
This means, then, that the evaluative portrayal of the subject
"Family" is almost entirely negative. Statistical analyses of the
results, then, demonstrate for the book Sing Down the Moon that
Native American subjects are positively evaluated relative to
Non-Native Americans. Additionally, it is inferred that the Navajo
institutional subject, "Family," and the Native American character subject, "Tall Boy," are substantively significantly more negatively
evaluated than other Native American subjects. More importantly,
however, the portrayals of Native American subjects are essentially
descriptive in style. Conversely, the portrayals of Non-Native
Americans are uniquely more evaluative than those for Native
Americans or than expected. As such, this book presents the appearance of having a positively evaluated portrayal of Native
Americans, but in fact is simply an unusually negatively evaluated
portrayal of Non-Native Americans. 87
Book Number Three (#3): The Spirit is Willing (New York: Macmillan and Company, Publishing, 1974).
The third book analyzed in this study is Betty Baker's
The Spirit is Willing. The book is 136 pages long including the cover page, and is listed as being fifth grade and up reading level by Wilson's Children's Catalog (1981:440). While there are numerous potential subjects in the book, only nine are measured and analyzed.
It is noted, however, that the page frequencies for the subjects
"Chinese" (#8) and "Apache Sam" (#9) are not large enough to meet the minimum sampling requirement of ten pages. They are included in the study because both are crucial to the plot of the story.
Secondly, the subject "Apache Sam" provides an additional Native
American character subject, while the subject "Chinese" provides an additional ethnic group for comparison. More importantly, it is noted that no Native American subject, except the "Mummy" subject, appears with great frequency so that measurement of the subject "Apache Sam" adds to the overall sample size for Native
Americans in ethnic comparisons.
It is seen from Table 9, using the same null hypothesis
(H ) and alpha error, or level of significance, calculated X = 148.8 with df = 16. The table X^ = 32.0 and is less than the .01 df=16 2 calculated X and as such the null hypothesis is rejected. This means that the distributional differences found in Table 9 are statistically significant for the subjects in the book Table 9
The Spirit Is Willing (Book #3): Observed Word Frequencies
Term Direction Subjects Totals
Portia Carrie Non-Indians Mummy Mexicans Indians Geronimo Chinese Apache Sam (#1 ) (#2) (#3) (#4) (#5) (#6) (#7) (#8) (#9)
+ 15 21 12 3 2 2 1 1 4 61 (4 .21%) (3.93%) (2.43%) (1 .88%) (1.37%) (2.33%) (2 .38%) (1 .12%) (2,.70%) (2.97%)
0 326 501 461 146 135 55 36 82 129 1871 (91 .57%) (93.64%) (93.51%) (91 .25%) (92.47%) (63.95%) (85 .71%) (92.13%) (87..16%) (91.05%)
- 15 13 20 11 9 * 29 5 6 15 123 (4 .21%) (2.43%) (4.06%) (6 .88%) (6.16%) (33.72%) (11 .90%) (6.74%) (10..13%) (5.98%)
Totals 356 535 493 160 146 86 42 89 148 2055 (99 .99%) (100.00%) (100.00%) (100 .01%) (100.00%) (100.00%) (99 .99%) (99.99%) (99..99%) (100.00%)
Ratios 1.0000 1 .6154 .6000 .2727 .2222 .0690 .2000 .1667 .2667 .4959 +: -
. X2 = 148.8 df = 16 X2 p<.001 01 df=16 = 32,0
1 *—indicates statistical significance for individual cell distribution
00 CX) 89
The Spirit is Willing. Evaluative and descriptive term frequencies, then, are directly related to or dependent upon who the subjects are (H ). Furthermore, Table 9 demonstrates that the negative 3. term frequency distribution for the subject "Indians" (#6) is individually statistically significant. This means that the observed term frequency in this one cell alone accounts for the significance in the distribution of terms and to which subjects they apply.
The negatively biased portrayal of the subject "Indians," therefore, also is statistically and substantively significant in comparison to the other subjects within this book.
Since this chi-square analysis is statistically significant, it is of interest to further analyze the results. First of all,
Table 9 lists the ratios of positive to negative words that provide some additional interesting results. The ratios range from a high of 1.6154 for the subject "Carrie" (#2) to a low of .0690 for the subject "Indians" (#6). This means, again, that for the subject
"Carrie" the source utilizes 162 positive words for every one hundred negative terms, which is a substantively significantly positive portrayal of this subject. Conversely, the source uses only seven positive words for every one hundred negative terms to evaluatively portray the subject "Indians." This is clearly a negative portrayal of the subject "Indians," which is of even greater interest when combined with the fact that the subject is also statistically 90 significantly negatively portrayed. Generally, the ratios for the other subjects indicate and substantiate the trend to more positively evaluate Non-Native Americans than Native Americans by the source in this book, The Spirit is Willing. Secondly, subjects are further analyzed for this book, as presented in Table 10, by scope or page frequency, depth or word frequency, and statistical scores for the
Coefficient of Evaluation and Evaluativeness Score.
Table 10 The Spirit is Willing (Book #3): Page and Word Frequencies, Coefficient of Evaluation, and Evaluativeness Score
Coefficient Evaluativgness Subject Page Word of Evaluationc Score Number Subject Frequency Frequency % %
1 Portia 93 (3)* 356 (3)* 50 8 2 Carrie 136 (1 ) 535 (1 ) 62 6 3 Non- Indians 134 (2) 493 (2) 37 7 4 Mummy 60 (4) 160 (4) 21 9 5 Mexicans 17 (5) 146 (6) 18 8 6 Indians 15 (6) 86 (8) 6 36 7 Geronimo 10 (7) 42 (9) 17 16 8 Chinese 7 (9) 89 (7) 14 8 9 Apache Sam 8 (8) 148 (5) 21 13 100 (F) a—Coefficient of Evaluation = where F = total favorable F + U terms and U = total unfavorable terms 100 (Total Evaluative Words) b—Evaluative Score = Total Words
*—indicates ranking 91
It is seen that for this book, The Spirit is Willing, the two frequencies are ranked the same for the first four subjects.
The subject of interest, however, is "Apache Sam" (#9) who is ranked eighth on page frequency with only eight pages, but is ranked fifth on the total words frequency just behind subject number four,
"Mummy," which appears on nearly eight times as many pages. Clearly, the subject "Apache Sam," who appears only in the conversations of others, received little scope in the book, yet the depth of the portrayal is extensive, indicating a latent portrayal of the subject.
Another trend found in these frequency rankings is that non-Anglo subjects (#4-#9) are all found to be secondary characters or groups in the book, although several are important to the plot. A parallel trend is observed in the statistical Coefficient of Evaluation scores, which demonstrates that these same subjects (#4—#9) are all negatively skewed (<25%), with the subject "Indians" (#6) being the most negatively evaluated subject in the book with six percent. It is interesting to note that the subject group "Non-
Indians" (#3) is also negatively evaluated, although not substantively significant (i.e., 37% > 25%). Additionally, an important trend is noted in the Evaluativeness Scores where Native American subjects
(#4, #6, #7, and #9) rank highest in the proportion of evaluative terms. Furthermore, the portrayals of Native American subjects
"Indians" (#6), "Geronimo (#7), and "Apache Sam" (#9), are more 92 evaluative than expected (>10%). The largest Evaluativeness Score found is for the subject "Indians" (#6) which is thirty-six percent, indicating that more than one-third of the portrayals of "Indians" are judgmental.
In looking at the data distribution in Table 9 for the book
The Spirit is Willing, it is seen that the subjects "Portia" (#1),
"Carrie" (#2), "Non-Indians" (#3), "Mummy" (#4), "Mexicans" (#5), and "Chinese" (#8), all have, as expected, a preponderance of descriptive terms (>90%). Conversely, three of the four Native
American subjects are less descriptive than expected: (1) "Indians"
(#6), 63% < 90%; (2) "Geronimo" (#7), 86% < 90%; and (3) "Apache Sam"
(#9), 87% < 90%. Moreover, as a group subject, "Indians" are portrayed by less than two-thirds descriptive terms, which means that more than one-third of the portrayals of "Indians" in this book are evaluative. Further analysis of the types of terms used for each subject would be interesting. Table 11, therefore, ranks each subject by percentage frequency of type of term.
The rankings in Table 11 indicate some very interesting trends. First, Native American subjects are the top four ranked subjects for negative evaluations. The group subject "Indians"
(#6) is the most negatively evaluated, while the more ambiguous character subject "Mummy" (#4) is the least negatively evaluated
Native American subject. Second, Native American subjects are the Table 11 The Spirit is Willing (Book #3): Positive, Negative, and Neutral Word Rankings
Rank Positive Rank Negative Rank Neutral
1 (1 )* Portia 1 (6) Indians 1 (2) Carrie (4.21%) (33.72%) (93.64%)
2 (2) Carrie 2 (7) Geronimo 2 (3) Non-Indians (3.93%) (11.90%) (93.51%)
3 (9) Apache Sam 3 (9) Apache Sam 3 (5) Mexicans (2.70%) (10.13%) (92.47%)
4 (3) Non-Indians 4 (4) Mummy 4 (8) Chinese (2.43%) (6.88%) (92.13%)
5 (7) Geronimo 5 (8) Chinese 5 (1 ) Portia (2.38%) (6.74%) (91.57%)
6 (6) Indians 6 (5) Mexicans 6 (4) Mummy (2.33%) (6.16%) (91.25%)
7 (4) Mummy 7 (1 ) Portia 7 (9) Apache Sam (1.88%) (4.21%) (87.16%)
8 (5) Mexicans 8 (3) Non-Indians 8 (7) Geronimo (1.37%) (4.06%) (85.71%)
9 (8) Chinese 9 (2) Carrie 9 (6) Indians (1.12%) (2.43%) (63.95%)
*—indicates the subject number 94 bottom four ranked subjects for neutral or descriptive evaluations.
Closer observation further demonstrates that the negative and neutral term rankings for Native American subjects are inversely related. A third result from these rankings concerns the other two non-Anglo or minority group subjects, "Mexicans" (#5) and "Chinese"
(#8). These two group subjects are ranked last in positive evalu ations and between Native American and other Non-Native American subjects in negative judgments, yet are ranked fairly high for neutral portrayals. That is, the portrayals for the subjects
"Mexicans" and "Chinese" are both more descriptive than all Native
American subjects and one Non-Native American character subject,
"Portia" (#1). A further interesting result is found for the subject "Apache Sam" (#9), which is based upon a smaller sampling than the other subjects, yet the subject is ranked third on both positive and negative evaluations and number seven on neutral words.
This corroborates the results, discussed above, concerning the unusually evaluative, though ambivalent, portrayal of the subject
"Apache Sam."
Statistical analysis of the results deomonstrates that in the book The Spirit is Willing, Native American subjects—especially the subject "Indians" (#6)—are both substantively and statistically negatively portrayed. In addition to this biased degree of quality, the portrayals of Native Americans are also proportionately more 95 evaluative. Native Americans, furthermore, are more negatively
portrayed than are all Non-Native Americans, including "Mexicans"
(#5) and "Chinese" (#8).
Book Number Four (#4): Squaw Man's Son (New York: Atheneum Publishers, 1978).
The fourth book analyzed in this study is Evelyn Sibley
Lampman's Squaw Man's Son. The book is 173 pages long, including
the cover page, and is listed as being fifth grade and above reading
level by Wilson's Children's Catalog (1981:496). The subjects in
the book are too numerous to evaluate individually and, therefore, many individual character subjects are included in their respective group categories or subjects. As such, there are four Native
American individual character subjects, two Native American group
subjects, one Non-Native Aemrican individual subject, and two
Non-Native American group subjects for a total of nine subjects actually measured for this book.
In Table 12, as with the previous book analyses, the 2 statistic of chi-square (X ), which tests to see if the terms used are independent of the subjects, is presented. Once again, this is done to more closely determine if the differences across the cells exist because the terms are related to the subjects, or if
the differences exist simply because of chance alone or because
of sampling error. Chi-square tests utilize the following null Table 12
Squaw Man's Son (Book #4): Observed Word Frequencies
Term Direction Subjects Totals
Captain Scarface Billy Jack Father Mother Indians Soldiers Modocs Non-Indians Charley (#1 ) (#2) (#3) (#4) (#5) (#6) (#7) (#8) (#9)
19 16 17 1 1 9 9 7 85 (2.52%) (3.58%) (3.16%) (3.06%) (0.55%) (0.31%) (1.51%) (1.34%) (2.01%) (2.09%)
718 406 504 187 148 314 560 642 318 3797 (95.10%) (90.83%) (93.68%) (95.41%) (81.32%) (96.02%) (94.12%) (95.82%) (91.12%) (93.55%)
18 25 17 3 * 33 12 26 19 24 177 (2.38%) (5.60%) (3.16%) (1.53%) (18.13%) (3.67%) (4.37%) (2.84%) (6.88%) (4.36%)
Totals 755 477 538 196 182 327 595 670 349 4059 (100.00%) (100.01%) (100.00%) (100.00%) (100.00%) (100.00%) (100.00%) (100.00%) (100.01%) (100.00%)
Ratios 1.0556 .6400 1.0000 2.0000 .0303 .0833 .3462 .4737 .2917 .4802
2 2 X = 125.1 df = 16 X q1 = 32.0 p<.001
*—indicates statistical significance for individual cell distribution 97 hypothesis, alternative hypothesis, and level of significance or alpha error of oC = .01 :
H : Evaluative and descriptive terms are not related to the subjects.
H : Evaluative and descriptive terms are related to the subjects.
For this particular book, Squaw Man's Son, it is found that 2 calculated X =125.1 with df = 16. Using the above null hypothesis 2 and alpha level, the table chi-square is X ^ = 31.9. Since 2 2 the critical X value is less than the calculated X value the null hypothesis is rejected. This means that the evaluative and descriptive term distributional differences across the cells in
Table 12 are statistically significant for the subjects in this book. Furthermore, it is seen from Table 12 that the negative cell distribution or portrayal for the subject "Indians" (#5) is individually statistically significant. This means that the observed negative term frequency in this one cell alone accounts for most of the significance in the distribution of terms. In comparison to the other subjects within this book, then, the fact that the subject "Indians" is negatively evaluated to the degree that it is, is uniquely significant.
Since the chi-square analysis is statistically significant, it is of interest to further analyze the results. First of all, of related interest, are the ratios of positive to negative 98
evaluations found in Table 12 for each subject and the book as a whole. It is seen that the highest ratio is for the subject "Mother"
(#4) at 2.0000, while the lowest is observed for the subject group
"Indians" (#5) at .0303. This latter ratio is extremely significant
because of its very low value, and also because the negative cell of this ratio is statistically significant. Furthermore, the negative evaluation for the subject "Indians" at eighteen percent is three times greater than the next largest evaluative portrayal, which is the negative evaluation of the subject "Scarface Charley"
(#9) at almost seven percent. These ratios indicate that for the subject "Mother" the source uses two hundred positive terms for every one hundred negative words, whereas for the subject "Indians"
the source utilizes three positive terms for every one hundred negative words to portray the respective subjects. As would be expected, the ratio for the major character subject "Billy" (#1) is very close to being balanced (1.0556); i.e., a ratio of 1.0000 would mean a balance of positive to negative evaluations or an ambivalent portrayal. Secondly, the subjects are further analyzed for this book as presented in Table 13, by frequency of pages mentioning the subject, the total frequency of descriptive and evaluative terms, the Coefficient of Evaluation, and Evaluativeness
Scores. 99
Table 13 Squaw Man's Son (Book #4): Page and Word Frequencies, Coefficient of Evaluation, and Evaluativeness Score
Coefficient Evaluativeness Subject Page Word of Evaluation ct Score D Number Subject Frequency Frequency % %
1 Billy 161 (1)* 755 (1)* 51 2 Captain Jack 71 (5) 447 (5) 39 9 3 Father 65 (6) 538 (4) 50 6 4 Mother 46 (8) 196 (8) 67 5 5 Indians 55 (7) 182 (9) 3 9 6 Soldiers 76 (4) 327 (7) 8 4 7 Modocs 119 (2) 595 (3) 26 Non- Indians 105 (3) 670 (2) 32 4 Scarface Charley 43 (9: 349 6) 23 9 100 (F) a—Coefficient of Evaluation = where F = total favorable F + U terms and U = total unfavorable terms 100 (Total Evaluative Words] b—Evaluativeness Score = Total Words
*—indicates ranking
The page and total word frequencies are indicative of the major characters and groups in the book—that is, it would be expected that the major characters would be ranked highest on both page and word frequencies. There are, however, some interesting discrepancies between the two measures for various subjects. The group subjects of "Indians" (#5), "Soldiers" (#6), and "Modocs" 100
C#7) all are portrayed in a narrower! depth or ranking (given in
parentheses next to the frequencies listed above) in terms of descriptive^andevaluative terms in comparison to their scope or
page-frequencies-. Secondly, both the subjects "Father" (#3)
(also Non-Indian) and "Non-Indians" (#8) have greater depth of coverage in words than would be predicted by their respective page frequencies. These differences in frequencies can be inferred as unconscious portrayals by the source. The page or appearance frequencies that are low for certain subjects present a manifest illusion of their being minor subjects, but are in fact latently much more important subjects because they are treated in greater depth. A third point of interest is the large difference in rankings for the Native American subject "Scarface Charley" (#9).
"Scarface Charley," having been mentioned on the minimum number of pages for inclusion in this analysis, ranks ninth in page frequency and sixth in total word frequency. Lastly, the.subject
"Soldiers" (#6) is notable because by page frequency the subject is considered a major identifiable group in the book, yet the total coverage by descriptive and evaluative words is ranked much lower than for page frequency. This discrepancy may be inferred, again, as unconscious treatment by the source. Manifestly it appears that the subject "Soldiers" plays an important part in the book, but latently the subject's portrayal is really limited to that of 101
a minor subject because of the lesser depth or word
frequency.
While none of the Coefficient of Evaluation scores
for subjects are skewed in the positive portrayal direction,
several are skewed in the negative direction: (1) "Indians" (3%);
(2) "Soldiers" (8%); and (3) "Scarface Charley" (23%). This means,
again, that the larger the score the more favorable the portrayal,
while conversely, the smaller the score the less favorable or
more unfavorable the portrayal. As a percentage of favorable
terms to total evaluative words, the Coefficient of Evaluation
scores range from zero to one hundred percent. As such, the scores
for a normal distribution would be expected to be between twenty-
five and seventy-five percent, and scores outside this range can
be considered skewed and, therefore, significant. Additionally,
scores between fifty and seventy-five percent are in the positive
direction but ambivalent, while scores between twenty-five and
fifty percent are in the negative direction but ambivalent.
Coefficient of Evaluation scores, then, range from a low of three
percent for the subject "Indians" (#5) to a high of sixty-seven
percent for the character subject "Mother" (#4). In general, group subjects are subject to evaluative terms relatively more often than character subjects by the source, and Native American group subjects (#5, #7) are negatively evaluated relatively more 102 often than Non-Native American group subjects (#6, #8).
Interestingly, the two most negatively evaluated subjects are
"Indians" (#4) and "Soldiers" (#6), which, it is noted, are frequently portrayed as antagonists and protagonists in the various medias. The Evaluativeness Score, or the percentage of value judgments expressed by the source about a subject relative to the total words, provides another statistic on the subjects. From
Table 13, then, it is observed that the subject group "Indians"
(#5) has twice as many evaluative judgments (19%) than any other subject. Additionally, the two Native American character subjects,
"Captain Jack" (#2) and "Scarface Charley" (#9), both have the second greatest number of evaluative judgments (9%).
In looking at the distribution of scores in Table 12 for
Squaw Man's Son it is clear that the overwhelming number of words used in association with subjects are neutral or descriptive terms.
The percentage of descriptive or neutral terms for the subject
"Indians" (#5), however, is to be noted as being lower than predicted (81 .32% < 90%) for books in general. As such, it is of interest to analyze the types of terms used to portray subjects.
Table 14, therefore, ranks the subjects for each type of term and for each character.
The rankings in Table 14 indicate several interesting trends and results. First, the subject "Captain Jack" (#2) is 103
Table 14
Squaw Man's Son (Book #4): Positive, Negative, and Neutral Word Rankings
Rank Positive Rank Negative Rank Neutral
1 (2)* Captain 1 (5) Indians 1 (6) Soldiers Jack (18.13%) (96.02%) (3.58%)
2 (3) Father (9) Scarface (8) Non- (3.16%) Charley Indians (6.88%) (95.82%;
3 (4) Mother [2) Captain (4) Mother (3.06%) Jack (95.41%) (5.60%)
4 ( 1) Billy (7) Modocs 4 (1) Billy (2.52%: (4.37%) (95.10%)
Scarface [6) Soldiers 5 (7) Modocs Charley (3.67%) (94.12%) (2.01%)
(7) Modocs (3) Father (3) Father (1 .51%) (3.16%) (93.68%
(8) Non- (8) Non- Scarface Indians Indians Charley (11.34%) (2.84%) (91.12%)
Indians M ) Billy Captain (0.55%) (2.38%) Jack (90.83%)
9 (6) Soldiers [4) Mother Indians (0.31%) (1.53%) (81.32%;
*—indicates the subject number 104 ranked first in positive evaluations. Second, Native American subjects, including "Captain Jack," compose the top four rankings for negative evaluations. Third, like book number three, the top three ranked negatively evaluated subjects (#5, #9, #2) are inversely the bottom three ranked descriptive subjects (#9, #2, #5).
Contrary to results discussed above, the rankings in Table 14 clearly indicate that the subject "Soldiers" (#6) is the most descriptive portrayal in the book Squaw Man's Son, although the evaluative portion of the portrayal is essentially negative.
Lastly, it is found that the specific Native American group subject "Modocs" (#7), in comparison to the general subject group
"Indians" (#5), is more descriptively portrayed. Furthermore, the subject "Modocs" is fairly evenly balanced in rankings across types of terms and is more balanced in its evaluative portrayal than the subject "Indians." It is inferred, therefore, that the source is more evaluative about "Indians" in general than for the specific group "Modocs."
Statistical analysis of the results demonstrates that for the book Squaw Man's Son, Native American subjects—especially the subject "Indians" (#5)—are negatively portrayed both substantively and statistically speaking. Generally, portrayals of Native American subjects are more evaluative than those of Non-Native American subjects, although there are exceptions for both (e.g., "Mother" 105
(#4) and "Father" (#3)). Lastly, although both the subject
"Indians" (#5) and the subject "Soldiers" (#6) have significantly
low Coefficient of Evaluation scores, the subjects are diametrically
neutrally described. That is, the subject "Soldier" is the most
descriptive portrayal in the book, while the subject "Indians" is
the most evaluative portrayal in the book.
Book Number Five (#5): Sarah Bishop (Boston: Houghton Mifflin Company, 1980).
The fifth book to be analyzed in this study is Sarah Bishop,
which is also by Scott O'Dell. The book is 186 pages long, including
the cover page and the author's forward, and is listed as being sixth grade and up reading level by Wilson's Children's Catalog
(1982:62). There are more than twenty-eight characters and groups
in the story, but only five main subjects are measured and analyzed.
Subjects are grouped to obtain the maximum scope and depth without
great variance, yet allow for a sample and valid analysis of the
book. Initially nine subjects were selected, but it was later
decided that three of the subjects were minor enough to be included
in the subject "Non-Indians." The subjects, then, include one
Native American character subject, one Native American group subject,
one Non-Native American character subject, and two Non-Native
American group subjects.
It is seen from Table 15, using the same null hypothesis Table 15
Sarah Bishop {Book #5): Observed Word Frequencies
Term Direction Subjects Totals
British Sarah Indians Non-Indians Longknifes (#1 ) (#2) (#3) (#4) ' (#5)
+ 12 8 3 5 8 36 (1,.85%) (0 .95%) (1 .19%) (0.56%) (1.63%) (1.15%)
0 630 826 244 867 466 3033 (96,.92%) (97 .75%) (96 .44%) (97.53%) (95.10%) (96.99%)
- 8 11 6 17 16 58 (1,.23%) (1 .30%) (2 .37%) (1.91%) (3.27%) 11.85%)
Totals 650 845 253 889 490 3127 (100 .00%) (100 .00%) (100 .00%) (100.00%) (100.00%) (99.99%)
Ratios 1.5000 .7273 .5000 .2941 .5000 .6207 +: -
X2 = 15.3 df = X2 = 20 P< .10 8 .01 df=;8 -°
O on 107
(H ) and alpha error, or level of significance, that calculated
X2 = 15.3 with df = 8. The table X2 = 20.0, which is •UI at-o0 greater than the calculated X 2 and as such the null hypothesis is not rejected. This means that the overall distributional differences found among terms used to describe and evaluate subjects in this book do not vary significantly at the .01 (alpha) level.
The differences, however, are nearly significant at the .05 level, and are significant at the .10 level.
Although the chi-square analysis is rejected at the .01 level, it is noted that this level of significance is very rigid for an exploratory study such as this one. Normally an alpha level of .10 in an exploratory study can be considered acceptable.
In any case, it is of interest to further analyze the results for this book, Sarah Bishop. First of all, it is observed that Table 15 also lists the ratios of positive to negative terms, which provides some interesting results. It is observed that the ratios of positive to negative terms range from a high of 1.5000 for the subject group "British" (#1) to a low of .2941 for the subject group "Non-Indians" (#4). This means that for the subject "British" the source utilizes 150 positive words for every one hundred negative terms, which is a substantively significantly positive portrayal of the subject "British." Conversely, for the subject "Non-Indians" the source uses only twenty-nine positive terms for every one 108 hundred negative words. Both Native American subjects (#3, #5) have twice as many negative terms with a positive to negative ratio of .5000. Although most ratios are less than balanced, the ratio for "Non-Indians" is unusually low, which tends to indicate a substantively significantly negative portrayal of this subject in Sarah Bishop. Secondly, subjects are analyzed for this book, as presented in Table 16, by scope or page frequency, depth or word frequency, and statistical scores for the Coefficient of
Evaluation and Evaluativeness Score.
Table 16 Sarah Bishop (Book #5): Page and Word Frequencies, Coefficient of Evaluation, and Evaluativeness Score
Coefficient Evaluativeness Subject Page Word of Evaluation3 Score Number Subject Frequency Frequency % %
1 British 121 (3)* 650 (3)* 60 3 2 Sarah 183 (1 ) 845 (2) 42 2 3 Indians 20 (5) 253 (5) 33 4 4 Non- Indians 151 (2) 889 (1 ) 23 3 5 Longknifes 41 (4) 490 (4) 33 5 100 (F) a—Coefficient of Evaluation = — rr-, where F = total favorable r + u terms and U = total unfavorable terms 100 (Total Evaluative Words) b—Evaluativeness Score = Total Words
*—indicates ranking 109
The page and word frequencies, again, indicate which subjects are the major characters or groups in the book. It is observed that Native American subjects (#3, #5) are minor subjects, being ranked fourth and fifth on both the page and word frequencies, indicating a balance in depth and scope of their poartrayals, which is substantively significant to this study. The subject "British"
(#1 ) also has a balanced portrayal, ranking number three on both frequencies. The subject "Sarah" (#2) and "Non-Indians" (#4) are reversed in their rankings, but the difference is not that great considering the broadness of the latter subject; that is,
"Non-Indians" encompasses many characters and groups.
Once again it is observed that none of the Coefficient of
Evaluation scores for the subjects are positively skewed, although the score for the subject group "British" (#1) is ambivalent in the positive direction. Four of the five scores, conversely, are in the negative direction, with the score for the subject "Non-
Indians" (#4) being negatively skewed (23% < 25%), which therefore yields a mean negative Coefficient of Evaluation for Sarah Bishop of twenty-six percent, (see Appendix G). Furthermore, Native
American subjects (#3, #5) are observed to be consistently evaluated in the negative direction with both having a Coefficient of
Evaluation score of thirty-three percent, while Non-Native
American subjects (#1, #2, #4) exhibit a broad range of Coefficient 110 of Evaluation Scores (23%-60%). The Evaluativeness Score, however,
provides a substantively significant finding in that none of the subjects are very evaluatively portrayed as all scores are- less
than ten percent. This finding is reinforced by looking at
Table 15, from which it is observed that all subjects are over whelmingly descriptively portrayed as predicted (>90%). Regardless, it is noted that Native American subjects are more evaluatively
portrayed than are Non-Native Americans. It is further observed
that the portrayal of the character subject "Longknifes" (#5) is more evaluative than that for the group subject "Indians" (#3).
Conversely, portrayal of the Non-Native American character subject
(#2) is less evaluative than that for Non-Native American group subjects (#1, #4).
In looking at the data distribution in Table 16 once again, it is noted that for the book Sarah Bishop all subjects
have, as predicted, a preponderance of descriptive terms (>90%).
Additionally, because the book was published after 1975 it is
predicted that subjects will be more neutrally portrayed by the use of a higher percentage of descriptive terms. This proposed result is found to be true in Sarah Bishop for all subjects: (1) "Sarah"
(97.75% > 95%); (2) "Longknifes" (95.10% > 95%); (3) "British"
(96.92% > 95%); (4) "Non-Indians" (97.53% > 95%); and (5) "Indians"
(96.44% > 95%). It is noted, nonetheless, that Native American 111 subjects have the least descriptive portrayals, or rather that the portrayals are the most evaluative. Further analysis of the types of terms used for each subject would be interesting. Table 17, therefore, ranks each subject by percentage frequency of terms.
Table 17 Sarah Bishop (Book #5): Positive, Negative, and Neutral Word Rankings
Rank Positive Rank Negative Rank Neutral
;i)* British (5) Longknifes (2) Sarah (1 .85%)- (3.27%) (97.75%)
Longknifes (3) Indians (4) Non-Indians (1.63%) (2.37%) (97.53%)
(3) Indians (4) Non-Indians ! 1 ) British (1.19%) (1.91%) (96.92%)
Sarah \2) Sarah (3) Indians (0.95%) (1.30%! (96.44%)
[4) Non-Indians [1) British (5) Longknifes (0.56%) (1.23%) (95.10%)
*—indicates the subject number
In looking at Table 17 some very interesting and substantively significant trends stand out. First, as with previous books analyzed in this study, the two Native American subjects
(#5, #3) are ranked first and second in negative terms. Secondly, it is observed again that the descriptive or neutral rankings for 112
Native American subjects are both inversely ranked (#3, #5) from the negative term rankings and ranked fourth and fifth, or lowest, in descriptive portrayals. These results, then, tend to substan tiate other results discussed above. Another interesting trend is observed for the subject "British" (#1), which has been shown to have a positive Coefficient of Evaluation. The subject "British" ranks first in positive words, last in negative terms, and in the middle for neutral words, which combine to present a very positive evaluative portrayal. That is, holding the fact that the portrayal is considered descriptive, that portion of the portrayal that is evaluative is given the illusion of being very positive. This is accomplished through the process of utilizing more positive terms and less negative words for the subject "British" in relationship to the other characters. This is the same process employed for
Native American subjects, which allows the source to covertly evaluatively portray subjects. That this technique is used, is best substantiated by the trend found in Sarah~ Bishop and other books to inversely correlate negative and positive portrayaLs for
Native American subjects. This result is corroborated by a last result observed in Table 17. While the portrayals of Native
Americans, as well as all other subjects, are uniquely descriptive, the subjects also rank high on both positive and negative evalu ations. This means, therefore, that the book Sarah Bishop, while 113 being overtly descriptive or having very small percentages of evaluative portrayals, actually portrays Native Americans negatively as indicated by these trends.
Substantive and partial statistical analysis of the results demonstrates that in the book Sarah Bishop Native Americans are negatively portrayed relative to Non-Native Americans. More significant, however, is the fact that it is the only book in this study that meets the predicted results concerning increased descriptive portrayals. Furthermore, it is shown that, despite inproved evaluativeness, the source still portrays Native Americans the same way as the other sources in this study. Lastly, it is interesting that the Native American character subject "Longknives"
(#5) has the most evaluative portrayal, rather than the group subject "Indians" (#3) as generally found in the other books.
Book Number Six (#6): Where the Buffaloes Begin (New York: Frederick Warne and Company, Inc., 1981).
The last book to be analyzed, Where the Buffaloes Begin, was written by Olaf Baker but copyrighted by Frederick Warne and Company,
Incorporated. The book in unpaged (unpj, but is forty-four pages long, including the front and back cover pages, and is listed by Q Wilson's Children's Catalog under fiction as being at the second to fourth grade reading level''^ (1982:52). The book contains two human subjects, which are measured and analyzed, and one nonhuman 1U
subject—buffaloes—which was measured but is not analyzed here
because it proved to be the only subject of its type in this study and is outside the definitions, or scope, provided in the
procedures discussion in chapter two. As with the first book,
Akavak, an Eskimo Journey, this source presents all Native American subjects, but in a we-they dichotomy. There is, then, one "we"
character subject and one "they" group subject in the book. It
is noted, qualitatively, that this dichotomy is exaggerated by the fact that the "we" subject is an identifiable character subject
"Little Wolf" (#1), while the "they" subject is an impersonal, but specified, group subject "Assiniboines" (#2).
It is seen from Table 18 that in the chi-square analysis for this particular book, using the same null hypothesis (H ) and 2 alpha error or level of significance, X =10.1 with df = 2. The 2 2 chi-sqare table shows that X Evaluative and descriptive term frequencies, then, are directly related to, or dependent upon, the subjects (H ) and who the cL subjects are. -- Since chi-square analysis is statistically significant, it is of interest to further analyze the results. First of all, Table 18 Where the Buffaloes Begin (Book #6): Observed Word Frequencies Term Direction Subjects Totals Little Wolf Assiniboines (#1 ) (#2) + 7 3 10 (5 .34%) (2.78%) (4.18%) 0 118 87 205 (90 .08%) (80.56%) (85-77%) - 6 18 24 (4 .58%) (16.67%) (10.04%) Totals 131 108 239 (100 .00%) (100.01%) (99.99%) Ratios 1 .1667 .1667 .4167 +:- 2 X = 10.1 df = 2 X2 9 2 P< 01 .01 df=2 = " - 116 it is observed that Table 18 lists the ratios of positive to negative words, which are very interesting. The obvious result is the substantively significantly large difference in the ratios. The obvious trend is the negativeness and imbalance in the portrayal of the subject group "Assiniboines" {42), and the more balanced portrayal of the character subject "Little Wolf" (#1). The ratio of positive to negative evaluative terms for the subject "Little Wolf" is found to be 1.1667, which is a nearly balanced portrayal with the source using 116 positive words for every one hundred negative words. Conversely, the ratio of positive to negative terms for the subject "Assiniboines" is substantively significantly skewed in the negative direction with a ratio of .1667, which means that for every sixteen positive terms the source utilizes one hundred negative words to evaluate the subject "Assiniboines." Subjects are also analyzed, as found in Table 19, by word and page frequencies, and by Coefficient of Evaluation and Evaluativeness Scores. In looking at the subject portrayals it is seen that subject number one, "Little Wolf," the major character subject in the story, has 3.67 times the scope in appearances or page frequencies as the other subject, "Assiniboines." Conversely, the depth or word frequency for "Little Wolf" is only 1.12 times that of "Assiniboines." More importantly, however, is the 117 significantly higher proportion of words to pages for the "they" subject "Assiniboines" (#2), which is eighteen to one, than for the "we" subject "Little Wolf" (#1), which is found to be just under six to one. This difference is, again, inferred to be an unconscious and latent evaluative portrayal of the "they" subject "Assiniboines." It is noted, however, that both subjects rank the same on both frequency counts, which manifestly and latently indicates balanced portrayals between subjects. Yet, the ratios of words to pages, pages to pages, and words to words, indicate that the portrayals are covertly im'balanced between the "we" and the "they" subjects. The Coefficient of Evaluation scores also indicate a substantial difference between the we-they subjects, where the subject group "Assiniboines" is substantively signifi cantly skewed toward a very negative evaluation or portrayal (14% < 25%), whereas the subject character "Little Wolf" is found to be ambivalent but in the positive direction (52% > 50%). The scores, it is remembered, are expected to normally range from twenty-five to seventy-five percent, with fifty percent being perfectly ambivalent in evaluativeness. Scores on either side of the range are considered, thereby, substantively significant. Scores greater than fifty percent indicate positive evaluative portrayals and scores less than fifty percent indicate negative portrayals. While the range for ambivalence is left ambiguous, 118 it is generally conceived of as including scores between thirty-three and sixty-six percent. Thus a score of thirty-eight percent is ambivalent in the negative direction, and the score fifty-two percent is ambivalent in the positive direction, while the score fourteen percent is significantly negative. The second analytical technique, the Evaluativeness Score, additionally demonstrates that the source uses nearly twice the number of judgmental terms to portray the "they" subject "Assiniboines" (19%) as for the "we" subject "Little Wolf" (10%), both of which are, furthermore, substantively significantly greater than predicted for a book published after 1975 (>5%). More signifiant and relevant to this study, however, is that the Evaluativeness Scores statistically substantiate that portrayals of Native Americans, both "we" and "they" subjects, are again becoming significantly evaluative. The portrayals in this book, Where the Buffaloes Begin, are the second most evaluative in this study. Another trend inferred from these results is that books (i.e., Akavak, an Eskimo Journey (Book #1), Where the Buffaloes Begin (Book #6)) in which sources employ we-they dichotomies with Native American subjects only are substantively significantly more evaluative than books (i.e., Sing Down the Moon (Book #2), The Spirit is Willing (Book #3), Squaw Man's Son (Book #4), Sarah Bishop (Book #5)) that employ Native American-Non-Native American ethnic comparisons. 119 Table 19 Where the Buffaloes Begin (Book #6): Page and Word Frequencies, Coefficient of Evaluation, and Evaluativeness Score Coefficient Evaluativgness Subject Page Word of Evaluation Score Number Subject Frequency Frequency % % 1 Little Wolf 22 (1 )* 131 (1)* 52 10 2 Assiniboines 6 (2) 108 (2) 14 19 100 (F) a—Coefficient of Evaluation = —= 77-, where F = total favorable r + U terms and U = total unfavorable terms 100 (Total Evaluative Words) b—Evaluativeness Score = Total Words *—indicates ranking The distribution of scores in Table 18 for Where the Buffaloes Begin shows that the "we" subject "Little Wolf" (#1 ) has a larger percentage of descriptive words than the "they" subject "Assiniboines" (#2). It is also clear that, although the subject "Little Wolf" barely has a greater percentage of neutral words than proposed for books in general (90.08% > 90%), both major subjects have smaller percentages of descriptive portrayals than predicted for books published after 1975 (#1: 90.08% < 95%; #2: 80.56% < 95%). The percentage difference for the "they" subject is almost fifteen percent and, therefore, is substantively signifi cant. This difference between the we-they subjects in positive to 120 negative terms, once again, stands out. These results, then, suggest further analysis of the terms by type, found in Table 20, that ranks subjects by frequency percentages. Table 20 Where the Buffaloes Begin (Book #6): Positive, Negative, and Neutral Word Rankings Rank Positive Rank Negative Rank Neutral 1 (1)* Little Wolf 1 (2) Assiniboines 1 (1) Little Wolf (5.34%) (16.67%) (90.08%) 2 (2) Assiniboines 2 (1) Little Wolf 2 (2) Assiniboines (2.78%) (4.58%) (80.56%) *—indicates the subject number In looking at Table 20 several by now familiar trends are observed. First, it is observed that the "we" subject "Little Wolf" (#1) ranks first for positive and neutral evaluations and last for negative evaluations. Conversely, the subject "Assiniboines" (#2) ranks first in negative terms and last in positive and neutral terms. Secondly, and more importantly, it is seen that the rankings for the "they" group, like for Native Americans and "they" subjects in other books, are inversely ranked for negative and neutral portrayals. While both portrayals are found to be substantively significantly evaluative and negative, these results by rank 121 substantiate that the "they" subject "Assiniboines," which is a specific group subject, also is the most biased portrayal in the book and, indeed, the second most evaluative in this study. Statistical analyses of the results, then, deomonstrate that the book, Where the Buffaloes Begin, significantly portrays Native American subjects and their way of life negatively. The source, by using all Native American subjects and a we-they dichotomy, reverses the general trend to more descriptively portray Native Americans. The portrayals, however, are found to fit the patterns of other books as well. That is to say, the negative portrayal of the "they" subject "Assiniboines" (#2) is inversely related to the descriptive portrayal. Similarly, through other stylistic techniques, the source also compounds the evaluative portrayal for a latent, covert negative portrayal of the "they" subject "Assiniboines." Comparison of Books This section of the results chapter will look at distri butional differences between books, having observed how Native American and Non-Native American subjects are portrayed within individual books. Data from Tables 3, 6, 9, 12, 15, and 18 have been summated and grouped by either: (1) ethnicity (Native American and Non-Native American) for the books Sing Down the Moon 122 (Book #2), The Spirit is Willing (Book #3), Squaw Man's Son (Book #4), and Sarah Bishop (Book #5) (see Table 21); or (2) dichotomy ("we" and "they") for Akavak, an Eskimo Journey (Book #1) and Where the Buffaloes Begin (Book #6) (see Table 22). Subjects are also classified by either ethnicity or dichotomy in summarizing the Coefficient of Evaluation and Evaluativeness Scores (see Appendix G). Books number one and number six, Akavak, an Eskimo Journey and Where the Buffaloes Begin, do not contain ethnic separation per se and are dichotomized, therefore, into "we" and "they" groups. In Where the Buffaloes Begin, the "they" group comprises a second specified Native American group (Assiniboines) who are the antagonists to the "we" subject who is from an unnamed Native American group. In Akavak, an Eskimo Journey, the "they" group encompasses a variety of concepts within the Eskimo belief system, while the "we" group includes both Eskimo character subjects. If these term distributions are found to be statistically significant, then ethnicity and dichotomy will be held constant to analyze term distributional differences across books (see Tables 23 and 2k). Similarly, books will be held constant to analyze ethnic and dichotomy differences in term distributions (see Tables 25 and 26). As with previous tables, each of these tables contain: (1) positive, neutral, and negative cell frequencies with the frequency percentage in parentheses; (2) the ratio of positive to negative words for each 123 cell; and (3) the chi-square test of significance, which includes both the table and calculated chi-square values, the degrees of freedom, and the greatest level of significance (e.g., p<.001). A statistical test of significance is made for the data 2 distribution in Table 21 using chi-square (X ), which tests to see if the terms used in portrayals of Native American and Non- Native American subjects are independent of these subjects. This is done, again, to more closely determine if the differences across cells exist because the terms are related to the subjects, or if the differences exist simply because of chance or sampling error. As such, the null hypothesis (H ), the alternative hypothesis (H ), O 3 and level of significance or alpha error (oc) are established as follows: H : Evaluative and descriptive terms are not related to the o subjects.k • 4. H : Evaluative and descriptive terms are related to the a subjects. oC = .01. Assuming this null hypothesis (H ) and level of significance or alpha level {<& ), it is seen from Table 21 that calculated X2 = 203.7 with df = 14. Since table X2^ = 29.1, which is less than the calculated X 2 , the null hypothesis is rejected and the observed differences are statistically significant. Additionally, two separate cells within the table are also found Table 21 Observed Frequencies by Book and Ethnicity Term Direction Books/Ethnicity Totals Sing Down the Moon The Spirit is Willing Squaw Man 's Son Sarah Bishop (#2 ) (#3) (#4) (#5 I NA1 NNA2 NA NNA NA NNA NA NNA + 52 8 10 51 58 27 11 25 242 (2.11%) (2.24%) (2.29%) (3.15%) (2.30%) (1.76%) (1.48%) (1 .05%) (2.00%) 0 2295 317 366 1505 2337 1460 710 2323 11313 (93.22%) (88.80%) (83.94%) (92.96%) (92.59%) (95.11%) (95.56%) (97.44%) (93.80%) _ 115 32 * 60 63 129 48 22 * 36 505 (4.67%) (8.96%) (13.76%) (3.89%) 15.11%) (3.13%) (2.96%) (1 -51%) (4.19%) Totals 2462 357 436 1619 2524 1535 743 2384 12060 (100.00%) (100.00%) (99.99%) (100.00%) (100.00%) ( 100.00%) (100.00%) l(100.00%) (99.99%) Ratios .4522 .2500 .1667 .8095 .4496 .5625 .5000 .6944 .4792 x2 = 203.7 df = 14 x2 1 p<.001 .01 df:=14 = 29- 1—NA stands for Native Americans 2—NNA stands for Non-Native Americans *—indicates statistical significance for individual cell distribution 125 to be statistically significant; that is, each cell's difference between observed and expected frequencies is large enough to be 2 greater than X .. =29.1. This means that the observed .01 df=14 greater than expected frequency of negative terms to portray Native Americans in book number three, The Spirit is Willing by Betty Baker, is by itself statistically significant. Similarly, the observed lower than expected frequency of negative terms to portray Non- Native Americans in book number five, Sarah Bishop by Scott O'Dell, is by itself statistically significant. This latter result is particularly noteworthy in that the frequency distributions for Sarah Bishop (Book #5) by themselves are found not to be statisti cally significant. When compared with other books, however, it is found that the low percentage of negative evaluations used by the source to portray Non-Native Americans is indeed significant. This infers that the negative portrayal of Non-Native Americans is significantly less than expected, which would cause, thereby, the seemingly balanced portrayal of Native Americans to be latently more negative. Further comparison shows that the observed frequency of all evaluative terms for both Native Americans and Non-Native Americans in Sarah Bishop (Book #5) are less than expected, meaning that the book is less evaluative than would be expected. Conversely, it is found that observed frequencies of all evaluative terms for Native Americans and positive terms for Non-Native Americans in 126 The Spirit is Willing (Book #3) are greater than expected, meaning that the book is more evaluative than expected. This indicates that the portrayal of Native Americans in The Spirit is Willing (Book #3), then, is manifestly negatively judgmental, while the portrayal of Native Americans in Sarah Bishop (Book #5) is latently negatively judgmental because of the statistically less than expected negative portrayal of Non-Native Americans, but is manifestly descriptive. Since the chi-square analysis is found to be significant, it is of interest to see if there are any other patterns or trends. First of all, the majority of books listed in Table 21 are found to contain descriptive portrayals (>90%) for both Native Americans and Non-Native Americans. The descriptive portrayal of Native Americans in The Spirit is Willing (Book #3), however, is less than expected (83.94% < 90%), as is the descriptive portrayal of Non-Native Americans in Sing Down the Moon (Book #2) (88.80% < 90%). These differences are substantiated by the Evaluativeness Scores (see Appendix G). In looking at the ratio of positive to negative terms for these two categories it is observed that their ratios are significantly low. It is additionally observed that the ratios for all groups are less than balanced (1.0000), indicating that all evaluations are in the negative direction. The ratios range from a low of .1667 for Native Americans in The Spirit is Willing 127 (Book #3) to a high of .8095 for Non-Native Americans in the same book. Furthermore, it is found from these ratios that the portrayals of Native Americans in The Spirit is Willing (Book #3), Squaw Man's Son (Book #4), and Sarah Bishop (Book #5) are more negative than the portrayals of Non-Native Americans. Conversely, the portrayals— while not positive portrayals—of Native Americans in Sing Down the Moon (Book #2) are less negative than those for Non-Native Americans. Lastly, it is observed that the differences in ratios within The Spirit is Willing (Book #3) between Native American and Non- Native American subjects is substantively significant. Similarly, the difference in ratios within Sing Down the Moon (Book #2) between Native Americans and Non-Native Americans is observed to be substan tively large, if not significant. A statistical test of significance for the data distribution 2 in Table 22 yields X =134.6 with df = 6. Assuming the same null hypothesis (H ) and level of significance or alpha level as stated 2 above, the table X ,„ ^ = 16.8, which is less than the calculated .01 df=6 X2 . This means the null hypothesis is rejected and the observed differences in Table 22 are statistically significant. Additionally, it is found that the differences in the neutral and negative cells of the "they" category in Akavak, an Eskimo Journey (Book #1) are statistically significant on their own accord. This indicates that the observed less than expected frequency of descriptive terms Table 22 Observed Frequencies by Book and We-They Term Direction Books/We-They Totals Akavak, an Eskimo Journey Where the Buffaloes Begin (#1) (#2) They We They We + 4 50 3 7 64 (4.04%) (4.06%) (2.78%) (5.34%) (4.08%) 0 * 39 1004 87 118 1248 (39.39%) (81.49%) (80.56%) (90.08%) (79-49%) * 56 178 18 6 258 (56.57%) (14.45%) (16.67%) (4.58%) (16.43%) Totals 99 1232 108 131 1570 (100.00%) (100.00%) (100.01%) (100.00%) (100.00%) Ratios .0714 .2809 .1667 1.1667 .2481 +:- X2 = 134.6 df = 6 X2 = 16.8 p<.001 .01 df=6 *—indicates statistical significance for individual cell distribution 129 and the observed greater than expected frequency of negative terms used to portray the "they" subjects in Akavak, an Eskimo Journey (Book #1) are each separately statistically significant, making the portrayal very negatively judgmental. Since the chi-square analysis is significant, it is of interest to see if there are many other patterns or trends. First of all, in looking at Table 22 the most striking result is the phenomenal level of the negative evaluations in the portrayal of the "they" subjects (16.67% and 56.57%). Secondly, it is observed that only the "we" category in Where the Buffaloes Begin (Book #6) has a frequency of descriptive terms greater than predicted (90.08% > 90%). It is further noted that the difference between we-they positive term percentages for Akavak, an Eskimo Journey (Book #1) is nearly zero (4.06%^ 4.04%), but the difference for Where the Buffaloes Begin (Book #6) is nearly three percent. It is substantively significant that the percentage of positive evaluations for "they" subjects in Where the Buffaloes Begin (Book #6) is less than in Akavak, an Eskimo Journey (Book #1), while for "we" subjects positive evaluations increased. These trends infer that portrayals of "they" subjects have become less positive over time, whereas portrayals of "we" subjects have become more positive. These results are more substantively significant when it is observed that the percentage of negative evaluations for both categories have decreased over time. 130 Next, in looking at Table 22 it is also seen from the ratios of positive to negative words that the "they" subjects have more negative portrayals and that, more importantly, the portrayal of the "we" subjects in Where the Buffaloes Begin (Book #6) is in the positive direction (1.1667 > 1.0000). Indeed, based on the ratio of positive to negative evaluations of subjects, the portrayals of the "we" subjects in Where the Buffaloes Begin (Book #6) are the only positive evaluations in this study. These differences, then, substantively significantly indicate an extreme polarity in the portrayals of "we" and "they" subjects in book number six, Where the Buffaloes Begin by Frederick Warne and Company, Incorporated. Similarly for book number one, Akavak, an Eskimo Journey by James Houston, while the differences are not as significant, the "they" subjects are relatively more negatively portrayed than the "we" subjects—although both are very negatively portrayed. These results, again, imply that differences in we-they portrayals have increased or gotten worse over time. Having found considerable statistical significance for both Tables 21 and 22, analyses are make (see Tables 23-26) to try to determine if this statistical significance is due more to book differences or to ethnic/dichotomatic differences. Table 23 shows the distributions for the books Sing Down the Moon (Book #2), The Spirit is Willing (Book #3), Squaw Man's Son (Book #4), and Table 23 Observed Frequencies by Book with Ethnic Subjects Term Direction Books Totals Sing.Down the Moon The Spirit is Willing Squaw Man's Son Sarah Bishop (#2) (#3) (#4) (#5) + 60 61 85 36 242 (2.13%) (2.97%) (2.09%) (1.15%) (2.00%) 0 2612 1871 3797 3033 11313 (92.66%) (91.05%) (93.55%) (96.99%) (93.80%) — 147 123 177 * 58 505 (5.21%) (5.99%) (4.36%) (1.85%) (4.19%) Totals 2819 2055 4059 3127 12060 (100.00%) (100.01%) (100.00%) (99.99%) (99-99%) Ratios .4082 .4959 .4802 .6207 .4797 +: - 2 X = 90.6 df = 6 X!oi df=6 = 16' 8 p<.001 *—indicates ststistical significance for individual cell distribution 132 Sarah Bishop (Book #5), while holding ethnicity constant. The calculated X2 = 90.6 with df = 6, and the table X2^ = 16.8 using the same null hypothesis (H ) and level of significance or 2 alpha level (oC = .01). Since the calculated X value is greater 2 than the critical X value the null hypothesis is again rejected, meaning that the distributional differences in evaluative judgments among books shown in Table 23 are statistically significant. It is seen, however, that the observed lower frequency of negative terms in book number five, Sarah Bishop, is also statistically significant, although it does not solely account for this significance 2 of X . These results suggest, therefore, that book number five, Sarah Bishop, does indeed show a different pattern when ethnicity is held constant. Since the chi-square analysis is significant, it is of interest to see if there are any other patterns or trends. First, it is observed that the books are all more descriptive than evalu ative (i.e., >90%). Similarly, it is observed in Table 23 that by holding ethnicity constant the differences in evaluative portrayals for the books Sing Down the Moon (Book #2), The Spirit is Willing (Book #3), and Squaw Man's Son (Book #4) are very minimal, while book number five, Sarah Bishop, again stands out, particularly the low negative portrayal. In looking at just evaluative terms, it is observed that the ratios for positive to negative words are 133 all substantially less than 1.0000 (which would be a balanced portrayal). This means that all evaluative portrayals are more negative than positive, ranging from a low of forty-one positive words for every one hundred negative terms for book number two, Sing Down the Moon, to a high of only sixty-two positive words for every one hundred negative terms to evaluatively portray subjects in book number five, Sarah Bishop. Evaluative portrayals, based upon ratios of positive to negative words, and descriptive portrayals, based upon percentages of neutral words, are ranked by book as follows: Evaluative Descriptive Book Title Ratio Rank Percentage Rank 2 Sing Down the Moon .4082 4 92.66 3 3 The Spirit is Willing .4959 2 91.05 4 4 Squaw Man's Son .4802 3 93.55 2 5 Sarah Bishop .6207 1 96.99 1 It is seen from the above rankings that book number five, Sarah Bishop, has both the highest evaluative rank or positive to negative terms ratio, and the highest ranking for percentage of descriptive words. Conversely, book number three, The Spirit is Willing, has the second highest positive to negative terms ratio, but the lowest percentage of descriptive terms, while both books number two, Sing Down the Moon, and number four, Squaw Man's Son, rank higher in use of descriptive terms than their ratios of positive to negative terms. Generally, 134 then, very little of substantive significance is found when ethnicity is held constant. Table 24 shows the distributions for books Akavak, an Eskimo Journey (Book #1) and Where the Buffaloes Begin (Book #6) while holding the we-they dichotomy constant. It is found that the calculated X2 = 8.4 with df = 2, and the table X2 „ = 9.2 .01 df=2 using the same null hypothesis (H ) and level of significance or 2 alpha level ( = .01). Since the calculated X value is less than 2 the critical X value, the null hypothesis is not rejected, meaning that the term distributional differences are not significant and terms are independent of books number one and number six (Akavak, an Eskimo Journey and Where the Buffaloes Begin, respectively). It is noted from Table 24, however, that the term distributional differences are in fact significant at the level of significance or alpha level of oC = .02. As such, it is valuable to further analyze the data on book differences, holding we-they constant. First, looking at the descriptive portrayal percentages it is seen that neither book is very descriptive, although there havs been improvement over time. Similarly, evaluative portrayals are greater than expected (i.e., >10%) although, again, both positive and negative evaluations, and hence the ratios of positive to negative words, have improved over time. Generally, then, very little of significance is found holding we-they constant. Nonetheless, it is observed that books Table 24 Observed Frequencies by Book with We-They Subjects Term Direction Books Totals Akavak, an Eskimo Journey Where the Buffaloes Begin (#1 ) (#6) + 54 10 64 (4.06%) (4.18%) (4.08%) 0 1043 205 1248 (78.36%) (85. 77%) (79.49%) 234 24 258 (17.58%) (10. 04%) (16.43%) Totals 1331 239 1570 (100.00%) (99. 99%) (100.00%) Ratios .2308 .4167 .2481 +:- 2 X = 8.4 df = 2 X!oi df=2 = 9"2 p<.02 136 having Native American dichotomies of we-they subjects are far more evaluative than books having Native American and Non-Native American ethnic subjects. Since the results from holding ethnicity and dichotomy constant (see Tables 23 and 24) demonstrate very little substantive or statistical significance between books, except for the very broad tendency for books to apparently be improving. Table 25, then, shows the distributions for Native Americans and Non-Native Americans while holding books constant. A statistical test of 2 significance yields X = 39.3 with df = 2, while the table 2 2 X df-2 = Since the calculated X value is greater than the 2 critical X value the null hypothesis is rejected, meaning that the differences in Table 25 are significant and that the types of terms are dependent upon ethnicity. More importantly, it is found that the observed lower rate of negative terms used in portrayals of Non-Native Americans and the observed greater frequency of negative terms found in portrayals of Native Americans are both individually statistically significant so that each cell could account for the. statistical significance of X 2 in Table 25 by itself. This means that the portrayals of Non-Native Americans in these books—Sing Down the Moon (Book #2), The Spirit is Willing (Book #3), Squaw Man's Son (Book #4), and Sarah Bishop (Book #5)— taken as a whole are significantly (p<.001) less negative than Table 25 Observed Frequencies by Ethnicity Term Direction Ethnicity Totals NA1 NNA2 + 131 111 242 (2.12%) (1.88%) (2.00%) 0 5708 5605 11313 (92.59%) (95.08%) (93.80%) * 326 * 179 505 (5.29%) (3.04%) (4.19%) Totals 6165 5895 12060 (100.00%) (100.00%) (99.99%) Ratios .4018 .6201 .4792 X2 = 39.3 df 2 x p<.001 = !oi df=2 = 9,2 1—NA stands for Native Americans 2—NNA stands for Non-Native Americans *—indicates statistical significance for individual cell distribution 138 expected, while the portrayals of Native Americans in these same books, taken together, are significantly (p<.001) more negative than expected. Since the chi-square is significant, it is of further interest to see if there are any other patterns or trends. First, it is again noted that the ratio of positive to negative judgments are both less than 1.0000, indicating generally negative evaluations of both Native Americans and Non-Native Americans, however, the ratio of positive to negative words is less for Native Americans than for Non-Native Americans (.4018 < .6201). Secondly, in looking at Table 25 it is observed that both Native Americans and Non-Native Americans, holding books constant, are more descriptive than evaluative as was predicted (>90%). In terms of evaluative words only, however, it is found that Native Americans are evaluated more frequently. Lastly, it is interesting to note that Native Americans have a larger percentage of positive portrayals than Non-Native Americans, which is substantively significant. Generally, ethnicity is found to have both statistical and substantive significance, when books are held constant, upon the portrayal of subjects. The difference is, furthermore, individually significant for negative evaluations or judgments. This means that if the subject is Native American the portrayal is signifi cantly negative, whereas the portrayal of Non-Native Americans is 139 significantly less negative. In sum, then, portrayals of Native Americans are negatively biased because the subjects are Native Americans, and not because of book or source biases. Table 26 shows the distribution for the dichotomy of we-they subjects while holding the books constant. From the table it is observed that a test of significance provides a calculated X2 = 64.8 with df = 2, while the table X2^ ^ = 9.2. 2 2 Since the calculated X value is greater than the critical X value, the null hypothesis is rejected, which means that the relationships in Table 26 are statistically significant. Additionally, it is found that the observed greater frequency of negative evaluations of "they" subjects are independently statistically significant (p<.001). Since the chi-square is significant, it is of further interest to see if there are any other patterns or trends. First, it is observed that the portrayals of both "we" and "they" subjects are substantively significantly less descriptive than would be prediected (i.e., >90%). The descriptive portrayals of "they" subjects, furthermore, are substantively significantly less than the descriptive portrayals for "we" subjects (60.87% < 82.32%). Secondly, it is observed that the negative portrayal is substantively significantly more evaluative for "they" subjects than for "we" subjects (35.75% > 13.50%). A third observation is that while the Table 26 Observed Frequencies by We-They Term Direction We-They Totals They We + 7 57 64 (3.38%) (4.18%) (4.08%) 0 126 1122 1248 (60.87%) (82.32%) (79.49%) — * 74 184 258 (35.75%) (13.50%) (16.43%) Totals 207 1363 1570 (100.00%) (100.00%) (100.00%) Ratio .0946 .3098 .2481 +I — 2 X = 64.8 df 2 x p<.001 = !oi df=2 = 9'2 *—indicates statistical significance for individual cell distribution 141 positive portrayals of both "we" and "they" subjects are fairly equal, the ratio of positive to negative words clearly indicates a near absence of positive evaluations for "they" subjects. Lastly, it is observed that books using dichotomatic comparisons are significantly more evaluative and hence negative in their portrayals of subjects, particularly "they" subjects. Generally, then, "they" subjects are, both substantively and statistically, significantly negatively portrayed. While "we" subjects are also substantively significantly evaluative, in comparison to "they" subjects they are descriptive. In analyzing book and ethnic differences several trends over books, and thereby time, have been suggested. As stated in the introduction of this chapter and in the proposed results for this study, it is of interest to see if patterns over time are to be found based on the data from the books in this study. To facilitate this analysis the books have been collapsed into three time periods: (1) Time 1 (T^): 1968-1970; (2) Time 2 (T^): 1974-1978; and (3) Time 3 (T ): 1980-1981. The earliest book in the sample is Akavak, -an Eskimo Journey (Book #1 ), published in 1968, and the most recent book is Where the Buffaloes Begin (Book #6), copyrighted in 1981. Both of these books, it is noted, encompass dichotomatic rather than ethnic comparisons. Nonetheless, these dichotomatic categories have been included with the ethnic 142 groupings, so that the "we" subjects are included with Non-Native Americans and the "they" subjects are incorporated with Native Americans. The theoretical justification for this is that both "they" subjects and Native American subjects are the assumed "others" in all the stories except Sing Down the Moon (Book #2) and Squaw Man's Son (Book #4) where similar dichotomies exist; e.g., between the Navajos ("we") and other Native Americans ("they") in Sing Down the Moon (Book #2). These additional differences between "we" and "they" Native American subjects are, however, beyond the scope of this study. If, however, there is interest in looking at the strictly compatible ethnic comparisons only, this is accomplished simply by ignoring the first and last groups of data presented in Table 30 in Appendix G. The other four books, with ethnic comparisons, were published in 1970 (Sing Down the Moon (Book #2)), 1974 (The Spirit is Willing (Book #3)), 1978 (Squaw Man's Son (Book #4)), and 1980 (Sarah Bishop (Book #5)). Therefore, books number one, Akavak, an Eskimo Journey, and number two, Sing Down the Moon, will collapse into Time 1 ; books number three, The Spirit is Willing, and number four, Squaw Man's Son, will collapse into Time 2; and books number five, Sarah Bishop, and number six, Where the Buffaloes Begin, will collapse into Time 3. Before analyzing these chronological trends, however, 143 it is of interest to look at patterns across books. It will be recalled that it is predicted that books published between 1960 and 1970 will be more evaluative than books published between 1971 and 1982. Furthermore, it is predicted that books published after 1975 will be more descriptive than books published between 1960 and 1974. In this study there are two books published between 1960 and 1970 (Akavak, an Eskimo Journey (Book #1) and Sing Down the Moon (Book #2)), and four books published between 1971 and 1982 (The Spirit is Willing (Book #3), Squaw Man's Son (Book #4), Sarah Bishop (Book #5), and Where the Buffaloes Begin (Book #6)). In taking the mean scores (see Appendix G), patterns over these time periods can be analyzed by the categories Native American/"they", Non-Native American/"we", and the total book. In doing so, the following patterns are found for Native American subjects: (1) the percentage of positive evaluations decreases slightly; (2) the percentage of negative evaluations decreases greatly, but still remains high at ten percent for books published between 1971 and 1982; (3) the percentage of neutral descriptions greatly increases, but is still less than predicted (i.e., >90%); (4) the Coefficient of Evaluation barely increases and remains substantively significantly negative; and (5) the Evaluativeness Score clearly decreases, but remains greater than expected (12% > 10%). Conversely, it is found for Non-Native American 1 44 subjects and the book as a whole that : (1) the percentage of positive evaluations remains the same; (2) the percentage of negative evaluations are greatly reduced; (3) the percentage of neutral descriptions increases, but is still slightly less than the predicted ninety-five percent; (4) the Coefficient of Evaluation for Non-Native Americans significantly increases, although it is still ambivalent in the negative direction, while for the total book the score only slightly increases; and (5) the Evaluativeness Scores are more than cut in half. Analysis of books published after 1975 (see Appendix G) found very little change from these results. Furthermore, the results of this analysis of books published after 1975 refute the propositions that books would be more descriptive (i.e., >95%) and, conversely, less evaluative (i.e., <5%). Only the portrayal of Non-Native Americans comes close with ninety-four percent of the portrayals being descriptive. Generally, then, there are very few changes observed in portrayals across books within nominal time periods. In consideration of these minimal across book findings, it now remains to see if there are any patterns or trends over time. Table 27, then, presents the mean percentages for positive, neutral, and negative terms, Coefficient of Evaluation scores, and Evaluativeness Scores for Native American/"they" subjects, Non-Native American/"we" subjects, and the total books for each Table 27 Mean Scores Over Time Mean Term Direction Time 1: 1968-1970 Time 2: 1974-1978 Time 3: 1980-1981 NA1/They NNA2/We Books NA NNA Books NA/They NNA/We Books % % % % % % % % % 3 3 3 2 3 3 2 3 3 66 85 86 88 94 92 88 94 91 31 12 11 9 4 5 10 3 6 Mean Coefficient 19 21 24 23 40 33 24 47 34 of Evaluation Mean Evaluativeness 34 15 15 12 12 Score 1—NA stands for Native Americans 2—NNA stands for Non-Native Americans 146 of the three designated time periods. To facilitate interpretation of Table 27 scores, the evaluative and descriptive word frequencies are graphed separately in Figure 1 for Native American, Non-Native American, and total book portrayals over the three time periods as represented by the books of this study. Figure 1, then, visually expresses (1) the trends in use of evaluative and descriptive words to portray Native American/"they" and Non-Native American/"we" subjects in books, and (2) the trends to evaluate and/or describe subjects by books in general. It is seen from Table 27 and Figure 1 that the results provide some interesting trends. First of all, it is seen that portrayals of Native Americans and Non-Native Americans and the books themselves are more descriptive than evaluative. Related to this observation it is found that between Time 1 and Time 2 descriptive portrayals all increased, with the greatest increase for Native Americans and the least increase for the book as a whole. Conversely, between Time 2 and Time 3 descriptive portrayals for Native Americans and Non-Native Americans remained constant to the same, while for books in general descriptive portrayals became more evaluative between Time 2 and Time 3. Furthermore, it is observed that portrayals of Native Americans are less descriptive, or more evaluative, than Non-Native Americans or the books in general. More importantly, however, are comparisons with the 147 Native Non-Native Americans Americans Total Books •100r (0) (0) (0) 80 60 40- Figure 1 Evaluations: (+)(0)(-) T = Time 1 (1968-1970); T2 = Time 2 (1974-1978; T3 = Time 3 (1980-1981) 148 expected scores for descriptive or neutral portrayals. In Time 1 all categories are observed to be less than the expected descriptive portrayal frequency, although Non-Native Americans and books are within five percent. At Time 2, while Non-Native Americans and books are observed to be more descriptive than expected, Native Americans remain more evaluative than expected. As a function of increased expectations and the general absence of changes in descriptive portrayals, it is found that portrayals for all three categories at Time 3 are less descriptive than expected, with the greatest difference observed for Native Americans and the least for Non-Native Americans. A last note on neutral evaluations or descriptions is that between Time 1 and Time 2 portrayals of Native Americans, relative to Non-Native Americans and books, have greatly improved, but are still more evaluative than expected. In looking at the results for positive evaluative portrayals in Table 27 and Figure 1, the clearest trend is that such portrayals for Non-Native Americans and books remain constant over all three time periods and are both fixed at three percent. This substan tively significant result becomes more important when it is observed that the percentage of positive evaluations of Native Americans is equal to those of Non-Native Americans and books at Time 1. Conversely, at Time 2 and Time 3 the percentage of positive portrayals for Native Americans is slightly less than for Non- Native Americans and books. 149 In analyzing the results for negative evaluations in Table 27 and Figure 1, several substantively significant results are found. Most interesting is the observation that negative evaluations are generally inverse functions of the neutral descriptions. This is substantively significant because both types of evaluative terms are expected to be a function of neutral or descriptive terms, not just negative (or positive) terms. Another clear result is that between Time 1 and Time 2 negative evaluations generally decreased, with the greatest decrease for Native Americans and the least decrease for books. Very significant to this study, however, is the finding that between Time 2 and Time 3 negative evaluations of Native Americans and books increased, while negative evaluations of Non-Native Americans continued to decrease so that the percentage of postive and negative evaluations at Time 3 are equivalent. Comparisons of the results to the expected results indicate that at Time 1 all categories are more negatively evaluative than expected, but that the difference for Native Americans is very significant. At Time 2 Native Americans are just slightly less negatively portrayed than expected, while for Non-Native Americans and books the percentage of negative evaluations is considerably lower than expected. This trend toward improvement, however, reverses at Time 3 for Native Americans and books to become more negatively 150 evaluative than expected, while for Non-Native Americans the trend toward less negative evaluations continues and is less than expected. Furthermore, it is seen that negative portrayals for Native Americans are more than twice as evaluative as for Non-Native Americans and books across all three time periods. Figure 2 graphically portrays the mean Coefficient of Evaluation scores from Table 27 for Native Americans, Non-Native Americans, and books in each time period. The most obvious result found is that no score at any time is over fifty percent, which means that all scores are in the negative direction. Even more importantly this means that no portrayals, least of all for Native Americans, are greater than seventy-five percent positively evaluated as predicted in the proposals for this study. Another very clear finding is that Native Americans, Non-Native Americans, and books are all substantively significantly negatively evaluated at Time 1, with Native Americans the most negatively evaluated, books the least negatively evaluated, and Non-Native Americans in between. Between Time 1 and Time 2 marked improvement is noted for Non-Native Americans, who become the least evaluative, and for books. Such increases reflect greater ambivalence for these two categories, particularly for Non-Native Americans. Conversely, for Native Americans between Time 1 and Time 2 the Coefficient of Evaluation increased a few percentages, but remains significantly 151 1 00- Positive Evaluations 80- 60- Ambivalent Non-Native Evaluations Americans 40- Total Books Native Americans Negative Evaluations T T T 1 2 3 Figure 2 Coefficient of Evaluation Over Time T = Time 1 (1968-1970); T? = Time 2 (1974-1978); T3 = Time 3 (1980-1981) 152 lower than expected. Between Time 2 and Time 3 similar trends and results are observed. Additionally, it is seen that over time the Coefficient of Evaluation increases the greatest percentage for Non-Native Americans and the least for Native Americans. It is interesting to note that between Time 2 and Time 3 the Coefficient of Evaluation levels off for Native Americans and books, but continues a steady increase for Non-Native Americans. Similarly, it is observed that over time the Coefficient of Evaluation for books tends to be more positive, but parallels the portrayals of Native Americans. The portrayal of Non-Native Americans, therefore, is seen to continually improve substantively more than for Native Americans and books. Conversely, the portrayal of Native Americans is seen to remain substantively significantly negative, while Non-Native Americans and books increasingly become more ambivalent, over the time periods of this study: 1968-1970, 1974-1978, and 1980-1981. Lastly, Figure 3 presents the mean Evaluativeness Scores from Table 27 for Native Americans, Non-Native Americans, and books for each time period. , The most obvious trend is that the evalu ativeness of all portrayals improved between Time 1 and Time 2. The greatest decrease in evaluativeness is for Native Americans with the percentage of evaluativeness at Time 2 being about two-thirds of what it was at Time 1. Substantively significantly more important 153 50 40 30 20 Native Americans 10- Total Books Non-Native Americans "1 Figure 3 Evaluativeness Scores Over Time T = Time 1 (1968-1970); T2 = Time 2 (1974-1978); T3 = Time 3 (1980-1981) 154 is the observation that at Time 2 portrayals of Non-Native Americans and books are less evaluative than expected (6% < 10%; 8% < 10%), as proposed by this study, while the portrayal of Native Americans remains more evaluative than proposed (12% > 10%). Between Time 2 and Time 3 a very interesting trend occurs in that the evaluativeness of Native Americans stabilizes at twelve percent while Non-Native Americans and books unexplainably become slightly more evaluative. Furthermore, because of this leveling off, all three are more evaluative at Time 3 than proposed (i.e., >5%), with portrayals of Native Americans remaining almost twice as evaluative as Non- Native Americans. Table 27 and Figures 1, 2, and 3 clearly demonstrate that portrayals of Native Americans are, and remain, more negative and evaluative than for Non-Native Americans and books in general. Second, the results clearly show that such negative and evaluative portrayals of Native Americans remain consistently greater than would be expected over time. The results presented over time demonstrate that between Time 1 (1968-1970) and Time 2 (1974-1978) children's fiction made strides toward improving books. This is observed from the general patterns toward more descriptive portrayals overall, and more ambivalent evaluative portrayals, although the patterns are much less so for Native Americans than for Non-Native Americans and book in general. Conversely, between 155 Time 2 (1974-1978) and Time 3 (1980-1981) a general leveling off is observed, except for the Coefficient of Evaluation for Non- Native Americans that continues to steadily improve. This is to say, then, that at best, portrayals of subjects have remained the same since 1978, particularly for Native Americans. It can be inferred, however, that overall there has been a regression tendency, i.e., portrayals have gotten worse since 1978. Regardless of the position, the fact is that there has been no improvement per se since 1978, particularly in the already highly negative and overly evaluative portrayals of Native Americans in children's fiction 11 as inferred from this study's results. Clearly, then, the data results indicate why some studies infer positive change in the portrayals of Native Americans and others find consistency or continuing negative portrayals. Studies analyzing just trends through changes in frequency counts, will measure the definite improvements observed over time, but then fail to recognize that despite this improvement the portrayals remain more negative and evaluative in comparison to Non-Native Americans and books as a whole (or to the proposed criteria or predicted standards). Conversely, studies utilizing randomly selected categories will measure the negativeness and/or evaluativeness of portrayals of Native Americans, which are then compared to a proposed standard, and thereby find that portrayals of Native Americans remain 156 consistently biased because they fail to measure relative changes. As such, therefore, it is of interest to compare the qualitative findings to these quantitative results. Comparison of Quantitative Results with Quantitative Results Since the quantitative findings and quantitative results have already been presented above, the results of the semi quantitative method will be presented first. The primary measurement, and the only one of interest here, is the thematic frequency that is part of the ECO Administrative Data Instrument. Measurement of themes proved to be a difficult task as each book tended to incor porate slightly different themes. Thus, frequencies are very rough and are considered estimates only. The presence or absence of themes, however, at this point is more important. A complete list of themes observed in each book, along with rough estimates of their frequency of occurrence, is provided in Appendix H. (It is noted that due to library loan constraints there was not enough time to measure either themes or other categories from the ECO Administrative Data Instrument for Akavak, an Eskimo Journey.) The most obvious finding of this data measurement is the decline of theme appearance and frequency for both the 1980 and 1981 books (Sarah Bishop (Book #5) and Where the Buffaloes Begin (Book #6)). The most persistent themes are the "Noble Savage" and "Indians and 157 the Land." The themes with the greatest frequency rates within books, however, are: (1) "Indian Warriors;" (2) "Chiefs;" and (3) "Indian-White Relations." It is also found that measuring themes produces a large number of categories peculiar to each separate book, but with no category through which all books can _ be compared. This means that the thematic measurement—at least for small samples—is generally inappropriate. The qualitative and quantitative measures can now be compared, having presented the very minimal findings of the qualitative measures. First of all, it is found that the qualitative measurements usually yield negative critiques of the books. The exception is Sarah Bishop (Book #5) which is perceived to be manifestly a descriptive book with basically balanced or ambivalent evaluative portrayals, a fact that substantiates the quantitative results. The thematic measurement, however, demonstrates that both of these methods miss several very latent but old themes in the book Sarah Bishop (Book #5): (1) "Indians as the White Man's Helper" and (2) "Indians and the Land." Conversely, the qualitative and thematic analyses measure the portrayal of Native Americans in Sing Down the Moon (Book #2) as being latently negative through subtle impressions and themes, while the quantitative results indicate an ambivalent portrayal of Native Americans and a more negative portrayal of Non-Native 158 Americans. Similarly, the qualitative analysis of Akavak, an Eskimo Journey (Book #1) measures the portrayal of Native Americans to be quite good, while the quantitative ECO Analysis demonstrates significant levels of biased or judgmental evaluations of both "we" and "they" Native American subjects. The qualitative thematic analyses of Squaw Man's Son (Book #3) indicate that the book is significantly evaluative. Themes such as "Indian Warriors," "Chief," "War Dances/Councils," "Intertribal Warfare," and "Smelly Indians" appear with high frequency, while impressionistic techniques simplify, obliterate, and generally malign the portrayals of Native American subjects in this book, Squaw Man's Son (Book #3). Yet the ECO Analysis measures the composite portrayal as being ambivalent, although more negative than for Non-Native Americans in the same book. The qualitative measurement of The Spirit is Willing (Book #2) finds the book to be significantly negative in its portrayal of Native Americans and ambivalent to positive in the evaluative portrayals of Non-Native Americans. The quantitative and thematic analyses, as with other books, verify or collaborate this finding. Similarly, the qualitative analysis of Where the Buffaloes Begin (Book #6) perceives significant differences between the portrayal of the "Assiniboines" ("they") and that of the main character "Little Wolf" and his "village" ("we"). These differences are also found through the quantitative ECO Analysis of judgmental terms. The thematic analysis, however results in low frequencies of themes, although the very standard themes of "Noble Savage," "Warriors," "Indians and the Land," "Intertribal Warfare," and "Sneaky/Animal-like are all present in the story. Generally, then, each method of analysis results in a variety of findings that sometimes concur with each other, and other times do not, but each finds judgmental portrayals that the others fail to measure. Chapter 4 CONCLUSIONS The purpose of this study is to analyze value judgments about Native Americans made by sources—authors and publishers— in children's fiction over time. As such, the study analyzes how Native Americans and Non-Native Americans are portrayed in fiction through the expanded use of a known method to measure evaluative and descriptive words. Specifically, based upon previous research and theoretical constructs, the study expects the following results in response to questions proposed in the introduction (p. 40): 1 . That books published in the 1960s will be more evaluative than books published in the 1970s and 1980s. 2. That the evaluativeness of the books, based upon the sample average score, will be less than ten percent (<10%). 3. That the percentage of descriptive words for all subjects, groups, and books will be greater than ninety percent (>90%). 4. That books published after 1975 will have improved because of outside demands and, therefore, will have a descriptive percentage greater than ninety-five percent (>95%). 5. That books published after 1975, similarly, will have an even lower Evaluativeness Score of less than five percent (<5%). 6. That evaluative judgments, based upon other current research, will be favorable (i.e., >75%) toward Native Americans. 160 161 7. That judgments of Native Americans will be less favorable than those of others. 8. That the percentage of evaluative terms to total space devoted to Native Americans will be greater than that for Non-Native Americans. 9. That the percentage of negative evaluative judgments about Native Americans will be greater than five percent (>5%), while the negative evaluations for Non-Native Americans will be less than five percent (<5%). 10. That similar themes concerning Native Americans, found by Garcia in textbooks, will be found in fiction books. In looking at the results in Figures 1, 2, and 3, and Table 27 in chapter three, it is seen that books published between 1960 and 1970 are, as predicted, more evaluative than books published between 1971 and 1982. In looking at Table 30 in Appendix G, however, it is found that the more recent books, The Spirit is Willing (1974) and Where the Buffaloes Begin (1982), are both separately more evaluative than the book Sing Down the Moon, which was published in 1970. Similarly, the book Squaw Man's Son (1978) is seen to be as equally evaluative as Sing Down the Moon. A second related proposition is that the results will indicate that the Evaluativeness Scores will be less than ten percent (<10%). In looking back at Figure 3 and Table 27, and at Table 30 in Appendix G, it is observed that the expected results are not found. It is seen from Table 30 that the books Akavak, an Eskimo Journey (Book #1) and Where the Buffaloes Begin (Book #6) are both more evaluative than expected; i.e., book number one, 162 Akavak, an Eskimo Journey, is more than twice as evaluative as expected (22% > 10%), while book number six, Where the Buffaloes Begin, is substantively more evaluative than predicted (14% > 10%). It is further observed that both of these highly evaluative books involve dichotomies rather than ethnic comparisons, which means that no Non-Native Americans are involved. It is inferred, therefore, that books involving only Native American subjects are more evalu ative than books including other subjects. Book number six, Where the Buffaloes Begin, is a good example where a very positively- oriented ambivalent Native American subject is the protagonist, while a very negatively portrayed Native American group subject is the antagonist. The "they" group can thereby be justifiably negatively judged because it is Native Americans talking about and confronting Native Americans. An illusion of a positive portrayal is made by the extensive scope of the "we" subject while the negative portrayal of Native Americans slips unconsciously by, but is in reality latently perceived. The latent, covert portrayal of the subject "they" is largely ignored, then, due to the substantially positive, manifest, and overt portrayal of other Native American subjects as "we." This style thereby allows the source to perpetuate significantly negative portrayals of Native Americans and negative to ambivalent portrayals by the illusion of positive portrayals through varied levels of negative portrayals. 163 On the reverse side of these expected results is the proposition that the percentage or proportion of descriptive words for each subject and book will ;be greater than ninety percent (>90%); and furthermore, it is predicted that for books published after 1975 descriptive words will account for more than ninety-five percent (>95%) of the words used to portray a subject. This proposition does not hold up, which can be seen in Table 27 and Figure 1 in chapter three and Table 30 in Appendix G. Native Americans are descriptively portrayed less than ninety percent (<90%) in book number three, The Spirit is Willing (93.94% < 90%), and less than ninety-five percent (<95%) in book number four, Squaw Man's Son (92.59% < 95%). Conversely, Non-Native Americans are descriptively portrayed less than ninety percent (<90%) in only book number two, Sing Down the Moon (88.80% < 90%). It additionally is seen that both books with a we-they dichotomy (Akavak, an Eskimo Journey (Book #1) and Where the Buffaloes Begin (Book #6)) have, for all groups, descriptive portrayals that are less than ninety percent (<90%) (78.36% < 90% and 85.77% < 90%, respectively). Clearly, then, the only book with descriptive portrayals as predicted is book number five, Sarah Bishop, although it is noted that several descriptive portrayals are close to the predicted levels. Similarly, it is found from Table 30 in Appendix G and Figure 3 and Table 27 from chapter three, that two of the three books 1 64 published after 1975 did not have lower Evaluativeness Scores as predicted. Only book number five, Sarah Bishop, has an Evaluativeness Score less than five percent (<5%) (3% < 5%). The prediction of increased descriptive or neutral portrayals of Native Americans, and books with such portrayals, is again based upon purported improvement over the past ten to fifteen years by sources in portraying Native Americans. While Figure 3 suggests that progress has been made, it also indicates that a marked reversal is occurring, that books and portrayals are again becoming significantly evaluative. A fifth proposition predicted finding that evaluative judgments are favorable (i.e., >75%) toward Native Americans. Subsumed in this is the prediction that there is a tendency toward more favorable portrayals of Native Americans. Based upon the calculated Coefficient of Evaluation scores, as found in Table 30 in Appendix G and Figure 2 and Table 27 in chapter three, it is seen that no score is greater than seventy-five percent. This means that no portrayals of Native Americans, as well Non-Native Americans, are significantly positive, nor is there any trend toward positive portrayals for Native Americans. More important is the tendency of portrayals of Native Americans to fluctuate with the times or current events. Negative portrayed in the 1960s, awareness of such portrayals resulted in ambivalent, but in the negative direction, 165 portrayals by 1970. Paralleling contemporary conflict images, portrayals are again found to be negative in 1974. Between 1974 and 1980, increasing efforts for humanistic treatment are reflected in marginally improved portrayals. The new wave of conservativism of the 1980s, however, is seen to coincide with increased negative portrayals of Native Americans. Conversely, it is found that the portrayals of Non-Native Americans, which also have been consistently in the negative direction, do demonstrate a tendency toward more favorable portrayals. It is clearly observed, then, that judgments in portrayals of Native Americans are less favorable than for Non-Native Americans as predicted, except for book number two, Sing Down the Moon. Sing Down the Moon, published in 1970, may be inferred to be indicative of the times where Non-Native Americans were portrayed more negatively, while Native Americans were ambivalently portrayed. This more negative portrayal of Non-Native Americans thereby creates an impression that Native Americans are positively portrayed, when in fact the portrayal is still very much in the negative direction. Similarly, by looking at the Evaluativeness Scores for Native Americans and Non-Native Americans in Figure 3 and Table 27 in chapter three, it is observed that the percentages of evaluative to descriptive terms for Native Americans are indeed greater than for Non-Native Americans as predicted, except again for book number two, Sing Down the Moon, as observed in Table 30 (Appendix G). : > 166 An eighth proposition predicted finding that the percentage of negative evaluative judgments about Native Americans are greater than five percent (>5%), while the negative evaluation percentage for Non-Native Americans are less than five percent (<5%). The data results presented in Table 25 in chapter three, reveal that the observed percentage of negative evaluations for Native Americans is 5.29%, which is greater than five percent as predicted. The percentage for Non-Native Americans is found to be 3.04%, which is less than five percent as also predicted. Furthermore, and of note worthy importance, it is noted that these cells in Table 25 are both independently statistically significant. This indicates that portrayals of Native Americans are statistically more negative and evaluative than would be expected or predicted. From these findings, then, it is inferred that sources employ a style of writing whereby they: (1) portray Non-Native Americans more descriptively than Native Americans; (2) counter this with more evaluative terms about Native Americans than expected using both positive and negative terms; and (3) counter the first two stylistic methods by portraying Non-Native Americans with fewer than expected evaluative terms, especially negative terms. Thus, the subjects, which are statistically descriptive, are actually evaluatively portrayed within the guise of such descriptive portrayals. That is, within the statistically allowable range of evaluative terms, portrayals 167 of Native Americans are made statistically negative in comparison to Non-Native Americans, but appear to be neutral or ambivalent because of the averaging effect in the scope of the descriptive portrayal. In fact, the depth of the already negative evaluation (i.e., >5%) is compounded by the less than expected depth of the negative portrayals of Non-Native Americans, which thereby increases the effectiveness of the negative portrayal of Native Americans. Overtly, then, the portrayal of Native Americans is manifestly descriptive, but covertly the portrayal is manifestly negative. The last proposition predicted finding that similar themes concerning Native Americans, found by Garcia in textbooks, would be found in the fiction books analyzed in this study. While in no way an accurate picture of thematic frequencies, it is found that the standard themes used by Garcia are applicable, but not compre hensive enough, to the books in this study. Such a finding, however, is almost inherent to the books because all the stories are set in the eighteenth century or pre-Non-Native American contact period. Thus, it is difficult to avoid many of these themes—most of which imply negative portrayals—but some books are better than others. Books number one, Akavak, an Eskimo Journey, and number six, Where the Buffaloes Begin, which have no Non-Native American subjects, are both relatively low in thematic frequency but high in evaluative assertions frequency. Book number two, Sing Down the Moon, which 168 portrays Native Americans more positively than Non-Native Americans in terms of words, contains numerous negative themes. Despite these variances in uses of themes, no one theme is manifestly found in all of the books sampled. However, the "Noble Savage" theme appears in four of the books, as does the "Indians and the Land" theme. Both themes are inherent or latent to the other two books, but are not manifestly apparent. Other themes are also found through qualitative analysis to be inherent to the stories, but are immeasurable with the quantitative or semi-quantitative thematic methods available. Pratt's ECO Analysis methodology is found to be adaptable to the analysis of fiction books. The major difficulty with the technique of measurement is one found in all content analyses using words as the unit of measurement, and that is the problem of reliability. Many of the words encountered as part of the portrayals are not on Pratt's ECO Word List (see Appendix F), nor could synonyms always be used. Pratt's procedures were adhered to rigidly, but there still exists a degree of subjectivity to the scoring of words.. The sampling procedures used within the books proved to be functional but time consuming. It required an average of thirty-plus hours to measure the portrayals in each book, and initial reading and tracking of subjects required additional time. While the sample size may be statistically 169 questioned, it is evident that without assistant researchers, larger samples using the ECO Analysis would be unfeasible. Additionally, it is felt that while ECO Analysis eventually resulted in useful data, other methods and techniques are needed to analyze the portrayals of Native Americans and other subjects. The researcher has previously demonstrated that qualitative content analysis methodology critically is needed for measuring portrayals of Native Americans (Edwards, 1981). New content analysis methodology is needed to measure impressionistic styles of sources and to measure more structured aspects of content such as the dimensions of propa ganda and prejudice manifested in the content. This study, then, accomplishes the objectives of (1) applying ECO Analysis, previously used with textbooks, to children's fiction, (2) analyzing value judgments about Native Americans portrayed in fiction, and (3) determining how portrayals of Native Americans are made. It has been found that books became increasingly more descriptive until 1981, at which time the trend seemed to have reversed itself to perpetuating evaluative portrayals. Furthermore, it is found that while portrayals of Native Americans overtly and manifestly appear to be descriptive, they are in fact covertly evaluative and, as such, are significantly negative. It also has been found that portrayals of both Native Americans and Non- Native Americans are relative in that the Coefficient of Evaluation 170 scores are less than ambivalent; that the difference is in the degree of negativeness in portrayals. Additionally, this study shows that portrayals of Native Americans, in contrast to Non- Native Americans and the total books, are very irregular, which would account for the different findings concerning portrayals of Native Americans in literature. While trends toward improved portrayals are noted, it continues to be observed that such portrayals continue to be in the negative direction. Lastly, it- has been found that books on Native Americans only, are generally far more evaluative; and, more specifically, the "they" subjects are consistently, significantly, and negatively portrayed. That is, portrayals of Native Americans confronting other Native Americans provide sources freedom to develop overtly and manifestly negative portrayals of Native Americans through the averaging method of overtly and manifestly portraying other Native Americans as "we" groups in a more favorable fashion, but still in the negative direction. Similarly, negative portrayals are produced by sources through subtle illusions of negative themes, and thus, overwhelming the power of illusionary descriptive portrayals. In sum, portrayals of Native Americans tend to be more evaluative, more negative, and more judgmental than for Non-Native Americans. Through stylistic methods, reinforced by perpetuated themes, portrayals of Native Americans are found to be about the same in 1981 as in 1968, only more subtle, more latent, and harder to manifestly perceive. 171 In summary, this study found, in answer to the questions posed in chapter one, that for children's fiction: 1. Portrayals of Native Americans are generally evaluative while the portrayals of Non-Native Americans are generally neutral or descriptive. 2. Judgments made by sources are unfavorable toward Native Americans and are significantly negatively evaluative. 3. Generally, within books, evaluative judgments of Native American subjects are more negative than for Non-Native American subjects; while between books, evaluative portrayals are statistically significantly more negative for Native Americans than for Non-Native Americans, whose portrayals are statistically less negative than expected. 4. The frequency of evaluative terms to the total space devoted, or the Evaluativeness Scores, for Native Americans in each time period are thirty-four percent, fifteen percent, and fifteen percent, whereas for Non-Native Americans the Evaluativeness Scores for each time period are twelve percent, six percent, and eight percent. 5. The frequency of specific terms applied to portray a subject is indicated by the ratio of positive to negative terms. For Native Americans the ratio indicates that for every forty positive terms used there are one hundred negative terms used (.4018), whereas for Non-Native Americans sixty-two positive terms are used for every one hundred negative terms (.6201). This means that more negative terms are used than positive terms to portray all subjects, but more so for Native Americans. 6. Sources generally became less evaluative between Time 1 (1968-1970) and Time 2 (1974-1978), but remained constant for their portrayals of Native Americans while portrayals of Non-Native Americans and the books as a whole became more evaluative between Time 2 (1974-1978) and Time 3 (1980-1981). 7. Theme appearance and frequency is seen to decline over time. Persistent themes are the "Noble Savage" and "Indians and the Land," while the themes with the greatest frequency rates when they appear in books are "Indian Warriors," "Chiefs," and "Indian-White Relations." 172 Implications The results and findings of this study pose theoretical, methodological, and practical implications. Theoretically, this study indicates that children's fictional literature involving Native Americans as subjects clearly incorporates propaganda and other persuasive techniques and includes notions of prejudice. As such, the results and findings of this study need to be further studied to determine their accuracy. After more data is available, a theory can then emerge to describe and predict the effects of reading materials about Native Americans. More importantly, this study has shown the relevancy of theoretically based studies of literature, which thereby implicates the need for additional studies both in children's fiction and in other categories of literature. This study also indicates the need for greater methodological explorations in content analysis and the study of literature/media contents. This study reinforces the researcher's belief that the contents of literature affect the reader through covert, latent, and unconscious methods. Quantitative analyses are generally limited to measuring manifest portrayals while allowing some inferences into the latent content and/or effects. Thus, this study clearly indicates the need for additional qualitative methods for analyzing literature. Furthermore, it clearly points out the need to renew propaganda analysis and values analysis, as well as 173 the need to develop a formal method for analyzing media content for dimensions of prejudice or attitudes of prejudice. Additionally, this study makes clear the need to computerize ECO Analysis, to expand the ECO Word List, and to require multiple recorders or data collectors. Lastly, while the sampling procedures for this study proved to be sound, further procedures for using ECO Analysis with nonfiction literature is desperately needed because of the dearth of studies in the critical field of nonfiction. A final consideration is in the area of practical implications. Foremost is the need for an expanded ECO Word List to improve reliability. Secondly, there is a need for additional studies, but with larger sample sizes and more analysts. Perhaps most urgent is the need for further study of recent books to confirm the observed trends concerning the portrayals of Native Americans. Indeed, if the findings of this study are valid and reliable, then such needs to be both publically announced and challenged, because the current system of literature is institutionally perpetuating prejudiced attitudes by various means. While literature may not be the cause of such attitudes, this study shows that it certainly reinforces negative, highly evaluative, and often standardized portrayals of Native Americans through the use of standard themes and/or evaluative judgments about Native Americans. ENDNOTES 175 1 The term "American Indian" probably has more definitions than any other proper nomenclature, yet no other generic term for a group of peoples evokes such commonly held ideas. While the term "Native Americans" is not an adequate appellation either, it is less evaluative and less ambiguous in the sense that it is usually applied to two or more different cultures or societies rather than to individuals or one particular group. As such, the term "Native Americans" will be used descriptively like the term "European" for all groups and not for specific groups or individuals. Conversely, the term "Indian" will be used as an evaluative term that is ambiguously applied. (Also see footnote A for additional discussion of the use of the term "Indian"). 2 The ideas for this discussion are taken from Ritzer (1980), pp. 89-91. 3 The following definitions of awareness and perception are implicit to the discussion, but are explicitly stated here for continuity: Awareness: Having or showing realization, perception, or knowledge. Perception: The process through which organisms understand their environment by means of their senses. Perception is a process, then, to gain awareness. Conceptualization is the "minding" process between perception and awareness. Therefore, failure to perceive or notice a stimulus or something in the environ ment will disallow perception and awareness. Faulty perception may or may not be altered by the "minding" or conceptualization process. Regardless, the concept and vis-a-vis the perception is expressed through awareness, whether the awareness is valid or not. In comparing these two concepts to the communications model, perception may be seen as the receiver and awareness as the process of encoding and transmitting. 4 It should be pointed out that the research utilizes a more stringent or narrow definition for some descriptive words. In particular, anthropological terms are scored generally as evalu- atively because of their ability to make value judgments concerning the subject. Terms like "tribe," "band," "warrior," "chief," and so forth are scored as evaluative terms. Furthermore, the analyst negatively score the term "Indian" if it was used inappropriately. That is, the term "Indian" is not conceived of as being inherently negative, but is used generally in context to simplify, belittle, 176 generalize, and evaluate a subject. If the subject category is "Indians," then the term is treated as neutral or descriptive, whereas the term is scored negative when used to mean a known subject, i.e. Cap tain Jack." The scoring of anthropological words and similar terms is justifiable because the use of such terms serves to time-lock or freeze Native Americans into being solely people of the past and/or people inherently less than Non-Native Americans. The notion of inequality derives from the fact that anthropological terms also generally imply notions of social evolution, to wit tribes are politically less developed and therefore "inferior" somehow. These implications, thereby, latently load such words as judgmental or evaluative. 5In doing the analysis of the data and rereading Pratt's material (1971, 1972), problems concerning Pratt's second analytical technique, the Index of Evaluativeness, were found. First of all, his discussion of the Index of Evaluativeness was found to be conflicting. Pratt begins, "If the total number of lines or words of text devoted to a subject is divided by the number of evaluative terms applied to that subject," but then concludes "... the analyst can determine a score indicative of the expressed frequency with which value judgments are expressed about the subject relative to the amount of space devoted." (1972:25-26). The ratio Pratt proposed is: Total Words , „ „ , . IndexT of Evaluativenessn Total Evaluative Words Conversely, the ratio he actually intends in this conclusion is: Total Evaluative Words = Ratio of Evaluativeness Total Words This later statistic, then, was labeled the "Ratio of Evaluativeness" by the research. Further consideration, however, indicated that such an analytic score would be more useful if expressed as a percentage, which was labeled the "Evaluativeness Score": 100 (Total Evaluative Words) = Evaluatl>eness Score Total Words ^It is noted that a full discussion of these aspects is beyond the scope of this thesis. That is to say, examples of inaccuracies, omissions, and so forth will not be made. 177 7The edition used in this analysis did not have a cover or back page. g It should be noted that: The publisher has made every effort to determine the copyright of the original version of this story, which appeared in Volume XLII, No. 4, of St. Nicholas Magazine in February, 1915. The search has been to no avail (Baker, 1981:unp). It is furthermore noted that the copyright is actually listed under the publisher: Text Copyright © 1981 by Frederick Warne and Company, Inc. Illustrations Copyright © 1981 by Stephen Gummell (Baker, 1981:unp). This indicates, therefore, that the source of this book may in fact be considered to be Frederick Warne and Company, Incorporated, rather than Olaf Baker, who is the author in name only. q While the book is listed in many guides as fiction, it is not fiction in the literary sense, and is most appropriately catalogued—at the Missoula County Library at least—under the Dewey Decimal classification of 398.2 of the social science non- fiction section. 10 In reading the book the researcher found the reading level to be actually at the eighth grade level, rather than at the second to fourth grade level. 11 It is noted that book number five, Sarah Bishop, must be considered, at this point, an anomoly to the general pattern. For a graphic presentation of book differences see Appendix G. REFERENCES CITED 179 Abelson, Herbert I. 1959 Persuasion: How Opinions and Attitudes are Changed. New York: Springer Publishing Company, Inc. Agogino, George 1950 "The stereotype of the American Indian." Albuquerque: University of New Mexico, unpublished Master's thesis. Aitson, Richard 1980 "Some observations on contemporary Native American liter ature," Four Winds 1, 3(Summer):28-31, 80. Albig, William 1939 Public Opinion. New York: McGraw-Hill Book Company, Inc. Albrecht, M.C. 1956 "Does literature reflect common value?" American Socio logical Review 21:272-279> Allport, Gordon 1954 The Nature of Prejudice. New York: Addison-Wesley Publishing Company, Inc. 1968 "The historical background of modern social psychology," in G. Lindzey and E. Aronson (eds.), Handbook of Social Psychology, Second Edition (Boston: Addison-Wesley Publishing Company, Inc.), Vol. 1, pp. 1-8. Allport, Gordon and Leo Postman 1946 "An analysis of rumor," Public Opinion Quarterly 10, 2 (Winter):501-517. 1948 The Psychology of Rumor. New York: Henry Holt and (1947) Company. American Educator 1983 "The propaganda war," American Educator 7, 1(Spring):12-15. American Library Association 1905- The Booklist and Subscription Books Bulletin. Chicago: American Library Association. n.d. Library Journal. New York: R.R. Bowker Company. Andersen, Kenneth E. 1971 Persuasion: Theory and Practice. Boston: Allyn and Bacon, Inc. 180 Appelbaum, Ronald L. and Karl W.E. Anatol 1974 Strategies for Persuasive Communication. Columbus, OH: Charles E. Merrill Publishing Company. Asman, David 1983 "The hidden agenda," American Educator 7, 1(Spring):16-19, 31. Bachtold, Louise M., Mitzi Tanka, and Cynthia White 1977 "Changing perspectives in the study of children: comparison between research and popular literature," Psychological Reports 41:307-318. Bailey, Kenneth D. 1978 Methods of Social Research. New York: The Free Press. Bain, Reed 1939 "Verbal stereotypes and social control," Sociology and Social Research 23(May-June):431 -446. Baker, Donald G. 1974 "Identity, power and psycho-cultural needs: white responses to non-whites," Journal of Ethnic Studies 1:16-44. Banfield, Beryle 1975 "The more things change," Elementary English 52, 5(May): 715-716. Banks, J.A. 1969 "A content analysis of the black in textbooks," Social Education 33:954-957, 963. Barchs, F.F. 1959 "Communications content: analysis of the research, 1900-1958." University of Illinois, unpublished Ph.D. dissertation. Barghoorn, Frederick G. 1964 Soviet Foreign Propaganda. Princeton, NJ: Princeton University Press. Barron, Pamela Petrick 1981 "The characterization of Native Americans in children's and young adult's fiction, with a contemporary setting by Native American and Non-Native American authors: a content analysis." Florida State University, unpublished Ph.D. dissertation. 181 Barth, Ernest A.T. and Donald Noel 1972 "Conceptual frameworks for the analysis of race relations: an evaluation," Social Forces 50(March):333-348. Bataille, Gretchen M. and Charles L.P. Silet (Eds.) 1980 The Pretend Indians; Images of Native American in the Movies. Ames, Iowa: Iowa State University Press. Beggs, B. 1956 "Children's books help to mold attitudes," Instructor 66:33-34. Beisecker, Thomas D. and Donn W. Parson (Eds.) 1972 The Process of Social Influence. Englewood Cliffs, NJ: Prentice-Hall, Inc. Bekkedal, Tekkla K. 1973 "Content analysis of children's books," Library Trends 22(October):109-127. Bell, M.E., et. al. 1979 "Common factors in sex and race discrimination," Psychological Reports 44(September):831-834. Bennett, Stephen E. 1972 "On the existence of an underlying 'vertical structure' in whites' racial attitudes," Social Science Quarterly 53, 3(December):583-589. Berelson, Bernard 1952 Content Analysis in Communications Research. New York: The Free Press. Berelson, Bernard and Sebastian de Grazia 1947 "Detecting collaboration in propaganda," Public Opinion Quarterly 11, 2:244-253. Berelson, Bernard and Morris Janowitz (Eds.) 1966 Reader in Public Opinion and Communication, Second Edition. New York: The Free Press. Berelson, Bernard and Patricia J. Salter 1946 "Majority and minority Americans: an analysis of magazine fiction," Public Opinion Quarterly 10, 2(Summer):168-190. Berg, Philip L. 1975 "Racism and the Puritan mind," Phylon 36:1-7. 182 Berger, Arthur Asa 1982 Media Analysis Techniques. Beverly Hills, CA: Sage Publications. Berkhofer, Robert F.J. 1978 The White Man's Indian. New York: Alfred A. Knopf, Inc. Bernard, L.L. 1925- "A classification of environment," American Journal of 1926 Sociology 16:322-330. Berreman, J.V.M. 1940 "Factors affecting the sale of modern books of fiction: a study of social psychology." Stanford: Stanford University, unpublished Ph.D. dissertation. Bettelheim, Bruno and Morris Janowitz 1949 "Ethnic tolerance: a function of social and personal control," American Journal of Sociology 55(September): 137-145. 1964 Social Change and Prejudice Including Dynamics of Prejudice. New York: Harper and Row, Publishers. Bettinghaus, Erwin P. 1968 Persuasive Communication. New York: Holt, Rinehart and Winston, Inc. Beuf, Ann H. 1977 Red Children in White America. Philadelphia: University of Pennsylvania Press, Inc. Billington, Ray Allen 1966 "Bias in history textbooks," Education Digest 31, 8(April): 37-39. Bissell, Benjamin E. . 1923 "The American Indian in English literature of the eighteenth century." New Haven: Yale University, Ph.D. dissertation. 1925 The American Indian in English Literature of the Eighteenth Century. New Haven: Yale University Press. Blake, Robert and Wayne Dennis 1943 "The development of stereotypes concerning the Negro," Journal of Abnormal and Social Psychology 38:525-531 . 183 Blalock, H.M. 1967 Toward a Theory of Minority-Group Relations. New York: Wiley and Sons. Blatt, Gloria T. 1968 "The Mexican-American in children's literature," Elementary English 45(April):446-451. Bogardas, Emory S. 1925a "Measuring social distance," Journal of Applied Sociology 9:299-308. 1925b "Social distance and its origins," Journal of Applied Sociology 9:216-226. 1930 "A race-relations cycle," American Journal of Sociology 35(January):612-61 7. 1950 "Stereotypes versus sociotypes," Sociology and Social Research 34(March/April):286-291. Borden, George A. 1971 An Introduction to Human Communication Theory. Dubuque, Iowa: William C. Brown Company Publishers. Bowker, Lee H. 1972 "Red and black in contemporary American history texts: a content analysis," in Bahr, Chadwick, and Day (eds.), Native Americans Today: Sociological Perspectives (New York: Harper and Row, Publishers), pp. 101-110. Bowker, R.R. n.d. The Publishers Weekly. New York: R.R. Bowker Company. n.d. School Library Journal. New York: R.R. Bowker Company. Braroe, N.W. and E.E.. Braroe 1977 "Who's in a home: identity misapprehension on the Northern Plains," Diogenes 98(Summer):71-92. Brembeck, Winston L. and William S. Howell 1952 Persuasion: A Means of Social Control. New York: Prentice-Hall, Inc. Brewton, Berry 1958 Race and Ethnic Relations. Boston: Houghton Mifflin Company. 184 Brigham, John C. 1973 "Ethnic stereotypes and attitudes: a different mode of analysis," Journal of Personality 41, 2(June):206-223. Brizee, Robert L. 1954 "The stereotype of the Indian in the New Mexico press." Albuquerque: University of New Mexico, unpublished Master's thesis. Broderick, Dorothy 1973 Image of the Black in Children's Fiction. New York: R.R. Bowker Comp any. Brody, Celeste M. 1973 "Do instructional materials reinforce sex stereotyping?"' Educational Leadership 31, 2(November):119-122. Brown, Claude 1965 Manchild in the Promised Land. New York: Macmillan and Company, Publishers. Brown, W.O. 1934 "Culture contact and race conflict," in E.B. Reater (ed.), Race and Culture Contacts (New York: McGraw-Hill Book Company, Inc.), pp. 34-47. Bruner, Jerome S. 1941 "The dimensions of propaganda: German short-wave broadcasts to America," Journal of Abnormal and Social Psychology 36:311-337. Bryson, Lyman (Ed.) 1948 The Communication of Ideas (A Series of Addresses). New York: Institute for Religious and Social Studies (distributed by Harper and Row, Publishers). Burrow, T. 1924- "Social images versus reality," Journal of Abnormal and 1925 Social Psychology 19:232-235. Byler, Mary Gloyne 1973 American Indian Authors for Young Readers: A Selected Bibliography. New York: Association on American Indian Affairs. 185 Byler, Mary Gloyne 1974 "Image of American Indians projected by Non-Indian writers," School Library Journal 99(February):36-37. Carlson, Ruth Kearney 1971 To Humanize People or to Make the Hearts Strong. Atlantic City, NJ: International Reading Association. Cata, Juanita Opal 1977 "The portrait of American Indians in children's fictional literature." Albuquerque: University of New Mexico, unpublished Ph.D. dissertation. Caute, David, et. al. 1980 "Race, sex and class in children's books," New Statesmen 100(November 14):16—19; (November 21):16-18; (December 5): 28-29; (December 12)-.19-20; (December 19):46—47- Chambers, Dewey W. 1968 "How now, Dr. Doolittle?" Elementary English 45(April): 437-439, 445. Chapman, Abraham 1975 Literature of the American Indians: Views and Interpretations. New York: New American Library. Chein, I. 1951 "Notes on a framework for the measurement of discrimination and prejudice," in Marie Jahoda, M. Deutsch, and S.W. Cook, Research Methods in Social Relations (New York: Dryden Press), Vol. 1, pp. 382-390. Child, Irvin L. and Leonard W. Doob __ 1943 "Factors determining national stereotypes," Journal of Social Psychology 17(May):203-219. Churchill, Ward 1971 "'Media stereotyping' response to criticism," The Indian Historian 12, 2(Summer):63. Churchill, Ward, Norbert Hill, and Mary Ann Hill 1978 "Media stereotyping and native response," The Indian Historian 11, 4(December):45-56, 63. Cicourel, Aaron V. 1964 Method and Measurement in Sociology. New York: The Free Press. 186 Clark, Kenneth B. 1963 Prejudice and Your Child. Boston: Beacon Press. (1955) Clark, Kenneth B. and Mamie K. Clark 1940 "Skin color as a factor in racial identification of Negro preschool children," Journal of Social Psychology (SPSSI Bulletin) 11 :159-169 - Cohen, Bernard P. 1980 Developing Sociological Knowledge; Theory and Method. Englewood Cliffs, NJ: Prentice-Hall, Inc. Cole, Joan Hays 1975 "Institutional racism in a community mental health center." Wright Institute, unpublished Ph.D. dissertation. Condie, J.S. and J.W. Christiansen 1977 "Indirect technique for the measurement of changes in black identity," Phylon 38(March):46-54. Conkling, Robert 1975 "Expression and generalization in history and anthropology," American Ethnologist 2(May):239-250. Cornelius, Paul 1971 "Interracial children's books: problems and progress," Library Quarterly 41(April):106-127. Costo, Rupert (Ed.) and Jeannette Henry (Writer) 1970 Textbooks and the American Indian. San Francisco: The Indian Historan Society. Council on Interracial Books for Children 1974 "Ten quick ways to analyze children's books for racism and sexism," Interracial Books for Children Bulletin 5, 3:1-3. 1977a Stereotypes, Distortions and Omissions in U.S. History Textbooks. New York: Council on Interracial Books for Children. 1977b Unlearning "Indian" Stereotypes. New York: Racism and Sexism Resource Center for Educatiors, A Division of the Council on Interracial Books for Children. 187 Council on Interracial Books for Children n.d. "Examples of institutional racism in education." Mimeo graphed paper. n.d. Guidelines for Selecting Bias-Free Textbooks and Storybooks. (1979) New York: Council on Interracial Books for Children. Cronkhite, Gary 1969 Persuasion: Speech and Behavioral Change. Indianapolis: Bobbs-Merrill Company, Inc. 1976 Communication and Awareness. Menlo Park, CA: Cummings Publishing Company, Inc. Dale, Joanne 1970 "Teaching literature by and about minorities," Educational Leadership 28, 3(December):289-291 - Daniels, Roger and Harry H.L. Kitano 1970 American Racism: Exploration of the Nature of Prejudice. Englewood Cliffs, NJ: Prentice-Hall, Inc. Cawson, Mildred A. 1959 "Children learn as they read," Elementary English 36 (November):475-477. deCharms, R. and G.H. Moeller 1962 "Values expressed in American children's readers: 1900- 1950," Journal of Abnormal and Social Psychology 64:136-142. DeFleur, Melvin L, 1962 "Mass communication and the study of rumor," Sociological Inquiry 32:51-70. 1970 Theories of Mass Communication. New York: David McKay Company, Inc. DeFleur, Melvin L. and Frank R. Westie 1958 "Verbal attitudes, and overt-acts: an-experiment- on the salience of attitudes," American Sociological Review 23 (December):667-673. Deloria, Vine 1980a "American fantasy," in Gretchen M. Bataille and Charles L.P. Silet (eds.), The Pretend Indians (Ames, Iowa: Iowa State University Press), pp. ix-xvi. 188 Deloria, Vine 1980b "The American image in North America," in Gretchen M. Bataille and Charles L.P. Silet (eds), The Pretend Indians (Ames, Iowa: Iowa State University Press), pp. 49-54. Diamont, L. 1969 "Relationships between intelligence and parental attitude and race awareness in five-year-olds," Journal of Social Psychology 17:253-257. Dinnerstein, Leonard 1979 Natives and Strangers: Ethnic Groups and the Building of America. New York: Oxford University Press. Dixon, W.J. 1977 "Research on research revisited: another half decade of quantitative and field research on international organizations," International Organization 31(Winter):65-82. Doob, Leonard W. 1935 Propaganda: Its Psychology and Technique. New York: Henry Holt and Company. Dovring, Karin 1954 "Quantitative semantics in 18th century Sweden," Public Opinion Quarterly 18, 4:389-394. 1959 Road of Propaganda. New York: Philosophical Library. Dreyer, Sharon S. 1981 The Bookfinder: A Guide to Children's Literature About the Needs and Problems of Youth Aged 2-15. Circle Pines, Minn.: American Guidance Service. Eastman, Richard M. 1965 A Guide to the Novel. San Francisco: Chandler Publishing Company. Edwards, Allen L. 1940 "Studies of stereotypes: I. the directionality and uniformity of response to stereotypes," Journal of Social Psychology 12(November):357-366. Edwards, Karl 0. 1980 "Institutional discrimination and Native Americans." Missoula, MT: University of Montana, unpublished paper. 189 Edwards, Karl 0. 1981 "Content analysis: a review of a method." Missoula, MT: University of Montana, unpublished paper. 1983 "Thesis proposal." Missoula, MT: University of Montana, unpublished paper. Ehninger, Douglas 197A Influence, Belief, and Argument: An Introduction to Responsible Persuasion. Glenview, IL: Scott, Foresman and Company. Ehrlich, Howard J. and James W. Rinehart 1965 "A brief methodology of stereotype research," Social Forces A3(May):564-575. Ehrlich, Paul R. and S. Shirley Feldman 1977 The Race Bomb; Skin Color, Prejudice, and Intelligence. New York: Ballantine Books. Elkin, S.M. 1965 "Minorities in textbooks: the latest chapter," Teachers College Record 66, 6(March):502-508. Ellul, Jacques 1965 Propaganda: The Formulation of Men's Attitudes. New York: Alfred A. Knopf, Inc. Emmert, Philip and William D. Brooks (Eds.) 1970 Methods of Research in Communication. Boston: Houghton Mifflin Company. EPIE Institute 1975 Racism in Career Education Materials: How to Detect it and How to Counteract its Effects in the Classroom. New York: Racism and Sexism Resource Center for Educators and the" Council orf Interracial Books for Children. 1981 "Checking -isms in children's books (guidelines for selecting bias-free textbooks for children)," EPIEgram: Materials 9(Fall):5-6. Ewers, John C. 1972 "The static images," in Robert Iocopi (ed.), Look to the Mountaintop (H.M. Gonsha Company), pp. 107-109- 190 Fairchild, Hoxie Neole 1982a "The noble savage." New York: Columbia University, Ph.D. dissertation. 1928b The Noble Savage. New York: Columbia University Press. Faris, E. 1925 "The concept of social attitudes," Journal of Applied Sociology 9:404-408. Farris, C.E., et. al. 1980 "Self-concept formation in Indian children," Social Casework 61(October):484-489. Festinger, Leon and Daniel Katz (Eds.) 1953 Research Methods in the Behavioral Sciences. New York: The Dryden Press. Fiedler, Leslie A. 1968 The Return of the Vanishing American. New York: Stein and Day. Finney, R.L. 1926 "The unconscious social mind," Journal of Applied Sociology 10:357-365. Fischer, P.L. and J.V. Torney 1976 "Influence of children's stories on dependency, a sex-typed behavior," Developmental Psychology 12:489-490. Fishbein, M. 1967 Readings in Attitude Theory and Measurement. New York: Wiley and Sons. 1971 "Attitudes and the prediction of behavior," in Kerry Thomas (ed.), Attitudes and Behavior (Baltimore: Penguin Books), pp. 57-58. Fisher, Dexter (Ed.) 1977 Minority Language and Literature: Retrospective and Perspective. New York: Modern Language Association of America. Fisher, Frank L. 1965 "The influences of reading and discussion on the attitudes of fifth graders toward American Indians." Berkeley: University of California, unpublished Ed.D. dissertation. 191 Fisher, Frank L. 1968 "Influences of reading and discussion on the attitues of fifth graders towards American Indians," Journal of Educational Research 62:130-134. Flowerman, Samuel H. 1947 "Mass propaganda in the war against bigotry," Journal of Abnormal and Social Psychology 42:429-439. Fotheringham, Wallace C. 1966 Perspectives on Persuasion. Boston: Allyn and Bacon, Inc. Frasher, Ramona and Annabelle Walker 1972 "Sex roles in early reading textbooks," The Reading Teacher 25(May):741-749 - Frazier, E. Franklin 1957 Race and Culture Contacts in the Modern World. Boston: Beacon Press. Funk, Sandra G., et. al. 1976 "The perceived structure of American ethnic groups: the use of multidimensional scaling in stereotype research," Sociometry 39:116-130. Gaines, John Strother 1971 "The treatment of Mexican-American history in contemporary American high school textbooks." Los Angeles: University of Southern California, unpublished Ed.D. dissertation. Galvin, Kathleen M. and Cassandra L. Book 1972 Speech Communication. Skokie, IL: National Textbook Company. Garcia, Jesus 1978 "From bloody savages to heroic chiefs," Journal of American Indian Education 17(January):15-19. 1980 "The American Indian: no longer a forgotten American in U.S. history texts published in the 1970's," Social Education 44, 2(February):148-152, 164. Garcia, Jesus and Ricardo Garcia 1980 "Selecting ethnic materials for the elementary school," Social Education 44, 3(March):232-234, 236. 192 Gardner, R.C., D.M. Kirby, R.Y. Pablo, and Emma Santos Castillo 1975 "Ethnic stereotypes: the role of language," The Journal of Social Psychology 96:3-9. Gast, David Karl 1965 "Characteristics and concepts of minority Americans in contemporary children's fictional literature." Tempe: Arizona State University, unpublished Ed.D. dissertation. 1967 "Minority Americans in children's literature," Elementary English XLIV(January):12-23. 1970 "The dawning of the age of Aquarius for multi-ethnic children's literature," Elementary English XLVII(May): 661-665. George, Alexander L. 1959a Propaganda Analysis: A Study of Inferences Made from Nazi Propaganda in World War II. Evanston, IL: Row, Peterson. 1959b "Quantitative and qualitative approaches to content analysis," in Ithiel De Sola Pool (ed.), Trends in Content Analysis (Urbana: University of Illinois), pp. 7-32. Georgeoff, J.P. 1968 "The effect of the curriculum upon the self-concept of children in racially integrated fourth grade classrooms." Paper to the American Educational Research Association. Gerbner, George, Ole R. Holsti, Klaus Krippendorff, William J. Paisley, and Philip J. Stone 1969 The Analysis of Communication Content. New York: J. Wiley and Sons, Inc. Gezi, K.I. and B. Johnson 1970 "Enhancing racial attitudes through the study of black heritage," Childhood Education 46, 7:339. Gilbert, G.M. 1951 "Stereotype persistence and change among college students," Journal of Abnormal and Social Psychology 46:345-354. Gilliland, Hap 1980 Indian Children's Books. Billings, MT: Montana Council for Indian Education. 1 Gist, Noel P. 1932 "Racial attitudes in the press," Sociology and Social Research 17(September/October):25-36. Glick, Clarence E. 1955 "Social roles and types in race relations," in A.W. Lind (ed.), Race Relations in World Perspective (Honolulu: University of Hawaii Press), pp. 243-262. Golding, Peter and Graham Murdock 1980 "Theories of communication and theories of society," in Wilhoit and de Bock (eds.), Mass Communication Review Yearbook, Volume 1 (Beverly Hills: Sage Publications), pp. 59-76. Gonzalez, G.T. and K.F. Williams 1981 "Relationship between locus of control and sex-role stereotyping," Psychological Reports 49(August):70. Goode, William J. and Paul K. Hatt 1952 Methods in Social Research. New York: McGraw-Hill Book Company, Inc. Goodman, Mary Ellen 1964 Race Awareness in Young Children. New York: Anti- (1952) Defamation League of B'nai B'rith. Gordon, George N. 1969 The Languages of Communication: A Logical and Psycho logical Examination. New York: Hastings House Publishers 1971 Persuasion: The Theory and Practice of Manipulative Communication. New York: Hastings House Publishers. Gordon, Rosemary 1962 Stereotype of Imagery and Belief as an Ego Defense. Cambridge: Cambridge University Press. Gottschalk, Louis, "Clyde Klucktiohn, and Robert Angell 1947 The Use of Personal Documents in History, Anthropology and Sociology. New York: Social Science Research Council Gregory, Dick 1971 No More Lies: The Myth and Reality of American History. New York: Harper and Row, Publishers. 194 Gregory, W.S. 1941 "A study of stereotyped thinking: affective reactions to persons as the basis for judging their nationality," Journal of Social Psychology 13(February):89—1 02. Haley, Frances, Susan Hustleby, and Regina McCormick 1978 Ethnic Studies Handbook for School Librarians. Boulder, CO: Ethnic Studies Project for School Librarians, SSEC, Inc. Harail, Thomas Arthur 1970 "A cross-sectional examination of commitment to visual stereotypes, attitudes, and information about specific ethnic groups." Albuquerque: University of New Mexico, unpublished Ph.D. dissertation. Hamilton, David L. (Ed.) 1981 Cognitive Processes in Stereotyping and Intergroup Behavior. Hillsdale, NJ: Lawrence Erlbaum Associates, Publishers. Handlin, Oscar 1957 Race and Nationality in American Life. Boston: Little, Brown. Hansen, Donald A. and J. Herschel Parsons (Comps.) 1968 Mass Communication: A Research Bibliography. Santa Barbara, CA: The Glendessary Press. Harding, John, Harold Proshansky, Bernard Kutner, and Isidor Chein 1968 "Prejudice ana ethnic relations," in Lindzey and Aronson (eds.), The Handbook of Social Psychology, Second Edition (Menlo Park, CA: Addison-Wesley Publishing Company, Inc.), pp. 1-76. Harrah, David 1963 Communication: A Logical Model. Cambridge, MA: MIT Press. Harris, Helen L. 1973 "On the failure of Indian education," Clearing House 48 (December):242-247. 1974 "The bi-centennial celebration and the Native American," The Indian Historian 7, 4(Fall):5-8. Hartley, E.L. 1946 Problems in Prejudice. New York: King's Crown Press. 195 Haugh, O.M. 1952 "The relative effectiveness of reading and listening to radio drama as ways of imparting information and shifting attitudes," Journal of Educational Research 45:489-498. Haydon, C. 1978 "Union warns of dangers in myths about race and intelli gence," Times Educational Supplement 3, 299(September 22):4. Hazard, Lucy Lockwood 1937 The Frontier in American Literature. New York: Thomas Y. Crowell. Henry, Jeannette 1967 "Our inaccurate textbooks," The Indian Historian 1, 1 (December):21-24. Henry, Jeannette (Ed.) 1973 The American Indian Reader: Literature. San Francisco: The Indian Historian Press, Inc. Hesselbart, Susan 1975 "Education, ethnic stereotypes and question format," Sociology and Social Research 59(April):266-273. 1978 "Racial stereotypes, gender stereotypes and question format," Social Science Quarterly 59, 2(September): 388-346. Hester, Nellie 1975 "Textbook analysis and community action," Elementary English 52, 5(May):712-714- Holsti, Ole R. 1962 "The belief system and national images: a case study," Journal of Conflict Resolution 6:245-252. 1969 Content Analysis for the Social Sciences and Humanities. Reading, MA: Addison-Wesley Publishing Company, Inc. Hoover, Dwight W. 1976 The Red and the Black. Chicago: Rank McNally College Publishing Company. Hopper, Robert 1976 Human Message Systems. New York: Harper and Row, Publishers. 196 Hornbook Company 1960- The Hornbook Magazine. Boston: Hornbooks, Inc. Horton, John 1966 "Order and conflict theories of social problems as contemporary ideologies," American Journal of Sociology 71(May):701-71 3 - Houts, Kathleen C. and Rosemary S. Bahr 1972 "Stereotyping of Indians and blacks in magazine cartoons," in Bahr, Chadwick, and Day (eds.), Native Americans Today: A Sociological Perspective (New York: Harper and Row, Publishers), pp. 110-114. Hovland, Carl, Irving Janis, and Harold Kelly 1954 Communication and Persuasion. New Haven: Yale University Press. Hoy, James F. 1979 "The Indian through the eyes of The Cattleman," The Indian Historian 12, 2(Summer):41-46, 62. Iglitzin, Lynne B. 1972 Violent Conflict in American Society. San Francisco: Chandler Publishing Company. Jackson, E. 1944 "Effects of reading upon attitudes toward the Negro race," Library Quarterly 14(January):47-54. Janis, I.L. 1949 "The problem of validating content analysis," in H.D. Lasswell (ed.), Language of Politics: Studies in Quantitative Semantics (Cambridge, MA: MIT Press), pp. 55-82. Johnson, Carole Schul.te and Gloria R. Greenbaum 1980 "Are boys disabled readers due to sex role stereotyping?" Educational Leadership 37, 6(March):492-496. Johnson, E.C. 1911 A School History of Negro Race in America. New York: Goldman. Jones, George E., Jr. 1958 "The American Indian in the American novel (1975-1950)." New York University, unpublished Ph.D. dissertation. 197 Kane, Michael B. 1970a Minorities in Textbooks: A Study of Their Treatment in Social Studies Texts. Chicago: Quadrangle Books. 1970b Textbook Treatment of Other Minorities in America. Chicago: Anti-Defamation League of B'nai B'rith. Kaplan, Abraham 1943 "Content analysis and the theory of signs," Philosophy of Science 10:230-247. 1964 The Conduct of Inquiry: Methodology for Behavioral Science. San Francisco: Chandler Publishing Company. Karlins, M., T.L. Coffman, and G. Walters 1969 "On the fading of social stereotypes: studies on three generations of college students," Journal of Personality and Social Psychology 13:1-16. Katz, Daniel and Kenneth W. Braly 1933 "Racial stereotypes of one hundred college students," Journal of Abnormal and Social Psychology 28(0ctober/ December):280-290. 1935 "Racial prejudice and racial stereotypes," Journal of Abnormal and Social Psychology 30(July/September):175—193. Katz, Judy H. 1978 White Awareness: Handbook for Anti-Racism Training. Norman: University of Oklahoma Press. Keiser, Albert 1933 The Indian in American Literature. New York: Oxford University Press. Khatri, A.A. 1980 "Analysis of fiction—a method for intracultural and crosscultural study of family systems," Journal of Marriage and Family 42(Fall):197-203. Kinton, Jack 1974 American Ethnic Groups and the Revival of Cultural Pluralism. Aurora, IL: Social Science and Sociological Resource. Kirkness, Verna J. 1977 "Prejudice about Indians in textbooks," Journal of Reading 20(April):595-600. 198 Klapper, Joseph T. 1950 The Effects of Mass Media. New York: Bureau of Applied Social Research, Columbia University. Kluckhohn, Clyde 1960 Mirror for Man. Greenwich, Conn.: Fawcett Publications, Inc. Knowles, Louis L. and Kenneth Prewitt 1969 Institutional Racism in America. Englewood Cliffs, NJ: Prentice-Hall, Inc. Kolbe, Richard and Joseph C. LaVoie 1981 "Sex-role stereotyping in preschool children's picture books," Social Psychology Quarterly 44, 4(December):369-374. Kramer, B.M. 1949 "Dimensions of prejudice," Journal of Psychology 27:389-451. Krippendorff, Klaus 1980 Content Analysis: An Introduction to Its Methodology. Beverly Hills, CA: Sage Publications. Kuhn, Thomas S. 1970 The Structure of Scientific Revolutions, Second Edition. Chicago: University of Chicago Press. Kutner, B., C. Watkins, and P.R. Yarrow 1952 "Verbal attitudes and overt behavior involving racial prejudice," Journal of Abnormal and Social Psychology 47:649-652. Lambarth, John 1980 Social Psychology. New York: Macmillan and Company, Publishers. Landbreth, Catherine and Barbara Child Johnson 1953 "Young children's responses to a picture and inset test designed to reveal reactions to persons of different skin color," Child Development 24, 1(March):63-80. Lang, Gladys Engel and Kurt Lang 1981 "Mass communications and public opinion: strategies for research," in M. Rosenberg and R.H. Turner (eds.), Social Psychology: Sociological Perspectives (New York: Basic Books, Inc., Publishers), pp. 653-682. 199 LaRoque, Emma 1975 Defeathering the Indian. Agincourt, Canada: The Book Society of Canada Limited. Lasswell, H.D. 1927 Propaganda Technique in the World War. New York: Alfred A. Knopf. 1938 "A provisional classification of symbol data," Psychiatry 1:197-204. 1948 "The structure and function of communication in society," in Lyman Bryson (ed.), The Communication of Ideas (New York: Harper and Brothers), pp. 37-51. Lasswell, Harold D., Ralph D. Casey, and Bruce Lannes Smith 1935 Propaganda and Promotional Activities. Minneapolis: University of Minnesota Press. Lasswell, H.D., N. Leites, R. Fadner, J.M. Goldsen, A. Gray, I.L. Janis, A. Kaplan, D. Kaplan, A. Mintz, I. De Sola Pool, and S. Yakobson 1949 Language of Politics: Studies in Quantitative Semantics. Cambridge, MA: MIT Press. Lass-Woodfin, Mary Jo (Ed.) 1978 Books on American Indians and Eskimos: A Selection Guide for Children and Young Adults. Chicago: American Library Association. LaViolette, Forest and K.H. Silvert 1951 "A theory of stereotypes," Social Forces 29(March):257-262. Lazarsfeld, P.F., B. Berelson, and H. Gaudet 1940 The People's Choice: How the Voter Makes Up His Mind in Presidential Campaigns. New York: Columbia University Press. Leisy, Ernest E. 1930 An Interpretive History of American Literature. New York: (1929) Thomas Y. Crowell. Lester, Nancy B. 1982 "A system for analyzing characters' values in literary texts," Research in the Teaching of English 16, 4(December): 321-338. 200 Lew, T.T. 1923 "China in American school textbooks," Chinese Sociology and Political Science Review 7Supplement. Lieberson, Stanley 1961 "A societal theory of race and ethnic relations," American Sociological Review 26(December):902—910. Lindzey, Gardner (Ed.) 1954 Handbook of Social Psychology. Volume 1. Cambridge, MA: Addison-Wesley Publishing Company, Inc. Lindzey, Gardner and Elliot Aronson (Eds.) 1968 The Handbook of Social Psychology, Second Edition. Volume 1, Historical Introduction and Systematic Positions. Volume 2, Research Methods. Volume 3, The Individual in a Social Context. Volume 4, Group Psychology and Phenomena of Interaction. Volume 5, Applied Social Psychology. Menlo Park, CA: Addison-Wesley Publishing Company, Inc. Linn, Lawrence S. 1965 "Verbal attitudes and overt behavior: a study of racial discrimination," Social Forces 44(March):353-364. Lippmann, Walter 1922 Public Opinion. New York: Harcourt, Brace and Company. 1927 "Stereotypes as subjective environment," in K. Young (ed.), Source Book for Social Psychology (New York: Alfred A. Knopf), pp. 435-440. Lipsky, A. 1927 "The power of stereotyped words," in K. Young (ed.), Source Book for Social Psychology (New York: Alfred A. Knopf), 449-451. Litcher, J.H. and D.W. Johnson 1969 "Changes in attitudes toward Negroes of white elementary school students after use of multiethnic readers," Journal of Educational Psychology 60:148-152. Lopatka, R. 1965 "An analysis of the classroom periodical used in teaching of current events in Alberta high schools." University of Alberta, unpublished Master's thesis. 201 Lubbock, Percy 1957 The Craft of Fiction. New York: The Viking Press. Lucas, Phil 1980 "Images of Indians," Four Winds 1, 4(Autumn):68-77. Mack, Raymond W. and Troy S. Duster 1973 Patterns of Minority Relations. New York: Anti- (1963) Defamation League of B'nai B'rith. Malone, T.J. 1927 "History texts and American-Canadian amity," Historical Outlook 18:373-377. Manitoba Indian Brotherhood 1974 The Shocking Truth About Indians in Textbooks. Winnipeg, Manitoba: Manitoba Indian Brotherhood. Manley, Dorothy Shepard 1944 "Improving racial attitudes through children's books," Elementary English 21(November):267-269. Martin, C.L. and C.F. Halverson, Jr. 1981 "Schematic processing model of sex typing and stereotyping in children," Child Development 52(December):111 9-1134. McDiarmid, Garnet and David Pratt 1971 Teaching Prejudice. Toronto: The Ontario Institute for Studies in Education. McGraw-Hill Book Company 1975 "Guidelines for equal treatment of the sexes," Elementary English 52, 5(May):725-733. McGuire, William 1969 "The nature of attitudes and attitude change," in G. Lindzey and E. Aronson (eds.), The Handbook of Social Psychology, Second Edition. Volume 3 (Boston: Addison- Wesley Publishing Company, Inc.), pp. 136-314. Medicine, Bea 1979 "Hanto Yo: a new phenomenon," The Indian Historian 12, 2(Summer):2-5. 202 Meenes, Max 1943 "A comparison of racial stereotypes of 1935 and 1942," Journal of Social Psychology 17(May):327-336. Meltzer, H. 1939 "Nationality preferences and stereotypes of colored children," The Journal of Genetic Psychology 54:403-424. 1941 "The development of children's nationality preferences, concepts and attitudes," The Journal of Psychology 11: 343-358. Menefee, Selden C. 1936 "The effect of stereotyped words on political judgment," American Sociological Review 1(August):614-621. Merrill, John C. 1965 "How Time stereotyped three U.S. presidents," Journalism Quarterly 42:563-570. Merton, Robert K. 1967 "Manifest and latent functions," in R.K. Merton, On Theoretical Sociology (New York: The Free Press), pp. 73-138. Merton, Robert K. and Paul F. Lazarsfeld 1957 "Studies in radio and film propaganda," in R.K. Merton, Social Theory and Social Structure (New York: The Free Press), pp. 509-528. Miller, Clyde R. 1946 The Process of Persuasion. New York: Crown Publishers. Miller, Delbert C. 1977 Handbook of Research Design and Social Measurement, Third Edition. New York.: Longman. Miller, Randall M. (Ed.) _ _ 1978 Ethnic Images in American Film and Television. Phila delphia: Balch Institute. Miller, Wayne C. 1976 A Handbook of American Minorities. New York University Press. 203 Minnesota University Library Services Institute for Minnesota Indians 1970 American Indians; An Annotated Bibliography of Selected Library Resources. St. Paul, Minnesota State Department of Education, Indian Education Section. Moline, Jon 1983 "The moral of the story: why literature is still the best source for teaching values," American Educator 7, 1(Spring):25-31. Monti, D.J. 1979 "Biased and unbiased news: reporting racial controversies in the New York Times, 1960-July 1964," Sociological Quarterly 20(Summer).-399-409. Morland, J. Kenneth and Ellen Suthers 1980 "Racial attitudes of children: perspectives on the structural-normative theory of prejudice," Phylon XLI, 3(September):267-275. Morris, Jeane B. 1981 "Indirect influences on children's racial attitudes," Educational Leadership 38, 4(January):286-287. Mortensen, C. David 1972 Communication: The Study of Human Interaction. New York: McGraw-Hill Book Company, Inc. Mosher, E., et. al. 1941 Introduction to Responsible Citizenship. New York: Henry Holt and Company. Multi-Ethnic Media 1975 Selected Bibliographies in Print. Chicago: Office for Library Services to the Disadvantaged, American Library Association. Murphy, G. and R. Likert 1938 Public Opinion and the Individual. New York: Harper and Row, Publishers. Nash, Gary B. and Richard Weiss (Eds.) 1970 The Great Fear; Race in the Mind of America. New York: Holt, Rinehart, and Winston, Inc. 204 Nelson, John H. 1968 The Negro Character in American Literature. College (1926) Park, MD: McGrath_Publishing Company. Newcomb, Theodore Mead, Ralph H. Turner, and Philip E. Converse 1965 Social Psychology, the Study of Human Interaction. New York: Holt, Rinehart, and Winston, Inc. Noar, Gertrude n.d. Civil Rights and Civil Liberties. New York: Anti- Defamation League of B'nai B'rith. Noel, Donald L. 1968 "A theory of the origin of ethnic stratification," Social Problems 16(Fall):'157-172. North, Robert C., Ole R. Holsti, M. George Zaninovich, and Dina A. Zinnes 1963 Content Analysis: A Handbook with Applications for the Study of International Crisis. Northwestern University Press. Okanes, M.M. 1956 "Expected behavior as a predictor of attitude change through communication," Dissertation Abstracts 16:809-810. Orenstein, Alan and William R.F. Phillips 1978 Understanding Social Research: An Introduction. Boston: Allyn and Bacon, Inc. Osgood, C.E., W.H. May, and M.S. Miron 1975 Cross'Cultural Universals of Affective Meaning. Urbana: University of Illinois Press. Osgood, C.E., Sol Saporta, and J.C. Nunnally 1956 "Evaluative assertions analysis"," Litera 3:47-102. Osgood, C.E., G.J. Suci, and P.H. Tannenbaum 1957 The Measurement of Meaning. Urbana: University of Illinois Press. Oskamp, Stuart 1977 Attitudes and Opinions. Englewood Cliffs, NJ: Prentice- Hall, Inc. 205 Palmer, Edward Nelson 1956 "Culture contacts and population growth," in J.J. Spengler and O.D. Duncan (eds.), Population Theory and Policy (Genco, IL: The Free Press), pp. 410—415- Park, R.E. 1924a "The concept of social distance," Journal of Applied Sociology 8:339-344. 1924b "Experience and race relations," Journal of Applied Sociology 9:20-21. 1927 "Race prejudice as defense mechanism," in K. Young (ed.), Source Book for Social Psychology (New York: Alfred A. Knopf), pp. 494-497. Pate, Glenn S. 1981 "Research on prejudice reduction," Educational Leadership 38, 4(January):288-291• Payne, I.R. 1964 "Persuasibility and its relationship to affect, sex and type persuasive appeal," Dissertation Abstracts 24:3439. Pecoraro, Joseph 1971 "The effect of a series of special lessons on Indian history and culture upon the attitudes of Indian and Non-Indian students." Boston University School of Education, Ed.D. dissertation. 1972 "The effect of a series of special lessons on Indian history and culture upon the attitudes of Indian and Non-Indian students," Journal of Education 15, 4(February): 70-78. Peltola, Bette J. 1979a "Choosing the Newberry and Caldecott medal winners," Top of the News 36, 1:43-47. 1979b "Newberry and Caldecott medals: authorization and terms," Top of the News 36, 1:48-54. Pierce, C.M., et. al. 1977 "Experiement in racism: TV commercials," Education and Urban Sociology 10(November):61-87. 206 Poussaint, A. (U.S. Commission on Civil Rights) 1970 What Students Perceive. Clearinghouse Publications. Pratt, David 1969a "An instrument for measuring evaluative assertions concerning minority groups and its application in an analysis of history textbooks approved for Ontario schools." University of Toronto, unpublished "Ph.'D. dissertation. 1969b "An instrument for measuring evaluative assertions concerning minority groups and its application in an analysis of history textbooks approved for Ontario schools," Dissertation Abstracts 32/03-A:1402. 1971 "Value judgments in textbooks: the Coefficient of Evaluation as a quantitative measure," Interchange 2, 3:1-14. 1972 How to Find and Measure Bias in Textbooks. Englewood Cliffs, NJ: Educational Technology Publications. Prewitt, Kenneth, et. al. n.d. "Institutional racism in American society: a primer." Mimeographed collection of essays by graduate students at Stanford University. Price, A. Grenfell 1950 White Settlers and Native Peoples. Melbourne: Georgian House. Queen, Rensee 1975 "Genderation of the self," Elementary English 52, 5(May): 717-720. """ ' " "" Quinn, Arthur H. 1951 The Literature of the American People. New York: Appleton- Century-Crofts, Inc. Raab, Earl and Seymour M. Lipset 1968 Prejudice and Society. New York: Anti-Defamation League (1959) of B'nai B'rith. Radke, Marian and Jean Sutherland 1949 "Children's concepts and attitudes about minority and majority American groups," Journal of Educational Psychology 40:449-468. 207 Razran, Gregory 1950 "Ethnic dislikes and stereotypes: a laboratory study," Journal of Abnormal and Social Psychology 45(January):7-27. Reardon, Kathleen Kelley 1981 Persuasion: Theory and Context. Beverly Hills: Sage Publications. Reddick, L.D. 1934 "Racial attitudes in American history textbooks of the South," Journal of Negro History 19:225-265. Renninger, C.A. and J.E. Williams 1966 "Black-white color connotations and race awareness in preschool children," Perceptual and Motor Skills 22:771-785. Rice, S.A. 1926 "'Stereotypes': a source of error in judging human character," Journal of Personnel Research 5:268-276. Ritzer, George 1980 Sociology: A Multiple Paradigm Science, Revised Edition. Boston: Allyn and Bacon, Inc. Rokeach, Milton 1968 Beliefs, Attitudes and Values. San Francisco: Jossey- Bass Publishers. Rollins, Charlemae 1941 We Build Together. Chicago: NCTE. Rollins, Charlemae (Ed.) 1967 We Build Together. Champaign, IL: NCTE. Roloff, Michael and Gerald R. Miller (Eds.) 1980 Persuasion: New Directions in Theory and Research. Beverly Hills, CA: Sage Publications. Root, E.M. 1962 Brainwashing in the High Schools. New York: The Devin- Adair Company. Rose, Peter I. 1964 They and We: Racial and Ethnic Relations in the United States. New York: Random House. 208 Rosenberg, Max 1975 "Evaluate your textbooks for racism, sexism!" Educational Leadership 31, 2(November):1 07-109. Rosenberg, M.J., C.I. Hovland, W.J. McGuire, R.P. Abelson, and J.W. Brehm _ _ 1960 Attitude Organization and Change: An Analysis of Consistency Among Attitude Components. New Haven: Yale University Press. Rosenberg, Morris and Ralph H. Turner (Eds.) 1981 Social Psychology: Sociological Perspectives. New York: Basic Books, Inc., Publishers. Rosengren, Karl Erik 1980 "Mass media and social change: some current approaches," in Wilhoit and de Bock (eds.), Mass Communication Review Yearbook (Beverly Hills: Sage Publications), pp. 168-180. 1981 Advances in Content Analysis. Beverly Hills: Sage Publications. Rosi, Eugene J. 1964 "How 50 periodicals and the Times interpreted the test ban controversy," Journalism Quarterly 41:545-556. Rosnow, Ralph L. and Gary A. Fine 1976 Rumor and Gossip: The Social Psychology of Hearsay. New York: Elsevier. Rosnow, Ralph L. and Edward J. Robinson 1967 Experiments in Persuasion. New York: Academic Press. Rothman, Jack 1977 Issues in Race and Ethnic Relations; Theory, Research, and Action. Itasca, IL: F.E. Peacock-Publishers, Inc. Royce, J. 1927 "Antipathy and prejudice," in K. Young (ed.), Source Book for Social Psychology (New York: Alfred A. Knopf), pp.. 483-485. Running, John 1980 ^'Images of the Northern Plains," Four Winds 1, 3(Summer): 18-28. ' 209 Ruth, R.W. 1969 "The effects of black studies on the Negro fifth grade students," Journal of Negro Education 38(Fall):435-439. Saenger, Gerhart and Samuel Flowerman 1954 "Stereotypes and prejudical attitudes," Human Relations 7:217-238. Sampson, Edward E. (Ed.) 1964 Approaches, Contexts, and Problems of Social Psychology. Englewood Cliffs, NJ: Prentice-Hall, Inc. Sanders, Ronald 1973 "Literature of the Indian: a critique," in Jeannette Henry (ed.), The American Indian Reader: Literature (San Francisco: The Indian Historian Press, Inc.), pp. 219-244. Schlachter, Gail A. 1977 Minorities and Women: A Guide to Reference Literature in the Social Sciences. Los Angeles: Reference Service Press. Schnell, Thomas R. and Judith Sweeney 1975 "Sex role bias in basal readers," Elementary English 52, 5(May):737-742. Schuman, H. and J. Harding 1964 "Prejudice and the norm of nationality," Sociometry 27: 353-371• Schwartz, Howard and Jerry Jacobs 1979 Qualitative Sociology: A Method to Madness. New York: The Free Press. Scott, Kathryn P. 1980 "Sexist and non-sexist materials: what impact do they have?" The Elementary School Journal 81 , 1:46-52. Sex Desegregation Assistance Center n.d. "Identifying bias in instructional materials." Mimeo graphed copy of training program. Shannon, Claude and Warren Weaver 1962 The Mathematical Theory of Communication. Urbana, IL: University of Illinois Press. 210 Shaughnessy, Tim 1978 "White stereotypes of Indians," Journal of American Indian Education 17-(-January )-:-2Q^24. Shaw, Elaine 1974 "The treatment of the American Indian in children's-fiction, 1960-1970." Southern Connecticut State College, unpublished Master's thesis. Shaw, Marvin E. and Philip R. Costanzo 1970 Theories of Social Psychology. New York: McGraw-Hill Book Company, Inc. Sherif, Muzafer 1935 "An experimental study of-stereotypes," Journal- of Abnormal and Social Psychology 29(January-March):371—375- Sherif, Muzafer and Carolyn Sherif 1956 An Outline of Social Psychology, Revised Edition. New .York :. Harper and-Row, Publishers. Shibutani, Tamotsu 1966 Improvised News: A Sociological Study of Rumor. New York: Bobbs-Merrill Company, Inc. Shirley, F.L. 1966 "The influence of reading on the concepts, attitudes, and behavior of tenth, eleventh, and twelfth grade students." University of Arizona, unpublished Ph.D. dissertation. Shuey, Audrey, et. al. 1953 "Stereotyping of Negroes and whites: an analysis of magazine pictures," Public -Opinion Quarterly 17:281-288. Siegel, Sidney 1956 Nonparametric Statistics for the Behavioral Sciences. New York: McGraw-Hill Book Company, Inc. Singh, Jane M. 1975 "Language education and ethnic children's literature," Elementary English 52, 5(May):721-724. Slawski, Carl J. 1981 Social Psychological Theories: A Comparative Handbook for Students^ Glenview, IL: _ Scott, Foresman and Company. 211 Smith, Bruce L., Harold D. Lasswell, and Ralph D. Casey 1946 Propaganda, Communication and Public Opinion. Princeton: Princeton University Press. Smith, Jacklyn 1978 "'Unlearning' Indian 'stereotypes': a classroom experience," Interracial Books for Children Bulletin 9, 7:14-15. Smith, M.B. 1947 "The personal setting of public opinions: a study of attitudes toward Russia," Public Opinion Quarterly 11: 502-523. Smith, N.B. 1948 "Some effects of reading in children," Elementary English 30, 5(May):271-278. Smith, Raymond G. 1978 The Message Measurement Inventory: A Profile for Communi cation Analysis. Bloomington, IN: Indiana University Press. Smith, R.J., et. al. 1980 "When is a stereotype a stereotype?" Psychological Reports 46(April):643-651. Snider, James G. and Charles E. Osgood (Eds.) 1969 Semantic Differential Techniques. Chicago: Aldine Publishing Company. Social Science Education Consortium, Inc. 1975 Ethnic Heritage Studies Materials Analysis Instrument. Boulder, CO: SSEC. Spiller, Robert E. 1957 The Cycle of American Literature. New York: The Free Press. Stabler, J.R. and E.E. Johnson 1971 "Children's perception of black and white boxes as a reflection of how they regard their own and other's racial membership." Paper presented at American Psychological Association, Washington, D.C., September, 1971. Stabler, J.R., E.E. Johnson, and S.E. Evans 1970 "The measurement of children's self concepts as related to racial membership." Georgia State University, unpublished paper. 212 Stanton, William 1960 The Leopards' Spots: Scientific Attitudes Toward Race in America, 1815-1859- Chicago: University of Chicago Press. Steensland, Anna Lee 1971 "American Indian culture: promises, problems and possibilities," English Journal 60(December):1145-2000. 1974 "Indian writers and Indian lives," Integrated Education 12, 4(November-December) : 3-7_. Stevenson, Harold W. and Edward C. Stewart 1958 "A developmental study of racial awareness in young children," Child Development 24, 3(September):399-409. Stone, P.J., D.C. Dunphy, M.S. Smith, and D.M. Ogilvie 1966 The General Inquirer: a Computer Approach to Content Analysis. Cambridge: MIT Press. Stryker, Sheldon 1981 "Symbolic interactionism: themes and variations," in M. Rosenberg and R.H. Turner (eds.), Social Psychology: Sociological Perspectives (New York: Basic Books, Inc., Publishers), pp. 3-29. Sullivan, Marjorie and John S. Goodell 1975 "Media use in the study of minorities," The Emporia State Research Studies XXIV, 2(Fall):1-63. Swayne, John n.d. "Discrimination." Mimeographed paper by the Head Start Deputy, Nez Perce CAP. Takaki, Ronald T. 1979 Iron 'Cages:' "Race" arid" Culture" in Nineteenth-Century America. New. York: Alfred A. Knopf, Inc. Tannenbaum, P.H. 1953 "Attitudes towards source and concept as factors in attitude change through communictions," Dissertation Abstracts 13:1288. Tenkate, Herman F.C. 1970 "The Indian in literature," The Indian Historian III, 1(Summer):23-32. 213 TeSelle, Sallie (Ed.) 1973 The Rediscovery of Ethnicity. New York: Harper and Row, Publishers. Thomas, Kerry 1971 " Attitudes and Behavior. Baltimore: Penguin Books. Thomas, William I. 1904 "The psychology of race prejudice," American Journal of Sociology 9, 5(March):593-611. 1927 "Race prejudice as a form of isolation," in K. Young (ed.), Source Book for Social Psychology (New York: Alfred A. Knopf), pp. 499-501. Trager, H.G. and M.R. Yarrow 1952 They learn what they live. New York: Harper and Row, Publishers. Troy, Alice Anne 1972 "The Indian in adolescent literature, 1930-1940 vs. 1960— 1970." University of Iowa, unpublished Ph.D. dissertation. 1978 "The Indian in adolescent novels," The Indian Historian 8, 4(Winter):32-35. Turner, Jonathan H. 1982 The Structure of Sociological Theory, Third Edition. Homewood, IL: The Dorsey Press. United States Commission on Civil Rights 1977 Window Dressing on the Set: Women and Minorities in Television. Washington, D.C.: Government Printing Office. 1980 Characters in Textbooks, a Review of the Literature. Washington, D.C.: Government Printing Office. United States Department of Health, Education, and Welfare 1975 The North American Indian: A Bibliography of Community Development. Washington, D.C.: Government Printing Office. 1977 Identifying Discrimination: A Review of Federal Anti- Discrimination Laws and Selected Case Examples. Washington, D.C.: Government Printing Office. 214 United States Department of the Interior, Bureau of Indian Affairs 1972 An Annotated Bibliography of Young People's Fiction of American Indians. Washington, D.C.: Office of Education Programs, United States Bureau of Indian Affairs. 1973 An Annotated Bibliography of Young People's Books on American Indians. Washington, D.C.: Office of Education Programs, United States Bureau of Indian Affairs. University Microfilms International 1977 North American Indians: A Dissertation Index. Ann Arbor, MI: University Microfilms. University of Montana, Native"American Studies"Program n.d. "Definitions." Mimeographed copies. n.d. "Guide to self-awareness experience." Xeroxed copy. Vander Zanden, James W. 1983 American Minority Relations, Fourth Edition. New York: Alfred A. Knopf, Inc. Van Til, William 1975 Prejudiced—How Do People Get That Way? New York: Anti- (1970) Defamation League of B'nai B'rith. Vaughan, Graham M. 1963 "Concept formation and the development of ethnic awareness," The Journal of Genetic Psychology 103:93-103. Walcutt, Charles C. 1966 Man's Changing Mask. Minneapolis: University of Minnesota Press^ Walker, P.A. 1971 "The effects of hearing selected children's stories that portray blacks in a favorable manner on the racial attitudes of groups of black and white kindergarten children." University of Kentucky, unpublished Ph.D. dissertation. Waples, Douglas, Bernard Berelson, and Franklyn R. Bradshaw 1940 What Reading Does to People. Chicago: University of Chicago Press. 215 Wasserman, Paul 1976 Ethnic Information Sources of the United States: A Guide to Organizations, Agencies, Foundations, Media, Commercial and Trade Bodies, Government Programs. . . Detroit: Gale Research Company. Watts, W.A. 1967 "A relative persistence of opinion change induced by active compared to passive participation," Journal of Personality and Social Psychology 5:4-15. Webb, Eugene J., Donald T. Campbell, Richard D. Schwartz, and Lee Sechrest 1966 Unobtrusive Measures; Nonreactive Research in Social Science. Chicago: Rand McNally. Weeks, A.D. 1927 "How stereotypes arise," in K. Young (ed.), Source Book for Social Psychology (New York: Alfred A. Knopf), pp. 440-443. Weigel, Russell H., James W. Loomis, and Matthew J. Soja 1980 "Race relations on prime time television," Journal of Personality and Social Psychology 39> 5:884-893. Weiss, Walter 1961 "The effect of a communication on attitude change and scale judgments," Journal of Abnormal and Social Psychology 62:133-140. 1968 "Effects of the mass media of communication," in Lindzey and Aronson (eds.), The Handbook of Social Psychology, Second Edition (Menlo Park, CA: Addison-Wesley Publishing Company, Inc.), pp. 77-195. Westphal, R. 1977 The Effects of a Primary-Grade Level Inter and Ethnic Curriculum on Racial Prejudice. San Francisco: R and E Research Associates. Whilton, John B. and Arthur Larson 1964 Propaganda: Towards Disarmament in the War of Words. Doobs Ferry, NY: Oceana Publications for the World Rule of Law Center, Duke University. 216 White, Ralph K. 1947 "Black boy: a value analysis," Journal of Abnormal and Social Psychology 42 :_44JD-461_. Wilhoit, G. Cleveland and Harold de Bock (Eds.) 1980 Mass Communication Review Yearbook. Volume 1. Beverly Hills: Sage Publications. Williams, J.E. 1966 "Connotations of racial concepts and color names," Journal of Personality and Social Psychology 3:531-540. Williams, J.E. and C.D. Edwards 1969 "An exploratory study of the modification of color and racial concept attitudes in preschool children," Child Development 40:737-750. Williams, J.E. and J.K. Roberson 1967 "A method for assessing racial attitudes in preschool children," Educational and Psychological Measurment 27: 671-689. Wilson, H.W. 1956 Children's Catalog, Ninth Edition. New York: H.W. Wilson Company. 1957 Standard Catalog for High School Libraries, Seventh Edition. New York: H.W. Wilson Company. 1960 Fiction Catalog, Seventh Edition. New York: H.W. Wilson Company. 1961 Children's Catalog, Tenth Edition. New York: H.W. Wilson Company. 1962 Standard Catalog for High School Libraries, Eighth Edition. New York: _ H.W. Wilson Company^ 1965 Junior High School Library Catalog, First Edition. New York: H.W. Wilson Company. 1966 Children's Catalog, Eleventh Edition. New York: H.W. Wilson Company. 1967 Standard Catalog for High School Libraries, Tenth Edition. New York: H.W. Wilson Company. 217 Wilson, H.W. 1968 Public Library Catalog, Fifth Edition. New York: H.W. Wilson Company. 1970 Junior High School Library Catalog, Second Edition. New York: H.W. Wilson Company. 1971a Children's Catalog, Twelfth Edition. New York: H.W. Wilson Company. 1971b Fiction Catalog, Eighth Edition. New York: H.W. Wilson Company. 1972 Standard Catalog for High School Libraries, Tenth Edition. New York: H.W. Wilson Company. 1973 Public Library Catalog, Sixth Edition. New York: H.W. Wilson Company. 1975a Fiction Catalog, Ninth Edition. New York: H.W. Wilson Company. 1975b Junior High School Library Catalog, Third Edition. New York: H.W. Wilson Company. 1976 Children's Catalog, Thirteenth Edition. New York: H.W. Wilson Company. 1977 Standard Catalog for High School Libraries, Eleventh Edition. New York: H.W. Wilson Company. 1978a Public Library Catalog, Seventh Edition. New York: H.W. Wilson Company. 1978b Standard Catalog for High School Libraries, Supplement to the Eleventh Edition. New York: H.W. Wilson Company. 1979 Standard Catalog for High School Libraries, Supplement to the Eleventh Edition. New York: H.W. Wilson Company. 1980a Fiction Catalog, Tenth Edition. New York: H.W. Wilson Company. 1980b Junior High School Library Catalog, Fourth Edition. New York: H.W. Wilson Company. 218 Wilson, H.W. 1980c Public Library Catalog, Supplement to the Seventh Edition. New York: H.W. Wilson Company. 1 980d Standard.Catalog Ior_High School Libraries, Supplement to the Eleventh Edition. New York: H.W. Wilson Company. 1981a Children's Catalog., Fourteenth .Edition. New York: H.W. Wilson Company. 1981b Fiction Catalog, Supplement to the Tenth Edition. New York: H.W. Wilson Company. 1981c Junior High..School Xib.rary .Catalog, Supplement to the Fourth Edition. New York: H.W. Wilson Company. 1981 d Public, Library Catalog, Supplement to the Seventh Edition. New York: H.W. Wilson Company. 1981 e Standard Catalog for High School Libraries, Supplement to the Eleventh Edition. New York: H.W. Wilson Company. 1982 Children's Catalog, Supplement to the Fourteenth Edition. New York: H.W. Wilson Company. Wilson, Raymond 1980 "Native Americans in college textbooks," The Indian Historian 13, 2(June):44-47. Woodbury., Marda 1979 Selecting Materials for Instruction: Issues and Policies. Littleton, CO: Libraries Unlimited, Inc. 1980a Selecting Materials for Instruction: Media and the Curriculum. LittletonCO: Libraries Unlimited, ...Inc. 1980b Selecting Materials for Instruction: Subject Areas and Implementation.Littleton., CO: Libraries Unlimited, Inc. Woods, Marilyn 1979 Proceedings on^Discrimination Against Indians in Montana: Towards New Directions in Human Relations. Great Falls, MT: College of Great Falls. 219 Wright, Lyle 1939 "Propaganda in early American fiction," Papers of the Bibliographical Society of America XXX1II:98-106. Yancey, Anna V. and Jane M. Singh 1975 "A study of racial attitudes in white first grade children," Elementary English 52, 5(May)*.734-736. Yee, A.H. and M.J. Fruth 1971 "Do black studies make a difference in ghetto children's achievement and attitudes?" Journal of Negro Education 42:33. Yetman, Norman R. and C. Hoy Steele 1975 Majority and Minority: The Dynamics of Racial and Ethnic Relations, Second Edition. Boston: Allyn and Bacon, Inc. Young, E.F. 1925 "What is race prejudice?" Journal of Applied Sociology 10:135-140. Young, Kimball 1927a "Personality and the mental patterns of the group," in K. Young (ed.), Source Book for Social Psychology (New York: Alfred A. Knopf), pp. 419-421. 1927b "Prejudice: an outgrowth of subjective environment," in K. Young (ed.), Source Book for Social Psychology (New Young: Alfred A. Knopf), pp. 482-483. 1927c "Types of prejudice," in K. Young (ed.), Source Book for Social Psychology (New York: Alfred A. Knopf), pp. 505-507. Zeligs, Rose 1936 "Racial attitudes of children as expressed by their concepts of races," Sociology and Social Research 21:361-371. 1938 "Tracing racial attitudes through adolescence," Sociology and Social Research 23:45-54. 1952 "Children's concepts and stereotypes of Norwegian, Jew, Scotch, Canadian, Swedish, and American Indian," Journal of Educational Research 45, 5(January):349-360. APPENDIX A 221 DISSERTATIONS Behen, Dorothy M.F. 1951 "The captivity story in American literature, 1577-1826: an examination of written reports in English, authentic and ficticious, of experiences of white men captured by the Indians north of Mexico." Chicago: University of Chicago, unpublished Ph.D. dissertation. Blaine, Harold A.H. 1936 "The frontiersman in American prose fiction: 1800-1860." Cleveland: Case Western Reserve University, unpublished Ph.D. dissertation. Bleasby, George 1951 "The frontier in Cooper's Leatherstocking Tales." Pittsburgh: University of Pittsburgh, unpublished Ph.D. dissertation. Brenzo, Richard Allen 1973 "Civilization against the savage: the destruction of Indians in American novels 1823-1854." Milwaukee: University of Wisconsin, unpublished Ph.D. dissertation. Buntin, Arthur Roy 1961 "The Indian in American literature, 1680 to 1760." Seattle: University of Washington, unpublished Ph.D. dissertation. Cauthers, Janet Helen 1974 "The North American Indian as portrayed by American and Canadian historians, 1830-1930." Seattle: University of Washington, unpublished Ph.D. dissertation. Clark, Harold Edward 1955 "Fenimore Cooper's Leatherstocking Tales: a problem in race." Indiana, unpublished Ph.D. dissertation. Clark, Robert Thomas, Jr. 1932 "Herder and the noble savage." Stanford: Stanford University, Ph.D.dissertation. Pub: Herder: His Life and Thought. Berkeley: University of California Press, 1955. 222 Clausen, Dennis Monroe 1971 "The search for identity in Cooper's 'Leatherstocking Tales': nature as the determinant of American character." Riverside: University of California, unpublished Ph.D. dissertation. Coen, Rena Neuman 1969 "The Indian as the noble savage in nineteenth century American art." Duluth: University of Minnesota, unpublished Ph.D. dissertation. Crane, Fred Arthur 1952 "The noble savage in America, 1815-1860: concepts of the Indian, with special reference to the writers of the Northwest." New Haven: Yale University, unpublished Ph.D. dissertation. Deflyer, Joseph Eugene 1974 "Partition Theory: patterns and partitions of consciousness in selected works of American and American Indian authors." Lincoln: University of Nebraska, unpublished Ph.D. disser tation. Dippie, Brian William 1970 "The vanishing American: popular attitudes and American Indian policy in the nineteenth century." Austin: University of Texas, unpublished Ph.D. dissertation. Evans, James Leroy 1967 "The Indian savage, the Mexican bandit, the Chinese heathen— three popular stereotypes." Austin: University of Texas, unpublished Ph.D. dissertation. Gaston, Edwin Willner 1959 "The early novel of the Southwest: a critical study." Texas Tech., unpublished Ph.D. dissertation. Green, Rayna Diane 1973 "The only good Indian: the image of the Indian in American vernacular culture." Bloomington: Indiana University, unpublished dissertation. Gribskov, Margaret Elsie T.H. 1973 "A critical analysis of textbook accounts of the role of Indians in American history." Eugene: University of Oregon, unpublished Ph.D. dissertation. 223 Hardman, Marion Payzant 1939 "Terror in American prose fiction to 1835." Duluth: University of Minnesota, unpublished Ph.D. dissertation. Harris, John West, Jr. 1928 "The glorification of American types in American literature from 1775 to 1825." North Carolina, unpublished Ph.D. dissertation. Hazard, Lucy Lockwood 1925 "The frontier in American literature." Berkeley: University of California, Ph.D. dissertation. Pub: The Frontier in American Literature. New York: Thomas Y. Crowell, 1927. Heilman, Robert Bechtold 1935 »xhe English novel, 1760-1800, and the American Revolution." 2 vols. Howard University, Ph.D. dissertation. Pub: America in English Fiction 1760-1800: The Influences of the American Revolution. University of Louisiana Press, 1937. House, Kay Seymour 1963 "James Fenimore Cooper's American characters." Stanford: Stanford University, unpublished Ph.D. dissertation. Irving, Donald Charles 1969 "James Fenimore Cooper's alternatives to the Leatherstocking hero in the frontier romances." Bloomington: Indiana University, unpublished Ph.D. dissertation. Jones, Henry Broadus 1924 "The death song of the 'noble savage': a study in the idealization of the American Indian." Chicago: University of Chicago, unpublished Ph.D. dissertation. Lewis, Merrill Embert 1968 "American frontier history as literature: a study in the historiography of George Bancroft, Frederick Jackson Turner and Theodore Roosevelt." Salt Lake City: University of Utah, unpublished Ph.D. dissertation. Lloyd, Peter 1975 "The emergence of a racial prejudice towards the Indians in seventeenth-century New England, some notes on an explanation." Ohio State University, unpublished Ph.D. dissertation. 224 Macphee, Laurence Edward 1967 "Patterns of action and imagery in the Leatherstocking Tales." Rutgers University, unpublished Ph.D. dissertation. Marovitz, Sanford Earl 1968 "Frontier conflicts: villain, outlaws, and Indians in selected 'Western* fiction 1799-1860." Duke University, unpublished Ph.D. dissertation. McKay, James 1980 "Hamlin Garland's 'The Book of the American Indian: A View from the Inside.'" Seattle: University of Washington, unpublished Ph.D. dissertation. McLaurin, Nancy Delia 1958 "A study of the southern frontier in prose fiction prior to 1868." South Carolina, unpublished Ph.D. dissertation. Meade, James Gordon 1971 "The 'Westerns' of the East: narratives of Indian captivity from Jeremiad to Gothic novel." Northwestern University, unpublished Ph.D. dissertation. Meyer, William Claus 1972 "The development of myth in the Leatherstocking Tales of James Fenimore Cooper." Ball State University, unpublished E.Ed, dissertation. Moore, Jack Bailey 1963 "Native elements in American magazine short fiction, 1741- 1800." North Carolina, unpublished Ph.D. dissertation. Morgan, Paul 1954 "The treatment of the Indian in Southwestern literature since 1915: a study in primitivism." Austin: University of Texas, unpublished Ph.D. dissertation. Movalli, Charles Joseph 1972 "Pride and prejudice: James Fenimore Cooper's frontier fiction and his social criticism." University of Connecticut, unpublished Ph.D. dissertation. Nigliazzo, Marc Anthony 1973 "Faulkner's Indians." Albuquerque: University of New Mexico, unpublished Ph.D. dissertation. 225 Paine, Gregory Lansing 1924 "James Fenimore Cooper as an interpreter and critic of America." Chicago: University of Chicago, Ph.D. dissertation. Pub: "The Indians of the Leatherstocking Tales," Studies in Philology XXIII(1926):1 6-39- Robinson, Linda Jane Rookwood 1968 "Henry W. Longfellow's 'Hiawatha': an American epic." New Orleans, LA: Tulane University, unpublished Ph.D. dissertation. Russell, Jason Almus 1924 "The Indian in American literature (1775-1875)." Cornell University, Ph.D. dissertation. Pub: "Thoreau, the interpreter of the real Indian," Queens University Quarterly XXXV, 1(1927):37—48; "Influence of Indian confederations on the American colonies," Journal of American History XXII(1929):27-124; "Narratives of Indian captivity," Education, LI, 2(1930:23-34. Schramm, Wilbur Lang 1932 "Studies in the longer narrative verse of America, 1775-1860." Iowa, Ph.D. dissertation. Pub: "Hiawatha and its predessors," Philology Quarterly XI, 4(1932):321-343. Shames, Priscilla 1969 "The long hope: a study of American Indian, stereotypes in American popular fiction, 1890-1950." Los Angeles: University of California, unpublished Ph.D. dissertation. Slotkin, Richard Sidney 1967 "Emergence of a myth: John Filson's Daniel Boone narrative and the literature of the Indian Wars, 1638-1848." Brown University, unpublished Ph.D. dissertation. Stein, Paul 1968 "In defense of truth: structure and theme in James Fenimore Cooper's novels." Cleveland: Case Western Reserve University, unpublished Ph.D. dissertation. Sullivan, Sherry Ann 1979 "The Indian in American fiction, 1820-1850." University of Toronto, unpublished Ph.D. dissertation. 226 Thigpen, Buelah Virginia 1981 "The Indians of the Leatherstocking Tales: a study of the noble and the ignoble savage." East Texas State University, unpublished Ph.D. dissertation. Tillson, Merl W. 1950 "The frontiersman in American drama: an analytical study of characters and plays reflecting the phenomenon of westward expansion." Denver: University of Denver, unpublished Ph.D. dissertation. Van Der Beets, Richard 1973 "The Indian captivity narrative: an American genre." Fresno, CA: University of the Pacific, unpublished Ph.D. dissertation. Van Keuren, M. Louise 1981 "American Indian responses and reactions to the colonists as recorded in seventeenth and eighteenth century American literature." University of Delaware, unpublished Ph.D. dissertation. Van Norman, Carrie Elter 1941 "An investigation of the concept of war in historical fiction written for children." Stanford: Stanford University, unpublished Ed.D. dissertation. Wasserman, Maurice M. 1954 "The American Indian as seen by the seventeenth century chroniclers." University of Pennsylvania, unpublished Ph.D.- dissertation. THESES Abbott, Hazel Belle 1924 "And; what of the Indian?, His literature and his treatment in our dramatic literature." New York: Columbia University, unpublished Master's thesis. Abbott, Rosalie 1964 "Imagery and theme in Willa Cather's Catholic novels." North Carolina, unpublished Master's thesis. 227 Agnew, Miriam 1937 "A contrast of Parkman's The Oregon Trail to similar works in the same period of the early West." Denver: University of Denver, unpublished Master's thesis. Anderson, Barbara Louise 1955 "The Southwestern Indian in contemporary American liter ature, 1920-1955; an annotated booklist for adolescent readers." Emporia: Kansas, unpublished Master's thesis. Arnold, Dovie Mae 1938 "American Indians in fiction." Austin: University of Austin, unpublished Master's thesis. Baird, Helen Delphine 1939 "The use and accuracy of accounts of the Lewis and Clark expedition in children's literature published from 1806 through December 1938." New York: Columbia University, unpublished Master's thesis. Ballotti, Geno A. 1955 "The Southwest Indian in fiction." Laramie: University of Wyoming, unpublished Master's thesis. Barre, Elizabeth Freeman 1941 "A study of the Indian in William Gilmore Simms' novels and short stories." South Carolina, unpublished Master's thesis. Bates, Lorna Doone 1934 "The noble savage in the works of Thoreau." Iowa, unpublished Master's thesis. Blessing, Edith Stephens 1941 "The North American Indian as presented in early chronicles." Albuquerque: University of New Mexico, unpublished Master's thesis. Blizzard, William Leland 1935 "The frontier press of Colorado in the Ute Indian uprising of 1879. A case study on public opinion." Kansas, unpublished Master's thesis. Bonney, India H. 1936 "Mary Austin's interpretation of the American Indian." Louisiana, unpublished Master's thesis. 228 Boyd, Jimmye Thomas 1962 "The early Kentucky frontier^ as represented in novels by Robert Montgomery Bird, James Lane Allen, Winston Churchill and Elizabeth Madox Roberts." Texas Christian University, unpublished Master's thesis. Bradley, Viola McKee 1944 "The contrast of Cooper's The Last of the Mohicans and Simms' The Yemassee." Alabama Poly, unpublished Master's thesis. Bryant, Edna 1935 "The Pocahontas theme in American literature." North Carolina, unpublished Master's thesis. Calhoun, Robert Francis 1959 "Indian characterization in Brown, Cooper and Simms." Virginia, unpublished Master's thesis. Cebollero, Carmen Rita 1963 "The noble savage: a comparative study of Melville and Gaugin." New Orleans, LA: Tulane University, unpublished Master's thesis. Collett, Beryl Bishop 1922 "The character of the California Indian as portrayed in California literature." Stanford: Stanford University, unpublished Master's thesis. Corkern, Florence R. 1940 "An evaluation of Southwest books for primary children." Austin: University of Texas, unpublished Master's thesis. Craig, Sarah Helen 1933 "Comparative, study of the novels, dealing with the Southwest, of Gertrude Atherton and Mary Austin." Ohio State University, unpublished Master's thesis. Crelly, Frances Lois 1955 "What a newspaper reveals about a tri-ethnic community." Boulder: University of Colorado, unpublished Master's thesis. Cronyn, George William 1917 "The frontier in American literature prior to the Civil War." New York: Columbia University, unpublished Master's thesis. 229 Dabney, E. Jeanne 1963 "Romantic and realistic conceptions of the Indian in American literature." Laramie: University of Wyoming, unpublished Master's thesis. Duke, Mary Fender 1942 "The Indian in American literature since 1920." Southern Methodist, unpublished Master's thesis. Dusold, Lenore 1970 "A study of the portrayal of the American Indian in selected children's fiction." New York: Long Island University, unpublished Master's thesis. Engel, Grace Margaret 1937 "Pocahontas in American literature." New York: Columbia University, unpublished Master's thesis. Farmer, Freda M. 1953 "The changing attitudes of American authors toward the Indians." Ball State University, unpublished Master's thesis. Field, Nadine Eaton 1957 "Portraiture of the American Indian in the novels of Cooper, Bird, and Simms." Tulsa: Oklahoma State University, unpublished Master's thesis. Finkelstein, Naomi J. 1966 "Rodolphe Bresdin and 'The Noble Savage.'" New York: Columbia University, unpublished Master's thesis. Fox, Velda 1927 "The development of the Pocahontas story in American literature, 1607-1927." State University of Iowa, unpublished Master's thesis. Frost, Melvin Jesse 1965 "Saints and savages: the Puritan's conception of the Indians." Wisconsin, unpublished Master's thesis. Garwin, Elaine Judith 1961 "An analysis of the treatment of the American Indian in juvenile fiction." Chicago: University of Chicago, unpublished Master's thesis. 230 Gentry, Dorothy Delores 1954 "The treatment of the American Indian in the American novel, 1929-1953." Washington, D.C.: Howard University, unpublished Master's thesis. Gill, Eloise Wise 1935 "The changing attitude toward the Indian in American literature." Louisiana State University, unpublished Master's thesis. Harms, Tina Cecelia 1927 "A dictionary of characters in Cooper's Leatherstocking Tales." Kansas, unpublished Master's thesis. Hartung, Jean Buehler 1944 "An investigation of the literature concerning the Spanish missions of California and Southwest." Denver: University of Denver, unpublished Master's thesis. Herbert, Jeanne Clark 1961 "The growth of realism in the treatment of the Southwestern Indian in fiction since 1900." Tempe: University of Arizona, unpublished Master's thesis. Hoagland, Robert John 1930 "A comparison of the Indians of James Fenimore Cooper and those seen by Indian agents, travellers, and scientists." New York: Columbia University, unpublished Master's thesis. Hopcraft, Margaret Lindsay 1943 "Attitudes towards the Indian as found in American literature, 1700-1800." Albuquerque: University of New Mexico, unpublished Master's thesis. Huntington, Margaret Alice 1924 "The treatment of the Indian in American literature previous to 1860." Illinois, unpublished Master's thesis. Jacob, Isabelle Griffith 1937 "The West in American literary magazines, 1851-1883." Boulder: University of Colorado, unpublished Master's thesis. Jones, Lucy Thweatt 1925 "A comparative study of the American Indian in prose fiction, from Edgar Huntley to Ramona." Virginia, unpublished Master's thesis. 231 Jordan, Howell H., Jr. 1966 "The changing character of Natty Bumppo in the Leather- stocking Tales." New York: Columbia University, unpublished Master's thesis. Kahn, Dorothy 1955 "An analysis of minority groups." Chicago: University of Chicago, unpublished Master's thesis. Kennedy, Lawrence Michael 1967 "The Colorado press and the red man; local opinion about Indian affairs, 1859-1870." Denver: University of Denver, unpublished Master's thesis. Lemons, William Everett 1950 "John G. Neihardt's conception of the Plains Indian." Boulder: University of Colorado, unpublished Master's thesis. Meier, Alice Emma 1915 "The American Indian in German poetry." Illinois, unpublished Master's thesis. Monroe, Barbara Ann 1956 "An investigation of fictional books for children which deal with the American Indian in the United States." Cornell University, unpublished Master's thesis. Montgomery, Esther D. 1927 "A study of the Indian in American fiction, 1820-50." Iowa, unpublished Master's thesis. Morris, Mabel Marie 1926 "The democratic influences in Charles Brockden Brown's Treatment of the Indian." State University of Iowa, unpublished Master's thesis. Napier, Georgia Pierce 1970 "A study of North American Indian characters in twenty selected children's books." Little Rock: University of Arkansas, unpublished Master's thesis. Nasci, Gaetano F. 1963 "The image of the American Indian in the visual arts to 1800." Syracuse: University of New York, unpublished Master's thesis. 232 Nelson, Horatia Dodson 1932 "Indian characters and customs as portrayed in the novels of James Fenimore Cooper." Ohio State University, unpublished Master's thesis. Paine, Gregory Lansing 1920 "Realism in the novels of James Fenimore Cooper." Chicago: University of Chicago, unpublished Master's thesis. Paxson, Frances D. 1927 "The Indian in American drama." State University of Iowa, unpublished Master's thesis. Prewitt, Dorothy E. 1923 "North American Indians in English poetry." Iowa, published Master's thesis. Rader, Alice Dresser 1928 "The American Indian in the American novel and short story." Kansas, unpublished Master's thesis. Richmond, Gordon Donald 1962 "A study of Minnesota newspaper coverage of the Sioux uprising from August 20 to October 3, 1862." Duluth: University of Minnesota, unpublished Master's thesis. Robertson, Jennie R. Gould 1951 "The picture of the Indian in fiction with South Dakota setting." South Dakota, unpublished Master's thesis. Scott, Elizabeth 1946 "A study of the chronicles of Indian captivity in the Southwest." Austin: University of Texas, unpublished Master's thesis. Smith, Garland Garvey 1923 "The Indian in American fiction before 1850." Southern Methodist, unpublished Master's thesis. Sullivan, Cecille Gerard 1925 "The Indian as treated by Cooper and Simms." New Haven: Yale University, unpublished Master's thesis. 233 Timmons, David R. 1970 "Elements of prejudice toward Negroes and Indians as found in daily newspapers of Seminole County, Oklahoma, 1950-1959." Norman: University of Oklahoma, unpublished Master's thesis. Wells, Helen G. 1942 "Navaho of the painted desert in American fiction." Los Angeles: University of Southern California, unpublished Master's thesis. White, Clara Mae 1946 "A study of James Fenimore Cooper's social teachings on American democratic culture." Texas A & I, unpublished Master's thesis. Wilson, Marjorie H. 1966 "A comparative evaluation of the material on the American Indian as presented in current high school American history textbooks." Northern State College, South Dakota, unpublished Master's thesis. Young, William Donald 1952 "The Indian character in American plays, 1808-1860." Stanford: Stanford University, unpublished Master's thesis. APPENDIX B REFERENCES ON NATIVE AMERICANS Abel, Midge B. 1973 "American Indian life as portrayed in children's literature, Elementary English 50, 2(February):202-208. Allen, Paul G. 1973 "Symbol and structure in Native American literature," College Composition and Communication 24(October):267-270. 1976 "Sacred Hoop: a contemporary Indian perspective on American Indian literature," Cross Currents 26(Summer): 144-163. American Association of School Librarians 1970a "Multi-ethnic media: selected bibliographies," School Libraries (Winter):49-57. 1970b "Multi-ethnic media: selected bibliographies," School Libraries (Summer):44-47. 1971 "Multi-ethnic media: selected Bibliographies," School Libraries (Summer):42-48. 1972 "Multi-ethnic media: selected bibliographies," School Libraries (Summer):47-54. American Indian Historical Society 1968 "Common misconceptions about American Indians." Mimeo graphed paper, San Francisco. American Library Association 1970 "Guidelines for the evaluation of Indian materials for adults," adapted from American Indians: An Annotated Bibliography of Selected Library Resources. Minneapolis: University of Minnesota Library Services Institute for Minnesota Indians. 1972 "American Indians," Booklist 69(October):176—181 . Annelli, Catherine M. 1978 "American Indian people and children's resources," in Esther R. Dyer (comp.), Cultural Pluralism and Children's Media (Chicago: The American Library Association), pp. 13-26. 236 Ashliman, D.L. 1977 "The American Indian in German travel narratives and literature," Journal of Popular Culture 10, 4(Spring): 833-839- Axtell, James and William C. Sturtevant 1980 "The unkindest cut, or who invented scalping?" The William and Mary Quarterly, Series 3, 37, 3(July): 452-472. Baden, J., et. al. 1981 "Myths, admonitions and rationality: the American Indian as a resource manager," Economic Inquiry 19(January): 132-143. Bahr, Howard M. and Bruce A. Chadwick, 1972 "Contemporary perspectives on Indian Americans: a review essay," Social Science Quarterly 53:606-618. 1974 "Conservatism, racial intolerance and attitudes toward racial assimilation among whites and American Indians," Journal of Social Psychology 94:45-56. Bahr, Howard M., Bruce A. Chadwick, and Joseph H. Strauss 1972 "Discrimination against urban Indians in Seattle," The Indian Historian 5, 4(Winter):4-11. Baird, Bruce 1976 "Is it worth the climb?: programing efforts for Native Americans." Public Telecommunications Review 4, 4(July/ August):50-52. Baker, Donald G. 1972 "Color, culture and power: Indian-white relations in Canada and America," Canadian Review of American Studies 3:3-20. Ballard, Louis W. 1969 "Cultural differences: a major theme in cultural enrichment," The Indian Historian 2, 1(Spring):4-7. Barker, James N. 1808 The Indian Princess or La Belle Sauvage. Philadelphia. Bean, Lowell John 1969 "The language of stereotype distortion inaccuracy," The Indian Historian 2, 3(Fall):6-11. 237 Beaver, H. 1978 "Transcendental savage," Times Literary Supplement 3992:1115. Beer, David F. 1969 "Anti-Indian sentiment in early colonial literature," The Indian Historian 2, 1(Spring):29-33, 48. Berkman, Brenda 1976 "The vanishing race: conflicting images of the American Indian in children's literature, 1880-1930," North Dakota Quarterly XLIV(Spring):31-40. Bernstein, G.M. 1979 "Robert Altman's Buffalo Bill and the Indians; or Sitting Bull's history lesson: a self-portrait in celluloid," Journal of Popular Culture 13(Summer):17-25. Berry, Brewton 1960 "The myth of the vanishing Indian," Phylon 21:51-57. Bevis, William 1974 "American Indian verse translations," College English, (March):693-702. Bierhorst, J. 1975 "American Indian verbal art and the role of the literary critic," Journal of American Folklore 88(October):401-408. Black, Robert Abner 1965 "A content analysis of 81 Hopi Indian chants." Indiana University, unpublished Ph.D. dissertation. Blank, Ruth (comp.) 1977 What Shall Our Children Read? A Selected Bibliography of Native American Literature for Young People. San Jose, CA: San Jose Indian Center, Inc. Boettler, Fred W. 1970 "Topics commonly ignored," The Indian Historian 3, 3 (Summer):22. Bonney, Rachel A. 1978 "'Indian Reservation' (life today on the Northern Plains)," Simulation and Games 9, 2(June):221-226. 238 Bosmajian, Haig A. 1973 "Defining the 'American Indian': a case_study in the language suppression," The Speech Teacher 22, 2(March): 89-99. Boutwell, Richard C., et. al. 1973 ""Red apples," Journal of American Indian Education 12, 2:11-14. Howd, David 1978 "Eight prevalent myths about Indian education," Education Canada 18, 2(Summer):4-7, 47. Brandon, William 1971 "American Indian literature," The Indian Historian 4, 2 (Summer):53-55. Brown, Richard William 1978 "Characteristics and concepts of American Indians in children's fictional literature published between 1963 and 1973." Temple University, unpublished Ed.D. dissertation. Brumble, H. David 1980 "Anthropologists, novelists and Indian sacred materials," The Canadian Review of American Studies 11, 1(Spring): 31-48. Cane, Suzanne S., et. al. 1970 Selected Media About the American Indian for Young Children, K-3. Boston: Massachusetts State Department of Education. Carleton, Phillip D. 1943- "The Indian captivity," American Literature 15(March 1943— 1944 January 1944):169-180. Carlson,"Tlewis ~H~.~ arid""George A. Colburn (Eds.I 1972 In Their Place: White America Defines Her Minorities 1850-1950. New York: John Wiley and Sons, Inc. Caselli, Ron " ~ 1970 "Historical repression and the Native American," The Indian Historian 3, 4(Fall):44-45. Cawelti, John G. 1974 "Savagery, civilization and the Western hero," in Jack Nachbar (ed.), Focus on the Western (Englewood Cliffs, NJ: Prentice-Hall, Inc.), pp. 57-63. 239 Chamberlin, J.E. 1975 The Harrowing of Eden; White Attitudes Toward Native Americans. New York: Seabury Press. Clemmer, Richard 0. 1978 "Take three steps backward: a review," The Indian Historian 11, 3(Summer):47-57. Coldevin, Gary C. 1976 "Some effects of frontier television in a Canadian Eskimo community," Journalism Quarterly 53, 1(Spring):34-39• Collier, Peter 1970 "The red man's burden," Ramparts 8, 8(February):26-38. Conway, Thomas G. 1972 "Public interest in the Indian," The Indian Historian 5, 1(Spring):37-44. Cook, Liz 1977 "American Indian literature in servitude," The Indian Historian 10, 1 (Winter):3-6. Costo, Rupert 1977 "Fact from fiction," The Indian Historian 10, 1(Winter): 31-36. Deegan, Dorothy Yost 1951 The Stereotype of the Single Woman in American Novels, (1969) A Social Study with Implications for the Education of Women. New York: Octagon Books. Doerr, Donald J. 1972 "On literature," The Indian Historian 5, 1(Spring):55. Dugan, Kathleen 1976 "Recent books, on the Indian in America," Cross Currents 26(Summer):196-201. Dyer, Esther R. (Comp.) 1978 Cultural Pluralism and Children's Media. Chicago: American Library Association. Easy, P. 1978 "Treatment of American Indian materials in contemporary American poetry," Journal of American Studies 12(April): 81-93. 240 Elkin, Frederick 1974 "The psychological appeal for children of the Hollywood B_Western," in Jack Nachbar (ed.)j Focus on the Western (Englewood Cliffs, NJ: Prentice-Hall, Inc.), pp. 73-77. Etulain, Richard W. 1974 "Cultural orgins of the Western," in Jack Nachbar (ed.), Focus on the Western (Englewood Cliffs, NJ: Prentice-Hall, Inc.), pp. 19-24. Falkenhagen, Maria, Carole Johnson, and Michael A. Balasa 1973 "The treatment of Native Americans in recent children's literature,11 Integrated Education 11 (July):58-59. Feagin, Joe R. and Randal Anderson 1973 "Intertribal attitudes among Native American youth," Social Science Quarterly 54:117-131. Fisher, Alice Poindexter 1979 "The transportation of tradition: a study of Zitkala Sa and Mourning Dove, two transitional American Indian writers." City University of New York, unpublished Ph.D. dissertation. Fisher, Laura 1974 "All chiefs, no Indians: what children's books say about American Indians," Elementary English 51, 2(February): 185-199. Flaste, Richard __ 1974 "Why doesn't anybody ever play Jew?" The Missoulian, Saturday, October 12, 1974, p. 8A. (From the New York Times News Service.) Folsom, James K. 1974 "Westerns as social and political alternatives," in Jack Nachbar (ed.). Focus on the Western (Englewood Cliffs_, NJ: Prentice-Hall, Inc.), pp. 81-83. Forst, John 1935 "Indien und die Deutsche^literatur von 1900 bis 1923." New York, Ph.D. dissertation. Pub: Leipzig: R. Noske, 1934. Gabriel, Ralph H. 1929 The Pageant of America: The Lure of the Frontier: A Story of Race Conflict. New Haven: Yale University Press. 241 Gere, Anne R. 1976 "New thinking for old stereotypes," Elementary English 53, 3(March):245-247. Grose, Burl Donald 1979 "'Here come the Indians': an historical study of the representations of the Native American upon the North American stage, 1808-1969." Columbia: University of Missouri, unpublished Ph.D. dissertation. Gura, P.F. 1977 "Thoreau's Maine woods Indians: more representative men," American Literature 49(November):366-384. Haberly, D.T. 1976 "Women and Indians: The Last of the Mohicans and the captivity tradition," American Quarterly 28(Fall):431-443. Halpin, Glennelle, Gerald Whiddon, and Thomas Whiddon 1980 "The relationship of perceived parental behaviors to locus of control and self-esteem among American Indian and white children," The Journal of Social Psychology 3, 2(August):189-195. Hamm, V.M. 1975 "Greeks and Indians: a study in mythic syncretism," Thought 50(December):351-366. Harriman, Willis W. 1928 "Our first American dramatist," The Granite Monthly (Manchester, N.H.), April-May, 1928 (Vol. 60, nos. 4 and 5). Harris, J. 1978 "Moulting of the plumed serpent: a study of the relationship between the novel and three contemporary tales," Modern Language Quarterly 39(June):154-168. Hartman, Hedy 1976 "A brief review of the Native American cinema," The Indian Historian 9, 3(Summer):27-29- Henderson, B. 1980- "Searchers: an American dilemma," Film Quarterly 34 1981 (Winter 80/81):9-23. 242 Heuterman, T.H. 1980 "Racism in frontier journalism: a case study," Journal of the West 19(April):46-50. Herbst, Laura 1975 "That's one good Indian: unacceptable images in children's novels," Top of the News 31(June):192-198. Hill, Elbert Ray 1973 "Tales and trials: children's stories and cultural alienation among the Winnebago." Lincoln: University of Nebraska, unpublished Ph.D. dissertation. Hirschfelder, Arlene B. 1975 "The treatment of Iroquois Indians in selected American history textbooks," The Indian Historian 8, 2(Fall):32-39. Hofdahl, Grayce 1971 "Is self-determination an answer to problems of American Indians?" The Delta Kappa Gamma Bulletin 33, 2(Winter): 54-58. Hooten, Richard J. 1973 "Race, skin color, and dress as related to the American Indian stereotype," Dissertation Abstracts 33A(January): 3822A. Horsman, Reginald 1975 "Scientific racism and the American Indian in the mid- nineteenth century," American Quarterly 2:152-168. Howard, James H. 1975 "The culture-area concept: does it diffract anthropological light?" The Indian Historian 8, 1(Spring):22-26. Hoxie, F.E. 1979 "Red man's burden," Antioch Review 37(Summer):326-342. Hymes, D. 1977 "Discovering oral performance and measured verse _in American Indian narrative," New Literary History 8(Spring): 431-457, 521-534. Indian Historical, Commentary 1977 "An Indian psychiatrist looks at racism," The Indian Historian 10, 4(Fall):52-53. 243 Inkster, Norma 1976 "An annotated bibliography of recent fiction about Native Americans for young people in grades 7-12," in Ken Donelson (ed.), Adolescent Literature Revisited After Four Years (Tempe, AZ: Arizona English Teachers Association). Jaimes, Marie Annette 1982 "Towards a new image of American Indian women," Journal of American Indian Education 22, 1(October):18-32. Jerome, John L. 1971 "On textbooks," The Indian Historian 4, 2(Summer):57. Johnson, Clifton 1973 "The Indians: first victims of American racism," Crises 80, 3(March):80-83. Johnson, Warren Bertrum 1962 "The content of American colonial newspapers relative to international affairs, 1704-1763." Seattle: University of Washington, unpublished Ph.D. dissertation. Kahl, June (Comp.) 1976 Non-Stereotyped Indian Literature: A Bibliography. ERIC Doc. 128795, August. Kehoe, John W. 1975 "A role play of Native Indian-white value differences," History and Social Science Teacher 11, 1(Fall):45—51 . Kerr, B. 1978 "Novel as sacred text; N. Scott Momaday's myth-making ethic," Southwestern Review 63(Spring):172—179- Kitses, Jim 1974 "The Western:, ideology and archetype," in Jack Nachbar (ed.), Focus on the Western (Englewood Cliffs, NJ: Prentice-Hall, Inc.), pp. 4-72. Knight, Oliver Holmes 1959 "Reporting by accredited newspaper correspondents of Army campaigns against hostile Indians in the American West." University of Wisconsin, Ph.D. dissertation. Pub: Following the Indian Wars. Norman: University of Oklahoma Press, 1960. 244 Koster, J. 1976 "American Indians and the media," Cross Currents 26 (Summer):164-171 . Krefft, J.H. 1978 "Possible source for Faulkner's Indians: Oliver La Farge's Laughing Boy," Tulane Studies in English 23:187-192. Kunz, D. _ _ 1978 "Lost in the distance, of winter: James Welch's Winter in the Blood," Critique 10, 1:93-99. Kupperman, K.O. 1977 "English perceptions of treachery, 1583-1640: the case of the American savages," The Historical Journal 20(June): 263-287. Larson, Sidner J. 1977 "James Welch's Winter in the Blood," The Indian Historian 10, 1(Winter):23-26. Lattin, V.E. 1979 "Quest for mythic vision in contemporary Native American and Chicano fiction," American Literature 50(January): 625-640. Leechman, Douglas 1940 "The popular concept of the 'Red Indian' as revealed in literature." Ottawa: University of Ottawa, unpublished Master's thesis^ : 1941 "The 'Red Indian' of literature: a study in the perpetuation of error." Ottawa: University of Ottawa, unpublished__Ph.D, dissertation. Lester, Joan 1972 "The American. Indian: a museum's eye view," The Indian Historian 5,_ 2(Summer):25-31 . Lewis, Robert W. 1971 "Hemingway," The Indian Historian 4, 2(Summer):56-57. 1973 "English and American Indian studies," The Indian Historian 6, 4(Fall):32-37, 54. 245 Lincoln, K. 1978 "(Native) American poetries," Southwestern Review 63 (Autumn):367-384. Livingston, Richard 0. 1973 "Carol Burnett TV show degrades the American Indian," The Indian Historian 6, 2(Spring):23. Lurie, Nancy 0. 1966 "The enduring Indian," Natural History LXXV(September): 10-22. 1974 "Forked tongue in cheek or life among the noble civilages," The Indian Historian 7, 2(Spring):28-40, 52. 1976 "The will-o'-the-wisp of Indian unity," The Indian Historian 9, 3(Summer):19-24. Lutz, Hartmut 1979 "The image of the American Indian in German literature," Interracial Books for Children Bulletin 10, 2:17-18. Mackinock, Rey 1971 "The plight of the Native American," School Library Journal 17(September):46-51. Mallam, R. Clark 1973 "Treatment of the Indian in public school texts," Journal of American Indian Education 13(October):14—19 - Manitoba Indian Brotherhood 1974 The Shocking Truth About Indians in Textbooks. Winnipeg, Manitoba: Manitoba Indian Brotherhood. Mansfield-Kelley, Deane 1979 "Oliver La Farge and the Indian women in American literature." Austin, TX: University of Texas, unpublished Ph.D. dissertation. Martin, Calvin and Steven Crain 1974 "The Indian behind the mask at the Boston Tea Party," The Indian Historian 7, 1(Winter):45-47. Martin, J.C. 1978 "Locus of control and self-esteem in Indian and white students," Journal of American Indian Education 18 (October):23-29. 246 Mathur, Mary E. Fleming 1970 "The tale of the lazy Indian," The Indian Historian 3, 3(Summer):1 A—18. McKay, Sandy, et. al. n.d. "The attitudes of Toronto students towards the Canadian Indians." Glendon College, York University, unpublished paper. McLaughlin, William G. 1976 "A note on African sources of American Indian racial myths," Journal of American Folklore 89(July):331 -335. McNickle, D'Arcy 1970 "American Indians who never were," The Indian Historian 3, 3(Summer):4-7• Medicine, Bea 1971 "The anthropologist as the Indian's image-maker," The Indian Historian 4, 3(Fall.): 27-29. Mikkelsen, Leatrice (Ed.) 1968 "Exhibit shows new trend against stereotyped art," The Indian Historian 1, 3(Summer):20-21. Monkman, L. 1979 "Richardson's Indians," Canadian Literature 81(Summer): 86-94. Morgan, James and Marilyn Morgan 1969 "About textbooks," The Indian Historian 2, 1(Spring): 39-40. Morris, C. 1979 "Indians and other aliens: a Native American view of science fiction," Extrapolation 20(Winter):301-307. Nachbar, Jack (Ed.) 1974 Focus on the Western. Englewood Cliffs, NJ: Prentice- Hall, Inc. Narang, H.L. 1974 "Improving reading ability of Indian children," Elementary English, pp. 190-192. 247 Nash, Gary B. 1976 "Whither Indian history?" Journal of Ethnic Studies 4, 3(Fall):69-76. Nelson, Mary 1972 "Problems Indian students face," The Indian Historian 5, 2(Spring):22-24. Newton, Ray 1976 "Native Americans and the mass media." Paper presented at the Annual Conference of the Western Social Science Association (18th), Arizona State University, Tempe, AZ. Newtown, George Allyn 1979 "Images of the American Indian in French and German novels of the nineteenth century." Yale University, unpublished Ph.D. dissertation. O'Donnell, Holly 1974 "Cultural bias: a many-headed monster," Elementary English 51, 2(February):181 -184, 214. Officer, James 1970 "A new image for the great white father," Midcontinent American Studies Journal 11:5-19- Onstad, Gwen 1971 "A talk with some Native Americans," Personnel and Guidance Journal 50, 2(0ctober):103-108. Ortiz, Alfonso 1977 "Some concerns central to the writing of 'Indian' history," The Indian Historian 10, 1(Winter):17-22. Otis, Morgan 1977 "Textbooks and the people known as American Indians," The Indian Historian 10, 4(Fall):40-46. Paine, Gregory Lansing 1926 "The Indian in the Leatherstocking Tales," Studies in Philology XXIII, 1(January):16-39- Panzarella, Robert and Ansley LaMar 1979 "Attitudes of blacks and whites toward Native American revolutionary tactics for social change," Human Relations 32, 1(January):69-75. 248 Pearce, Roy Harvey 1953 The Savages of America: A Study of the Indians and the Idea of Civilization. Baltimore: John Hopkins University Press. Popp, James A. 1975 "An examination of children's books on the American Indian," BIA Education Research Bulletin 3, 1(January):10-23. Ramsey, W.M. 1980 "Moot points of Melville's Indian Hating," American Literature 52(May):224-235. Ritt, Leonard G. 1979 "Some social and political views of American Indians," Ethnicity 6, 1(March):45-72. Roesler, Max 1978 "Enough is enough . . . ," The Indian Historian 11, 3: 25-28, 57. Rogers, Robert 1971 Ponteach or the Savages of America: A Tragedy. New York: (1766) B. Franklin. (Originally published: London: John Millan.) Ryan, John Paul ----- 1975 Cultural Diversity and the American Experience: Political Participation Among Blacks, Appalachians and Indians. Beverly Hills/London: Sage Publications. Ryan, Marah E. 1908 Indian Love Letters. Chicago: A.C. McClurg and Company. Salter, John R., Jr. " 1981 "Letter to a younger Indian (and to our Anglo friends)," Integrated Education 18, nos. 1-4(Spring):91 —93. Sando, Joe S. 1971 "White-created myths about the Native American," The Indian Historian 4, 4(Winter):10—11. Sawyer, Thomas E. 1976 "Assimilation versus self-identity: a modern Native American perspective." An address to the Western Social Science Association Conference (18th), Arizona State University, Tempe, AZ. 249 Scullin, Michael 1969 "Reviewing the mass media," The Indian Historian 2, 1 (Spring):49-50. Seyersted, Per 1974 "The Indian in Knickerbocker's New Amsterdam," The Indian Historian 7, 3(Summer):14-28. Shaughnessy, Tim 1976 "The attitudes of selected educational groups in Arizona toward Indians." Tempe, AZ: Arizona State University, unpublished Ph.D. dissertation. Sheehan, Bernard W. 1969 "Indian-white relations in early American: a review essay," The William and Mary Quarterly, Series 3, 26, 2(April): 267-286. Sheppard, George K. 1971 "The effects of an audio-visual presentation of fifth graders' attitudes toward American Indians." Utah State University, unpublished E.Ed, dissertation. Simmons, William S. 1979 "Conversion from Indian to Puritan," New England Quarterly 52(June):197-218. 1981 "Cultural bias in the New England Puritan's perception of Indians," The William and Mary Quarterly, Series 3, 38, 1(January):56-72. Sluman, Norma 1967 "The text book Indian," Toronto Education Quarterly 4, 4:10-12. Smith, June Smeck 1971 "Library service to American Indians," Library Trends 20(October):223-238. Spang, Alonzo 1971 "Understanding the Indian," Personnel and Guidance Journal 50(October):97-102. Sprague, Arthur William, Jr. 1972 "Attitudinal changes in secondary school students as a result of studying an ethnohistory of the Kiowa Indians." Ohio State University, unpublished Ph.D. dissertation. 250 Squires-PoPour, Wallace 1974 "Reverse prejudice: a new Indian ideal?" The Indian Historian 7, A(Fall):35-36. Stengar, Ernest 1970 "I was born an Indian," The Indian Historian 5, 1(Summer):17. Stoodt, Barbara D. and Sandra Ignizio 1976 "The American Indian in children's literature," Language Arts 53, 1(January):17-19. Swanson, Charles H. 1977 "The treatment of the American Indian in high school history texts," The Indian Historian 10, 2(Spring):28-37. Tauran, R.H. 1967 "The influences of reading on the attitudes of third graders towards Eskimos." University of Maryland, unpublished Ph.D. dissertation. Thornton, Russell and Mary K. Grasmick 1979 "Sociological study of American Indians: a research note on journal literature," Ethnicity 6, 4(December):299-305. Tomkins, J.P. 1981 "No apologies for the Iroquois: a new way to read the Leatherstocking novels," Criticism 23(Winter):24-41. Townsend, Mary Jane 1976 "Taking of the war bonnet: American Indian literature," Language Arts 53, 3(March):236-244. Trigger, B.G. 1980 "Archaelogy and the image of the American Indian," American Antiquity 45(0ctober):662-676. United States Department of Health, Education, and Welfare, Public Health Service 1977 "Reality of Myths?" Indians in the Red. Washington, D.C.: Government Printing Office. United States, Smithsonian Institute 1975 Indian Images, Photographs of North American Indians, 1847-1928. Washington, D.C.: Government Printing Office. 251 University of Montana, Native American Studies n.d. "Custer died for your sins." Mimeographed copy of comments on the book by Deloria. Vanderburgh, Rosamond M. 1968 The Canadian Indian in Ontario's School Texts: A Study of Social Studies Textbooks, Grades 1 through 8. Port Credit, Ontario: University Women's Club of Port Credit, Study Group on the Canadian Indian arid Eskimo. Vaughan, Alden T. 1965 New England Frontier: Puritans and Indians, 1620-1675. Boston: Little, Brown. Velie, A.R. 1978 "Cain and Abel in N. Scott Momaday's House Made of Dawn," Journal of the West 17(April):55-62. Vogel, Virgil J. 1968 "The Indian in American history textbooks," Integrated Education 6, 3(May-June):16-32. Warrior, Clyde 1964 "Which one are you? Five types of young Indians." Mimeographed copy from Native American Studies, University of Montana, Missoula, MT. Wax, Murray L. 1968 "The white man's burdensome 'business': a review on the change and constancy of literature on the American Indian," Social Problems 16, 1(Summer):106-113. Weinkauf, M.S. 1979 "Indians in science fiction," Extrapolation 20(Winter): 308-320. Wenrick, Jon S. 1975 "Indians in Almanacs (1793-1815)," The Indian Historian 8, 4(Winter):36-42. White, John Kennardh 1972 "An experiment with time," The Indian Historian 5, 4 (Winter):31-40. Whiteman, Henrietta n.d. Notes on Winter in the Blood. Xeroxed copy of notes from class lecture. 252 Whiteman, Henrietta n.d. "Pave-esh-iv-Good Day." Unpublished paper on House Made of Dawn. Wilkinson, Gerald 1974 "Colonialism through the media," The Indian Historian 7 3(Summer):29-32. Wilson, Charles R. 1976 "Racial reservations: Indians and blacks in American magazines, 1885-1900," Journal of Popular Culture 10, 1(Summer):70-79. Wissler, Clark, Constance Lindsay Skinner, and William Wood 1925 Adventures in the Wilderness, Volume 1 in the Pageant of America. New York: Yale University Press. Work, James C. 1974 "Indian literature: a plea for formality," The Indian Historian 7, 4(Fall):36-38. Yost, James 1976 "Traces of the native tongue," The Indian Historian 9, 1(Winter):39-40. Young, Kimball 1927 "White-American Indian prejudice," in K. Young (ed.), Source Book for Social Psychology (New York: Alfred A. Knopf), pp. 528-533. Zamora, Mario D. 1973 "Moral, immoral science, the case for cultural anthropology," The Indian Historian 6, 2(Spring):24-28. APPENDIX C 254 THE PRINCIPLES OF PROPAGANDA 1. Principle of the intention of the propagandist. In intentional propaganda, the propagandist is aware of his interested aim; in unintentional propaganda, he does not appreciate the social effect of his own actions. 2. Principle of perception. The propagandist makes his stimulus- situation stand out from its competing ground. 2a. Perceptual principle of auxiliary attitudes. The propa gandist makes his stimulus-situation outstanding through the arousal of auxiliary attitudes. 2b. Perceptual principle of repetition. The propagandist repeats his stimulus-situation to increase the probability that it will be perceived. 2c. Perceptual principle of simplification. The propagandist simplifies his stimulus-situation to bring it within the range of perception. 3. Principle of the type of propaganda. The propagandist employs any one or all of the following types of propaganda: revealed, delayed revealed, and concealed propaganda. 3a. Principle of revealed propaganda. In revealed propaganda the propagandist enables people to perceive his aim through direct suggestion. 3b. Principle of delayed revealed propaganda. In delayed 255 revealed propaganda the propagandist reveals his aim only after he has aroused related attitudes. 3b1. Temporal principle of delayed revealed propaganda. In delayed revealed propaganda the propagandist enables people to perceive his aim at a moment when that aim can be integrated into the previously aroused related attitudes. 3c. Principle of concealed propaganda. In concealed propa ganda the propagandist refrains from stating his aim and integrates through indirect suggestion the aroused related attitudes into a new attitude which predisposes people toward that aim. 4. Principle of related attitudes. In the process of suggestion, the propagandist arouses related attitudes that are instru mental in bringing about the desired integration. 4a. Principle of related dominant attitudes. The propa gandist employs attitudes that are already dominant as related attitudes or he arouses related attitudes that remain dominant over a period of time. 4b. Principle of related central attitudes. The propagandist arouses related attitudes that are central attitudes. 4c. Principle of related auxiliary attitudes. The propa gandist arouses auxiliary attitudes that also function as related attitudes. 256 4d. Principle of variation. The propagandist varies the content of his stimulus-situation, in order to arouse related attitudes in different people and, by changing their stereotypes, to construct new attitudes in others through positive suggestion. 5. Principle of the desired integration. The propagandist secures a desired integration that predisposes people toward his aim. 5a. Principle of the type of integration. The desired inte gration is either a central or a segmental attitude. 5b. Principle of action. The propagandist secures a desired integration that leads to action. 6. Principle of the sphere of unpredictability. Before the desired integration is achieved between the related attitudes and, except in the case of concealed propaganda, the compre hension of the propagandist's aim and before it leads to action, there is a sphere of unpredictability due to the temporal character of the propaganda, the presence of competing propagandists, and the complexity of the personalities in the group with which the propagandist must deal. 6a. Principle of the auxiliary submissive attitude. The propagandist reduces the sphere of unpredictability by restricting the mental field through the arousal of a submissive attitude toward a stimulus-situation which 257 has prestige and the effect of which is a tendency toward increased suggestibility. 6a1. Principle of positive social value. The propa gandist includes within his stimulus-situation objects and persons with positive social value. 6a2. Principle of the impression of universality. The propagandist produces an impression of universality. 6a3. Principle of selection of propaganda. When the prestige of the propagandist or of the stimulus- situation is not diminished by the revelation of his aim, revealed propaganda is employed; when the prestige is diminished by that revelation, concealed propaganda is employed; when it is diminished by an immediate but not by a subsequent revelation, delayed revealed propaganda is employed. 6b. Principle of indicating the paths of action. In revealed and delayed revealed propaganda, the propagandist reduces the sphere of unpredictability by indicating the paths of action to which the desired integration may lead. 6c. Principle of reinforcing. The propagandist reduces the sphere of unpredictability by preventing the desired integration from remaining latent or from disintegrating. 6c1. Reinforcing principle of repetition. The propagandist repeats the same or similar stimulus-situations. 6c2. Principle of additional related pre-existing attitudes. The propagandist arouses other pre-existing related attitudes. 6c3. Principle of additional related new attitudes. The propagandist forms new attitudes through positive suggestion by changing people's stereotypes. 6d. Principle of limitation. The propagandist reduces the sphere of unpredictability by producing the initial, relatively stable integration. Principle of counter-propaganda. The propagandist uses counter-propaganda when conflicting attitudes tend to prevent the desired integration from emerging. 7a. Principle of negative suggestion in counter-propaganda. In counter-propaganda, the propagandist uses negative suggestion to render conflicting attitudes ineffective. 7b. Principle of positive suggestion in counter-propaganda. In counter-propaganda, the propagandist uses positive suggestion to form new related attitudes that will counteract conflicting attitudes. Principle of persuasion. The propagandist uses persuasion as a supplementary method. 8a. Prestige principle of persuasion. The propagandist employs persuasion on people with prestige toward whom later a submissive attitude is directed by other people. Temporal principle of persuasion. Persuasion supplements propaganda at a crucial moment to bring about the desired integration and action among certain people. APPENDIX D 261 REFERENCES ON CONTENT ANALYSIS Albig, W. 1938 "The content of radio programs, 1925-1935," Social Forces 16:338-349. Anderson, D.C. and W.W. Sharrock 1979 "Biasing the news: technical issues in media studies," Sociology 13:367-385. Armstrong, Robert Plant 1959 "Content analysis in folkloristics," in Ithiel De Sola Pool (ed.), Trends in Content Analysis (Urbana: University of Illinois Press), pp. 151-170. Berelson, Bernard 1952 Content Analysis in Communication Research. Glenco, IL: The Free Press. 1954 "Content analysis," in Gardner Lindzey (ed.), Handbook of Social Psychology, Volume 1 (Cambridge: Addison- Wesley Publishing Company, Inc.), pp. 488-522. Broom, L. and Shirley Reece 1955 "Political and racial interest: a study in content analysis," Public Opinion Quarterly 19:5-19. Boder, D.P. 1940 "The adjective-verb quotient: a contribution to the psychology of language," Psychological Record 3:310-343. Budd, Richard W. and Robert K. Thorp 1963 An Introduction to Content Analysis, Including Annotated Bibliography. Iowa City: State University of Iowa. Budd, Richard W., Robert K. Thorp, and Lewis Donohew 1967 Content Analysis of Communications. New York: Macmillan and Company, Publishers. Bush, Chilton R. 1960 "A system of categories for general news content," Journalism Quarterly 37, 2:206-210. 262 Cancian, F.M. and B.C. Ross 1979 "Mass media and the women's movement: 1900-1977," Journal of Applied Behavioral Sciences 17, 1:9-26, 81. Carney, T.F. 1970 "Content analysis," Listener 84:479-481. Cartwright, Dorian 1953 "Analysis of qualitative material," in Leon Festinger and David Katz (eds.), Research Methods in Behavior Sciences (New York: Dryden Press), pp. 421-470. Casty, Alan (Comp.) 1968 Mass Media and Mass Man. New York: Holt, Rinehart and Winston, Inc. Cicourel, Aaron V. 1964 "Historical materials and content analysis," in A.V. Cicourel, Method and Measurement in Sociology (New York: The Free Press), pp. 142-156. Cohen, Felix 1964 "The reconstruction of hidden value judgement," in Lyman (1954) Bryson, Louis Finkelstein, R.M. Maclver, and Richard McKeon (eds.), Symbols and Values (New York: Cooper Square Publishers, Inc.; originally, New York: Conference on Science, Philosophy and Religion), pp. 545-561. Colby, Benjamin N. 1966a "The analysis of culture and the patterning of narrative concern in texts," American Anthropologist 68, 2, 1 (April):374-388. 1966b "Cultural patterns in narrative," Science 151:793-798. Doling, J. 1978 "Use of content analysis in identifying the determinants of house prices," Urban Studies 15(Fall):89-90. Dovring, Karin 1954 "Quantitative semantics in 18th century Sweden," Public Opinion Quarterly 18, 4:389-394. Driessen, Jon 1974 "Content analysis bibliography, SOC 613.2 (Winter Quarter)." Mimeographed copy. 263 Duffell, Molly Lynn 1957 "Content analysis methods applied to literary criticism: plays of Christopher Marlowe." Missoula, MT: Montana State University, unpublished Master's thesis. Edwards, Karl 0. 1981 "Content analysis: a review of a method." Missoula, MT: unpublished paper. Garraty, John A. 1959 "The application of content analysis to biography and history," in Ithiel De Sola Pool (ed.), Trends in Content Analysis (Urbana: University of Illinois Press), pp. 171-188. Geller, A., D. Kaplan, and Harold Lasswell 1942 "An experimental comparison of four ways of coding editorial content," Journalism Quarterly 19:362-370. George, Alexander L. 1959 "Quantitative and qualitative approaches to content analysis," in Ithiel De Sola Pool (ed.), Trends in Content Analysis (Urbana: University of Illinois Press), pp. 7-32. Gerbner, George (Ed.) 1969 The Analysis of Communication Content; Developments in Scientific Theories and Computer Techniques. New York: John Wiley and Sons, Inc. Gettschalk, Louis A. 1969 The Measurement of Psychological States Through Content Analysis of Verbal Behavior. Berkeley: University of California Press. Holsti, Ole R. 1968 "Content analysis," in Lindzey and Aronson (eds.), The Handbook of Social Psychology, Second Edition (Menlo Park, CA: Addison-Wesley Publishing Company, Inc.), pp. 596-692. 1969 Content Analysis for the Social Sciences and Humanities. Reading, MA: Addison-Wesley Publishing Company, Inc. 264 Janis, I.L. 1949 "The problem of validating content analysis," in H.D. Lasswell, et. al., Language of Politics: Studies in Quantitative Semantics (Cambridge, MA: MIT Press), pp. 55-82. Kaplan, Abraham 1943 "Content analysis and the theory of signs," Philosophy of Science 10:230-247. Kobre, Sidney and Juanita Parks 1954 "The newspapers cover a murder case," Journalism Quarterly 31, 3:311-318. Kracauer, Siegfried 1952 "The challenge of qualitative content analysis," Public Opinion Quarterly 16(Winter):631-642. Krippendorff, Klaus 1980 Content Analysis: An Introduction to its Methodology. Beverly Hills, CA: Sage Publications. Lantz, Herman R., Eloise C. Snyder, Margaret Britton, and Raymond Schmitt 1970 "A content analysis of colonial magazines," in Thomas Lasswell, John Burma, and Sidney Aronson (eds.), Life in Society (Glenview, IL: Scott, Foresman and Company). Lasswell, Harold D., Nathan Leites, and Associates 1949 Language of Politics. New York: George W. Stewart, Publisher, Inc. Lasswell, Harold D., Daniel Lerner, and Ithiel De Sola Pool 1952 The Comparative Study of Symbols. Stanford: Stanford University Press. Lasswell, Harold D., Paul Lewis, Joseph W. Martin, and Joseph Goldsen 1942 "The politically significant content of the press: coding procedures," Journalism Quarterly 19, 1(March):12-23. Lazarsfeld, Paul F., Bernard Berelson, and Hazel Gaudet 1944 The People's Choice: How the Voter Makes Up His Mind in a Presidential Campaign. New York: Columbia University Press. 265 Mahl, George F. 1959 "Exploring emotional states by content analysis," in Ithiel De Sola Pool (ed.), Trends in Content Analysis (Urbana: University of Illinois Press), pp. 89-130. North, Robert Carver, et. al. 1963 Content Analysis; A Handbook with Applications for the Study of International Crisis. Evanston: Northwestern University Press. Osgood, Charles E. 1959 "The representational model and relevant research methods," in Ithiel De Sola Pool (ed.), Trends in Content Analysis (Urbana: University of Illinois Press), pp. 33-88. Osgood, Charles E., W.H. May, and M.S. Miron 1975 Cross Cultural Universals of Affective Meaning. Urbana: University of Illinois Press. Osgood, Charles E., Sol Saporta, and J.C. Nunnally 1956 "Evaluative assertion analysis," Litera 3:47-102. Osgood, Charles E., G.J. Suci, and P.H. Tannenbaum 1957 The Measurement of Meaning. Urbana: University of Illinois Press. Pool, Ithiel De Sola 1959 Trends in Content Analysis; Papers. Urbana: University of Illinois Press. Rosengren, Karl Erik (Ed.) 1981 Advances in Content Analysis. Beverly Hills, CA: Sage Publications. Saporta, Sol and Thomas A. Sebeok 1959 "Linguistics and content analysis," in Ithiel De Sola Pool (ed.), Trends in Content Analysis (Urbana: University of Illinois Press), pp. 131-150. Schultz, William 1958 "On categorizing qualitative analysis," Public Opinion Quarterly 22, 4:503-515. Stephenson, William 1963 "A critique of content analysis," Psychological Record 13:155-162. 266 Stone, Philip J., et. al. 1966 The General Inquirer; A Computer Approach to Content Analysis. Cambridge, MA: MIT Press. Symanski, R. 1976 "Manipulation of ordinary language," Association of American Geographers, Annals 66(December):605-61 A. University of Chicago Law Review 1948 "Content analysis—a new evidentiary technique," University of Chicago Law Review 15:910-925. Wiseman, Jacqueline and Marcia S. Aron 1970 "Content analysis," in Wiseman and Aron, Field Projects for Sociology Students (Cambridge, MA: Schonkman Publisher), pp. 113-125. APPENDIX E 268 ECO Analysis Score Sheet/Instrument Source: Score: Title: Subject: Page Type* Terra Direction Page Type* Term Direction 1 . 23. 2. 24. 3. 25. 4. 26. 5. 27. 6. 28. 7. 29. 8. 30. 9. 31 . 10. 32. 11 . 33. 12. 34. 13. 35. 14. 36. 15. 37. 16. 38. 17. 39. 18. 40. 19. 41 . 20. 42. 21 . 43. 22. 44. *—i.e., Verb, Adverb, Noun, Adjective 100 (F) 100 x + = = F + U + ECO Analysis Comparison Instrument Source: Rating Title: Subjects: Qualitative Review: 270 ECO Analysis Administrative Data Instrument Sample Group: Catalog Source: Vol._ Year Sample Number: Random Sample No. Page Source: Title: Publisher: Copyright: Subjects: 1 . 6. 11 . (list) 2. 7. 12. 3. 8. 13. 4. 9. 14. 5. 10. 15. 1.0 TRENDS AND FREQUENCIES 1.1 LENGTH/SPACE 1.1.1 Total number of pages in the book: 1.1.2 Total number of pages per subject: 1 . 6 11. 2. 7 12. 3. 8 13. 4. 9 14. 5. 10 15. 1.1.3 Total number of illustrations (i.e., maps, drawings, pictures! Positive (+): Neutral (0): Negative (-): 1.1.4 Total number of illustrations on each subject: + 0 + 0 1 . 6. 11 . 0 2. 7. 12. 3. 8. 13. 4. 9. 14. 5. 10. 15. 271 1.2 TEMPORAL SETTING 1.2.1 Is the story time explicitly given? Yes No 1.2.2 At what time does the story begin? NA 1.2.3 At what time does the story end? NA 1.2.4 If the story time is not explicitly given, is it presented implicitly? (i.e., illustrations, descriptions of the environment, events in the story): Yes No 1.2.5 What is the appropriate story time as suggested by the source? Pre-Columbian 1800 1900 Present Invasion 1.3 SOURCE 1.3.1 Has the source provided evidence of professional research? Yes No What? How? 1.3.2 Has the source written other materials on Native Americans? Yes No Number Level Type 1.3.3 Is the source of Native American ancestry? Yes No Nation/Tribe 272 1.4 THEMES 1.4.1 What themes are presented in the story?-'r. Themes: Frequency 1 . Noble Savage .ioi: . 2. White Man's Helper "hi/ 3. Indian Maiden Inc. 4. Red Varmint ?.er: 5. Chief "h" 6. Protestor Pre 7. Indian/Non-Indian Relations '.v, 8. Meeting of Two Cultures Mo v. 9. Indians and the Land Err 10. Contemporary Issues Co- 11. Other: (list) APPENDIX F 274 ECO Word List able + conscientious + fat - achievement + conspirator - fearful - admirable + corrupt - fearless + advanced + courageous + feeble - adventurous + courteous + ferocious - afraid - coward - fierce - agile + crafty 0 fiery 0 alert + criminal - fine + amazing + cruel - foolish - ambitious + cultured + foreign 0 angry - daring + free + ardent* 0 dauntless + friendly + attractive + dear + furious - audacious 0 dedicated + genius + backward - delicate + gentle + bad - delightful + gentleman + barbarian - dependent 0 gifted + beautiful + deserter - glorious + bickering - determined + good + bitter - devoted + goodwill + blind - devout + gracious + blood-thirsty - dictator - grave 0 boastful - dirty - great + bold 0 disgruntled - greedy - brave + dishonest - handsome + bright + disloyal - happy + brilliant + distinguished + hard 0 butcher - drunk - hardworking + calm + eager + hardy + capable + elegant + haughty - careful + eloquent + healthy + charitable + enduring + heathen - charming + energetic + helpful + cheap - enthusiastic + hero + chivalrous + experienced + honest + civilized + expert + honorable + clean + extremist - horde - clever + failure - horrible - cold - fair + hospitable + colorful + faithful + hostile - common 0 false - howling - complex 0 famous + humble + idealistic + outrage - ruthless - idle - outstanding + sacrificial 0 ignorant - pagan - sad - imaginative + panic-stricken - savage - impatient - patient + scheming - important + patriotic + scholarly + independent + peaceful + selfish - industrious + pillager - sensitive + inferior - pious 0 serious + infidel - pitiless - shrewd 0 ingenious + pleasant + shrieking - insane - pleasurable + simple 0 insolent - plotting - sincere + inspired + plunderer - skillful + inspiring + polite + slaughter - intelligent + poor 0 slow - interesting + popular + smelly - jealous - primitive - soft + just + problem - splendid + kind + promising + strange 0 late - proper* 0 striking + lazy - proud + strong + liar - prowling - successful + lively + pure + sullen - lovely + quality + superb + loving + quarrelsome - superior + loyal + quick + suspicious - lurking - quiet + swarm - magnificent + raiding - sweet + martyr 0 reasonable + sympathetic + massacre - rebel* 0 talented + mean - reckless - tenacious 0 menacing - remarkable + terrible - merciless - renegade - terrified - mistaken* 0 renowned + terrifying - mob - resentful - terrorist - moderate* 0 resourceful + thief - modest + respected + threatening - murderer - respectful + thrifty + mutinous - revengeful - tireless + natural + rich 0 tolerant + nice + right + tough* 0 noble + rioter - traitor - normal + robber - treacherous - notable + rough - trickery - 276 troublesome - true + trustworthy + ugly - uncivilized - undisciplined - uneducated - unfriendly - unreliable - unselfish + unskilled - untrustworthy - useful + vain - valiant + valuable + venerable* 0 vicious - victorious + vigorous + violent - warlike - warm + wasteful - weak - well-known + wild - wise + wonderful + worthy + wrong - zealous* 0 *—"Note: The rationale for the development of this list is reported in Apendix I. The values are based on judgments by students in Grades 11 and 12 and in a teacher education program. An asterisk beside a word indicates that there was significant disagreement between the two groups regarding the value to be assigned to the word" (Pratt, 1972:39; emphasis in the original). APPENDIX G 278 Table 28 Coefficient of Evaluation Scores Subjects Books 1 2 3 4 C 6 #1 51 .35 53.85 16,.67 50.00 43..90 60,.00 #2 39.02 14 .29 27,.19 61 .76 12..50 42,.11 #3 50.00 — — 6,.67 37.50 34..50 33,.33 #4 66.67 — — — — 21 .43 18.. 64 22..73 #5 2.94 -•— — — 18.18 56..25 33..33 #6 7.69 — — —-- 6.45 20..00 — — #7 25.71 — — — — 1 6..67 —-- — — #8 32.14 — — — — 14 .29 — — — — #9 22.58 —-- —-- 21 .05. — — --— Native Americans 31 .02 — — — — 14 .29 31 ..14 33..33 They — — 14,.29 6,.67 —-- — — — — Non-Native Americans 36.00 —-- — — 44,.74 20..00 40. 00 We — — 53,.85 21 ..93 — — — — — — Total Book 32.44 29..41 18..75 33,.15 28..99 38..30 100 (F) Coefficient of Evaluation = F + U where F = total favorable terms U = total unfavorable terms 279 Table 29 Evaluativeness Scores Subjects Books 1 2 3 4 5 6 % % % % % % #1 5 10 20 8 7 3 #2 9 9 17 6 6 2 #3 6 — 61 7 6 4 #4 5 — — 9 8 3 #5 9 — — 8 8 5 #4 4 — — 36 11 — #7 6 — — 14 — — #8 4 — — 8 — — #9 9 — — 13 — — Native Americans 7 — — 16 7 4 They — 19 61 — — — Non-Native Americans 5 — — 7 11 3 We — 10 19 Total Book 14 22 100 (Total Evaluative Words) Evaluativeness Scores = Total Words Table 30 Summary of Data by Book, Ethnicity, and We-They Terra Direction Books Akavak, an Eskimo Journev Sing Down the Moon The Spirit is Willing Souaw Man's Son, Sarah Bishop Where the Buffaloes Begin Book #1 Book #2 Book #3 Book *4 Book #5 Book 16 They We Total NA1 NNA2 Total NA NNA lota) NA NNA Total NA NNA Total They We Total i * 4 50 54 52 8 60 10 51 61 58 27 85 1 1 25 36 3 7 10 (4.04%) (A.06%) (4.06%) (2.11%) (2.24%) (2.13%) (2.29%) (3.15%) (2.97%) (2.30%) (1.76%) (2.09%) (1.48%) (1.05%) (1.15%) (2.78%) (5.34%) (4.18%) 0 • 1CC4 1043 2295 317 2612 366 IVJc, 1871 2337 1460 3797 7^0 2323 3033 87 118 205 ' (39.39%) (81 (78.36%) (93.22%) (88.80%) (92.66%) (63.94%) (92.96%) (91.05%) (92.59%) (95.11%) (93.55%) (95.56%) (97.44%) (96.99%)i (80.56%) (90.08%) (85.77%) | • 56 178 234 115 32 147 • 60 63 123 129 48 177 22 • 36 58 18 6 24 I (56.57%) (14.45%) (17.58%) (4.67%) (8.96%) (5.21%) (13.76%) (3.89%) (5.98%) (5.11%) (3.13%) (4.36%) (2.96%) (1.51%) (1.85%) (16.67%) (4.58%) (10.04%) Coefficier.t of Evaluation 7% ?2% 19% 31% 20% 29% it* 45% 33% 31% 36% 32% 33% 41% 38% 14% 54% 29% Evaluativeness Score 61 % 19% 22% 7% 11% 7% 16% 7% 9% 7% 5% 7% 4% 3% 3% 19% 10% 14% 1--NA stands for Native Americans 2--NNA stands for Non-Native Americans •—indicates statistical significance for individual cell distribution APPENDIX H 282 THEMES Theme 1970 1974 1978 1980 1981 Noble Savage 4 4 17 2 White Man's Helper 1 — 21 Indian Maiden Red Varmint — 10 7 Indian Warrior 19 -- 73 Chief 14 6 47 Protestor Indian-White Relations 51 5 160 Meeting of Cultures 33 — 5 Indians and the Land. 1 — 12 2 2 Contemporary Issues Object of Display 9 Traitor to Whites 3 Lazy Indian 10 1 Knowledgeable/Good Worker 1 "Freed" Indian 2 1 15 Heathen Indian 3 Ghosts/Spells/Witchcraft 15 27 3 Dumb Indian 3 War Dancing/Council 4 — 18 Poor Material Life 2 — 13 Insensitive 2 Joking 1 Different Habits, etc. 5 — 10 1 Intertribal War 8 — 10 2 Animal-like 3