The effects of training in racial identity stage theory on attitudes and perceptions of whites toward blacks

MISSING IMAGE

Material Information

Title:
The effects of training in racial identity stage theory on attitudes and perceptions of whites toward blacks
Physical Description:
v, 108 leaves : ; 29 cm.
Language:
English
Creator:
Heath, Anne Elizabeth
Publication Date:

Subjects

Subjects / Keywords:
Race relations -- Psychological aspects -- United States   ( lcsh )
Genre:
bibliography   ( marcgt )
non-fiction   ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 1992.
Bibliography:
Includes bibliographical references (leaves 100-106).
Statement of Responsibility:
by Anne Elizabeth Heath.
General Note:
Typescript.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
aleph - 001751586
notis - AJG4522
oclc - 26529155
System ID:
AA00003271:00001

Full Text








THE EFFECTS OF TRAINING IN RACIAL IDENTITY STAGE THEORY
ON ATTITUDES AND PERCEPTIONS OF WHITES TOWARD BLACKS

















By

ANNE ELIZABETH HEATH


A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL
OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT
OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY

UNIVERSITY OF FLORIDA

1992


JlVERSITY OF FLORIDA r ,ES












ACKNOWLEDGEMENTS

I would like to thank the chairperson of my committee,

Dr. Greg Neimeyer, for his patience and assistance as I

completed this dissertation, and Dr. Franz Epting for his

encouraging comments. I would also like to thank the other

members of my committee, Drs. Mary Fukuyama, W. Max Parker,

and Robert Ziller, for their encouragement and enthusiasm

for the study, and their personal dedication to the field of

multicultural issues in psychology and counselor training,

which was inspiring and helped motivate me to continue.

Special thanks also go to Dr. Monica Biernat whose

contribution to the theory and methodology of this study was

invaluable.

I would also like to thank my research assistant, Cheri

Marmaroosh, for her hard work and dependability in

collecting, encoding, and developing the preliminary matrix

for the data set.

Finally, I want to thank my close friend and colleague,

Dr. Beree Darby, who provided a model of tenacity and

professionalism when striving for goals, and whose belief in

me often outlasted my own.













TABLE OF CONTENTS






ACKNOWLEDGEMENTS . . .. ii

ABSTRACT . . . V

CHAPTERS

I INTRODUCTION . . ... 1
Purpose of the Study . . 2
Importance of the Study . 3

II REVIEW OF THE LITERATURE . 5

Goals and Procedures of Multicultural Counselor
Training . . 6
Racial Identity Stages . 7
Effects of Multicultural Counselor Training 12
Attitudes and Perceptions of Whites Toward Blacks 14
Ingroup/Outgroup Perception and Stereotypes 17
Effects of Perceived Out-group Homogeneity. 20
Improving Intergroup Relations . .. .22

III METHODS . .... ... 24

Subjects . . ... .24
Design and Procedures . .. .25
Dependent Variables . ... 28
Hypotheses . . ... .. .34

IV RESULTS . . ... .37

Perceived Distribution . .. 37
Absolute Difference Score . .... 43
Modern Racism Scale . .. 44
Posttest Only Measures .. . .... 45
Post Hoc Analyses . ... 49


iii













V DISCUSSION .. . ... 57

Perceived Distribution. ... ... 58
Absolute Difference Score . .. 63
Modern Racism Scale. .. . .. .69
Posttest Only Measures . 72
Training Receptivity and the Dependent Variables 75
Conclusion . . .. 75

APPENDICES

A DEBRIEFING . .. 80

B STAGES OF MAJORITY IDENTITY DEVELOPMENT .. 81

C STAGES OF MINORITY IDENTITY DEVELOPMENT .. .85

D RESPONSE SHEET . . .. .88

E DEPENDENT MEASURES . . 91

F PILOT STUDY FOR GENERATING ATTRIBUTES .. .97

G PILOT STUDY FOR SELECTING VIGNETTES .. 98

REFERENCES .. . . 100

BIOGRAPHICAL SKETCH . . 107












Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy

THE EFFECTS OF TRAINING IN RACIAL IDENTITY STAGE THEORY
ON ATTITUDES AND PERCEPTIONS OF WHITES TOWARD BLACKS

By

Anne Elizabeth Heath

May, 1992

Chairman: Greg J. Neimeyer, Ph.D.
Major Department: Counseling Psychology

The purpose of this study was to examine the effects of

Minority and Majority Identity Development Training on the

attitudes and perceptions of Whites toward Blacks. Three

treatment groups, one receiving Minority and Majority

Identity Development training, another receiving Minority

Identity Development training, and the third a no-treatment

control group, were tested on a pre-, post-, and follow-up

test basis using several measures of prejudice and

stereotyping. Results indicated that subjects who received

Minority Identity training in conjunction with Majority

Identity training increased in stereotyping and anti-Black

attitudes. Subjects who received Minority Identity training

alone decreased or remained the same in their attitudes

toward Blacks. Possible reasons for these effects are

discussed and future research in this area is suggested.











CHAPTER I
INTRODUCTION

As a result of an increased awareness of multicultural

concerns in the counseling field, many approaches for

training counselors to be effective in cross-cultural

situations have been proposed. An approach that has

recently gained a great deal of attention is to introduce

counselor trainees and professionals to the stage theories

of racial identity. For example, the American College

Personnel Association (ACPA), a division of the American

Association for Counseling and Development (AACD), recently

completed a five year program of regional training sessions

for counselors and university administrators using the

"Stage Model for Minority/Majority Group Member Identity

Awareness" as the foundation for understanding and fostering

intercultural relations. To date, no research has been

conducted to test the impact of identity stage theory

training on trainees.

Social psychologists have also developed models for

understanding the relationship between majority and minority

perceptions and attitudes. One perspective that has

attracted considerable attention is the "Outgroup

Homogeneity Hypothesis." According to this hypothesis

people tend to perceive out-group (i.e. minority) members as

1







being more homogeneous in their characteristics than in-

group (i.e. majority) members. Several social psychologists

consider group (or category) differentiation as the central

component of the stereotype concept (e.g. Flannagan, Fried,

& Holyoak, 1986; Linville, Salovey, & Fischer, 1986). That

is, the more one is able to perceive many types within a

given category and to distinguish among category members,

the less one is stereotyping.

Research on the effects of perceived homogeneity has

shown that undifferentiated thinking facilitates more

extreme evaluations of individual group members (Linville,

1982; Linville & Jones, 1980), stronger inferences from the

behavior of one member to the group as a whole (Quattrone &

Jones, 1980), and outgroup discrimination (Wilder, 1978).

Conversely, Pettigrew (1981) suggested that perceptions of

variety within a group may cause a greater degree of

interactionall flexibility" in intergroup relations. Thus,

promoting differentiated thinking, as is done when teaching

people about stages of racial identity, may be a useful

strategy for altering stereotypes and combating prejudice.

Purpose of the Study

The purpose of this study was to examine the effects of

Minority and Majority Identity Development training on

Whites' attitudes and perceptions of Blacks. The general

questions addressed were as follows: Does training Whites

about racial identity development affect stereotyping and








prejudice of Blacks? Does training content have a

differential effect on Whites' stereotyping and prejudice of

Blacks?

Importance of the Study

This study has important implications regarding

interracial relations on college campuses. The recent

increase in racial conflicts that have plagued college

campuses in the last few years make the results of this

study directly applicable to a college student population.

Additionally, counselors and university administrators are

in need of models and techniques that foster positive

contact and communication between groups in general, and

between Blacks and Whites in particular. Finally, and what

is considered to be primary impetus for this study, trainers

in the multicultural counseling field have come to recognize

that, compared to other racial groups in the United States,

White Americans are less likely to have had the opportunity

(or necessity) to learn how to work effectively with

culturally and racially different peoples. Many counseling

graduate students and professionals are receiving mandatory

(covertly or overtly) multicultural training. Little

research has been conducted to see what effect this training

has on attitudes and beliefs. By exploring the impact of

racial identity development training on White people's

attitudes and perceptions of Black people, a contribution

can be made toward understanding the variables involved in







4

designing effective multicultural counselor training

programs.












CHAPTER II
REVIEW OF THE LITERATURE

Training recommendations for multicultural counseling

have been generated from the 1975 Austin Conference and the

1978 Dulles Conference. Recommendations included the notion

that education and training programs should evaluate their

graduates regarding cross-cultural competency, and that

continuing professional development should occur beyond the

advanced degree (Sue, Bernier, Durran, Feinberg, Pedersen,

Smith, & Vasquez-Nuttal, 1982). Furthermore, a

recommendation generated from proceedings at the 1973

American Psychological Association Vail Conference boldly

stated that it should be regarded as unethical to conduct

cross-cultural counseling without being trained or judged

competent in the specialty (Korman, 1974). In this chapter

I will discuss goals, procedures, and effects of

multicultural counselor training, with a special focus on

the current role of racial identity stage theory in

counselor training. Then I will discuss some of the

contributions made by social psychologists toward

conceptualizing, measuring, and changing attitudes of Whites

toward Blacks.








Goals and Procedures of Multicultural Counselor Training

Among the specific training suggestions made by

multicultural counselor educators is that direct exposure of

counselors to culturally different populations should be

incorporated into cross-cultural training (e.g. Korchin,

1980; Pedersen, Holwill & Shapiro, 1978; Vontress, 1981).

Copeland (1983) specifies that a supervised practicum in a

setting providing direct exposure to different cultures

would be desirable once the trainee has had simulation

training (e.g. role-play). Exposure to cultural and racial

diversity is not strictly relegated to clients and

multicultural training courses. For example, Smith (1982)

concludes that increasing minority faculty not only

increases the credibility of graduate training programs in

the eyes of the ethnic minority students, but may serve to

counterbalance the forces of stereotyping in the eyes of

White students and faculty.

Several authors suggest that training programs should

include at least four components (Copeland, 1983; Helms,

1984; Sue, 1981): a) consciousness raising,

b) cognitive/didactic instruction c) affective/empathic

development, and d) counseling skills training. Carney and

Kahn (1984) have devised a developmental training model that

promotes and evaluates changes in knowledge, attitudes, and

behaviors. The model consists of five stages of

development, each responding to an appropriate training







environment. Several other models have been developed that

also emphasize a trainee's level of cultural awareness (e.g.

Arredondo-Dowd & Gonsalves, 1980; Copeland, 1982; McDavis &

Parker, 1977; Parker, Valley, & Geary, 1986; Sue, 1983).

Most recently, Margolis and Rungta (1986) suggested that the

best, and most feasible training model to implement, is one

where specialized knowledge (of ethnic minorities and other

special populations) would be integrated into a single

course, insuring that relevant knowledge, awareness, and

skills are generalized across populations.

Racial Identity Stages

An area of multicultural counselor training that has

gained considerable attention in the field of multicultural

counseling is the recognition that differences exist within

a particular racial or ethnic group (not just between racial

groups). Counselors are warned about the dangers of

stereotyping and are encouraged to treat each client as an

individual, not merely as a representative of a particular

culture or race (Smith, 1977; Sue, 1981). Models of racial

identity present one method of conceptualizing differences

among members of a particular minority group.

The past two decades have witnessed a significant

number of studies that have focused on descriptions and

explanations about proposed changes in Black people's racial

identity. In the 1970s much of this attention was

influenced by the work of Sherif and Sherif (1970), who








suggested that acceptance of one's blackness is not

developed in a vacuum. They postulated that initial

feelings of inferiority and shame, which develop from

growing up in a White-emphasized society, lead to guilt and

rage. Eventually, Blacks develop self-confidence and a

positive self-concept.

Influenced by the seminal work of Sherif and Sherif

(1970), Thomas (1971) proposed, a more comprehensive five-

stage developmental model in which Blacks move from a state

of confusion about their Black identity and dependence on

the dominant White society for self-definition, to one of

accepting their position as Black persons in a White world.

Similarly, Cross's (1971) four-stage model, known as the

"Negro-to-Black Conversion Experience," postulates that one

evolves from a self-perception in which blackness is

degraded, to a self-perception in which one is firmly secure

with one's blackness. Several other theorists have

developed similar stage models about Black racial identity

(Jackson, 1975; Williams, 1981) and identity of other

minority groups (Atkinson, Morton, & D.W. Sue, 1979).

Cross's model has received considerably more research

attention than any other model (e.g. Parham & Helms, 1981,

1985a, 1985b; Ponterotto & Wise, 1987). One study in

particular stands out from among the rest, lending support

for the notion that racial identity development is one of

the most important current issues in the field of







multicultural counseling. In this study, the relationship

between Black racial identity development and cross-cultural

counseling was highlighted (Parham & Helms, 1981). This

study was the most referenced empirical article in the

multicultural counseling literature as identified in a

recent 5-year content analysis of nearly 4,000 citations

used in racial/ethnic minority-focussed counseling research

(Ponterotto & Sabnani, 1989). Additionally, Cross's (1971)

"Negro-to-Black Conversion Experience" was found to be the

third-most-referenced conceptual article in the multi-

cultural counseling literature.

Clearly, racial/ethnic identity in general, and Black

racial identity in particular is an important contemporary

issue in the field of multicultural counseling. Indeed, the

April 1989 issue of The Counseling Psychologist is devoted

entirely to exploring Black racial identity. Although

models of racial identity stages are a basic component of

multicultural counselor training, and research has focused

almost exclusively on minority populations, particularly

Blacks, a recent surge in the literature is challenging this

focus. The principle behind the movement is basic to

counseling theory--counseling is a relationship; and in

multicultural counseling it is predominantly a relationship

between a White counselor and a racial/ethnic minority

client. Ponterotto (1988) states it well:

Although examining and incorporating the racial
identity development levels of minority clientele into








the counseling process is of great importance, so too
is the examination of the racial identity and
consciousness levels of the counselors, who are
oftentimes White. It seems that a whole arena of the
counseling process has been neglected. The counseling
encounter is a dyadic process in which both
interlocutors contribute to the success of the
encounter. What pragmatic use will the counselor's
understanding of the client's racial consciousness
level foster, if the counselor himself or herself is
racially 'unconscious.' (p. 149)


The idea that White counselors need to examine their

tendency to be "culturally encapsulated" (Wrenn, 1962) or

racially blind (Katz & Ivey, 1977) is not new to the field

of counseling. Additionally, the APA Executive Goals

Committee of Division 17 (Counseling) endorsed the guideline

that counselors should be aware of their own values and

biases and how they may affect minority clients (Smith,

1982).

A new and significant development in multicultural

counselor training is the recognition that White people

progress through similar developmental stages as Black

people and other racial/ethnic minorities. For example,

Helms (1984) developed a five-stage model of White racial

consciousness. These stages are Contact, Disintegration,

Reintegration, Pseudo-Independence, and Autonomy. Helms

explained that unlike Blacks, Whites are the dominant race

in this country and can choose environments that permit them

to remain at a particular stage of racial consciousness.

Therefore it is possible that each stage can culminate in

either a positive or negative resolution -- positive being







associated with greater personal adjustment and better

interpersonal relationships with people of other races. A

similar stage theory was developed by Jackson and Hardiman

(1983) and Corvin and Wiggins (1989). All three models

recognize that for Whites to develop a healthy sense of

racial identity, awareness must occur on two issues: (1)

awareness of an identity as a member of the White race; and

(2) awareness of racism as part of the dominant White

culture.

Suggestions for facilitating the development of White

racial identity have been made by several authors. For

example, Helms (1984) has created a cognitive development

model that responds to the trainee's level of racial

consciousness. She proposes that trainees start with

historical and sociocultural information, then proceed to

skill building, cognitive and affective self-awareness, and

finally to cultural emersion. This model also takes into

account the cultural predisposition of the minority client.

As such, it is an interactional model that can be used for

investigating cross- and same-race counseling processes.

Corvin and Wiggins (1989) created an antiracism

training model for White professionals. Drawing on their own

experiences as antiracism trainers, they combined Hardiman's

(1982) first two stages and integrated them into a training

framework. Parallelling Helms' ideas, they conceptualized

four developmental stages based on the belief that in order








for a White person to be able to develop multicultural

competence, one must first recognize, assess, and understand

oneself as a member of a particular racial group.

Additionally, one must also become aware of one's own racism

and take active steps to change. Their training procedures

involve increasing racial self-awareness, and understanding

the existence and impact of racism on other Whites, non-

Whites, and on oneself. The procedures are intended to help

trainees integrate their experiences by providing a

framework for understanding their level of racial attitude

development and guidance for continued development.

Ridley (1989) has written on the special multicultural

training needs of White professionals. He delineated

specific types of behavior exhibited by White counselors

that need to be understood and discontinued because they

systematically produce adverse consequences for ethnic

minority clients: (1) color blindness, (2) excessive color

consciousness, (3) failure to recognize or effectively deal

with cultural transference, (4) cultural counter-

transference, (5) ambivalent counseling motives, (6) evoking

pseudotransference, and (7) misinterpreting client

nondisclosure.

Effects of Multicultural Counselor Training

Despite the evident need and interest for multicultural

counselor training, multicultural counselors view existing

efforts as experimental, scarce (Lefley, 1985), and lacking





13

direction because of the haphazard approach and low priority

given to multicultural counselor training. Criticism has

also been voiced regarding the inadequate dissemination of

information of successful training programs, resulting in

unnecessary duplication of efforts and a lack of

sophisticated critical analysis of existing programs

(D.W. Sue, 1981).

Very little research has been conducted to test the

effects of any aspect of multicultural counselor training.

However, observations and exploratory studies of graduate

students enrolled in "counseling ethnic minorities" classes

indicate that a process of attitude change takes place. For

example Ponterotto (1988) observed that most of the White

students in his multicultural counseling course pass through

four stages of "majority" development, similar to the stages

described by Helms (1984). Although his findings are based

on personal experience and not empirical testing, Ponterotto

suggests that students enter class naively and perhaps

racist, and emerge accepting and appreciating their own and

other races. Research conducted with graduate counseling

students enrolled in similar courses at another university

indicated that trainees were more conflicted (Fukuyama,

Metzler, & Heath (1988) following the completion of the

class.

Lefley (1985) discussed the major unresolved issues for

cross-cultural training. Among the questions she addressed








were a) who should be trained and at what level, b) can

cross-cultural training usefulness be demonstrated

empirically, c) and how can its relevance to positive

outcome in the field be evaluated? Added to this list is

the basic question specifically related to this study:

Given the role of racial identity stage theory as a central

component of multicultural counselor training, what is the

impact of racial identity training on attitudes and

perceptions of trainees?

The lack of adequate control groups and the effects of

demand characteristics lead to the conclusion that there is

a strong need to begin systematic empirical research to test

the effects of the various components of multicultural

counseling courses on the attitudes and perceptions of

counselors-in-training. To date, no research has been

conducted to test the impact of racial identity stage theory

training on attitudes and perceptions of trainees, despite

the fact that it is a primary component of multicultural

counseling courses and training seminars.

Attitudes and Perceptions of Whites Toward Blacks

Some of the questions multicultural counselors have

about White people's attitudes and perceptions of Black

people, and how training experiences may affect these

attitudes and perceptions, can be answered by social

psychologists. Social psychologists are the major

contributors to the psychology of intergroup relations and








have provided a rich mine of empirically based theoretical

models of how people process and act upon information about

social groups.

For example, research conducted at the National Opinion

Research Center (NORC) between 1940 and 1978 (Greeley &

Sheatsley, 1971; Hyman & Sheatsley, 1956, 1964; Taylor,

Sheatsley, & Greeley, 1978), and the National Advisory

Committee of Civil Disorders (Campbell, 1971) is responsible

for much of the data on White's attitudes toward Blacks.

The NORC surveys conducted in 1942, 1956, 1963, 1970, 1972,

and 1976 indicate an egalitarian trend in the expressed

racial attitudes of Whites toward Blacks. Research has

shown that White's stereotypes of Blacks have become more

favorable and more similar to White's stereotypes of Whites.

Unfortunately, other research indicates that these

egalitarian trends in racial attitudes and stereotyping may

be more superficial than real. For example McConahay,

Hardee, and Batts (1981) found that many of the prejudice

items used in the opinion surveys are highly susceptible to

social desirability influences. Subjects will often change

their responses on these items in order to appear more

egalitarian than they are. Moreover, research on

stereotyping indicates that self-report measures are often

systematically distorted to appear less biased (Sigall &

Page, 1971).






16

Research support has been found for several conceptual

models that have been developed to understand racial

prejudice as it is manifested today. For example, Gaertner

and Dovidio (1986) suggest that many White Americans

experience a particular type of ambivalence in which there

is conflict between feelings and beliefs associated with a

sincerely egalitarian value system, and unacknowledged

negative feelings and beliefs about Blacks. The negative

affect these aversivee racists" have for Blacks is not

hostility or hate, rather it involves "discomfort,

uneasiness, disgust, and sometimes fear" (p. 63). These

authors report evidence showing that Whites do not exhibit

bias against Blacks when norms prescribing appropriate

behavior are clear; but Whites do discriminate when norms

are ambiguous or conflicting.

Gaertner and Dovidio (1986) offer a sociological

perspective and explain that prejudice and racism is

perpetuated by the structure of society. Beliefs about

relative status and power are embedded in social roles and

norms. In the most important aspects of American life,

Whites currently have advantages relative to Blacks (e.g.

standard of living, educational achievement, socioeconomic

status, life expectancy, infant mortality). Therefore,

even if people genuinely attempt to reject the
socially less desirable stereotypes and characteristics
of Blacks, it may be difficult for even the most well-
intentioned White person to escape the development of
negative beliefs concerning Blacks and to avoid








feelings of superiority and relative good fortune
over the fact that they are advantaged. (p. 65)


A similar conclusion was recently reached by Devine (1989).

She found that high- and low-prejudiced White students

equally generate negative stereotypes about Blacks.

However, if low-prejudiced Whites are allowed to monitor

their thoughts they actively reject negative stereotypes of

Blacks.

Ingroup/Outgroup Perception and Stereotypes

Social psychologists have developed various ways of

conceptualizing and measuring stereotypes. Traditionally,

stereotypes have been defined as a set of traits that

characterize a group of people. Stereotypes were also

considered to be inherently bad or deviant, and were usually

equated with the concept of prejudice. The vast majority of

research on stereotypes is descriptive and uses Katz and

Braly's (1933, 1935) checklist method. Subjects are given a

list of traits and asked to check the traits that are most

typical of a particular group. Using this procedure, a

stereotype is defined as those characteristics on which

there is consensus among the subjects and is referred to as

a "social stereotype."

Currently there is considerable disagreement among

researchers about whether or not stereotypes are bad by

definition, and whether stereotypes constitute individual or

consensual sets of beliefs (Ashmore & Del Boca, 1981).








Alternatives to the descriptive method of stereotype

measurement, and a cognitive orientation that views

stereotypes and stereotyping as natural cognitive processes,

have generated research that more fully explores the

origins, nature, and effects of stereotypes. For example,

McCauley, Stitt, and Segal (1980) propose that stereotypes

are not just consensual characterizations of a social group

based on a fixed set of adjectives. They criticize this

approach stating that the major flaw in the checklist

methodology is that it does not provide a measure of

individual stereotyping. Instead, they define stereotypes

as "those generalizations about a class of people that

distinguish that class from others" (p. 197). In essence,

stereotyping is differential prediction based on group

membership. Additionally, McCauley, Stitt, and Segal (1980)

view stereotype predictions as probabilistic rather than

all-or-none. Thus, someone who believes that all American's

are rude would be stereotyping, but someone who believes

that Americans are more likely to be rude than other people

of the world is also stereotyping.

Linville and her colleagues (Linville & Fisher, 1986;

Linville, Fisher, & Salovey, 1989) offer another perspective

on stereotypes and stereotyping. Instead of examining

differences between groups (e.g. how Americans differ from

people in general), they look at within group differences

(e.g. how many different types of Americans one can think






19

of), and consider the lack of category differentiation to be

the primary ingredient of stereotypes. Since individuals

are more familiar with groups to which they belong, they

will be more differentiated in their thinking about their

own group than about other groups. In other words, people

tend to perceive outgroup members as being more homogeneous

in their traits and behavior than ingroup members. This

tendency is referred to as the "Outgroup Homogeneity

Hypothesis" (Quattrone & Jones, 1980).

Linville et al. (1989) propose two statistical

properties of perceived distribution of the characteristics

of social category members: attribute differentiation and

perceived variability. Attribute differentiation refers to

the probability of distinguishing between two group members

in terms of a given attribute. A differential measure can

be used to detect stereotyping in the sense of over-

generalization and a failure to make distinctions among

group members. Perceived variability refers to the extent

to which category members are viewed as being widely

dispersed about the mean of the attribute. Very low and

very high measures of variability reflect two different

types of stereotyping; the former reflects very little

dispersion, while the latter reflects a tendency to see

group members as lying at either of two polar extremes.








Effects of Perceived Out-aroup Homogeneity

Several researchers have studied the impact that

perceived homogeneity has on attitudes and perceptions. For

example, Linville and Jones (1980) provided evidence

suggesting that because people have more complex schema

regarding in-groups than out-groups, judgments of outgroup

members will be more extreme or polarized than judgments of

in-group members. Perceived out-group homogeneity also

seems to be related to out-group discrimination. Wilder

(1978) found that subjects discriminated when the out-group

was perceived to be unanimous in its behavior but reacted

more favorably when the out-group was more individuated.

Quattrone and Jones (1980) hypothesized that the

tendency to generalize from the behavior of a group member

to the group as a whole is facilitated by perceived

homogeneity of the group. In their study, students from two

rival universities, Princeton and Rutgers, saw a videotape

of either a Princeton or Rutgers student making a decision

during a psychology experiment (e.g. to wait alone or with

other students). After viewing the choice of the student in

the video, subjects were asked to estimate the proportion of

students from that same university who would make the

identical decision as the student in the video. Overall,

the estimated proportions were greater when the student was

an outgroup member. That is, Princeton students generalized

more strongly to the behavior of the Rutgers population







after observing the choice of the Rutgers participant than

they did to the Princeton population after observing the

choice of the Princeton participant.

Among the factors Quattrone and Jones concluded to be

important when taking into consideration the assessment of

an individual's perception of a group's variability are (1)

the number of subgroups believed to compose the group, (2)

the extent to which characteristics consistent with the

group's stereotype are attributed to the average member of

the group, and (3) direct ratings of the group's

variability.

Nisbett, Krantz, Jepson, and Kunda (1983) replicated

and extended Quattrone and Jones study using University of

Michigan students. Subject were told that the videotapes

were either University of Michigan or Ohio State University

students. As Quattrone and Jones reasoned, so had Nisbett

and his colleagues: "generalizations about groups from the

behavior of its members are mediated by assumptions about

variability of group members, it should be possible to

manipulate those assumptions and therefore to influence the

degree of generalization" (pp. 350-351). Nisbett

hypothesized that requiring subjects to contemplate the

central tendencies of a university population, before

viewing the choice behavior, would cause an increase in the

salience of perceived population homogeneity, thus leading

subjects to generalize more. As expected, inducing subjects









to contemplate central tendencies increased the degree of

generalization.

Several researchers have shown that simply categorizing

people into distinct groups is sufficient to arouse

intergroup bias (see reviews by Brewer, 1979; Hogg, Abrams,

& Patel, 1988; Stephan, 1985; and Wilder, 1986). Similarly,

having out-group members respond as individuals rather than

as a group (Wilder, 1978), or recategorizing ingroup and

outgroup members into one superordinate group, or as

separate individuals (Gaertner, Mann, Dovidio, Murrell, &

Pomare, 1990) reduces intergroup bias.

Another factor that can increase ingroup/outgroup bias

is ingroup cohesiveness. Cohesiveness among ingroup members

is related to the effects of similarity, in that increased

similarity among ingroup members leads to greater

liking of ingroup members and increased outgroup bias

(Billig & Tajfel, 1973).

Improving Interqroup Relations

Social psychologists have made a major contribution to

the examination of techniques for improving intergroup

relations. The most well known approach for modifying

intergroup relations is contact.

Many other approaches have been devised to reduce

prejudice by using group interaction settings. Studies have

used techniques adapted from group therapy, while others

have used group discussion and other types of structured






23

groups. The majority of these techniques have proved to be

successful (Stephan, 1985).

Courses or segments of courses on intergroup relations

in school settings are among the most widely advocated

methods for improving intergroup relations and are the

approach most clearly related to the present study.

Unfortunately, these studies suffer from some of the same

limitations as mentioned in the multicultural counseling

research: poor control groups, alternative explanations for

positive results (especially demand characteristics), lack

of follow-up data, and a lack of comparison of alternative

techniques so that there is little information on what types

of course materials are most effective (Stephan, 1985).

The purpose of this study was to investigate the

effects of Minority and Majority Identity Development theory

training on White students' stereotypes and prejudice of

Blacks. The subjects, design and procedures, dependent

variables, and hypotheses are discussed in the next section.












CHAPTER III
METHODS

Subjects

The subjects for this study were drawn from the

population of White undergraduate students enrolled in

introductory psychology courses. There were a total of 66

subjects participating in the study. Three subjects were

dropped from the study because they were significantly older

than the others. The remaining 63 consisted of 38 females

and 25 males. Ages ranged from 18 to 21. Table 1 presents

the number of subjects by group and time.

TABLE 1

Subject Frequency by Group and Time



Pretest Posttest Follow-up

Minority/ 21 21 21
Majority
Training

Minority 15 15 14
Training

No 27 25 22
Treatment
Control








Design and Procedures

This study is a 3 (group) x 3 (time) mixed factorial

design. Subjects signed-up for one of three groups: (1)

Treatment Group 1: Minority and Majority Identity stages,

(2) Treatment Group 2: Minority Identity stages, and (3)

No-treatment control group.

The dependent measures involved a series of paper-and-

pencil questions, the majority of which were administered to

each subject on a pre-, post-, and follow-up basis, one week

prior to the treatment, immediately after the treatment, and

three weeks after the treatment, respectively. Two other

dependent measures were administered on a posttest basis

only. No-treatment control subjects were administered the

same questions during the same time intervals as the two

treatment groups.

To control for demand characteristics, a research

assistant administered the dependent measures and told the

subjects that she was conducting a study separate from the

researcher administering the training sessions (this author)

in an effort to maximize the assignment of credit by asking

subjects to participate in two studies back-to-back. The

research assistant told subjects that she was doing a

statistical project on test-, re-test reliability, and

encouraged them to answer the questions as honestly as

possible for all three administrations of the questionnaire.

Also, to prevent subjects from changing their answers, each







26
page of the dependent measures was handed out and completed

one at a time. The entire packet of dependent measures

required 20 minutes to complete.

After the follow-up, dependent measures were collected,

subjects were debriefed, and the nature and purpose of the

study was revealed. Questions regarding the use of

deception were answered (see Appendix A).

Treatment Groups

Both treatment group training sessions were conducted

by this author and began by telling subjects that the

purpose of the study was to look at the usefulness of a

models) used when training therapists, administrators, and

business managers to become more effective when working with

ethnic minorities. At the end of the training session

subjects were asked to rate the usefulness of the models)

for understanding racial identity.

Treatment Group 1 subjects were introduced to the

Majority Identity Development Stages and the Minority

Identity Development Stages as outlined by the American

College Personnel Association's training program. Each

subject was given a packet that included a general

description of each stage of identity development, followed

by examples of behaviors and attitudes and common statements

indicative of the stages (see Appendices B and C). Not

included in that packet, but presented orally, were the

advantages and disadvantages associated with each stage.







Subjects were told to pay close attention during the

didactic instruction because they were going to have to

place individuals seen in a 30 minute video into the

appropriate stage of identity development. The video was a

copy of a PBS special about Rutgers University Black and

White student's attitudes toward each other. It included

detailed interviews with three Black students and two White

students. Each interviewed student exhibited attitudes and

behaviors that could be associated with a different stage of

racial identity.

Subjects were given a Response Sheet (see Appendix D)

that included the name of the students seen in the video,

followed by the list of identity stage titles. They were

encouraged to take notes on their Response Sheet and told

that after they completed the sheet they would be asked to

share their matches and comments. The training session

concluded with a discussion that clarified the model by

reviewing the rationale for the correct Response Sheet

answers. The session took 1.5 hours.

Treatment Group 2 consisted of the same procedure as

the first, without the training in Majority (White) Identity

Development. This session lasted 1 hour. Both groups were

administered the dependent measures immediately after the

training session and again three weeks later.








Control Group

Subjects in the control group were not given any

treatment. They were given the same pre-, post- and follow-

up dependent measures as the treatment groups.

Dependent Variables

The dependent measures included three measures of

prejudice and four measures of stereotypes. The three

measures of prejudice were the Modern Racism Scale (MRS),

the Central Tendency of the Perceived Distribution Score

(M), and Desire for Contact (DFC). The four measures of

stereotyping included the Absolute Difference Score (AD),

Member-to-Group Inference (MTGI), Probability of

Differentiation (PD), and Perceived Variability (Var). Two

of these measures were administered on a posttest only

basis: Member-to-Group Inference and Desire for Contact

(see Appendix E).

Pilot Study

The majority of the dependent variables necessitated

the use of 8 attributes. A pilot study was designed and

implemented to select the specific attributes (see Appendix

F). One hundred students enrolled in an introductory

psychology class were asked to generate stereotypes for

Blacks and Whites. A sample of 20 of these students was

randomly selected and the 4 most frequently cited positive

and negative responses were used for the dependent measures.








The pilot was also designed to select the vignettes

used for creating the dependent measure used only during the

posttest: Member-to-Group Inference and Desire for Contact

(see Appendix G). The two most positively and negatively

rated vignettes were selected.

Perceived Distribution (Pd. Var, M)

Two measures of Whites' stereotypes and one measure of

prejudice were generated by asking subjects to assess

perceived distributions of Blacks for eight bipolar

attributes selected from the pilot test results. They were

asked to distribute 100 Black college students over seven

levels of each attribute (as in Linville, Fisher, & Salovey,

1989). For example, subjects were asked to form perceived

distributions of the "friendliness of UF Black students" by

distributing 100 Black students over seven levels of the

attribute friendliness: very unfriendly, unfriendly,

somewhat unfriendly, neutral, somewhat friendly, friendly,

and very friendly. Therefore, the subjects' task was to

estimate what proportion of the Black students at UF are

very friendly, and so forth.

Probability of Differentiation (Pd)

The two stereotype measures generated represent the

differentiation and variability of the perceived

distribution. For the first, let (p1,...,pm) represent a

perceived distribution for an attribute having m discrete

levels, where p, is the subjectively perceived proportion of








category members described by level i of the attribute in

question.

Attribute differentiation refers to the probability of

distinguishing among group members in terms of a given

attribute. The Probability of Differentiation score, Pd,

provides a direct measure of this property

Pd = 1 S,=Ii,.p

and is one measure of stereotype used in this study.

Operationally defined, Pd reflects the probability that two

randomly chosen group members will be perceived to differ in

terms of the attribute in question. The Probability of

Differentiation score assumes its minimum possible value if

one attribute level has a probability of 1.0 and all other

levels have a probability of 0. In this case, Pd = 0. That

is, there is a 0% chance that the perceiver will distinguish

between two category members in terms of this attribute

because all members of the category are perceived to be the

same with respect to this attribute. With a seven point

scale, as was used in the present study, Pd assumes its

maximum possible value if all attribute levels have a

probability of 14.28. In this case Pd would equal .86.

Perceived Variability (Var)

Perceived Variability is the other stereotype measure

generated from the subject's perceived distributions of

attributes. Operationally defined, Perceived Variability

refers to the degree to which the members of a group are







perceived to be widely dispersed (about the mean) in terms

of the attribute in question. The variance of a perceived

distribution provides a direct measure of this property

Var = EiI,mPi(Xi M)2

where X is an interval (or ratio) scale attribute, Xi is the

scale value of the ith level of the attribute, the number of

levels of the attribute is discrete, and M denotes the mean

of the perceived distribution. In this study, the minimum

and maximum possible values for Var are 0.0 and 9.0

respectively. For example, when Var = 9.0, it represents

the most widely dispersed variability about the mean of the

attribute in question because the distribution is perfectly

bimodal. That is, the category members were equally divided

and put at each extreme end of the seven point scale. Both

differentiation and variability have been shown to be

conceptually distinct properties of a perceived distribution

(Linville et al., 1989).

Central Tendency of the Perceived Distribution (M)

The Central Tendency (M) of the Perceived Distribution

Score was used to indicate the overall level of favorability

(or prejudice) toward Blacks and is defined by

M = i=lmPiXi

Using a seven point scale, M ranges from 1 to 7. For

example, if when forming perceived distributions of the

attribute "friendliness," a subject decides that all 100

Black students are in the seventh category "very friendly,"








then M would equal 7, indicating the highest level of

favorability.

Absolute Difference Score (AD)

Whites' stereotypes of Blacks were also measured using

a modified version of McCauley and Stitt's (1978) diagnostic

ratio. This measure defines stereotypes as "differential

trait attributions or differential predictions based on

group membership information" (McCauley, Stitt, & Segal,

1980, p. 197). In this study an "Absolute Difference Score"

was used instead of a ratio score (see Biernat, 1991) and

was calculated as

P(trait 1 I Black students) P(trait 1 White students)l

For example, subjects were asked to pretend they had met 100

Black students, and of those 100, how many would be

friendly. The same question was asked regarding 100 White

students. A stereotype exists if one sees Blacks as having

more or less of a trait than White students.

Modern Racism Scale (MRS)

The Modern Racism Scale (McConahay, Hardee, & Batts,

1981) is designed to measure anti-Black attitudes in a

nonreactive fashion. It has proven to be useful in

predicting a variety of behaviors including voting patterns

and reactions to busing. Subjects indicate their agreement

with each of the items on the 5-point rating scale that

ranges from -2 (disagree strongly) to +2 (agree strongly).

The instrument consists of 22 items, seven of which are used







to calculate the racism score (numbers 1,6,7,9,12,15, and

19). Item number 6 is reversed scored. Thus, the Modern

Racism Scale ranges from -14 (low prejudice) to +14 (high

prejudice). The scale has good reliability (Cronbach's

alpha .83).

Posttest Only Measures

Two other measures were taken for all three groups

during the posttest session only: Member-to-Group Inference

and Desire for Contact. The former may be conceptualized as

a measure of stereotype behavior while the latter may be

viewed as a measure of prejudice. Two stereotype measures

were generated by asking subjects to determine a group

stereotype based on the behavior of one or more members of

the group. For example, subjects were presented with a

vignette in which a Black student desired to have a meeting

with Black and White students to discuss interracial

relations. Subjects were asked what percentage of Black

students at the University of Florida would be interested in

such a meeting.

Three prejudice measures required subjects to indicate

their interest in participating in interracial activities

and relationships. For example, regarding the same vignette

referred to above, subjects were asked how interested they

would be in participating in the interracial discussion (see

Appendix E).








Hypotheses

Both treatment groups will be exposed to an identity

stage model that facilitates the perception of variety

within the racial category of Black Americans.

Stereotyping, as it is defined by Linville and her

colleagues (1986) should decrease for subjects receiving

Minority Identity training. More specifically, the

probability of distinguishing between two Blacks in terms of

a given attribute (Pd) should increase as a result of

training. Also, the extent to which Blacks are viewed as

being widely dispersed about the mean of a given attribute

(Var) should also increase as a result of training. The

resulting two hypotheses are:

1. Treatment Group 1 and 2 will have significantly higher

posttest Probability of Differentiation scores (Pd)

than the control group.

2. Treatment Group 1 and 2 will have significantly higher

posttest Perceived Variability scores (Var) than the

control group.

McCauley and Stitt (1980) view the perception of

between group differences as the primary component of

stereotyping. The more that subjects see themselves

(Whites) as similar to Blacks on the various attributes the

less they will be stereotyping. Minority Identity training

may provide subjects with the opportunity to see how some

Blacks are similar to Whites and to themselves.








Additionally, subjects who receive Majority Identity

training in conjunction with Minority Identity training,

will notice that Whites and Blacks go through similar stages

of racial identity development. The third hypothesis is as

follows:

3. Treatment Group 1 and 2 will have significantly lower

posttest Absolute Difference Scores (AD) than the

control group.

Member-to-Group Inference (MTGI) is the third dependent

variable that measures stereotyping. Quattrone and Jones

(1980) found that the tendency to generalize from the

behavior of one group member to the group as a whole is

facilitated by perceived group homogeneity. Minority

Identity training increases perceived group heterogeneity

and should therefore lead to a decrease in Member-to-Group

Inference responses. MTGI will be measured on a posttest

basis only. The fourth hypotheses is therefore:

4. Treatment Groups 1 and 2 will have significantly lower

Member-to-Group Inference scores than the control

group.

Prejudice involves an intricate interplay of social,

emotional and cognitive sources. Negative affect towards

Blacks is related to discomfort, uneasiness, and fear

(Gaertner & Dovidio, 1986). Furthermore, stereotyping, or

undifferentiated thinking, has been found to facilitate

discrimination (Wilder, 1978). Therefore, it is possible









that Minority Identity training may not only lead to a

decrease in stereotyping of Blacks but also to a decrease in

discomfort, uneasiness, fear, and the negative feelings

associated with discrimination. Based on this rationale,

the following three hypotheses were created for the three

dependent measures of prejudice:

5. Treatment Group 1 and 2 will have significantly lower

posttest prejudice scores, as measured by the Central

Tendency of the Perceived Distribution score (M), than

the control group.

6. Treatment Group 1 and 2 will have significantly lower

posttest Modern Racism Scale (MRS) scores than the

control group.

7. Treatment Groups 1 and 2 will have significantly lower

prejudice scores, as measured by Desire for Contact

(DFC), than the control group.

No hypotheses were be made concerning the change in

attitudes from posttest to follow-up test because the long

term effects of training, if any, are unpredictable.











CHAPTER IV
RESULTS

To adjust treatment effects for any pretest differences

between groups, a repeated measures analysis with a

covariate for time 1 pretestt) scores was performed along

dependent measures in a 3 (Minority and Majority Identity

training, Minority Identity training, no treatment control)

x 3 pretestt, posttest, follow-up test) design. Dependent

variables included Probability of Differentiation (Pd),

Perceived Variability (Var), Central Tendency of the

Perceived Distribution (M), Absolute Difference Score (AD),

and the Modern Racism Scale (MRS). In addition, two

dependent measures were administered on a posttest only

basis (Member-to-Group Inference and Desire for Contact).

One-way analyses of variance were performed along each of

these measures.

Perceived Distribution (Pd. Var. M)

Probability of Differentiation (Pd)

It was hypothesized that there would be greater

posttest means of the Probability of Differentiation scores

(Pd) for both Treatment Groups compared to the control

group. Results of the 3 (group) x 3 (time) analysis of

covariance (ANCOVA) revealed no significant differences.

Neither the predicted group effect, F(2,52)= 1.88, R>.16,

37








nor the time x group interaction, F(4,102)= 1.13, p>.34,

reached significance, and for this reason results did not

support these hypotheses, (see Table 2 for means and

standard deviations).

TABLE 2

Means and Standard Deviations for Probability of
Differentiation



Pretest Posttest Follow-up Test



Group N M S.D. M S.D. M S.D.



Minority/ 20 .68 .14 .68 .11 .72 .09
Majority
Training

Minority 14 .69 .11 .72 .11 .74 .07
Training

No 22 .73 .08 .75 .09 .77 .07
Treatment
Control


Note. Larger numbers indicate a higher probability that two
randomly chosen Black students were perceived to differ
along attributes. Numbers in table reflect raw means.

Perceived Variability (Var)

It was hypothesized that there would be higher posttest

means of Perceived Variability (Var) for the two Treatment

Groups compared to the control group. As with the Pd, the

ANCOVA on the Var scores did not yield significance.

Neither the analysis for group, F=(2,52)=1.92, R>.15, nor

the predicted time x group interaction, E(4,102)=1.91,





39

E>.11, reached significance, and for this reason results did

not support hypotheses (see Table 3 for means and standard

deviations).

TABLE 3

Means and Standard Deviations for Perceived Variability



Pretest Posttest Follow-up Test



Group N M S.D. M S.D. M S.D.



Minority/ 20 2.38 .92 2.23 .91 2.48 .82
Majority
Training

Minority 14 2.41 .61 2.70 .62 2.61 .62
Training

No 22 2.75 .78 2.78 .65 2.82 .62
Treatment
Control


Note. Larger numbers indicate less stereotyping and
represent the degree to which subjects saw Black students as
being widely dispersed (about the mean) in terms of an
attribute. Numbers in table reflect raw means.

Central Tendency of the Perceived Distribution (M)

The Central Tendency of the Perceived Distribution (M)

is used as a measure of favorableness or prejudice. In

order to assess the hypothesis concerning the Central

Tendency of the Perceived Distribution in a manner that

would be interpretable, separate 3 (group) x 3 (time)

ANCOVAS were performed for the positive (e.g. athletic,

intelligent) and the negative (e.g. lazy, aggressive)








attributes. Additionally, because of the relationship

between M and Var, if M has any significant changes between

groups or across time along the same parameters as Var, then

those particular results for M will not be reliable.

Positive M

It was hypothesized that there would be a decrease in

prejudice, as measured by the Central Tendency of Perceived

Distribution scores, for both Treatment Groups as compared

to the control group. However, no significant main effect

for group was found, F(2,52)=1.72, R>.18, and clarifying

simple effect tests for a significant time x order

interaction F(4,102)=3.33, E<.01 revealed insignificant

between group differences on the posttest, F(2,56)=2.27,

R>.1, and follow-up test scores, F(2,52)=2.37, R>.1.

Simple effects tests for the within subjects variable

yielded a significant increase in favorableness across time,

F(2,104)=3.72, R<.05, for subjects who did not receive

training. However, because the control group simple effects

test for time was also significant for the Var,

E(2,104)=4.03, R<.025, the significant control group effects

for M are not interpretable. Simple time effects tests for

the other two groups were not significant,F(2,104)=.841,

R>.2, F(2,104)=.809, R>.2, for Treatment Group 1 and 2

respectively.









TABLE 4

Means and Standard Deviations for the Positive Attributes of
the Central Tendency of Perceived Distribution Score



Pretest Posttest Follow-up Test



Group N M S.D. M S.D. M S.D.



Minority/ 20 3.19 .42 3.25 .50 3.47 .42
Majority
Training

Minority 14 3.21 .49 3.13 .47 3.23 .45
Training

No 22 3.18 .64 3.38 .62 3.27 .60
Treatment
Control


Note. Larger numbers indicate greater prejudice toward
Black students. Numbers in table reflect raw means.

Negative M

The hypothesis that the two treatment groups would have

lower posttest Central Tendency of Perceived Distribution

scores than the control group was not supported for Negative

M. Results of the 3 x 3 ANCOVA revealed no main effect for

group, F(2,52)=1.30, E>.28. And, as was the case with the

Positive M, a significant time x group interaction effect,

E(4,102)=2.39, R<.05, did not lead to significant simple

order effects for posttest, E(2,56)=1.64, p>.2, nor follow-

up test, E(2,52)=2.08, R>.13.








Across time, both treatment groups decreased and then

increased in favorableness, with subjects receiving Minority

Identity training alone reaching significance,

F(2,104)=3.13, 2<.05. A Tukey's (R<.05) pairwise comparison

revealed that the decrease in favorableness from pretest

(M=4.12) to posttest (M=3.95) was not significant, but the

increase in favorableness (or return to pretest levels) from

the posttest to follow-up test (M=4.24) was significant (see

Table 5 for means and standard deviations).

TABLE 5

Means and Standard Deviations for the Negative Attributes of
the Central Tendency of the Perceived Distribution Score



Pretest Posttest Follow-up Test



Group N M S.D. M S.D. M S.D.



Majority/ 20 3.88 .77 3.69 .71 3.79 .64
Minority
Training

Minority 14 4.12 .43 3.95" .52 4.24b .38
Training

No 22 3.98 .41 3.99 .44 3.89 .56
Treatment
Control


Note. Smaller numbers indicate greater prejudice toward
Black students. Means with different superscripts differ
from each other by at least p<.05. Numbers in table reflect
raw means.









Absolute Difference Score (AD)

It was hypothesized that there would be a decrease in

posttest Absolute Difference Scores (AD) for Treatment

Groups 1 and 2 that significantly differed from the control

group. The repeated measures ANCOVA revealed nonsignificant

main effects for group, F(2,51)=2.59, p<.08, and time x

group, F(4,100)-2.18, R<.07 (see Table 6 for means and

standard deviations).

TABLE 6

Means and Standard Deviations for the
Absolute Difference Score



Pretest Posttest Follow-up Test



Group N M S.D. M S.D. M S.D.



Minority/ 22 17.99 5.84 13.58 6.52 14.91 7.34
Majority
Training

Minority 14 18.88 5.83 10.75 6.80 10.00 4.64
Training

No 21 13.18 5.06 9.79 7.24 9.26 6.84
Treatment
Control


Note. Larger numbers indicate greater stereotyping, i.e.
seeing Black students as having more or less of a trait than
White students. Numbers in table reflect raw means.








Modern Racism Scale (MRS)

It was hypothesized that there would be a decrease in

the scores of the Modern Racism Scale for both Treatment

Groups as compared to the control group. The repeated

measures ANCOVA revealed no significant main effects for

group, E(2,52)=1.27, E>.2; however, there was a significant

time x group interaction, F(4,102)=2.80, E<.02. An analysis

of variance on posttest scores yielded significance,

F(2,56)=4.65, R<.01. Subsequent pairwise comparisons

revealed that subjects receiving Majority and Minority

Identity training were significantly more negative toward

Blacks (M=-3.6) than were subjects who received training on

Minority Identity only (M=-7.1) and control group

subjects,(M=-7.3), P,.05, Tukey's. Differences between

groups were not significant during the follow-up test.

A significant time effect, F(2,104)=5.43, R<.01, was

reached for subjects who received Majority and Minority

Identity training, revealing a significant increase in

prejudice from pretest (M=-4.9) to posttest (M=-3.6), and

then significantly decreasing in prejudice from posttest to

follow-up test, (M=-5.5, Tukey,s, R<.05), returning to

pretest levels. Treatment Group 2 subjects decreased in

prejudice after training but the difference was not

significant, E(2,104)=.336, R>.2 (see Table 7 for means and

standard deviations).








TABLE 7

Means and Standard Deviations for the Modern Racism Scale



Pretest Posttest Follow-up Test



Group N M S.D. M S.D. M S.D.



Majority/ 20 -4.9c 4.6 -3.6'bc 4.1 -5.5d 3.9
Minority
Training

Minority 14 -6.1 4.2 -6.6a 3.5 -6.3 3.9
Training

No 22 -6.9 4.3 -7.1b 4.6 -7.3 5.1
Treatment
Control


Note. Larger numbers (i.e. more positive) indicate greater
negative attitudes toward Blacks. Means with a common
single-letter superscript differ from each other by at least
R<.05. Numbers in table reflect raw means.

Posttest Only Measures

An analysis of variance (ANOVA) was performed on each

question assessed by the dependent measures that were

administered during the posttest session only. Simple

effects measures were performed when appropriate.

Member-to-Group Inference (MTGI)

Two questions relevant to the study addressed Member-

to-Group Inference. It was hypothesized that Treatment

Groups 1 and 2 would have significantly lower MTGI scores

than the control group. A one-way analysis of variance








yielded a tendency toward a significant main effect,

F(2,60)=2.81, R>.06, for the question (number I.a., Appendix

E) that began with the vignette "A Black student asked to

meet with Vice President Art Sandeen to express his desire

to get Black and White students together to discuss ways to

improve relations between Blacks and Whites on campus."

Subjects were asked what percentage of UF Black students

would want to be involved in such a meeting (see Table 8 for

means and standard deviations).

The second question (number III.d., Appendix E) yielded

a significant main effect, E(2,60)=3.83, R<.02, and began

with the vignette "A Black student was feeling angry and

picked a fight with a White student for no apparent reason,"

and asked subjects "What percentage of Black students at UF

do you believe would exhibit the type of behavior as the

Black student in this episode?" A Tukey's pairwise

comparison yielded significant differences (p<.05) between

means of Treatment Group 1 (M=27.1) and the means for both

Treatment Group 2 (M=15.6) and the control group (M= 17)

along this question. In other words, subjects receiving

Minority and Majority Identity training were more negatively

stereotypical in their view toward Black students than

subjects who received Minority Identity training only and no

training.








TABLE 8

Means and Standard Deviations for Member-to-Group Inference



Questions



I.a. III.d.



Group N M S.D. M S.D.



Minority/ 21 39.9 17.4 27.12 16.7
Majority
Training

Minority 16 45.5 16.5 15.66 9.2
Training

No 26 53.6 23.3 17.0b 14.8
Treatment
Control


Note. Larger numbers for I.a. indicate a positive MTGI,
while large numbers for III.d. indicate a negative MTGI.
Means with different superscripts differ from each other by
at least R<.05.

Desire for Contact (DFC)

Three questions asked during the posttest session only

addressed subjects' interest in becoming involved in an

interracial dialogue and having friends who are Black

students. It was hypothesized that Treatment Group 1 and 2

would have significantly lower prejudice scores, as measured

by DFC, than the control group. One-way analyses of









variance yielded significant main effects for two of these

questions (see Table 9 for means and standard deviations).

TABLE 9

Means and Standard Deviations for Desire for Contact



Questions



I.b IV.f. IV.g.



Group M S.D. M S.D. M S.D.



Minority/ 3.8a 1.64 3.1" 1.48 2.1 1.46
Majority
Training

Minority 2.1" 1.18 1.9b 1.18 3.2 1.28
Training

No 2.8 1.44 2.1c 1.33 2.9 1.5
Treatment
Control


Note. Larger numbers for I.b. and IV.f. indicate less
desire for contact with Black students. Smaller numbers for
IV.g. indicate less desire for contact. Means with a common
single-letter superscript differ by at least E<.05.

Significant group differences were found, F(2,57)=5.89,

R<.004, for the question (number I.b., Appendix E) that

began with the vignette "A Black student asked to meet with

Vice President Art Sandeen to express his desire to get

Black and White students together to discuss ways to improve

relations between Blacks and Whites on campus," and asked






49
subjects "How interested would you be in participating in a

discussion like this?" A Tukey's pairwise comparison

indicated that subjects who received Minority and Majority

Identity training (Treatment Group 1) were significantly

less interested in participating in the interracial

discussion (M=3.8) than subjects who received Minority

Identity training only (M=2.1).

Significant group differences were also found,

F(2,57)=4.68, R<.01, for the question (IV.f.) that asked,

"In general, how interested would you be in becoming friends

with Black students at UF?" A Tukey's pairwise comparison

indicated that subjects who received Minority and Majority

Identity training (M=3.1) were significantly less interested

in becoming friends with Black students than were subjects

who received Minority Identity training only (M=1.9) or no

training (M=2.1).

No significant differences were found for the question

(IV.g.) that asked subjects, "If you were to have 10 very

close friends from UF by the time you graduate, how many of

them would you like to have that were Black," F(2,52)=2.87,

E>.06.

Post Hoc Analyses

Training Receptivity and the Dependent Variables

This study required that two groups receive training in

Majority and/or Minority Identity Development. As part of

the training, subjects were asked to indicate on a response









sheet into which identity stage students seen in a video

belonged. Subjects in Treatment Group 1 had to respond to

five video-taped students while Treatment Group 2 subjects

responded to three of the students. The range of incorrect

answers was from 0 to 2. No subject picked all incorrect

responses, but compared to Treatment Group 1, a higher

percentage of Treatment Group 2 subjects picked all correct

responses (see Table 10).

TABLE 10

Percentage of Subjects Answering Incorrectly
on the Response Sheet



None Incorrect One Incorrect Two Incorrect



Minority/ 45% 40% 15%
Majority
Training

Minority 73% 20% 6%
Training



It is possible that subjects who had difficulty

matching individuals with their appropriate stage were less

receptive to training because of anti-Black attitudes. A

post hoc decision was made to assess whether there was a

relationship between the ability to learn the racial

identity stages, as measured by response sheet answers, and

the dependent variables. The percentage of correct answers

("Learn" variable) was calculated for each subject and








correlated with the pretest scores of the dependent

variables. There was a significant trend toward a positive

correlation between Positive M and the Learn variable,

r=.39, P>.08, and a trend toward a negative correlation

between Negative M and the Learn variable, r=-.39, g>.08 for

Treatment Group 1. That is, subjects higher in prejudice

may have been less receptive to training while those with

positive feelings toward Blacks were more receptive to

training. No other correlations for either treatment group

were close to significance.

Training Relevance of the Attributes Used in the Dependent
Variables

It is possible that non-significant results on measures

of Pd, Var, Positive and Negative M, and AD were related to

the relevance of the attributes used in these measures to

the training received by subjects. For example, the

attributes "aggressive," "clickish," "friendly," and

"prejudiced" (see Appendix E) were addressed or implied

during the racial identity training, and used to generate

the Pd, Var, Positive and Negative M, and AD. Conversely,

the attributes "athletic," "intelligent," "lazy," and

"trustworthy" were not addressed during the training but

were also used in calculating the above dependent variables.

To explore the possibility that attribute relevance may have

led to not finding a treatment effect where there might have

been one, separate ANCOVAs were performed for relevant and








non-relevant attributes of the Pd, Var, Positive and

Negative M, and AD.

Probability of Differentiation (Pd)

Results of the ANCOVA for the relevant attributes

revealed a significant order effect, F(2,52)=3.17, R=.05,

and a non-significant time x order effect, F(4,104)=1.21,

R>.31. A follow-up test (Tukey's, E<.05) for the

significant order effect yielded significant differences

between the overall means of Treatment Group 1 (M=.696) and

the control group (M=.77), indicating that overall,

Treatment Group 1 subjects stereotyped more (differentiated

less) than control group subjects.

Results of the ANCOVA for the non-relevant attributes

of the Pd did not yield significant order, F(2,52)=.72,

R>.49, nor time x order effects, F(4,104)=.46, R>.76.

Perceived Variability (Var)

The ANCOVA on the relevant attributes of the Var

revealed a significant order, F(2,52)=4.33, R=.01, and a

significant time x order effect, F(4,104)=4.15, R<.004.

Simple effects test for order revealed a significant

posttest difference, F(2,56)=7.76, p=.001, with Treatment

Group 1 (M=2.0) demonstrating significantly more

stereotyping (less perceived variability) than Treatment

Group 2 (M=2.6) and the control group (M=2.8). There were

no significant between group differences for time 3 (follow-

up test), F(2,52)=.79, E>.46.








Simple effects tests for the within subjects variable

yielded a significant overall time effect for Treatment

Group 1, F(2,104)=5.25, R<.01, with the increase in

stereotyping from pretest (M=2.41) to posttest (M=2.0)

reaching significance. The changes across time for

Treatment Group 2, F(2,104)=.21, R>.2, and the control

group, F(2,104)=1.49, E>.2, did not reach significance.

Results of the ANCOVA for the non-relevant attributes

of the Var did not yield significant order, E(2,52)=.47,

p>.63, nor time x order effects, F(4,104)=1.27, R>.29.

Positive M

The ANCOVA performed on the relevant attributes of

Positive M did not yield a significant order effect,

F(2,52)=2.10, E>.13; and the significant time x order

interaction, F(2,104)=2.73, R=.03, did not yield simple

order effects for the posttest, F(2,56)=2.40, p>.1, nor

follow-up test, E(2,52)=2.13, p>.13. Furthermore, the

simple effects test for the within subjects factor of the

time x order interaction did not yield significant effects

for Treatment Group 1, F(2,104)=1.03, R>.2, Treatment Group

2, F(2,104)=1.68, R>.1, nor the control group,

F(2,104)=2.75, R>.05.

The ANCOVA for the non-relevant attributes of the

Positive M did not reach significant order, F(2,52)=.90,

p>.41, nor time x order effects, F(4,104)=2.12, R>.08.








Negative M

The ANCOVA for the relevant attributes of Negative M

did not yield a significant order, F(2,52)=.95, R>.39, nor

time x order effects, F(2,104)=1.81, E>.13. Similarly, the

ANCOVA performed on the non-relevant attributes of the

Negative M did not yield significant order, F(2,52)=1.34,

R>.27, nor time x order effects, F(4,104)=2.24, E>.07.

Absolute Difference Score (AD)

Results of the ANCOVA on the relevant attributes

revealed a non-significant effect on order, F(2,51)=1.73,

E>.19, but a significant time x order interaction,

F(4,102)=2.61, R<.03. Clarifying simple effects tests

yielded a significant group effect on the follow-up test,

F(2,55)=3.56, E<.02. Subsequent pairwise comparisons

(Tukey's, R=.05) revealed that Treatment Group 1 subjects

stereotyped significantly more (M=15.5) than Treatment Group

2 subjects (M=9.04). Group differences were also

significant on the posttest, E(2,55)=6.48, p=.0008, but the

Tukey's pairwise comparison did not yield significant simple

effects.

The within subjects factor of the time x order

interaction also revealed significant changes across time.

Treatment Group 1 subjects had an overall change across

time, F(2,102)=15.26, R<.001, with a significant decrease in

stereotyping from pretest (M=20.75) to posttest (M=12.81).






55
The differences between pretest and follow-up test (M=15.5)

were also significant.

Treatment Group 2 subjects also changed significantly

across time on the Absolute Difference Scores that were

based on the relevant attributes, F(2,102)=24.99, R<.001.

Pairwise comparisons revealed that stereotyping decreased

significantly from pretest (M=20.98) to posttest (M=12.27)

and from pretest to follow-up test (M=9.04). Similar

results were obtained for control group subjects. They had

a significant overall change in stereotyping across time,

E(2,102)=4.86, R<.01, with the decrease from pretest

(M=14.07) to posttest (M=9.99) and from pretest to follow-up

test (M=10.5) reaching significance.

Results of the ANCOVA for the non-relevant attributes

of the Ad did not yield a significant order effect,

F(2,51)=2.97, R>.06, but the time x order interaction was

significant, F(4,102)=2.48, E<.04. Simple effects tests for

order revealed a significant group effect for the posttest,

F(3,55)=8.29, R<.0001, but the subsequent pairwise

comparison did not yield significant results. Simple effects

tests for order also revealed a significant group effect for

the follow-up test, F(2,51)=4.2, R<.009. Pairwise

comparisons revealed that Treatment Group 1 subjects

stereotyped significantly more (M=14.33) than control group

subjects (M=8.04). Treatment Group 2 subjects (M=10.96) did








not differ significantly from either the control group nor

Treatment Group 1 subjects.

The within subjects factor of the time x order

interaction revealed significant changes across time for

Treatment Group 1, F(2,102)=20.96, g<.001, with a

significant decrease in stereotyping from pretest (M=15.24)

to posttest (M=7.18) and a significant increase in

stereotyping (return to pretest levels) from posttest to

follow-up test (M=14.33).

Treatment Group 2 subjects also changed significantly

across time, F(2,102)=27.85, p<.001, with a significant

decrease in stereotyping from pretest (M=16.79) to posttest

(M=4.63) and a significant increase in stereotyping from

posttest to follow-up test (M=10.96). The differences in

stereotyping from pretest to follow-up test were also

significant, indicating that the increase in stereotyping

three weeks after training did not reach the same high

levels of stereotyping that existed before training.

Finally, significant changes in stereotyping were also

found for the control group, F(2,102)=15.98, p<.001. As

with Treatment Group 2, the control group significantly

decreased in stereotyping from pretest (M=12.3) to posttest

(M=4.8) and increased in stereotyping from posttest to

follow-up test (M=8.04). The differences in stereotyping

form pretest to follow-up test were also significant.











CHAPTER V
DISCUSSION

The primary focus of this study was to investigate the

effects of racial identity stage theory training on White

students' prejudice and stereotyping of Blacks. It was

hypothesized that racial identity training would decrease

prejudice and stereotyping on the posttest. Several

measures of prejudice and stereotyping were taken, yielding

findings that provided qualified support for the overall

idea that subjects who receive racial identity training

differ in their perceptions of Blacks than subjects who do

not receive such training. The findings also revealed that

different training content led to different results.

The strongest conclusion reached in the analyses of the

dependent variables for prejudice (Positive M, Negative M,

Modern Racism Scale, and Desire for Contact) is that

subjects who received training in Minority and Majority

Identity (Treatment 1) were more prejudiced toward Blacks

(immediately following training) than were subjects who

received Minority Identity training only (Treatment 2) and

no training. Support for this conclusion follows the

strength of the results yielded by the Modern Racism Scale

analysis. The time x group interaction for the MRS revealed

that, while Treatment Group 2 subjects had a slight decrease

57








in prejudice after training, Treatment Group 1 subjects

significantly increased in prejudice after training, and

that the level of prejudice reached by Treatment Group 1

subjects was significantly higher than either the prejudice

levels of Treatment Group 2 or of the control group.

The posttest results of Desire for Contact also support

the conclusion that training in Minority and Majority

Identity is not as effective in reducing prejudice as

training in Minority Identity only.

The stereotype measures requiring the use of various

attributes (Probability of Differentiation, Perceived

Variability, and Absolute Difference Score) yielded

significant results on the post hoc analyses when the

attributes relevant to the independent variable were

separated from those that were not. These measures, along

with Member-to-Group Inference, concur with the major

findings regarding prejudice, in that subjects who received

Minority and Majority Identity training (Treatment Group 1)

stereotyped more than subjects who received Minority

Identity training alone or no training. Below is a more

detailed interpretation and evaluation of the results as

they relate to the hypotheses and their corresponding

dependent measures.

Perceived Distribution (Pd, Var, M)

The three hypotheses that were based on the dependent

variables generated from the perceived distribution task








were not supported in this study. The perceived

distribution task, modeled on Linville and her colleagues'

(1986; 1989) method for assessing stereotyping behavior and

prejudice, required subjects to distribute 100 Black

students into seven categories along a given attribute

dimension. Calculated from this distribution are

Probability of Differentiation (Pd) and Perceived

Variability (Var), both measures of stereotyping, and the

Central Tendency of the Perceived Distribution (Positive M

and Negative M), measures of prejudice. It was

hypothesized, but not confirmed, that after receiving

Minority Identity training, with or without Majority

Identity training, subjects would decrease in stereotyping

and prejudice as determined by these three dependent

measures.

These dependent measures are of particular interest and

relevance to the study because the theoretical basis of the

distribution task and its corresponding measures closely

resembles the theoretical importance of Minority and

Majority Identity Development theories in the multicultural

counseling field. To illustrate, Linville et al. (1986)

make the assumption that "social categories evolve from

relatively general, undifferentiated structures to more

highly differentiated ones" (p. 166). Furthermore, they

state that greater category familiarity leads to greater

perceived differentiation and variability. Similarly, the








current emphasis in the multicultural counseling field on

racial identity development models stems in part from the

recognition that (1) to avoid stereotyping, counselors need

to be able to see differences among ethnic minority groups

and other underserved populations, and (2) Minority and

Majority Identity Development theories offer an approach for

understanding and increasing perceived group differentiation

in a manner that is directly applicable to counseling

(Helms, 1984; Parham, & Helms, 1981). Based on these points

it would be expected that subjects receiving Minority

Identity training (with or without Majority Identity

training) would acquire an increased capability for

perceiving differences among Black students. Since the Pd

and Var are direct measure of this property, the

nonsignificant results are puzzling.

Three possibilities might account for these findings.

First, the perceived distribution task may have been too

difficult to complete. In a study using Linville's

perceived distribution task, Judd and Park (1988) discovered

that subjects sometimes had difficulty completing the task.

They reasoned that subjects had difficulty concentrating on

the shape of the distribution being formed because they were

so concerned with making the numbers add up to 100. For

this reason, it is possible that training could have had an

effect on Pd, Var, and M, but the numbers from which these






61

measures were generated were not an accurate representation

of subjects' perceived distribution.

It is also possible that, one of the measures, the Pd,

may not have served as an effective measure of dispersion.

Support for this possibility comes from Park and Judd

(1990). They examined how people form and use judgements of

variability and studied the validity and reliability of

several measures used to assess variability, including

Linville and her colleagues' perceived distribution task and

the accompanying Pd measure. Even when the difficulty of

completing the task was partially improved by simplifying it

(using colored dots of varying sizes instead of numbers),

Park and Judd (1990) concluded that the Pd, compared to

other measures of dispersion, was not a very sensitive

measure of perceived differentiation and failed to find an

effect where all other measures of dispersion had.

The third explanation for the nonsignificant results of

the Pd, Var, and Positive and Negative M, is that the

training did not influence perceived distributions because

of the particular attributes used for generating these

dependent variables. Theoretically, subjects taught about

the various stages of racial identity development should

have acquired a general understanding of the idea that there

are various differences among Black people. The specific

references to attitudes and beliefs of Blacks made during

the Minority Identity training should not be as important as








the more abstract content that emphasizes that there are

several differences among Blacks. However, the necessary

leap required to apply the concept of within group

differences (based on racial identity stages) to the

attributes used in the distribution task may have been too

great. Specifically, it is possible that the sensitivity of

the Pd, Var, and M would have been greater if only those

attributes that directly related to the attributes included

in the training were used in the distribution task. To test

this possibility, post hoc analyses targeted specific

dimensions of prejudice and stereotyping in the racial

identity training that corresponded with those attributes

used in calculating the dependent variables.

Results of the post hoc analyses for the relevant and

non-relevant attributes used to calculate the Pd, Var, and M

yielded findings that suggest that attribute relevance did

indeed affect the sensitivity of the Pd and Var, but not the

M. Although the results of the Pd post hoc analyses were

not strong, they showed that overall, Treatment Group 1

subjects stereotyped more than control group subjects. This

finding was not generated when the non-relevant attributes

were included in the initial analyses.

The post hoc analyses on the Var also yielded

significant results for the relevant attributes, indicating

that on the posttest, Treatment Group 1 subjects stereotyped

more than Treatment Group 2 and the control group subjects.








No group differences were found for the non-relevant

attributes but a time effect showed that Treatment Group 1

subjects increased in stereotyping immediately after

training. Separating the relevant and non-relevant

attributes from the Positive and Negative M did not lead to

significant effects.

Overall, these findings support the significant

findings yielded by the other dependent variables: the

training received by Treatment Group 1 was not only less

effective in reducing negative stereotypes as compared to

Treatment Group 2 and the control group, but actually led to

an increase in stereotyping across time.

Absolute Difference Score (AD)

The Absolute Difference Score was the only measure used

in this study that examined differentiation between groups.

Subjects' views toward Black students were examined in

relation to their views toward themselves (White students).

For each attribute, subjects were asked to determine how

many Black students and how many White students out of 100

would be characterized by the attributes. The difference in

predictions between Black and White students constituted a

measure of the extent of stereotyping. It was hypothesized

that subjects who received Minority Identity training, with

or without Majority Identity training, would stereotype less

than the control group in that they would see themselves

(i.e. White students) as less distinct from Black students.








Results did not support this hypothesis and no significant

differences were found until the attributes relevant to

training were separated from the non-relevant attributes in

the post hoc analyses.

As with the Pd, Var, and M, it was reasoned that a

treatment effect was being masked by the attributes used to

generate the AD score. When separated, results yielded

significant decreases in stereotyping across time for all

groups. However, not just a practice effect was found. The

analysis also yielded significantly higher stereotyping for

Treatment Group 1 subjects three weeks after training as

compared to Treatment Group 2 subjects (relevant attributes)

and control group subjects (non-relevant attributes). The

hypothesized decrease in AD scores for the groups receiving

training was not supported.

Perhaps the hypothesis for the AD was not supported

because of the effects of three confounding forces: liking,

contact, and the interaction of the two. Biernat (1991)

found that as subjects' liking increased, stereotyping, as

measured by the Absolute Difference Score, decreased. That

is, as subjects' interest in a group increased, they tended

to see fewer differences between the target group (e.g.

"greeks") and the comparison group (university students in

general). Although training did not significantly decrease

stereotyping as compared to the control group, there was a

decrease in AD scores for both groups receiving training.








It is possible that "liking" increased because of training,

a possibility that is supported by subjects' statements

during the training that they liked some of the Black

students introduced in the video tape.

Confounding effects may also follow from a second

finding of Biernat (1991). An increase in contact (number

of acquaintances) led to an increase in Absolute Difference

Scores. The explanation is that contact leads to a more

complex way of viewing a group and the potential for seeing

how the target group (Black students) differs for the

comparison group (White students).

The third point gleaned from Biernat's (1991) findings

adds to the already complex evaluation of the results

yielded by the AD. To summarize, an increase in contact

leads to an increase in liking, which in turn leads to a

decrease in seeing differences between groups. However, if

contact is a negative experience it leads to a decrease in

liking, which in turn should lead to an increase in

stereotyping. Some of the "contact" subjects received

during training was negative, as indicated by the negative

feelings expressed regarding one of the Black students in

the video. Contact with Black students, as had occurred

with the two treatment groups, may not only have affected

subjects' cognitively by increasing their perception of

distinctions between Blacks and Whites, but it may also have

affected them emotionally, and led to an overall decrease








(albeit tempered) in seeing differences between Blacks and

Whites.

Before exploring the other findings in the present

study, it might be useful to consider other reasons why the

primary dependent variables for stereotyping (Pd, Var, AD)

did not support the hypotheses.

The fundamental cognitive task required when one

attempts to avoid making stereotypical judgements is to pay

close attention to nonstereotypical characteristics, or

individuating information. Several researchers have found

that informational factors can lead subjects to pay closer

attention to individuating information about a target.

According to Fiske, Neuberg, Beattie, and Milberg (1987) and

Neuberg and Fisk (1987), automatic reliance on stereotypical

responses can be thwarted if individuating information is

inconsistent with categorical information. Subjects also

paid attention to individuating information when

motivational factors were implemented. For example, when

the performance of the target person led to a reward for the

subject, and when subjects where implored to strive for

accuracy, stereotyping decreased (Fiske et al., 1987;

Neuberg & Fiske, 1987).

In the present study, subjects were presented with

individuating information about Black students. However,

they were also presented with information that confirmed

some of the negative stereotypes of Blacks. Additionally,






67

no rewards and no efforts toward imploring sensitivity were

made in the current study.

Unfortunately, the primary findings in the present

study regarding the resiliency of stereotypes are consistent

with other literature. For example, Rothbart and John

(1985) created a model that examined the process of

stereotype change resulting from contact with individual

group members. They addressed the question of whether

information disconfirming a group stereotype obtained

through contact with a sample of group members generalizes

to the group as a whole. They concluded that it does not.

Their cognitive-processing model is based on the findings of

cognitive and intergroup perception research. They

postulated that when one is presented with individuating

information that is inconsistent with category information

(e.g. an engineer who is Black) the new information is not

encoded in memory by the superordinate category of "Blacks."

Instead, the perceiver more strongly associates the person

with the subordinate category "Engineer." Subsequent

requests to retrieve information about Blacks will therefore

not tap into the counterstereotypical example. This model

is confirmed by experiences of Black professionals who have

often heard their White colleagues say "When I look at you,

I don't see you as a Black person." The consciousness

raising retort of "Look real hard and maybe you'll notice"

illustrates the mechanism which Rothbart and John (1985)








claimed will lead to stereotype change. "Black

professional" must be associated with the superordinate

category "Black" and not with the subordinate category

"professional" for stereotypes to change. A direct test of

their model yielded significant results, supporting the

notion that atypical exemplars of categories were not

adequately represented in memory as legitimate category

members. Furthermore, when judgements were made about the

category, the atypical members were not taken into account.

The durability of stereotypes was also demonstrated by

Devine (1989). Her research was based on the premise that

learning about cultural stereotypes is inevitable for

everyone interacting with his or her culture. Belief in a

stereotype is a separate issue. Devine found that subjects

who were low in prejudice were able to inhibit the

expression of stereotypical thoughts. However, when low

prejudiced subjects were forced to make automatic responses

without conscious monitoring, stereotypical or prejudice-

like responses were generated. Good intentions could not

overcome the strength of involuntary stereotype effects.

Nelson, Biernat, and Manis (1990) also found that

despite a concerted effort to induce subjects to decrease or

eliminate their use of stereotypes when evaluating

individual group members, subjects were unable to do so.

Subjects were asked to judge the height of men and women,

and were explicitly warned not to let the stereotype about








height interfere with their judgement, because for every

male example of a given height there was a corresponding

female. Additionally, for the subject who demonstrated the

best judgement and sensitivity, a $50 reward was promised.

The two inducements did not alter subject reliance on the

stereotype about height. Men were still judged to be taller

than women. The only factor that had led to a decrease in

stereotyping was the matching of stereotypic with counter-

stereotypic characteristics. For example, a woman with a

typically male career goal was judged taller that a woman

with a typically female career goal. These results support

the cognitive-processing model developed by Rothbart and

John (9185) in that stereotypes are very difficult to

change, but when the superordinate (e.g. female) category is

emphasized along with the subordinate (e.g. male career),

some change is possible.

A final question which has yet to be addressed is why

the findings of the analyses on the AD, along with the

findings of the Pd and Var, indicate that Treatment Group 1

subjects stereotyped more after training than Treatment

Group 2 and control group subjects. These results were also

similar to the prejudice findings derived from the Modern

Racism Scale, and will be addressed below.

Modern Racism Scale (MRS)

The results of the Modern Racism Scale confirm that

racial identity training had an effect on subjects attitudes








toward Blacks and that different training led to different

results. Before exploring the specific findings it is

important to examine the theoretical basis of the MRS.

McConahay, Hardee, & Batts (1981) developed the theory of

modern racism that addresses the debate regarding the extent

to which racism had declined. They made a distinction

between "old fashioned racial beliefs," which everyone

recognizes as racism, and a new set of "symbolic beliefs"

that arose after the civil rights movement. Furthermore,

they proposed that, while belief in old fashioned racism has

declined, anti-Black affect has remained and has been

displaced onto new beliefs. They found that many Whites do

not recognize the racial basis of the new beliefs and

instead argue that they are based upon non-racial,

political, and social values. For example, by asking

questions regarding a subjects' belief in the continued

existence of discrimination, the right of Blacks to push

themselves into situations where they are not wanted, or the

extent to which Blacks are getting more money or attention

than they deserve, anti-Black attitudes are obtained.

The findings in this study demonstrated that subjects

who received Minority and Majority Identity Development

training increased in their anti-Black attitudes immediately

after training. In contrast, subjects who received Minority

Identity Development training alone slightly decreased in

anti-Black attitudes immediately after training. Two






71

questions arise: (1) Does Majority Identity training cause

Whites to increase their negative attitudes toward Blacks?

(2) Does the interaction of Minority and Majority Identity

training cause Whites to increase their negative attitudes

toward Blacks? Although only future research can provide

empirical answers to these questions, some speculative

inferences can be made. First, receiving Majority Identity

training may be threatening to most White people. Support

for this possibility can be found from the Majority Identity

Development model itself. For example, Ponterotto (1988)

stated that most students entering a university counseling

program are in the first stage of Majority Identity

development. He described the White person in this stage as

one who "has given little thought to multicultural issues or

to his or her role as a White person in a racist and

oppressive society" (p.151). Similarly, the Majority

Identity model presented in the present study, and presented

by official AACD trainers to university administrators and

counselors nationwide, states that in the first stage, White

people "do not perceive themselves as 'racial beings' and

tend to assume that racial and cultural differences are

individual matters, not social or political concerns" (see

Appendix B). It is very likely that if most White

counseling graduate students are in the first stage of

Majority Identity Development, White undergraduate students

are as well. Training in Majority Identity Development






72

requires trainees to face themselves as "racial beings" and

to explore the meaning of being White as it relates to Black

people. Having to face that one is White may lead to guilt

or depression (Katz & Ivey, 1977), and perhaps to an

increase in anti-Black attitudes (Helms, 1984). It is also

possible that an increase in anti-Black attitudes led to an

increase in negative stereotyping, thus yielding the results

of the Pd, Var, and AD.

By way of contrast, why would Minority Identity

Development training lead to a decrease in prejudice toward

Blacks? Three factors may be involved: (1) empathy toward

Black people may have developed once taught about the

struggle of becoming proud of who you are while living in a

society that does not value you; (2) focusing on Blacks and

their "problems," and not on White peoples' contribution to

their problems, is not very threatening and may even

encourage a rescuer-like response; and (3) the Minority

Identity Development theory provides a sympathetic

explanation for several negative stereotypes of Black

people.

Posttest Only Measures

The results of the MRS, Pd, Var, and AD are supported

by the results of the Member-to Group Inference scores and

the Desire for Contact scores. These dependent variables

demonstrated that subjects who received Minority and

Majority Identity training (Treatment Group 1) were more








prejudiced and negatively stereotypical toward Blacks than

subjects who received Minority Identity Development training

alone (Treatment Group 2) and those who received no training

at all.

Member-to-Group Inference (MTGI)

The significant results of the Member-to-Group

Inference (MTGI) scores demonstrate, once again, just how

durable stereotypes are. Additionally, the Member-to-Group

Inference results demonstrate that training in Majority

Identity training along with Minority Identity served to

stimulate anti-Black sentiments. The vignette upon which

this significant finding was based depicted a Black student

portraying a negative stereotypical behavior, i.e., picking

a fight with a White student. If either of the two

treatment groups had had a significant decrease in the MTGI

compared to the control group, then the conclusion that

stereotypes were reduced could be made. Instead, results

indicated that the Minority Identity trained group did not

differ from the control group in stereotype behavior, while

the Majority identity training group stereotyped more than

the control group. More specifically, the negative behavior

of one Black student was generalized to the population of

all Black students to a significantly higher level by

subjects who received Minority and Majority Identity

Development training, than by those who received only

Minority Training, or no training whatsoever. As stated






74

above, the negative impact of training on Treatment Group 1

subjects is probably due primarily to the negative feelings

evoked by the Majority Identity Development training.

Finally, it is important to note that no significant

between group differences existed on the MTGI response that

was based on the vignette that depicted a Black student in a

positive, counterstereotypical manner (asking the Vice-

President for an interracial meeting with White and Black

students). Since the more one holds counterstereotypical

beliefs and perceives variability within a group, the more

one would have endorsed this particular episode as typical

of Black students (Quattrone & Jones, 1980; Nisbett et.al.,

1983), the nonsignificant result of this MTGI score further

supports the idea that (negative) stereotyping did not

decrease as result of racial identity training.

Desire for Contact (DFC)

The DFC results concur with the results of the Pd, Var,

AD, MRS and MTGI, indicating that Minority Identity training

in conjunction with Majority Identity training led to

significantly less positive attitudes toward Black students

than Minority Identity training alone. Unlike the other

dependent measures, DFC provides a concrete measure of

attitude that is directly translated into behavioral

intention, in that it is operationally defined as the degree

to which subjects want to be involved in an interracial

dialogue and to have Black friends. Research on the








relationship between attitudes and behavior suggests that

the former does not necessarily predict the latter (e.g.

Wicker, 1969). Therefore, the results of the DFC provide

additional support that strengthens the results along the

other dependent measures.

Training Receptivity and the Dependent Variables

The results of the correlation between the ability to

learn the racial identity stage theories and the dependent

variables indicated that negative attitudes toward Blacks

may have been related to less receptivity to training while

favorable attitudes were related to greater receptivity to

training. This finding is important in that it appears as

though prejudice may interfere with training and that those

who may be in greatest need of training are least receptive

to it. Future research should address this possibility.

Conclusion

This was the first empirically based research study

that examined the effects of Minority and Majority Identity

Development training on attitudes and perceptions of

trainees. While Minority Identity training led to a

positive effect on attitudes, the combination of Majority

and Minority Identity training led to an increase in

negative attitudes toward Blacks. The implications of these

results for multicultural counselor training are serious and

warrant further study.






76

Minority and Majority Identity Development theories are

invaluable tools for helping counselors increase their

sensitivity and understanding of their clients. For the

White mainstream American counselor, it is crucial that he

or she not only understand what it means for a Black

American to be Black, but also what it means for a White

American to be White. Furthermore, a White counselor's

racial self-awareness and understanding of racism is

directly related to his or her potential effectiveness in

working with Black clients (see Helms, 1985; Parham & Helms,

1981).

Efforts are being made nationwide to incorporate

multicultural counselor training into graduate curriculum

and professional training seminars. Minority and Majority

Identity models are a significant component of such

programs. It is possible that multicultural counselor

training begins a process that involves an increase in

negative attitudes toward Blacks. The increase in anti-

Black attitudes may be a necessary stage that White people

go through when put into situations that involve the

exploration of racial identity and racism. The results

found in the studies of counseling graduate students

enrolled in multicultural counseling courses support the

idea that multicultural training is a process that may lead

to negative attitudes toward outgroups (Fukuyama, et al.,

1987; Ponterotto, 1988).








It seems important that multicultural trainers become

aware of such a process and construct training programs that

address the process with trainees. Furthermore, given the

results of the correlations between the ability to learn

racial identity stages and the dependent variables, it

becomes apparent that multicultural training must not only

address the process that is set into motion as a result of

training but also assess the differences between trainees

prior to training onset. While some trainees may enter

seminars and courses naively, yet willing to learn, others

may enter with negative attitudes that inhibit their own

learning process. How should training be conducted given

the variety of potential needs of training groups? Research

and experiences with multicultural training that are

successful in facilitating more positive attitudes and

behaviors must be conducted and shared.

Future Research and Limitations of the Study

This study raises important questions regarding the

effects of racial identity training. Future research must

explore more thoroughly the effects of Minority and Majority

Identity Development training on the attitudes and

perceptions of Whites, perhaps including more reliable

measures of stereotyping.

Given the negative impact of combining Majority with

Minority Identity training on attitudes, a closer, more

controlled examination should be made of the effects of






78

Majority Identity training alone. Several questions need to

be addressed. For example, should the Majority Identity

model be presented at the end of a semester-long

multicultural counseling course, after students have

explored the history and current manifestations of racism?

How should trainers handle the inevitable guilt and anger

experienced by White trainees when issues of oppression,

racism, and White racial identity arise? Ponterotto (1988)

gave an example of what happens: "students often feel angry

at me, the instructor, because I in essence have allowed

them to feel guilty and angry in class" (p. 152). Should

the expression of negative feelings and attitudes be

encouraged? What are the effects if they are or are not?

Future research also should address the effect of

teaching Majority and Minority Development to a group of

White and Black students. A racially and culturally mixed

classroom is the norm in many graduate counseling programs.

What impact does racial identity training have on Whites'

attitudes toward Blacks and toward other Whites, and on

Blacks' attitudes toward Whites and other Blacks, if they

are trained together versus separately? Some trainers who

have had successful results fighting campus racism have

found that the best way to change White students' anti-Black

attitudes is to have training programs aimed directly and

exclusively for White students.







Finally, future research should include empirically

controlled studies that examine the effects of Minority and

Majority Identity training on a population directly

generalizable to multicultural counseling trainees. This is

especially important given research that suggests that

motivational factors influence attitude change. It is

likely, for example, that more highly motivated graduate

students and professionals would be more responsive to

racial identity training than would the undergraduate

students used in the present study. Research that examines

the effectiveness of multicultural training will help

facilitate the creation of programs that lead to the desired

goal of such programs -- producing confident and effective

multicultural counselors.












APPENDIX A
DEBRIEFING

In the field of counseling psychology, training White
counselors to become more effective in counseling ethnic
minorities has been of great concern in recent years.
Multicultural counselor training often involves teaching
trainees about racial identity of Whites and Blacks. Racial
identity theory states that people of all races go through a
series of stages that have implications for one's behavior
and attitudes about oneself and other races. Very little
empirical research has been done to test the effects of
multicultural counselor training in general; and no research
has been done to test the impact of training in racial
identity theory specifically. The purpose of this project
is to study the effects of racial identity training on
attitudes and perceptions of Whites.

One of the theoretical underpinnings of this study is
the out-group homogeneity hypothesis, which states that
outgroups are viewed as being more homogeneous than
ingroups. Research has shown, for example, that White
people tend to view Black people as more homogeneous in
traits and attributes than White people. This is simply
another way on saying that people have a tendency to
stereotype outgroup members.

In studies measuring the effects of perceived
homogeneity, findings show that perceived homogeneity leads
to prejudice, discrimination, and the tendency to generalize
from the behavior of one outgroup member to the behavior of
all outgroup members. We are not sure what the effects of
racial identity training will be, but it is possible that
White people who have had training in racial identity theory
will be less likely to stereotype and more likely to be
willing to interact with Blacks than White people who have
not had such training. Perhaps teaching people about stages
of racial identity will also decrease one's tendency to be
prejudiced.

Thank you for participating in this study. If you
would like to have a copy of the results, please leave your
name and phone number with me.











APPENDIX B
STAGES OF MAJORITY IDENTITY DEVELOPMENT

The Stages of Majority Identity Development represent
the process by which majority members develop a healthy
sense of identity and sensitivity to minority concerns.
"Majority" groups are those groups in a society who, by
virtue of their group's control of economic, cultural and
other rewards, have had a negative effect on groups that
differ from the majority. Majority groups include such
groups as White Americans, Men, and non-handicapped people.
For the purposes of this study, we will be focussing on
White Americans as the majority group.

One does not generally "arrive" at the final stage and
remain at the level for all future times. For most people,
the experience of identity awareness is more like a loop
than a straight line. Levels of awareness may be
experienced both as linear progression from level to level
and as a recycling of a prior level based on new
experiences.

For example, one may develop a sense of "autonomy" at
the age of 20 but when new life experiences present
themselves one may experience a recycling of growth as one
goes through the process of synthesizing the changes with
one's former sense of identity.

Each stage is briefly described below. Behaviors and
attitudes associated with the stages are identified, and
some examples of statements associated with the stages are
listed.


I. CONTACT STAGE

In this stage majority members become aware of the
existence of minority members. White Americans do not
perceive themselves as "racial beings" and tend to
assume that racial and cultural differences are
individual matters, not social or political concerns.

Behaviors and Attitudes:
believes that everyone is the same, or some may believe
that minorities are inferior in some ways to majority
group members.








-has a naive curiosity about culturally different
people.
-an encounter with a minority group member is a minor
crisis.
believes in the "melting pot" theory of assimilation.

Common Statements:
"When I talk to you I don't think of you as Black."
"You can do whatever you want to do as long as you
don't do it around me."
-"Why are all the minority students sitting together?"


II. DISINTEGRATION STAGE

In this stage, one acknowledges that prejudice and
discrimination exist and are forced to view oneself as
a majority group member. Guilt may emerge as racial
differences become more apparent.

Behaviors and Attitudes:
may become immobilized by the enormity of the problem
suggested by minority-majority differences.
sees self as less prejudiced than other members of the
majority group.
-wants to be seen as an individual and not a member of
any group.
may attempt to protect minority members from negative
interactions with majority group members; will
challenge majority members about their behaviors and
attitudes.
may over identify with culture of the majority group.

Common Statements:
"My parents are very prejudice but I am not."
-"Although I am not responsible for the negative actions
of other majority group members, I will confront them
about their behaviors."
-"Most Whites are prejudiced towards minorities."


III. REINTEGRATION STAGE

In this stage, the majority member tends to focus on
how s/he has been socialized in a majority world. The
focus is less on oneself in comparison to minority
groups and more on oneself as a member of a majority
group.








Behaviors and Attitudes:
wants to focus on problems associated with own group.
wants to focus on belief that people are all the same
in important ways.
-thinks that too much attention is being placed on
cultural differences, and stresses instead the need to
foster climate that focuses on the strengths and
limitations of all cultures.

Common Statements:
-"Racism isn't the only problem; what about world
hunger?"
"I believe that quotas of any kind are wrong."
"We're not going to get very far unless other majority
group members join us in combating racism."


IV. PSEUDO-INDEPENDENCE STAGE

In this stage the person accepts minority group members
as a conceptual level and becomes interested in
understanding racial
differences. Their involvement with minority group
members tends to be limited to those who they perceive
to be similar to themselves.

Behaviors and Attitudes:
can articulate reasons for accepting minority group
members.
has friends who are members of minority groups.
tends not to be involved in any activity that supports
minority group concerns.
believes that discrimination is a problem of the
uneducated.

Common Statements:
"I accept all minority group members and believe that
we all should."
"Blacks have the same abilities as Whites."
"Racism is illogical."


V. AUTONOMY STAGE

This final stage is characterized by an individual
becoming knowledgeable about racial and cultural
similarities as well as differences. This person
accepts, respects, and appreciates both minority and
majority individuals.






84

Behaviors and Attitudes:
- seeks opportunities to involve themselves in cross-
cultural interactions.
-values diversity.
- respects and appreciates cross-cultural and inter-
racial interactions.

Common Statements:
- "I am actively involved in fighting racism."
- "I am a recovering racist."
- "We are all members of the same global community."
- "Discrimination and prejudice against any group has a
negative effect on us all."












APPENDIX C
STAGES OF MINORITY IDENTITY DEVELOPMENT

The following stages are typically experienced by
minority group members (Blacks, Hispanics, Asians, Native
American Indians, Women, etc.). Each stage is briefly
described below. Behaviors and attitudes associated with
the stages are identified, and some examples of statements
associated with each stage are listed.

When minority group members experience levels of
awareness, they may experience both a linear progression
from level to level and a recycling of a prior level based
on new experiences. One does not generally "arrive" at the
final stage and remain at the level for all future times.
For most people, the experience of identity awareness is
more like a loop than a straight line. For example, one may
develop a sense of "internalization" at the age of 20 but
when new life experiences present themselves one may
experience a recycling of growth as one goes through the
process of synthesizing the changes with one's former sense
of identity.


I. ACCEPTANCE STAGE

This stage is characterized by limited self-awareness
about differences, and dependence upon the majority
group for sense of worth.

Behaviors and Attitudes:
attitudes toward the world and self are determined by
the majority group's logic.
dislikes one's own group, emulates majority group.
accepts stereotypes of own group.
believes that assimilation is the most effective method
for problem solving.

Common Statements:
"We're all just people."
"Blacks are basically lazy." (said by a Black person)









II. RESISTANCE STAGE

In this stage a significant event creates receptivity
to a new identity.

Behaviors and Attitudes:

-openly challenges acts of discrimination against own
group.
intense search for own group history, identity begins.
reinterprets all events from one's own group
perspective.
experiences deepen the trauma of discrimination.
-separates self from other minority members who seem to
still be in the acceptance stage.

Common Statements:
-"I've discovered that my being Black makes a difference
to Whites."
"I was rejected because my skin was too dark."
"I met a Black man who was proud of being a Black man."


III. REDEFINITION STAGE

In this stage there is a transition from the old
identity to a new identity, and an emphasis on the
destruction of the old identity and a glorification of
the new identity.

Behaviors and Attitudes:
participates in political action groups, seminars,
awareness groups, etc.
undergoes a liberation from the majority group's values
and stereotypes.
-behaves as though majority group members are not human.
confronts the system.
person feels an overwhelming attachment to her/his
group.
gradually both the strengths and weaknesses of the
majority group and one's own group become visible.

Common Statements:
"Black is beautiful."
-"All White people are racist."
-"Black people must have their own organizations that
can protect Black people and ensure that they will be
dealt with equitably on the job."
"Only Blacks can really understand what it means to be
Black."





87

IV. INTERNALIZATION STAGE

The new identity is incorporated and the individual can
renegotiate with the majority.

Behaviors and Attitudes:
one behaves with a sense of inner security
one has compassion for all minority people and can
apply one's values to all "isms."
one demonstrates commitment and active participation in
making social change.

Common Statements:
"I can learn from both Whites and Blacks."
"I'll never change his mind but I can handle his
attitude."
"To be liberated as a Black man I must also confront my
own sexism."












APPENDIX D
RESPONSE SHEETS

(Treatment Group 1: Majority and Minority Stage Training)


As you view the video, determine into which stage each
student falls and put a check mark next to the stage name.


1. Avonne Abnathya

I Acceptance

II Resistance

III Redefinition

IV Internalization


2. Tim

____ I

___ II

III

IV

V


Farrell

Contact

Disintegration

Reintegration

Pseudo Independence

Autonomy


3. Donnese Cheatham

I Acceptance

II Resistance

III Redefinition

IV Internalization








4. Ben Brennan

I Contact

II Disitegration

III Reintegration

IV Pseudo Independence

V Autonomy


How useful do you think these models are for understanding
racial identity (circle the level that fits for you)?


5 4 3 2 1
very useful somewhat slightly neither not at all
useful useful useful useful
nor useless



(Treatment Group 2: Minority Stage Training)


As you view the video, determine into which stage each
student falls and put a check mark next to the stage name.


1. Avonne Abnathya

I Acceptance

II Resistance

III Redefinition

IV Internalization


2. Donnese Cheatham

I Acceptance

II Resistance

III Redefinition

IV Internalization








3. Sherie Pietranco

I Acceptance

II Resistance

III Redefinition

IV Internalization


How useful do you think this model is for understanding
Blacks (circle the level that fits for you)?


5 4 3 2 1
very useful somewhat slightly neither not at all
useful useful useful useful
nor useless













APPENDIX E
DEPENDENT MEASURES


Bar Code #

Name Age Sex

Phone Race/Ethnicity

Political Affiliation


(Modern Racism Scale)

Thank you for participating in this study. Below are a
number of opinion statements about public issues, politics,
and your beliefs about the world in general. You will agree
with some, disagree with some and have no opinion about
others. You are under no obligation to give and niger on
any item. However, we would like for you to indicate when
you do not have an opinion or when you do not wish to
answer, so please do not leave any questions blank.

Your replies will be completely confidential and there
are no right or wrong answers. We are interested only in
group averages and percentages.

Please use the following scale to indicate your degree
of agreement with each item:

+2 agree strongly
+1 agree somewhat
0 neither agree nor disagree, or no opinion
-1 disagree somewhat
-2 disagree strongly
X I do not wish to answer

1. Blacks shouldn't push themselves where they're not
wanted.

2. I would oppose a constitutional amendment aimed at
ridding the country of pornography and sexual
immorality.







3. Our society would have fewer problems if people had
less leisure time.

4. In a democratic society, the opinion of the majority
should always prevail.

5. Sex education should be taught in the public school
system of the United States.

6. It is easy to understand the anger of black people in
America.

7. Over the past few years, blacks have gotten more
economically than they deserve.

8. I am opposed to the United States establishing formal
diplomatic relations with the People's Republic of
China.

9. Over the past few years, the government and news media
have shown more respect to blacks than they deserve.

10.A distaste for work usually reflects a weakness of
character.

11.I would favor a constitutional amendment to permit
non-sectarian prayers and religious services in the
public schools.

12.Blacks are getting too demanding in their push for
equal rights.

13.The United States Senate should not ratify the SALT II
agreements with Russia.

14.I favor ratification of the ERA (Equal Rights
Amendment to the United States Constitution).

15.Blacks have more influence upon school desegregation
plans than they ought to have.

16.It is okay for a woman to ask a man out on a date.

17.Deep in my heart, I know that I am a white racist.


18.The United States Senate did the right thing when it
ratified the Panama Canal Treaty.

19.Discrimination against blacks is no longer a problem
in the United States.








20.It is easy to understand the anger of women in
America.

21.In a divorce, the woman should always receive custody
of the children.

22.I am usually shy whenever I attend a mixer or other
large party.




(First Half of Questions to Measure
Absolute Difference Score)

In this section the task is different. Note that the same
question is asked for each trait.

Pretend you have met 100 Black students attending UF


Of those 100
AGGRESSIVE:

Of those 100
ATHLETIC:

Of those 100
CLICKISH:

Of those 100
FRIENDLY:

Of those 100
INTELLIGENT:

Of those 100
LAZY:

Of those 100
PREJUDICED:

Of those 100
TRUSTWORTHY:


Black students, how many would you say are
out of 100

Black students, how many would you say are
out of 100

Black students, how many would you say are
out of 100

Black students, how many would you say are
out of 100

Black students, how many would you say are
out of 100

Black students, how many would you say are
out of 100

Black students, how many would you say are
out of 100

Black students, how many would you say are
out of 100









(Items Used to Measure Perceived Distribution)

This next task is a little more challenging. The same
traits you used on the previous page are also in this
section. This time, for each of the characteristics listed
below, estimate the percentage of UF Black students who fall
into each of the seven levels of each characteristic.

In other words, pretend you have just met 100 UF Black
students. How many fall into each level.

Assume that any given Black student falls into one and only
one level of the characteristic.

Write your estimate above each level on the scale below and
be sure that the percentages you assign to the different
levels of a characteristic add up to 100%.


very aggressive somewhat neutral somewhat unaggressive not at all
aggressive aggressive unaggressive aggressive

/ / / / / / / /
very athletic somewhat neutral somewhat unathletic not at all
athletic athletic unathletic athletic


very clickish somewhat neutral somewhat nonclickish not at all
clickish clickish nonclickish clickish


/ / / / / / / /
very friendly somewhat neutral somewhat unfriendly not at all
friendly friendly unfriendly friendly

/I / / I ///I I/
very intelligent somewhat neutral somewhat unintelligent not at
intelligent intelligent unintelligent intelligent

/ / / / / / / /
very lazy somewhat neutral somewhat nonlazy not at all
lazy lazy nonlazy lazy

/ / / / / / / /
very prejudiced somewhat neutral somewhat nonprejudiced not at all
prejudiced prejudiced nonprejudiced prejudiced

/ / / / / / / /
very trustworthy somewhat neutral somewhat untrustworthy not at all
trustworthy trustworthy untrustworthy trustworthy








(Second Half of Questions to Measure
Absolute Difference Score)


This final section is similar to one you have already done.
Note that the same question is asked for each trait. Please
answer as honestly as possible.

Pretend you have met 100 White students attending UF

Of those 100 White students, how many would you say are
AGGRESSIVE: out of 100

Of those 100 White students, how many would you say are
ATHLETIC: out of 100

Of those 100 White students, how many would you say are
CLIQUISH: out of 100

Of those 100 White students, how many would you say are
FRIENDLY: out of 100

Of those 100 White students, how many would you say are
INTELLIGENT: out of 100

Of those 100 White students, how many would you say are
LAZY: out of 100

Of those 100 White students, how many would you say are
PREJUDICED: out of 100

Of those 100 White students, how many would you say are
TRUSTWORTHY: out of 100



(Member-to-Group Inference and
Desire For Contact)

Below are four episodes occurring at UF, followed by
questionss. Read the episodes carefully and answer the
questions (s) that follow each episode as best as you can.
There are no right or wrong answers.

I. A Black student asked to meet with Vice President Art
Sandeen to express his desire to get Black and White
students together to discuss ways to improve relations
between Blacks and Whites on campus.

a. What percentage of Black students at UF do you believe
would want to be involved in such discussions: %