RACE BELIEF AND INTERPERSONAL
ATTRACTION

Thesis for the Degree of M. A.
MICHIGAN STATE UNIVERSITY
IAVDN JACKSON
1971

(I

     
      

LIB R A f: Y
MiChigan State
niversity

  

      
   

 
 

\
h

E: BIN‘DING By “5"
IIIIAB & WW
:1 RIIIIK BRIBERY INC.

 

i

1:
SINGERS t

. -.,.! meme“. I

 

LIL

 

ABSTRACT

RACE, BELIEF AND INTERPERSONAL ATTRACTION

BY

JAVON JACKSON

In an experiment purporting to study student opinion
on social issues, a Black or White accomplice presented
either a prepared "Radical" or AConservative" viewpoint
to the causes and solutions of big city racial problems
in the United States to White male gs who were all from
introductory psychology clases. The 80 gs were all
randomly assigned to one of four experiment conditions:
Black Radical, Black Conservative, White Radical, and
White Conservative. The scripts of the accomplices
sounded spontaneous and were rehearsed by all the
accomplices so that their vocal inflections and body
movements were similar. After the discussion, the gs
in each of the four conditions rated themselves and the
actors on an Interpersonal Check List. In order to find

out how favorably or unfavorably the gs rated each actor,

Javon Jackson

comparison ratings of the actors to other variables

were examined.

Seven comparative ratings were rated for mean

discrepancy scores. They were:

1.

2.

'The (S) self' compared to 'The ideal-self'
'The (§)' compared to 'How the g viewed the
actor'

'The ideal-self' compared to 'How the_§
viewed the actor'

'How the g viewed the actor' compared to 'How
the §_viewed the average college student'

'How the § rated how the actor viewed himself'
compared to 'How the S rated how the actor
viewed the s'

'The S‘s perception of himself' compared to
'How the S thinks that the actor viewed the g'
The pooled results of the summation discrepancy

SCOIGS

If the gs were, indeed, prejudiced, they should have

preferred the Conservative Speech to the Radical Speech,

and the White Race to the Black Race, and there should be

a Speech X Race interaction in which the gs most preferred

Javon Jackson

the White Conservative, followed by the White Radical
followed by the Black Conservative and, lastly, the
Black Radical as being most different.

Analyses of variance for the comparative ratings
were computed. Only on one comparative rating-—when the
'Ss compared how they felt the actors viewed themselves
compared to the gs rating how they felt the actors viewed
the Ss--were the results due to the Race of the accomplice
statistically significant. Contrary to expectations, no
results of the Speech variable reached statistical
significance. Thus, Radical or Conservative Speech had
no effect on the gs View of the actors. Only on one
comparative rating——when the SS compared the actors to
their ideal—self--were the results due to the interaction
of Race X Speech of the accomplice statistically significant.
The gs identified most with the Black Conservatives
followed closely by White Radicals and least with White
Conservatives and Black Radicals as ideal people.

The hypothesis that the Black Radical would be seen
as most significantly different from the Ss' perception
of themselves followed by the Black Conservative, White

Radical, and White Conservative was not supported.

Javon Jackson

The pooled results of the §s showed that the White
Radical not the White Conservative was most preferred
followed by the White Conservative, Black Conservative,
and, lastly, the Black Radical. H0wever, these differences
between rankings did not reach statistical significance.
One possible interpretation is that the gs rated the White

Radicals in terms of the way the gs would like to become.

DEDICATTON

This thesis is dedicated to my
parents, Equal Opportunity
Programs, and to better racial
understanding.

ii

ACKNOWLEDGEMENTS

First, I would like to thank Dr. Stollak, my committee
chairman, for his guidance and great perseverance in helping
me with my thesis.

I would, also, like to thank Dr. MeSse who aided me.
greatly in the analyses of the data.

And, I would like to thank Dr. Thorton for his scrutiny
and perusal of this thesis which were of invaluable service
to me.

Lastly, I would like to thank my advisor, Dr. Phillips,
and Dr. Allen for their kindness and understanding which

made it possible for me to accomplish this goal.

in

TABLE or CONTENTS Page
ztkaoWLEDerENTS . . . . . .‘ . . . . . . . . . :AAA
TEST or TABLES . . . I. . . . . . . . . g . . . v
IJST OF APPENDICES . . . . . . . . . . . . . . vi
Chapter

1. INTRODUCTION . . . . . . . -. . . . . 1

History . . . . . . . . . . . . 1
Statement of problem . . . . . . . . 58

2 ° IqETI‘IOD o o o o o o o o o o o o o o 62
Subjects . . . . . . . . . . . . 62
Accomplices . . . . . . . . . . . ' 63
Setting . . . . . . . . - . . . . ' 63
Des jugll 0 O O O O O O O O O O O O 63

Procedure . . . . . . . . . . '. . 65

I

4. ADISCUSSION . . . .‘ . - . - - -. L . - 77

5 o S Ub'lbm lax, o o o o o o o o o '0 o o o 84
BIBLIOGRAPY . . . . . . .A . ‘. . . . . . . . . 37

APPENDICES . . . . . . . . . . . . ., . . . . 97

iv

LIST OI.“ 'I‘x'aiiLllS

Table - Page

1. E:-:pcri.r:=.-;m‘al Design . . . . . . . . . 64

2. List of Comparative Ratings.- . , , ,

Appendix
A.
B0

C.

Scripts

Instructions

Table

10

LIST OF APPENDICB

Analyses of Variance

vi

Page
97

110

112

RACE, BELIEF AND INTERPERSONAL ATTRACTION

BY

I:
r. ‘.,3
-, VI

‘ \

JAVON‘JACKSON

A THESIS

Submitted to
Michigan State University
in partial fulfilment of the requirements
for the degree of

MASTER OF ARTS
Department of Psychology

1971

CHAPTER I

INTRODUCTION

This study was conducted in order to evaluate the
effects of Race and Belief on interpersonal attraction.
In the Race Congruency Theory, gs are attracted to people
of the same Race. In the Belief Congruency Theory, SS
are attracted to people with similar Beliefs. In this
study, a Black and a White actor gave both a Conservative
Speech and a Radical Speech so that four conditions were
tested--the White Conservative, White Radical, Black
Conservative, and Black Radical. A Race X Belief

interaction was predicted.

History

Social Distance

 

Social Distance Scales

Racial or ethnic membership has been emphasized as
the major determinant of interpersonal rejection and social
distance in previous theories of racial discrimination
(Adorno, Frenkel—Brunswik, Levinson, and Sanford, 1950).

For instance, Bogardus (1928) pioneered work with an
ordinal scale of social distance, in which a subject
indicated zero social distance by stating that he was
willing to marry a member of a particular ethnic group, and
maximum social distance by stating that he would exclude
such a person from the country, social distances scales
have been used in a large number of studies.

After Bogardus' (l928)use of the social distance
scale with ethnic groups it was used by a number of
researchers with other groups. For instance, Wilkinson
(1929) used it with occupations, and religions. Sartain
and Bell (1949) published equal appearing interval scale
values. Melikaian and Prothro (1952) obtained the social
distance of Leganese students towards national and
religious groups and found greater social distance towards
certain national groups than tOwards certain religious

groups. The reverse result had been obtained by Prothro

in 1935. Prothro and Miles (1953) used the Sartain and
Bell (1949) scale in the deep South. Ansari (1956) used
the scale in India and found a relationship between social
distance and stereotypes similar to that obtained by Katz
and Braly (1933) in the United States. Hunt (1956) used
the scale in the Philippines and found that his §s showed
little social distance towards Americans, Christians, and
whites, but very much social distance towards Jewish,
Moslem and Buddhist groups. Kirsh (1957) used the scale
in connection with voting behavior. Best and Sohner (1956)
used it with political, economic, racial, religious, and
nationality groups. They found very large social distances
towards fascists, anarchists, and communists, large
distances towards racial groups, small distances towards
religious groups, and very small distances towards national
groups. Hill (1953) has criticized the concept of social
distance and presented arguments in favor of the usage of
interviews and projective techniques for the determination
of prejudice, rather than objective indices of social dis-
tance, such as the Bogardus scale.

Triandis and Triandis (1960) found that in a study

of social distance of college students with respect to

f'“\

r
\*/

/‘\

various social objects, a factorial design with two levels
of value of race, social class, religion, and nationality
was employed and analyses of variance were computed on
social distance scores. For white gs race and social
class were found to be more important determinants of
social distance than religion or nationality. The relative
importance of the various factors, however, varied from
group to group.

Race according to Triandis and Triandis (1960) was
the most important determinant of social distance for white
lgs. Most of the prejudice towards Negroes was racial and
not social class prejudice. Negro gs do not experience
social distance towards whites, or persons with lower class
occupations, or persons of different religion, but if the
person in question has a combination of these "undesirable"
characteristics, e.g., he is a “white, Jewish, Portuguese,
unskilled worker," then Negro gs feel some social distance
as manifested in a refusal to accept such a person as a
close kin by marriage, or as an intimate friend.

Triandis and Triandis (1960) also found that race was
by far the most important factor determining social distance.

Furthermore, social class was a much more significant

determinant of social distance than was religion or
nationality. White gs were found to show more social
distance than Negro gs: high F score gs more than low F
score gs, women more than men, Catholics more than
Protestants because it seems that certain features of

the socialization practices of Catholic parents, partic-
ularly the clarity of their approach towards the Divine
Law, predispose their children to intolerance of ambiguity,
which in turn is correlated with prejudice, and both faiths
more than Jews, lower-class §s more than lower middle—
class gs (because it seems that certain lower-class child
rearing practices are likely to produce insecurity and
insecure persons are more likely to be prejudiced than
secure persons) and these more than upper middle-class

gs, and §s from Northern and Northeastern Europe more than
gs with ethnic backgrounds from Sbuthern and Eastern
Europe. The data are interpreted in terms of a theory of
prejudice that employed conformity, cognitive dissonance,

and insecurity as its main constructs.

Social Distance §§_a Function g£.Attitudina1 Dissimilarity

Contrary to this Race Congruency Theory position,

however, Rokeach (1960a) postulated that social distance

is a function of attitudinal dissimilarity. According to
his hypothesis, interpersonal attraction and choice occur
across racial lines when persons share similar beliefs,
attitudes, and opinions: on the other hand, rejection
occurs when there are differences in opinions, attitudes,
and beliefs. For Rokeach (1960a), belief congruence,
insofar as psychological processes are involved, is more
important than ethnic or racial membership in determining
social discrimination.

Other researchers testing determinants of interpersonal
attitudes found a number of lawful relationships between
the characteristics of a person and the behavioral in-
tentions and evaluations of this person by subjects. These
judgments are largely determined by cultural factors
(Triandis, 1964a: Triandis, Davis, & Takezawa, 1965: Triandis
& Triandis, 1960, 1962).

A multidimensional instrument which measures the
behavioral component of attitudes, called the behavioral
differential (Triandis, 1964b) may be used to measure the
behavioral intentions toward the stimuli, while the
semantic differential (Osgood, Suci, & Tannenbaum, 1957)
may be used to measure the evaluation, potency, and

preceived activity of the stimulus persons. With American

subjects, race was found to be a powerful determinant of
variance in social distance judgments: with Greek subjects
religion was very important: with German and Japanese
subjects occupation was the important determinant of variance.

Triandis, Fishbein, and Hall (1964) found high
correlations between the admiration (social acceptance)
factor of the behavioral differential and the evaluative
factor of the semantic differential in person perception.

In the Triandis, Loh & Levin (1966) study, 94_§s were
shown slides of either a Negro or a white young man, who
was either well dressed or poorly dressed and simultane-
ously heard a tape-recorded statement which was either in
favor or opposed to integrated housing and which was spoken
either in excellent or in ungrammatical English. The
stimuli formed a 2x2x2x2 factorial design. The evaluation
of the stimulus person was measured by the evaluative
factor of the semantic differential: the behavioral in-
tentions of SS toward the stimulus persons were measured
by 3 factors of the behavioral differential.

It was shown that liberal §s differed from nonliberal
§s in the relative weights they employed for the character-

istics race, dress, English, and opinion. Race was a

more important factor in the determination of the social
distance of prejudiced than in the determination of the
social distance judgment of unprejudiced subjects.

Furthermore, English and belief were the determinants
of the evaluation judgments, as well as the social
acceptance judgments on the behavioral differential. In
the case of admiration, English was the primary determinant
for all subjects, but the weight given to it differed be-
tween the subgroups. Prejudiced subjects gave some weight
to race and belief.

English, race, and dress, in that order, were important
in the determination of the judgments on the friendship
factor. 'The tolerant subjects gave no weight to race and
the prejudiced ones gave a substantial weight to that
characteristic as a determinant of their judgments.

These findings confirm previous research which
demonstrated that different aspects of the behavioral
intentions of subjects toward stimulus persons are
determined by different combinations of the characteristics
of these stimulus persons.

To understand prejudice in its full complexity, it
is necessary to think of a matrix, the rows of which are

defined by various "undesirable“ characteristics. In

addition to those determinant examined in this study,
subjects responded to stimulus persons who differed from
them in religion, age, sex, nationality, competence in
doing a job, degree of sociability, etc.

Rickard, Triandis, & Patterson, (1963): Triandis,
(1963) all investigated the social distance factors of
'undesirable characteristics in a stimulus person. In
summation, the studies showed that when all of the
information collected in these studies is placed together,
it becomes clear that the classes of behavioral intentions
mentioned above are distinct, and influenced by different
combinations of "undesirable" characteristics. “Admire
the ideas of“ is most sensitive to characteristics
indicative of status-—for example, kind of English spoken,
occupation-—and also to the opinions of the stimulus
persons: marital acceptance is sensitive to English, but
also to race and age: friendship acceptance is sensitive
to age, sex, religion, English, race, dress, and skin
color: subordination is likely when the stimulus person
is of a high status occupation: exclusion from the
neighborhood is primarily sensitive to race and English,

and slightly to religion: acceptance as an employee is

10

primarily sensitive to competence and disability, and

secondarily to race and sociability.

gggjudiced Attitudes that Q9 Not Lead E9 Prejudiced Behavior

 

 

According to Berg (1966), attitudes may not necessarily
lead to predicted behavior. In his study, a social context
was arranged in which verbal measures were expected to
predict behavior. A white g was placed in a conflict
between agreeing with a Negro confederate or a white con-
federate on autokinetic judgments. The results indicated
that the E Scale Negro items, F Scale items, and the Social
Distance Negro item did not relate to social agreement with
the Negro in either of 2 conditions of importance. The
Jewish gs were significantly less prejudiced than the
Protestant-Catholic gs on the E Scale Negro items. Jewish
gs disagreed more than Protestant-CathOlic §s with the
Negro confederate on autokinetic judgments. Therefore, it
seems that the common verbal measures of attitudes toward
Negro people maynot be good predictors of particular social
behavior toward a given Negro person. This may be espe-
cially true when the Negro person is not perceived as being

a representative member of his ethnic group. For example,

11

in Berg's (1966) study, a Negro who has the status of a
student at a major university may not be perceived as similar
to the category of people who are the object of prejudice.

In such social circumstances, increasing the instrumental
value of social rejection may not lead to the expression

of prejudice.

A further speculation is that the predicted social
behavior has to be relevant to the content and structure of
the attitudes. For example, agreement on autokinetic
judgments may not represent social behavior that is generally
related to the cognitive and affective aspects of attitudes
toward Negroes. Social and political opinions that are
more central to personality structure could be expected to
be more relevant when the prejudiced person is confronted
with the issue of agreeing or disagreeing with a Negro
person. Similarly, it may be that the verbal scales did
not measure an aspect of attitudes toward Negroes that is

central to personality structure.

Evidence Supporting Rokeach's Position

 

Other studies that support Rokeach's (1960) position

on attitude similarity—dissimilarity are Byrne (1969) and

12

Byrne, Nelson, & Reeves (1966) who found that the effect
of attitudinal material on attraction has been interpreted
in terms of effectance motivation. In these studies,
effectance is defined as a motive to cope effectively with
the environment by means of accurate perception, logical
thought processes, consistency, correct interpretation of
reality, etc. The similar attitudes of others serve to
satisfy this motive by providing consensual validation,
while the dissimilar attitudes of others frustrate this
motive via consensual invalidation. It has been found
that the steepness of the attitude-attraction function
increases as the subject's motivational level increases.
When effectance arousal is extremely high, however, there
is a reverse effect, and the slope is actually less steep
and the Y intercept higher than in a neutral situation.
This latter effect has been interpreted in terms of the
disorienting and differentiating qualities of high effectance
arousal, a condition also characteristic of schizophrenia.
In the Byrne, Griffitt, Hudgins, & Reeves (1969)
study, this was investigated further. The results showed
that since schizophrenics were functioning at a high level
of effectance arousal, they should have made less differ-

entiation between similar and dissimilar strangers than do

/\

l3

nonschizophrenics, but this finding was only partially
supported. That is, schizophrenic patients yielded a
linear attitude-attraction function which differed
significantly from that found with college undergraduates.
The divergence between these subjects and students is
meaningful, of course, only because the surgical patients,
alcoholic patients, and job corpsmen did not differ from
college students.

The results also showed that the clearest finding is
that the linear relationship between proportion of similar
attitudes and attraction is a phenomenon generalizable
beyond the college student. The tendency to make a
positive response to a stranger who expresses attitudes
similar to one's own and a negative response to one Who
expresses dissimilar attitudes is obviously relatively
general in the American population.

This supports previous studies which have shown that
the empirical law describing attraction toward a stranger
as a positive linear function of the proportion of the
stranger's attitudes which are similar to those of the
subject is well established in samples of undergraduate

students (Byrne and Clore, 1966: and Byrne and Nelson,

14

1965). This same empirical relationship has been found
among elementary and secondary school children (Byrne and
Griffitt, 1966). There is also evidence that female
clerical employees of the Bell Telephone Laboratories
respond differentially to two points along the similarity
continuum (Krauss, 1966).

Therefore, in samples unlike college undergraduates
in age, education, socioeconomic level, intelligence, and
adjustment, not only does a linear function hold between
the two variables, but the specific responses of the
subjects are predictable on the basis of a formula derived
on college students.

In summary, most studies of the past used social
distance scales to measure prejudicial attitudes of Whites
toward Blacks such that the greater the prejudiced at—
titudes of the Whites, the greater the distance these
Whites rated Blacks on social distance scales.

But, Rokeach (1960) believed that social distance was
a function of dissimilarity of attitudes. He argued that
belief was more important than race in rating people on
social distance scales while Triandis argued that race was
more important than belief in rating people on social

distance scales.

15

V'But, some studies of social distance have shown
that prejudiced attitudes do not lead to prejudiced
behavior if the content and structure of the attitudes
for rating are irrelevant to the gs.
In the present study, we also examined social distance.
We felt that prejudiced gs would show more social distance
toward Blacks than toward Whites even though both Blacks

and Whites may express the same attitudes to the SS.

Effects 9f Interpersonal Attraction

 

The accurate prediction of interpersonal attraction
and repulsion in dyadic relationships will undoubtedly
require that knowledge about several classes of independent
variables is obtained.

It has been suggested (Byrne, 1961a) that the degree
of attraction between two individuals is determined by
four classes of variables: the structural properties of
the environment which act to vary propinquity, the strength
of the characteristic affiliation motive of each individual,
generalization from previous learning with respect to the
overt stimulus properties of one another, and the number
of reciprocal rewards and punishments which occur during

their interactions.

16

Studies on the effect of interpersonal attraction
(1) on attitude similarity, (2) on propinquity, (3) on
socioeconomic similarity, (4) on threat reduction, (5)
race attraction, (6) on racial prejudice and assumed

dissimilarity of attitudes, are outlined below.

Attitude Similarity

 

A number of studies have found greater similarity

among friends than among nonfriends with respect to a
variety of issues (Bonney, 1946: Loomis, 1946: Newcomb,
1956: Precker, 1952: Richardson, 1940: Winslow, 1937).
A few studies of a more experimental nature also support
the notion of a relationship between attitude similarity
and interpersonal attraction (Jones and Daugherty, 1959:
Smith, 1957).

In the Byrne (1961a) study one group of Se received
attitude scales filled out exactly the same as theirs
had been, one received scales with exactly Opposite
views expressed, one received scales with similar opinions
on the most important issues and dissimilar on the least
important, and the fourth received scales with similar

opinions on the least important issues and dissimilar

l7

opinions on the most important.

He found that (a) a stranger who is known to have
attitudes similar to those of the subject is better liked
than a stranger with attitudes dissimilar to those of the
subject, (b) a stranger who is known to have attitudes
similar to those of the subject is judged to be more
intelligent, better informed, more moral, and better
adjusted than a stranger with attitudes dissimilar to
those of the subject, (c) but a stranger who is known to
have similar attitudes on issues important to the subject
and dissimilar attitudes on unimportant issues is not
better liked and is not evaluated more positively on the
other four variables which are the other student's
intelligence, knowledge of current events, morality, and
adjustment than a stranger for whom the reverse is true.

In fact, the Similar on Important Attitudes Group
(SIA) rated the “stranger" significantly more positively
than did the Similar on Unimportant Attitudes Group
(SUA) with respect to their personal feelings about him,
his morality, and his adjustment. On the other three
variables, which were desirability as a work partner,

intelligence, and knowledge of current events, the two

/"\

/~"\.

\J

18

groups did not differ as predicted. The results suggest
that the Personal Feelings scale is the most sensitive
measure of interpersonal attraction. With the other five
interpersonal judgment scales, additional factors
apparently contribute to the variance.

In summary, gs judged people who shared the Ss'
similar attitudes as being more attractive and more
intelligent than people who did not share the Ss'
similar attitudes.

In our study, we predicted that §s would like the
actors delivering the Conservative Speech better than the

actors delivering the Radical Speech.

Propinquity

Many studies in a diverse array of situations have
suggested that propinquity is of overwhelming importance
in determining who will interact with whom. Distance and
spatial arrangements have been found to influence friend-
ship choices in university housing projects for married
veterans (Caplow and Forman, 1950: Festinger, Schachter,
and Back, 1950), in the planned communities of American

suburbia (Whyte, 1956), among college undergraduates

19

(Lundberg and Beaxley, 1948: Lundberg, Hertzler, and
Dickson, 1949), in a college dormitory (Evans and Wilson,
1949: Newcomb, 1956), and within a college classroom
(Maisonneure, Palmade, and Fourment, 1952: Byrne and
Buehler, 1955).

Byrne (1961c) found that students became acquainted
with a greater proportion of 'seat neighbors' (those who
occupy neighboring seats in the same row) than of 'non-
neighbors' (all remaining students), if the contacts lasted
7 or 14 weeks, but not in 3.5 weeks of contact: (b)
Students did not become acquainted with a greater proportion
of 'cross—row neighbors' (those occupying neighboring seats
in adjacent rows) than of non—neighbors: It is clear that
the two types of neighboring seatshave very different
effects on their occupants: their physical distance is
quite different. It could be that the ease with which a
student can converse with his seat neighbor lessens the
probability that he will go through the motions necessary
to converse with cross—row neighbors and non—neighbors:

(c) The intensity of the seat-neighbor relationships ex-
ceeded that of the non-neighbor relationships in the
14-week condition, there was no difference in the 7-week

condition, and the reverse was true in the 3.5-week

\J

/\

1“\

20

condition: (d) The number of relationships per student
did not increase as the number of seat neighbors was increased.
In summary, §s tended to interact with people near
them although people at equidistances may have served
different functions in the relationship. Time seemed to
strengthen these interactions.
In our study, we predicted that the gs would rate
feeling closer to the White actors than the Black actors
because the SS would feel more relaxed and more familiar

around White people.

Status Similarity

 

There are sociometric data which suggest that
friendship choices within a group tend to be between
members of the same general socio—economic status (Bonney,
1949: Dahlke, 1953: Longmore, 1948: Lundberg and Beazley,
1948: Lundberg and Steel, 1938). One possible explanation
for such findings comes from Festinger's (1950, 1954)
theory of social comparison process which is not incom—
patible with reinforcement theory. He suggests that
individuals are attracted to others who are similar to

themselves with respect to both opinions and abilities.

/\\

21

Thus, groups are formed on the basis of such similarities,
and communication and comparison with those who are
divergent may be avoided. Festinger (1954) goes on to say:
"The segmentation into groups which are relatively alike
with respect to abilities also gives rise to status in a
society. And it seems clear that when such status dis-
tinctions are firmly maintained, it is not only members of
the higher status who maintain them. It is also important
to the members of the lower status to maintain them for it
is in this way that they can relatively ignore the
differences and compare themselves with their own group.
Comparisons with members of a different status group,
either higher or lower, may sometimes be made on a phantasy
level, but very rarely in reality.“

A number of investigations lend support to Festinger's
position. For example, Zander and Havelin (1960) found
that subjects in a group situation were most attracted to
others whose competence with respect to an experimental
task was closest to their own. In addition, many of the
affiliation studies indicate that subjects choose to be
with others similar to themselves in emotional state

(Zimbardo and Formica, 1963).

22

Both the sociometric findings and the social
comparison theory suggest the possibility that the effect
of attitude similarity-dissimilarity on attraction may
be a much more general phenomenon in that various other

types of similarity-dissimilarity also affect attraction.

Economic Similarity

The theory of social comparison processes suggests
that individuals are attracted to each other on the basis
of similarity in opinions, abilities, and emotional state.
Generalizing further in the present investigation,
attraction was hypothesized by Byrne, Clore, & Worchel
(1966) to be a function of similarity—dissimilarity in
economic status. In their study, a total of 84 gs was
divided into high and low economic status on the basis of
their responses to items dealing with spending money.
Three experimental conditions were devised in which gs
evaluated a stranger on the basis Of his or her responses
to the economic items and some attitudinal items. In 1
condition, low-status gs responded to a high-status
stranger, in a 2nd condition, high status gs responded

to a lowestatus stranger, and in a 3rd condition, high-

23

and lowestatusigs responded to strangers similar to
themselves. As hypothesized, attraction was signifif
cantly affected by similarity—dissimilarity of economic
status. It was found that the specific responses of §s
could be predicted on the basis of a law of attraction
formula derived in earlier work on attitude similarity-
dissimilarity.

Such that when Individual A receives positive
reinforcement from Individual B, A's attraction toward
B increases. Individual B constitutes a conditioned
stimulus for the positive affect evoked by the rein-
forcement. In the same general way, negative reinforcement
from Individual B results in negative affect and hence
dislike for B.

As in above mentioned studies, attitude similarity
and attitude dissimilarity are assumed to constitute
consensual validation and invalidation: such information
acts as positive and negative reinforcement with respect
to the need to be logical and accurate in interpreting
the stimulus world (Golightly and Byrne, 1964). The
rationale for attributing positive and negative rein-

forcement properties to information concerning economic

/'\

24

similarity—dissimilarity suggests a broader generalization.
It seems quite possible that in any type of social com-
parison with other human beings (e.g., Opinions, attitudes,
beliefs, values, abilities, traits, taste, income, overt
behavior, etc.) similarity is preferable to dissimilarity.
Consensual validation as evidence of one's accuracy and
reasonableness in dealing with one's environment may extend
to all behavior.

In summary, the gs rated people on their same
socio—economic status as being most attractive.

In our study, we predicted that the gs would rate the
White Conservative as most similar to themselves in beliefs

followed by the WR, BC, and BR as most dissimilar to the SS.

Threat Reduction

In a study by Pepitone and Kleiner (1957), it was
shown that low threat (i.e., low probability of a group
losing status) produces a greater increase in sociometric
attraction within the group than high threat (i.e., high
probability of a group losing status). The data suggested
the hypothesis that a reduction in threat would lead to
an increase in mutual attractiveness of group members for

having "improved" an existing state of affairs.

25

Groups of students were given two parts of a “test
battery," presumably for the purpose of comparison with
supervisory personnel in industry in group problem-
solving situations (Kleiner, 1960). After the first part,
the groups experienced one of two estimates of the prob-
ability of losing comparability with industrial personnel
and peer groups. After the second part, all groups
experienced one of two reductions in the probability of
loss due to improved performance through the contribution
of a confederate place in each group.

The results of the degree of threat reduction and
interpersonal attraction showed that the change in
attraction was greater in the large-reduction condition
regardless of the threat baseline from which the reduction
in threat occurred. There was a greater increase in the
attraction of the confederate in the large—reduction
condition, and this again was regardless of the baseline
from which the reduction occurred. The data on the §s'
reflected the effects of the threat conditions on their
liking for each other as well. The effect of threat
reduction from different baselines showed that the change

in attraction was larger when the reduction was from the

26

high threat baseline. There was a greater increase in the
social attraction of the confederate when the reduction in
threat was made from the high-threat baseline.

In testing the effectiveness of the confederate's
role, the effect of the magnitude of the threat reduction
is reflected, not in the increased importance of the
confederate, but in the gs' depreciation of each other in
the small—reduction condition. Apparently, the §s in the
small—reduction condition were holding each other
responsible for the relatively small improvement. There
was also a greater increase in the perceived importance of
the confederate when the threat reduction is made from the
high baseline than when it is made from the low baseline.

The generalized results have shown that (1) In all
conditions the confederate was perceived to be highly
important to the group performance: (2) "importance“
provided evidence for the effectiveness of the experi-
mental treatments: (3) The level of aspiration reflected
only the effect of the initial threat treatment, since
the level of aspiration measure failed to differentiate
between the large and small threat-reduction conditions,
but did differentiate between the high and low initial-

threat conditions: (4) The increment of perceived

27

attraction for the confederate varied directly with the
degree to which probability of loss was reduced: (5) The
increment of perceived attraction varied directly with the
initial level of probability of loss.

In summary, the gs rated people that reduced a great
deal of threat as being more attractive than a person
reducing a small amount of threat.

In our study, the gs should like the Conservative
Speech better than the Radical Speech because in the

Conservative Speech a great deal of threat is reduced.

Race Attraction

It has been reported in a number of investigations
that friendship choices or attraction ratings are
influenced by race. White gs of various ages in various
settings are found to be more attracted to other whites
than to Negroes when asked to make friendship choices
(Berkun and Meeland, 1958: Koch, 1946: Mann, 1958). The
importance of the interaction between race and racial
prejudice was recently demonstrated by Wong (1961). He
found that on the basis of minimal background data, 73

gs were asked to make several judgments about a stranger.

28

Attraction was measured by means of two scales of the
Interpersonal Judgment Scale (Byrne, 1961a). On this
instrument, gs indicated their personal feelings of liking
and disliking with respect to a stranger and also rate

his desirability as a work partner. As hypothesized,
highly prejudiced gs responded more negatively to a Negro
stranger than to one identified as white. The prediction
that'gs low in prejudice would not, respond differentially
on the basis of race was supported with respect to
Personal Feelings. On the WOrk Partner variable, they
actually responded more positively to a Negro than to a
white stranger.

In Summary, prejudiced gs responded more positively
to a White stranger than to a Black stranger even though
both parties' attitudes were unknown to the gs.

in our study, we predict that White actors would be
rated more positively than Black actors on traits unknown

to prejudiced gs.

Racial Prejudice and Assumed Dissimilarity g§_Attitudes
Both the balance throry of Heider (1958) and the
strain-toward-symmetry formulations of Newcomb (1953)

would predict that an individual who dislikes Negroes

'r

or o.

I».

1

0-.

.p-~v— -

29

should assume that they hold attitudes dissimilar from
himself. In Newcomb's terminology, if A (highly pre-
judiced S) has negative feelings toward B (Negro stranger)
and positive feelings toward X (e.g., classical music),
symmetry can be obtained by A if he assumes that B
dislikes classical music. Similarly, if a S low in
prejudice has positive feelings toward a Negro stranger
and positive feelings toward classical music, he achieves
symmetry by assuming that the Negro stranger also likes
classical music.

Rokeach and Rothman (1965), who reported supportive
evidence for Heider and Newcomb, compared and contrasted
the belief congruence model with the Osgood and Tannenbaum's
congruity principle.

The belief congruence model was compared and
contrasted with Osgood and Tannenbaum's congruity principle,
the main difference between the 2 principles being that
the former asserts that the outcome of cognitive inter-
action cannot be accurately predicted from a knowledge of
the evaluative meaning of the 2 elements judged in isolation,
while the latter asserts that it can. Quantitative
predictions generated by the 2 models were compared with

obtained results showing that the average absolute error

30

of the congruity model was 1.07 and the average absolute
error for the belief congruence model was .34, thus
suggesting that the latter model's predictions were
about 3 times as accurate as the congruity model's
predictions.

Congruence can be defined both in terms of similarity
and importance. Given two beliefs or subsystems of belief
equal in importance, the one more similar to our own is
the more congruent: conversely, given two beliefs or
subsystems perceived to be equally similar to our own, the
one judged as more important is the more congruent with
our own belief system.

There is some experimental support for the general
proposition. It has been found that individuals assume
greater similarity between themselves and liked others
than between themselves and disliked others (even in the
absence of actual differences) with respect to personality
traits (Fiedler, Warrington, & Blaisdell, 1952) and
values (Precker, 1953). In addition, Muraskin and
Iverson (1958) reported that as the social distance of
a minority group increased, subjects assumed greater
dissimilarity between themselves and members of that

group with respect to attributed social distance ratings.

31

Once interaction has begun, reciprocal reward and
punishment is proposed as the crucial determining factor.
It has been suggested (Newcomb, 1956) that attraction
between persons is a function of the extent to which
reciprocal punishments. A Special subclass of this
variable would be preceived similarity and dissimilarity
of the attitudes of two individuals. It can be assumed
that persons in our culture have well established learned
drives to be logical and to make a correct report of the
environment. Those who seem deficient in this respect
are generally categorized as being uninformed, of low
intelligence, immoral, and/Or as being out of contact with
reality. It is primarily through consensual validation
that people determine whether they or anyone else is
logical or correct in interpreting environmental events.
Hence, any time that another person offers one validation
by indicating that his percepts and concepts are
congruent with his, it constitutes a rewarding inter-
action and, hence, one element in forming a positive
relationship. Any time that another person indicates
dissimilarity between ones two notions, it constitutes

a punishing interaction and thus one element in forming

32

a negative relationship. Disagreement raises the
unpleasant possibility that one is to some degree stupid,
uninformed, immoral, or insane. An alternative possi-
bility is that it is the other person who is deficient

in one or more of these characteristics. Probably other
variables, such as the importance of the issue to each
individual, contribute to the effect (Byrne 1961).

As would be prediCted from Newcomb's (1953) A-B—X
model, prejudiced subjects not only dislike Negroes but
they erroneously assume attitudinal differences between
themselves and Negro strangers (Byrne and Wong, 1962).
Thus, symmetrical A-B—X relationships are formed, in
which those who dislike Negroes assume dissimilarity and
those who like Negroes assume similarity of attitudes.

In a further investigation, Byrne and‘Wong (1962) created
non-symmetrical conditions by presenting gs with in-
formation indicating that a stranger was either
completely similar to or completely dissimilar from
themselves in responding to a 26-item attitude scale.

It was found that attraction was primarily a function

of attitude similarity. That is, gs responded positively

to similar strangers and negatively to dissimilar

/‘\

33

strangers regardless of the race of the stranger or the
degree of prejudice of the g,

This finding was interpreted in terms of a reward
and punishment framework. Agreement about attitudes has
been interpreted as reward via consensual validation and
disagreement as punishment via consensual invalidation
(Byrne, 1961a: Byrne, 1961b: Byrne, 1962). As a high-
prejudiced g reads through an attitude questionnaire in
which a Negro stranger expresses views similar to his
own on 26 assorted items, it is assumed that this
experience functions as the administration of 26 rewards.
As a consequence, at least for a limited time-period and
with respect to one particular Negro, high-prejudiced gs
in the Byrne and Wong (1962) experiment indicated that
they liked the Negro stranger and would enjoy working with
him as a partner in an experiment.

Therefore, if prejudiced individuals were provided
with experiences in which equal status interactions with
Negroes involved interpersonal rewards rather than the
expected punishments, their expectancies and hence their
prejudiced beliefs should be altered.

It is hypothesized that interracial contact is

effective in reducing prejudice as a function of the

34

frequency of interpersonal rewards which occur in that
contact and of the proportion of rewarding interactions
in relation to punishing and neutral ones. Without
isolating the reward variables, several studies have
nevertheless, found that equal status contacts with
Negroes resulted in more positive interracial attitudes
among housewives dwelling in federal housing projects
(Deutsch and Collins, 1951: Wilner, Walkley, and Cook,
1952), students participating in a special seminar
experience (Smith, 1943), graduate seminar in leaderless
discussion groups (Mann, 1959), officers and enlisted
men in infantry platoons (United States War Department,
1947), merchant seamen (Brophy, 1956), and workers in var-
ious settings (Gundlach, 1956: Palmore, 1955).

In building a theory of interpersonal attraction,
the most inclusive independent variable is almost
certainly that of reciprocal reward and punishment.
Actually, the other three classes of variables appear to
be relevant only as they relate to reward and punishment.
Propinquity is important in determining attraction
because environmental variables can facilitate or inhibit

interaction, and without interaction, rewards cannot be

/\

35

given or received. The strong desire to form affiliative
relationships suggests that the expectancy of reward in
an interpersonal context is high. Similarly, attraction
or repulsion based on overt stimulus properties would
seem indicative of expectancies for rewards or for
punishments in the interaction (Byrne and Wong, 1962).
Therefore, interracial contact may be effective in
reducing prejudice because Negro-white interactions
involve rewards given directly by Negroes or rewards
administered by other individuals as a consequence of
interaction. In addition to the effects of attitude
similarity, acceptance of Negroes is brought about by
such rewarding interactions as judgmental support in an
Asch conformity situation (Malof and Lott, 1962).
Contrary to the previous finding, prejudice and
race did interact in influencing attraction (Byrne and
McGraw, 1964). Both lows and high-prejudiced gs
responded to white strangers on the basis of similarity
of attitudes, and loweprejudiced gs responded to Negro
strangers on the same basis. However, by those high in
prejudice, the Negro strangers were rated in a uniform

manner no matter what their attitudes. The reason for

36

the difference between this finding and that of Byrne
and Wong, (1962) was sought in a second experiment. It
was hypothesized that with the utilization of 26 similar
and 26 dissimilar attitudes, as in the Byrne and Wong,
(1962) investigation, the original results would be
replicated. It was also hypothesized that the use of
photographs in Experiment I exerted a negative influence
on the ratings of Negro strangers by high-prejudiced gs.
It was found that a stranger with 26 similar attitudes
evoked positive ratings and one with 26 dissimilar
attitudes evoked negative ratings from gs at each
extreme of the prejudice scale. However, those low in
prejudice were more positive toward Negroes than were
those high in prejudice. Contrary to expectations,
ratings of a stranger by both prejudice groups were more
positive when a photograph was present. It was concluded
that a.§ low in prejudice responds toward a Negro
stranger as toward a white stranger: attraction ratings
vary as a linear function of the proportion of similar
to dissimilar attitudes. A g high in prejudice, on the
other hand, will respond positively to a Negro stranger

providing that this stranger is completely similar to

37

himself concerning attitudes about a relatively large number
of topics: he will respond with indifference or dislike
toward a Negro stranger who departs from total similarity.

In summary, prejudice gs also falsely assumed that
their attitudes are dissimilar to the attitudes that Black
people have.

In our study, we felt that the prejudice gs would rate
the Black actors' attitudes as being more dissimilar to the

‘gs' than those of the White actors.

The Belief Versus Race Controvergy

Which is more important a person's belief or a
person's race? Rokeach's (1960) position supports a Belief
Congruence Theory whereas Triandis's (1961) position
supports a Race Congruence Theory. Both positions are
explored below.

Rokeach, Smith, and Evans (1960) stated that “insofar
as psychological processes are involved, belief is more
important than ethnic or racial membership as a determinant
of social discrimination" (p. 135). They argued that true
racial discrimination implies that the outgroup is
discriminated against, but that at the same time, the

ingroup is favored, and they provided evidence supporting

38

a belief-congruence theory of prejudice, which subsumes
racial and ethnic prejudice as a special case under
belief prejudice.

They asked gs to respond to a nine-point scale ranging
from "I can't see myself being friends with such a person”
to "I can very easily see myself being friends with such
a person." The stimulus persons used were persons of the
same or different race or religion as the gs, who had the
same or different beliefs as the gs. All possible
combinations of race and belief were used. The beliefs
dealt with God, communism, labor unions, socialized medicine,
etc. The writers stated that in the first of two studies
white gs making hypothetical choices were more accepting
of Negroes who agreed with them on important issues than
they were of whites disagreeing with them. This was true
both for a northern and a southern sample. In a second
study, they showed that Jewish children accepted Gentiles
agreeing with them to a greater extent than they did Jews
disagreeing with them. Most of the time the gs discrim-
inated on the basis of belief and not on the basis of
racial or ethnic group when they were given the oppor-
tunity to react to social stimuli differing simultaneously

on both characteristics.

39

Triandis (1961) rejected these findings on the
friendship variable because he felt that Rokeach's own
experimental confirmation of his hypothesis was due to
an artifact of his method of measuring prejudice. Had
Rokeach considered something more than friendship he would
not have Obtained his results. Triandis argued that
prejudice involved more than nonacceptance as a friend
because it involved negative behaviors as well. A concept
more general than prejudice is the concept of social
distance.

Since Rokeach's research technique used friendship
as the variable under investigation, and so inevitably
dealt only with a relationship involving small social
distance, prejudice and discrimination are much more
relevant to acceptance or rejection of a relationship
involving relatively large social distance, such as
acceptance of a person as a neighbor or as a student in
one's university. People do not exclude other people
from their neighborhood, for instance, because the other
people have different belief systems, but they do exclude

them because they are Negroes.

4O

Triandis (1961) felt that race rather than belief
congruence was the critical determinant. In his study,
belief congruence was measured by Morris' (1956) ”13
ways to live." Stimulus individuals in the study were
varied either has the same or different in race, religion,
occupational status and philosophy. The results showed
that race accounted for about four times as much of the
variance in social distance judgments as did belief
congruence, although all four main effects were highly
significant.

Rokeach (1961) replied to Triandis with the objection
that the long and involved passages of Morris' ways to
live were too vague and not sufficiently salient for the
.gs. He added that Triandis did not tell what his gs
remembered of these two paragraphs (not to speak of the
13). He doubts if Triandis's paragraphs were ego-involving.

Rokeach felt that the belief variable pitted against
race was vague, abstract, overintellectual, and complicated:
hence weak: And the polarities, "same“ and "different
philosophy," were nOt really Opposites, just different.
They were not Ways producing social conflict in everyday

life. They represent ideals by which "people say they

l“\

41

would like to live, and not how they do live" (Morris and
Jones, 1955). Thus, the difference between “same" and
"different philoSOphy" is a difference which, psychologically
and sociologically, makes little difference.

According to Rokeach, the major difference in method
was that Triandis had used vague descriptions of stimulus
persons, such as "Negro, same philOSOphy of life": while
Rokeach had used stimulus persons with specific beliefs,
such as "Negro who believes in God“ or "white atheist."
To make sure that his results were nor artifacts, Rokeach
repeated his research with eight widely differing issues:
God, communism, desegregation, etc. These issues were
socially salient and controversial, though not equally so.
On all eight issues, he found friendship ratings to be
more a function of belief than race, though not equally
so. These results hold in the South as well as in the
North.

He concluded that his position—that the more salient
a belief, the more will belief congruence override racial
or ethnic congruence as a determinant of social distance-

had not actually been challenged by Triandis' data.

42

Reconciliation g: the Belief Versus Race Controversy

 

 

Stein, Hardyck, and Smith (1965) presented a study
designed to reconcile these differences. In a test of
Rokeach's contention that prejudice is the result of
perceived dissimilarity of belief systems, 44 white
California 9th graders completed a "teenage social
distance scale" for 4 “stimulus teenagers,“ presented
as, respectively, white or Negro, and like or unlike the
respondent in values, in relation to gs' own responses
2 months earlier. .gs also indicated how friendly they
would feel toward each.

The results showed that gs felt most friendly towards
the white—like teenager, followed by the Negro-like,
white-unlike, and Negro-unlike. When white gs were given
no information at all about a Negro teenager, they
apparently assume that he is different from them in
values and react towards him accordingly. With belief
similarity held constant, the gs perceived that the white
stimulus teenagers were more like them, given identical
information about the whites and Negroes.

Strong “race effects" were obtained on "sensitive"

or controversial items by prevailing cultural standards

43

on the social distance scale, perhaps reflecting
institutionalized areas of prejudice, and on total social
distance scores when information concerning belief system
was not provided. They added that institutionalized
norms and large race effects exist in situations in which
there is both intimacy of contact and pressure of others.

Therefore, when gs are forced to evaluate stimulus
individuals in terms of their beliefs, then belief
congruence is more important than race. But when the
belief component is not provided, spelled out in con—
siderable detail, gs will react in racial terms on the
basis of assumptions concerning the belief system of
others, and of emotional or institutionalized factors.
The practical implications of these results are obvious.
If people of different races encounter one another under
conditions favoring the perception of belief congruence
(as, for example, in equal—status contacts), then racial
prejudice should be substantially reduced.

Although Stein et a1. obtained a strong effect on
belief congruence and a lesser race effect, their scale
was limited to positive items such as "invite home to
dinner,“ "have as member of social group,“ "work on

committee with,“ etc. Triandis' criticism of Rokeach

44

applies equally to Stein et a1. Prejudice involves negative
behaviors as well as the lack of positive behaviors.

Part of the confusion in previous discussions stems
from the assumption that prejudice is a unidiminsional
construct. Triandis (1964b) challenged this assumption
and presented the results of several factor analyses which
indicated that social acceptance (assumed here to be the
opposite of prejudice) is a multidimensional construct.

Triandis' (1964b) basic factors were obtained from
factor analyses of behavioral differential scales, and
were as follows:

Factor I. Formal Social Acceptance with Subordination
versus Formal Social Rejection, defined by high loadings
on items such as ”I would admire the ideas of," “I would
admire the character of,“ "I would cooperate in a political
campaign with,“ etc.

Factor II. Friendship Acceptance versus Friendship
Rejection defined by high loadings on "I would accept as
an intimate friend," "I would eat with," etc.

Factor III. Friendship Acceptance versus Friendship
Rejection defined by high loadings on "I would accept as

an intimate friend

 

45

Factor IV. Social Distance, defined by high loadings
on “I would exclude from my neighborhood,“ "I would exclude
from my country,“ etc.

Factor V. Subordination, defined by high loadings on
“I would obey,“ “I would not treat as a subordinate,“ etc.

Triandis' (1964b) analysis permitted the computation
of the relative variance accounted for by the race, age,
occupation, sex, and religion of the stimulus person as
determinants of responses on each of these factors. It
was found that occupation was, by far, the most important
determinant for Formal Social Rejection and for Subordination.
Incongruence of sex and age, followed by race and religion
were the primary determinants of variance for Friendship
Rejection. For Social Distance, the variance was almost
completely determined by race.

Thus, the findings of the studies described above can
be reconciled. Rokeach's (1961) data involved the friend-
ship factor: so did those of Stein et a1. (1965), Byrne
and Wong (1962), and other studies in which belief
dissimilarity emerged as the major variable. Triandis'
(1961) data were more general, since his Social Distance

scale included both friendship and negative behaviors.

46

His data were primarily determined by responses to the
item “exclude from the neighborhood," which about 55%

of his gs indicated they would do when the stimulus
person was a Negro. As a result, race explained most of
the variance.

Triandis and Davis (1965) offered additional data
on the influence of both belief and race dissimilarity
in determining social perception of stimulus persons.
Three hundred gs responded, on 12 semantic and 15 be-
havioral differential scales, to 8 stimulus persons
generated by all possible combinations of the
characteristics Negro—white, male-female, pro- or
con-civil-rights legislation. They also rated 35
political and civil rights issues on semantic differen-
tials. A factor analysis of gs, based on the covariances
of their responses to variables, defined 11 types of
gs. Two of these types were strongly prejudiced. One
type of‘g showed “conventional“ prejudice: they were
extremely sensitive to the race component when responding
to the stimulus persons. “gs of the other type showed
“belief" prejudice: they were particularly sensitive to

the beliefs of the stimulus person.

47

On the Formal Social Rejection factor, belief is
quite clearly the element which orders the pattern.

In contrast, the Friendship Rejection factor showed a
clear ordering by race, followed by belief. For the
Marital Rejection factor, the most important element is
sex, but the next most important element is race. For
the Social Distance factor, the most important element
is race, followed by belief. For Subordination, it is
also race, followed by belief. Thus, although both race
and belief are important, race is more important for
most of the factors of behavioral intentions, expecially
those behaviors that are more intimate: belief becomes
more important in the less intimate behaviors of the
Formal Social Rejection factor.

.gs high on Marital Acceptance versus Marital
Rejection rejected people primarily on the basis of
race: they also consider civil rights and many political
issues as quite unimportant. gs high on Subordination
showed belief prejudice and considered most issues as
very important. Thus, the gs' behavioral intentions were
a function of both personality and the type of behavioral

intentions involved.

48

The race and belief characteristics of stimulus
persons acquire different weights depending upon both
the nature of the items to which the gs are responding
and the type of g making the response.

Specifically, the Social Distance and Marital
Rejection factors were responded to in the manner
described by Triandis (1961), and Triandis and Triandis
(1960, 1962). The Evaluative and Formal Social Rejection
Factors were responded to more in the manner described
by Rokeach et a1. (1960): this was particularly true in
the case of unprejudiced gs. The Friendship Rejection
and Subordination cluster of behavior was responded to
either on the basis of race, or on the basis of belief,
depending upon whether the gs were high or low on general
prejudice. Those gs who were prejudiced responded to the
race component. Data show that both race and belief
were important in the case of highly prejudiced gs
responding to the Formal Social Rejection factor, but that
belief becomes less important in the other response
continua. This finding is consistent with the arguments

of Stein et a1. (1965).

49

Examination of the three clusters of behaviors
which were responded to differently revealed that the
Social Distance-Marital Rejection cluster included the
most intimate behaviors: the Friendship Rejection and
Subordination behaviors appearred to be intermediate in
intimacy and social commitment: Formal Social Rejection
included the least intimate behaviors, and this factor,
together with the semantic differential evaluative
factor, involved a minimum of social commitment.

In the case of the most intimate behaviors, all gs,
regardless of whether they were classified as prejudiced
or unprejudiced, gave social distance responses. In the
case of the least intimate behaviors, most gs gave
unprejudiced re5ponses. In the case of the behaviors
that were intermediate in intimacy, there was a clear
separation of the prejudiced from the unprejudiced gs,
as well as of the conventionally prejudiced from the
belief prejudiced gs.

In the case of intimate behaviors, Triandis'
arguments appear correct: in the case of nonintimate
behaviors, Rokeach's arguments appear correct: in the

case of behaviors intermediate in intimacy, both race

50

and belief are important. However, some gs behave more

as described by Triandis, and other gs more as described

by Rokeach. The first type, the conventionally prejudiced,
is distinguished from the second type, the belief prejudiced,
particularly on those behaviors that are intermediate in
intimacy because norms may play a smaller role in the case
of nonintimate behaviors. In this latter case, personality
and previous experience, as well as the total attitude
structure of the individual, may control most of the
variance of social rejection or acceptance.

According to Insko and Robinson (1967) there were,
however, at least two problems with Triandis and Davis'
(1965) study. First, the various items in several of the
factors appear fairly heterogeneous in implied intimacy
or social distance. (E.G., consider “admit as a tourist
in my country" and "accept as a close kin by marriage,"
both contained in Factor IV.)

Second, only one belief or attitude (favorability
toward civil rights) was manipulated. A more adequate
or thorough manipulation of belief possibly would have
produced a greater belief effect in Triandis and Davis'

(1965) college population.

51

Insko and Robinson (1967) attempted in two ways to
increase the generality of the evidence relating to the
belief versus race issue. First, the manipulation of
similarity-dissimilarity involved unselected beliefs as
well as beliefs about which of the gs perceived Negroes
as feeling strongly. In one condition.these latter
beliefs were selected on an individual §rbY<§ybaSiS: in
another condition,they were selected on a group basis so
as to represent the white conception or stereotype of
strongly held Negro beliefs.

Second, the study was conducted with southern junior
high school students. Since no one has previously
reported a test of Rokeach's theory with southern non-
college gs, the authors thought that it would be of some
interest to examine the race versus belief issue with gs
of this type. One scale Which Triandis and DaVid (1965)
interpreted as being more influenced by belief (semantic
differential assessment of attitude toward Negroes), one
scale influenced by both belief and race (Factor III), and
one scale more influenced by race (Factor IV) were used
for dependent variables.

The results indicated that both belief and race had

large significant effects on all three dependent variables

\.,

52

and that belief had a significant greater effect than did
race on the semantic differential assessment of attitude,
and that race had a nonsignificant greater effect than did
belief on the two factor scales. The Belief X Race
interactiOn on the two factor scales, however, indicates
that belief similarity is of somewhat greater importance
in determining reactions to white persons than to Negroes.

In addition, belief similarity was just as important
for beliefs that Negroes were perceived as holding strongly
as for unselected beliefs.

The semantic differential items obviously entail
minimal presence of norm-enforcing whites, while accepting
Negroes as close kin by marriage and eating with Negroes
involve both of these variables to a high degree. There
are, however, some difficulties with these variables as
ordering concepts. "Accept this person as an intimate
friend" would seem almost by definition to involve more
intimacy of contact than “eat with this persOn," and yet
"eat with this person“ has a more negative belief minus
race score. Perhaps this reflects the fact that traditional
southern culture tolerates friendship between Negroes and
whites, as long as Negroes stay in ”their place," but

does not tolerate eating with Negroes. This shows that

53‘

contact intimacy and presence of others may not be
completely adequate as operational criteria for speci-
fying the presence of norms. Since institutionalized
norms may override individual psychological propensities
to respond in terms of belief similarity, the results
infer that prejudicial reactions to race are still very
much a part of southern whites.

Pettigrew (1958) further has reported some data
indicating that conformity to norms has an effect on the
maintenance of prejudice in the south but not in the north.

Smith, Williams, & Willis (1967) confirmed Insko and
Robinson's (1967) findings. In the Smith et a1. (1967),
six samples (total N=307) of white and Negro gs from
northern, border, and southern states rates the friend-
ship acceptance of stimulus persons of specified race,
sex,and belief on 1 of 8 general or Negro-white issues.
For all samples except 1, belief cOngruence was more
important for acceptance than similarity of race: race,
in turn, was more important than similarity of sex. For
the southernmost white sample (Louisiana), race was the
most important factor by a slight margin due to the

institutionalized mores of the deep South, while sex

\F/

\ I

54

was again least important. Interactions were negligible,
except for a renegade effect observed in all three Negro
samples--members of the racial ingroup were consistently
penalized more for disagreeing than were members of the
outgroup. None of the samples exhibited "true“
discrimination (negative correlation between acceptance of
ingroup and outgroup), although the southern white sample
exhibited a significant tendency towards true racial
discrimination, while all three Negro samples showed
significant tendencies towards true belief discrimination.

McGrew (1969) also confirmed the Belief Congruency
Hypothesis. In his study, there were four independent
variables, each consisting of two levels: (a) Authoritarian
orientation (Left Authoritarian, liberal opinions vs.
Right Authoritarian, conservative opinions): (b) Sex
(Males vs. Females): (c) Debate Condition (Liberal vs.
Conservative): (d) Mode of Presentation (Television vs.
Audio-Only). The single dependent variable was the
extent to which gs agreed with either the liberal or
conservative position, as rated on a Semantic Differential-
Type Scale.

The major hypotheses were tested by means of two

debate situations which occurred between a Negro and

55

White male confederate. Eight issues were debated by
the speaker, four of which dealt with American foreign
policies and four with domestic policies.

It was hypothesized that Left and Right Authoritarians,
as measured by Rokeach's Opinionation Scale, would agree
with liberal and conservative positions respectively,
regardless of the ethnic membership or the person eSpousing
such views. Although the data supported Rokeach's
hypothesis that interpersonal attraction is a function of
attitudinal similarity, it was suggested that the degree
of intimacy and the social proscriptions governing it
limit the generality of the hypothesis.

If interpersonal attraction implies closeness
sanctioned by social custom, then the Belief Congruency
Hypothesis appears valid, but if such attraction implies
a relationship Which is socially taboo, then the Triandis
and Davis (1965) position appears valid.

The study by Willis and Bulatao (1967) suggests that
Triandis and Davis (1965) are right in a relative sense
but wrong in an absolute sense. In the Willis et a1.
(1967) study, the belief congruence theory of prejudice
was tested with friendship and marriage acceptance ratings

made by 160 Filipino college students. gs indicated opinions

56

on 10 issues and rated hypothetical friendship and marriage
candidates varying with regard to belief congruence
(agreement/disagreement with g) and ethnicity (Filipino,
Chinese, etc.). Anova revealed belief congruence to
account for over 20 times as much variance as ethnicity.
However, ethnicity too was significant in 17/20 instances,
with the remaining cases lending trend support. The
renegade effect was consistently observed and was sub-
stantially correlated (rho=.60, .64) with issue importance.
Generalization from these findings was, however,
severely limited by the fact that in all these studies
the social stimuli were “paper-and—pencil“ stimuli and
the discriminatory responses elicited were "paper—and-
pencil" responses. To overcome this limitation, Rokeach
and Mezei (1966) conducted three experiments in which gs
were given the opportunity to discriminate on the basis
of race or belief, or both, in real-life situations.
These experiments were all alike in basic design. A
naive g engaged four strangers, confederates of the
experimenter, in a group discussion about an important
or situationally relevant topic. Two of the confederates

were white and two were Negro. One white and one Negro

57

agreed with the gJ and one white and one Negro disagreed
with him. The g was then asked to state a preference for
two of the four confederates.

Two of these experiments again used college students,
but the third included as gs job applicants for positions
of janitor, laundry worker, and attendant in mental
hospitals. In all experiments,the choices, which were
made following a group discussion, were cognitively real
for the gs.

In each of the campus studies, 60 percent apparently
preferred partners differing from one another in both race
and belief. But, this was so of only 24 percent of the gs
in the field study: 60 percent in the field study chose two
partners with beliefs congruent with their own, one white
and one Negro.

The results also showed that (a) similarity of belief
is a considerably more frequent basis of choice than
dissimilarity of belief: (b) similarity of race is rarely
a basis of choice--considerably less often even than chance,
and no more frequently than dissimilarity of race: and
(c) similarity of belief is a considerably more frequent

basis of choice than similarity of race. Thus, the Belief

58

Congruence Principle was consistently the best predictor
of choices in all three experiments.

In summary, it appears that white gs like Blacks
who have many similar beliefs to themselves provided that
the Blacks do not share intimate behaviors with Whites,
especially where institutionalized norms prohibit such
interpersonal interactions. Most Whites who do not know
the beliefs of Blacks tended to judge Blacks negatively
and attributed dissimilar beliefs to the Blacks.

In our study, we predicted that the prejudice gs
would rate the Black Conservative more favorably than
the Black Radical: and, prejudiced gs would rate negatively
the Black Radical because the gs would feel that the Black

Radical holds dissimilar beliefs.

Statement 9; Problem

 

This study was conducted in order to evaluate the
effects of Race and Belief on interpersonal attraction.
In the Race Congruency Theory, gs are attracted to people
of the same Race. In the Belief Congruency Theory, gs
are attracted to people with similar Beliefs. In this

study, a Black and a White actor gave both a Conservative

59

Speech and a Radical Speech so that four conditions were
tested--the White Conservative, White Radical, Black
Conservative, and Black Radical. A Race X Belief inter-
action was predicted.

gs in each of the four conditions rated the actors
on an Interpersonal Check List. In order to find out how
favorably or unfavorably the gs rated each actor, comparison
ratings of the actors to other variables were examined.

If the gs were, indeed, prejudiced, they should prefer
the Conservative Speech to the Radical Speech, and the
White Race to the Black Race, and there should be a Speech
X Race interaction.

If the gs showed true racial discrimination, then
Race of actor should be significant regardless of Belief
of actor. Therefore, the gs should prefer the White Race.

If the gs showed true belief discrimination, then
Speech of actor should be significant regardless of Race
of actor. Therefore, gs should prefer the Conservative
Speech.

If the gs showed bOth true racial and belief
discrimination, then Race and Speech of actor should

predict a significant difference in preferences.

60

Therefore, the gs should prefer the White Conservative
followed by the White Radical, Black Conservative, and
Black Radical last.

In order to explore the possible effects of racial or
belief prejudice, the following hypotheses were tested:

Hypothesis I. gs will rate the Black Radical actor
most significantly different from the gs, themselves, on
the Interpersonal Check List Scores with Black Conservative,
White Radical, and White Conservative actors increasingly
like themselves in a race X Speech interaction.

Hypothesis II. gs will rate the Black actors more
significantly different from the gs, themselves, in the
Interpersonal Check List scores than the White actors.

Hypothesis III. gs will rate those who made a
Radical Speech more significantly different from themselves
than those Who made a Conservative Speech.

Hypothesis IV. The Black Radical actor will be seen
as significantly different from the gs' perception of the
average college student and the BC, WR, and WC as
increasingly closer to the gs' perception of the average
college student.

Hypothesis V. .§S will rate the Black Radical actor

viewing the gs as significantly different from the gs'

 

61

self—rating, i.e. the gs will view the BR as seeing the gs

as significantly different from the gs' perception of

themselves. The gs will also see the BC, WR, and WC as

increasingly perceiving them more like themselves.
Hypothesis VI. gs will rate the Black Radical actor

as significantly farther from the ideal-self of the gs

and the BC, WR, and WC as increasingly closer to the gs'

ideal-selves.

CHAPTER 2

METHOD

gubjects

The gs were 80 Caucasian males obtained from the
introductory psychology classes at Michigan State University
in April of 1969.2 The students received psychology credit

for their services.

 

1The data from this study were also used by Perlman
(1969) and Evans (1970) in analyses of the non-verbal and verbal
behaviors of the gs, respectively.

ng were only male Caucasians since the g_was not

interested, at this time, in the interaction effects
between a female g and a male accomplice.

62

63

Accomplices

 

Four male accomplices,3 two Black and two Caucasian
were hired from the theater department at Michigan State

University.

ggtting

A small room with a one way mirror, two chairs, and
a low table provided the experimental setting. The
chairs were soft backed and had arms. The chairs were
placed facing each other on a slight angle towards the

one way mirror approximately five feet apart.

.gesign

All accomplices learned both a Radical and Conservative
script. A copy of the scripts can be found in Appendix
A. One script presented a Radical vieWpoint about the

causes and solutions of racial disorders, the other script

 

3The two Black actors were Alan Smith and Donald
Trammel. The two Caucasian actors were Ray Price and Glen
Sussman. I would like to thank them all for their time
and splendid service.

64

was a Conservative viewpoint about the same topic. The
scripts were written so that each could be said by a Black
or Caucasian accomplice. The actors worked together in
learning the scripts and added pauses, “You know," “I
think,“ tone of voice, postural and gestural cues, etc.
As reported by the raters, in agreement with the E's
expectations, all scripts sounded spontaneous and almost
identical, no matter who was saying them.

gs were randomly assrgned to one of the two races of
accomplice, giving the Radical or Conservative viewpoint.
Therefore, there were four experimental conditions with
twenty gs to each condition. The experimental design is

diagrammed in Table 1.

TABLE 1

EXPERIMENTAL DES IGN

 

 

Viewpoint Race of Accomplice
Black Caucasian

 

 

Radical 20 20

Conservative 20 20

 

65

Instruction (see Appendix B) were read by one of two
Caucasian gs in a uniform manner so as to aVOid any test
effects.

The use of the selection of a number and letter to
determine the topic of discussion avoided giving the g
any insight into the true nature of the experiment. It
was feared that if the g was told directly that his
opinions to the causes and solutions of racial problems
were being studied, and if the g was in a dyad with a
Black accomplice, that the g may have correCtly surmised
that the Black g_is not a subject at all. A small box
containing 80 pieces of paper each listing a number and
letter was used for the supposed random selection of topic

and speaking order.

Procedure

That gs were told that the g was interested in studying
undergraduate opinions concerning social issues in the
United States. All accomplices were informally but neatly
dressed for the experimental sessions.

The accomplice and g were seated. The g did not know

that the accomplice was working for the E. Care was taken

66

to treat the accomplice at all times just as it he was any
other subject. For example, when reading instructions,
the E'lOOKed at both the g_and accomplice with equal
frequency. The g and accomplice were told that they were
to discuss one of several social issues the g had chosen.
One of them was asked to pick a piece of paper from the
small box. The g_then looked at a list of numbers and
told the g and accomplice that according to the letter
picked, they were to discuss the causes and solutions of
racial problems in the United States: specifically causes
and solutions of racial disorders in the big city, and
that according to the number picked, the accomplice was to
speak firSt.

The fig" (actually the accomplice) was asked to talk
for up to ten minutes (actually he talked for the length of
his script): then the real S talked for up to ten minutes
giving his opinions: finally they discussed their views
for an additional ten minutes. It was made clear that
the firSt time each spoke, that the other was not to
interrupt. To further insure that the accomplice's talk
did not appear prepared, the g and accomplice were given

about three minutes to think about what they wanted to say

fl.

.1 I. .1
. pr
r1

_
I.
Dr
.
\ ,
1 .
a.
. .
)
i a i L
.
u
L . ,
\l/
i It
1 . ,
«C . .
Ir
.1
-.
un. . .
._ .
x _
x
r)
L. l . _
a
_
uv )
a
—
I _ . FL I.
.I4 L
_
i
T.
c

67

before the accomplice began. If the experimental session
lasted over 35 minutes, the E re-entered the room and
informed the pair that their time was up. The g_was not
present in the room during the actual session.

‘When the discussion was over, each g was led into a
small office by the g, Another g_then gave instructions
to each g on how to fill out the Interpersonal Check List,
and he was always nearby if alg had any questions.

The Interpersonal Check List were then scored and,
the results of the comparative ratings are found in
Chapter 3.

The gs were asked to complete the Interpersonal
Check List under the following procedure:

In column 1, the g described himself. In column 2,
the g described his perfect or ideal self, how he would
like to be. In column 3, the g.described how he felt
about the actor. In column 4, the g described how he
feels the actor feels about himself (the actor). In
column 5, the g_described the average, typical college
student. In column 6, the g described how the actor felt
about his partner (the _s__).

Seven ratings were compared for discrepancy scores:

/‘\

\v/

68

'The (g) self' compared to 'The ideal—self'
'The (g)' compared to ‘How the g viewed the
actor'

'The ideal-self' compared to 'How the g viewed
the actor'

'How the g viewed the actor' compared to 'How
the g viewed the average college student'

'How the g rated how the actor viewed himself'
compared to 'How the g rated how the actor
viewed the g}

'The g's perception of himself' compared to
'How the g thinks that the actor viewed the g'
The pooled results of the summation discrepancy

scores

CHAPTER 3

RESULTS

The results of the discrepancy scores of the seven
comparative ratings are shown in the following Tables and
include the mean frequency of occurrence, standard

deviations, and mean totals:

69

70

Table 2

Discrepancy scores of the g.rating how the actor
viewed himself compared to the g's rating how the actor

viewed the g.

 

 

 

 

Behavior Speech Race Mean Totals
Comparison Black Caucasian of Speech
(4&6) Radical 37 .9543?) 27 .oo 32 .48

16.01=(S.D.) 15.52 15.76
Conservative 34.60 22.45 28.53
14.96 9.79 12.37
Mean Totals
of Race 36.28 24.73
15.49 12.65

 

An analysis of variance conducted on the data in Table

2 indicated tht the effect of Race reached statistical

significance (F=12 .41, p < .01).

See Appendix C for a

complete analysis of variance for the data in Table 2.

71

Table 3

Discrepancy scores of the ideal-self compared to

how the g Viewed the actor.

 

 

 

 

Behavior Speech Race Mean Totals
Comparison . Black Caucasian of Speech
(2&3) Radical 40.50432) 34.65 37.58

12.17=(S.D.) 12.48 12.32
Conservative 34.60 40.45 37.53
9.90 13.02 12.74

Mean Totals
of Race 37.55 37.55
11.03 12.75

 

An analysis of the variance conducted on the data in
Table 3 indicated that the interaction of Race X Speech
reached statistical significance (F=4.55, p<f.05). See
Appendix C for a complete analysis of variance for the data
in Table 3.

This interaction was explored through tests of simple
effects. These tests revealed no specific significant

simple effects comparisons.

F\

72

Table 4

Discrepancy scores of the_g comparing himself to the

ideal-self.

 

 

 

 

Behavior Speech Race Mean Totals
Comparison Black Caucasian of Speech
(1&2) Radical 3o.75=(§) 29.05 29.90

12.13=(S.D.) 18.29 15.21
Conservative 27.35 32.90 30.13
13.37 14.37 13.87
Mean Totals
of Race 29.05 30.97
12.75 16.33

 

An analysis of variance conducted on the data in Table

4 indicated that no results reached statistical significance.

See Appendix C for a complete analysis of variance for the

data in Table 4.

/'\

73

Table 5

Discrepancy scores of the g_rating himself compared

to the g rating the actor.

 

 

f.

 

 

Behavior Speech Race Mean Totals
Comparison BlaCk Caucasian of Speech
(1&3) Radical 43.20-(E) 36.55 39.88

10.25=(S.D.) 13.46 11.85
Conservative 39.45 43.25 41.35
14.72 11.63 13.18
Mean Totals
of Race 41.33 39.90
12.49 12.55

 

An analysis of variance conducted on the data in Table

5 indicated that no results reached statistical significance.

See Appendix C for a complete analysis of variance for the

date in Table 5.

74

Table 6

Discrepancy scores of the g rating the actor compared

to the g rating the average college student.

 

 

 

 

Behavior Speech Race Mean Totals
Comparison BlaCk Caucasian of Speech
(3&5) Radical 38.70::(32) 33.75 36.23

13.58=(S.D.) 15.28 14.43
Conservative 36.75 32.85 34.80
12.50 16.47 14.48
Mean Totals
of Race 37.73 33.30
13.04 15.87

 

An analysis of variance conducted on the data in Table

6 indicated that no results reached statistical significance.

See Appendix C for a complete analysis of variance for the

data in Table 6.

75

Table 7

Discrepancy scores of the g rating himself compared to

the_g rating how the actor viewed the g,

 

 

 

= “=1:
Behavior , Speech Race Mean Totals
Comparison Black Caucasian of Speech
(1&6) Radical 40.20432) 30.70 35.45
ll.7l=(S.D.) 12.92 12.32
Conservative 36.80 36.05 36.43
13.77 11.74 12.75
Mean Totals
of Race 38.50 33.38
12.74 12.33

 

An analysis of variance conducted on the data in Table
7 indicated that no results reached statistical significance.
See Appendix C for a complete analysis of variance for the

date in Table 7.

76

Table 8

The results of the summation discrepancy scores from

columns 1, 2, 3, 4, 5, and 6.

 

‘
-vv

 

 

Behavior Speech Race Mean Totals
Comparison Black Caucasian of Speech
(pooled) Radical 231.3043?) , 191.70 211.50

55.03=(S.D.) 71.18 63.10
Conservative 209.55 207.95 208.75
63.93 56.28 60.11

Mean Totals
of Race 220.43 199.83
59.48 63.73

 

An analysis of variance conducted on the data in Table
8 indicated that no results reached statistical significance.
See Appendix C for a complete analySis of variance for the

data in Table 8.

CHAPTER 4

DISCUSSION

The results of the effects of the race variable will
be discussed in decreasing order with the significant

finding discussed first.

Results of Race Variable

 

How the §§_Rated How the Actors Viewed Themselves Compared

 

to How the gs Rated How the Actors Viewed the SS

 

Only on one Comparative Rating (4&6) were the results
of Race statistically significant (p<:.01).

When the SS compared how they felt the actors viewed
themselves compared to the gs rating how they felt the
actors viewed the §s, a Race effect was noted. The gs
rated the White actors regardless of Speech as less

different from themselves as compared to Black actors.

77

78}

This can be interpreted to mean that the gs projected
to the actors the gs' prejudiced feelings. The gs might
not readily admit differences in character and personality
traits between themselves and Blacks. But, they could be
showing indirect hostility and differences by projecting
these negative feelings to the Black actors.. In this way,
the gs could be relieving or denying feelings of guilt,
hatred, or prejudice by rating the Black actors as rating
them, the gs, in a different way.

The gs made no distinction between the Black Radical
and the Black Conservative in the gs' ratings. This could
mean that regardless of how well a Black becomes Americanized
to the beliefs, values, and cultural mores of the 'White
Society' there might still be indirect mistrust and
hostility projected towards him. This suggests a caste
system that is difficult to break.

It appears that White gs did not in this situation
discriminate by Race. Although the White §s may feel that
Blacks are hostile to them: they, nevertheless, did not
perceive the Blacks in this situation as very different
from themselves. This would lead us to infer that the gs
in this experiment were not as prejudiced as we believed

them to be.

 

79

Other alternatives would be that this was an
inappropriate setting for testing the expression of
prejudicial attitudes because of the lack of institu-
tionalized norms supporting segregation and
discrimination since most university students are free
to associate with whomever they please without harasSment.
Or the inter-racial discussion was not in a sensitive
area for the gs. They were not threatened by the
statements made by the actors especially the statements
of future things to come. The gs may have perceived the
Black actors as college students and not as street Blacks
from the ghetto. Perhaps, the gs rated the Black actors
as the best representatives of their race. These gs
may have trusted the actors' judgment and did not fear
that these Blacks would destroy America. Or, there was
not sufficient time in the experiment for gs to become
more aroused and show their "true" feelings.

Perhaps, we would get significant results if we used
a naturalistic field setting instead of a closed door
discussion setting with the random pOpulation of both
sexes in an intimate racial conflict skit with the Black

Radical wearing Afro attire.

 

80

How the gs Rated the Actors Compared :3 the gs' Ideal-Self

 

 

Only on one Comparative Rating, (2&3) were the results
of the interaction of Race X Speech significant (p<(.05).

When the gs compared the actors to their ideal-self,

a Race X Speech interaction was noted. The gs viewed the
Black Radical and the White Conservative as being least
like the ideal—self of the gs and rated the White Radical
and the Black Conservative as being most like the ideal—
self of the gs. This interaction was explored through
tests of simple effects. These tests revealed no specific
significant simple effects comparisons.

The gs idealized most the Black Conservative and the
White Radical which at first may seem paradoxical. Because
these two types are the exact Opposites. Or are they?
Perhaps the White Radical Speech is equal in effect to a
Black Conservative Speech.

It appears that the gs liked the Black Conservative
because he was not hostile to them and did not wish to
destroy America.

The gs appeared to admire the White Radical because
he represented to them the new American who wanted to rid

America of racism and poverty. The gs seemed to deny or

rfl

81

disregard some of the vicious attacks on the destruction
of the White Race as being either stimulating rhetoric
or as being directed strictly at White racists.

The gs did not perceive the Black Conservative or
the White Radical as a threat: and, consequently, admired
what they exemplified.

On the other hand, the gs least admired the Black
Radicals and the White Conservatives who again seemed to
be opposites. Perhaps, the gs' low identification with
the Black Radical could be due to the gs feeling that the
militant Blacks might actually seek revenge for past and
present injustices. The gs could also fear revolution
and a new social order in which they, the gs might be
oppressed.

The gs identified with the White Conservative least,
possibly because to the gs the reactionary White Con—
servative represented blatant prejudice, repression, and
order without just law.

It appears that the Race effect is important when
gs did not know the true feelings of the Blacks. The gs
attributed hostile feelings to Blacks toward themselves,

the gs. This is congruent with the findings of Stein,

82

Hardyck, & Smith (1965). This also supports Triandis's
position of Race being more important than Belief.

But, when Race and Belief are both known, an
interaction occurs. A true Belief discrimination (Smith,
Williams, & Willis, 1967) would predict that Conservative
gs would prefer White Conservative and Black Conservative
actors while Radical gs would prefer White Radical and
Black Radical actors. Our results did not confirm this
prediction. Therefore, for this sample using this kind
of experimental procedure Rokeach's Theory of Belief
Congruency superordinate to Race Congruency was not
supported. It appears that Speeches are only meaningfully
interpreted not by their content but rather by the Race
of the speaker.

The White Radical and the Black Conservative were
interpreted as being non-threating to the gs while the
White Conservative and the Black Radical were perceived
as being somewhat more threatening to them as well as
possibly destructive to the American ideals and the

American way of life.

83

 

Results 2: gpeech Variable

None of the results concerning the Speech effect
were found to reach statistical significance. The mean
discrepancy scores of the Conservative and Radical
Speeches in the seven Comparative Ratings only varied

between zero and five points in each of the seven ratings.

CHAPTER 5

SUMMARY

This study was conducted in order to evaluate the
effects of Race and Belief on interpersonal attraction.

In an experiment purporting to study student opinion
on social issues, a Black or White accomplice presented
either a prepared "Radical“ or "Conservative" viewpoint
to the causes and solutions of big city racial problems
in the United States to White male gs who were all from
introductory psychology classes. The 80 gs were all
randomly assigned to one of four experiment conditions:
Black Radical, Black Conservative, White Radical, and
White Conservative. The scripts of the accomplices
sounded spontaneous and were rehearsed by all the
accomplices so that their vocal inflections and body
movements were similar. After the discussion, the gs
in each of the four conditions rated themselves and the

actors on an Interpersonal Check List test. In order to

84

85

find out how favorably or unfavorably the gs rated each

actor,

comparison ratings of the actors to other

variables were examined.

Seven comparative ratings were rated for mean

discrepancy scores. They were:

1.

20

'The Qg) self' compared to 'The ideal-self'
'The (g)' compared to 'How the g viewed the
actor'

'The ideal-self' compared to 'How the g viewed
the actor'

'How the g viewed the actor' compared to 'How
the g.viewed the average college student'

'How the g rated how the actor viewed himself'
compared to 'How the g rated how the actor
viewed the g}

'The gfs perception of himself' compared to
'How the g thinks that the actor viewed the g}
The pooled results of the summation discrepancy

scores

If the gs were, indeed, prejudiced, they should have

preferred the Conservative Speech to the Radical Speech,

and the White Race to the Black Race, and there should

86

be a Speech X Race interaction in which the gs most
preferred the White Conservative, followed by the White
Radical followed by the Black Conservative and, lastly,
the Black Radical as being most different.

Analyses of variance for the comparative ratings
were computed. Only on one comparative rating--when the
.gs compared how they felt the actors viewed themselves
compared to the gs rating how they felt the actors viewed
the gs—-were the results due to the Race of the accomplice
statistically significant. Contrary to expectations, no
results of the Speech variable reached statistical
significance. Thus, Radical or Conservative Speech had
no effect on the gs viewing the actors. Only on one
comparative rating--when the gs compared the actors to
their ideal-self--were the results due to the interaction
of Race X Speech of the accomplice statiStically significant.
The gs identified most with the Black Conservatives followed
closely by White Radicals and least with White Conservatives
and Black Radicals as ideal people.

The hypothesis that the Black Radical would be seen
as most significantly different from the gs' perception
of themselves followed by the Black Conservative, White

Radical, and White Conservative was not supported.

87

The pooled results of the gs showed that the White
Radical not the White Conservative was most preferred
followed by the White Conservative, Black Conservative,
and, lastly, the Black Radical. However, these
differences between rankings did not reach statistical
significance. One possible interpretation is that the
gs rated the White Radicals as how the gs are or would

like to become.

BIBLIOGRAPHY

BIBLIOGRAPHY

Ansari, Anwar. A study of the relation between group
stereotypes and social distance. Journal 9;
Educational Psychology, Baroda, 1956, $3, 28-35.

Berg, Kenneth. Ethnic attitudes and agreement with a
Negro person. Journal 9; Personality and
Social Psychology, 1966, g, (2), 215-220.

Berkun, M.,& Meeland, T. Sociometric effects of race
. and combat performance. Sociometry, 1958,
.21, 145—149.

 

Best, W. H.,& Sohner, C. P. Social distance and politics:
Some selected relationships. Sociology Social
Resources, 1959, 40, 339-342.

Bogardus, E. S. Immigration and Race Attitudes. Boston:
Heath, 1928.

Bogardus, E. E. A social Distance scale. Sociology
gocial Resources, 1933, 17, 265-271.

 

Bonney, M. E. A sociometric study of the relationship
of some factors to mutual friendships on the
elementary, secondary, and college levels.
Sociometry, 1949, 2, 21-47.

Brophy, I. N. The luxury of anti-Negro prejudice Public
Opinion Quarterly, 1956, 2, 456-466.

Byrne, D. Attitudes and attraction. In L. Berkowitz

(Ed.), Advances ig_Experimental Social Psycho-
logy. New York: Academic Press, 1969, g, 35-89.

88

\/

Byrne,

Byrne,

Byrne,

Byrne,

Byrne,

Byrne,

Byrne,

Byrne,

Byrne,

Byrne,

89

D. Interpersonal attraction and attitude similarity.
Journal g; Abnormal g_Socia1 Psychology, 1961a, ggJ
713-715.

D. Interpersonal attraction as a function of
affiliation need and attitude similarity. Human
Relations, 1961b, ;, 283—289.

D. Response to attitude similarity-dissimilarity
as a function of affiliation need. Journal 9§_
Personality, 1962, 39, 164-177.

D. The influence of propinquity and Opportunities
for interaction on classroom relationships. Human
Relations, 1961c, 13, 63-69.

 

D.,& Buehler, J. A note on the influence of
propinquity upon acquaintanceships. Journal 9;
Abnormal Social Psychology, 1955, §;, 147-148.

D.,& Clore, G.,& Worchel, P. Effect of economic
similarity-dissimilarity on interpersonal attraction.
Journal 2£_Personality and Social Psychology, 1966,
g, (2), 220-224.

D.,& Clore, G. Predicting interpersonal attraction
toward strangers presented in three different
stimulus modes. Psychonomic Science, 1966, (4),
239—240.

D.,& Griffitt, W. A developmental investigation of the
law of attraction. Journal 9f Personality g Social
Psychology, 1966, g, 699-702.

 

D.,& Griffitt, W.,& Reeves, K. Attitude similarity-
dissimilarity and attraction: generality beyond the
college sophomore. Journal_g§ Social Psychology,
1969, 12, 155-161.

D.,& McGraw, C. Interpersonal attraction toward
Negroes. Human Relations, 1964, i1, 201-213.

9O

Byrne, D.,& Nelson, D. Attraction as a linear function
of proportion of positive reinforcements.
Journal_g§ Personality g Social Psychology, 1965,
_1, 659—663.

 

 

Byrne, D.,& Wong, T. Racial prejudice, interpersonal
attraction, and assumed dissimilarity of attitudes.
Journal 9; Abnormal g_Social Psychology, 1962,

.gg, 246-253. ‘

 

 

Caplow, T. & Forman, R. Neighborhood interaction in a
homogeneous community. American Sociological
Review, 1950,_l§, 357-366.

 

Dahlke, H. Determinants of sociometric relations among
children in the elementary school. Sociometry,
1953,_;§, 327-338.

 

Davis, E.,& Grobstein, N. Multi-mode factor analysis of
interpersonal perceptions, ONR technical report
(in preparation) 1965.

Deutsch, M., & Collins, M. Interracial Housing.
Minneapolis: University of Minnesota Press, 1951.

 

Evans, A. “Expressions of Prejudice through Verbal
Behavior." Unpublished Master's thesis, Michigan
State University, 1970.

Evans, C.,& Wilson, M. Friendship choices of university
women students. Educational Psychology
Measurements, gJ 307-312.

 

 

Festinger, L. Informal social communication. Psychological
Review, 1950, gZJ 271-282.

 

Festinger, L. A theory of social comparison processes.
Human Relations, 1954, ZJ 117-140.

 

Festinger, L.,& Schachter, S. & Back, K. Social Iressures lg
Informal Groups. New York: Harper, 1950.

 

 

Golightly, G.,& Byrne, D. Attitude statements as positive
and negative reinforcements. Science, 1964, 146,
798-799.

91

Gundlach, R. Effects of on-the-job experiences with
Negroes upon racial attitudes of white workers
in union shops. Psychological Rsport, 1956,
2' 67-77.

Heider, F. The Psychology 9; Interpersonal Relations.
New York: Wiley, 1958.

Hill, M. Some problems of social distance in intergroup
relations. In M. Sherif & M. O. Wilson (Eds.),
Group Relations g£_the Crossroads. New York:
Harper, 1953, 285-307.

Insko, G.,& Robinson, J. Belief similarity versus race
as determinants of reactions to Negroes by
southern White adolescents: A further'test of
Rokeach's theory. Journal 3;: Personality g
Social Psychology, 1967, 1, 216—221.

Hunt, C. Social distance in the Phillippines.
Sociological Social Resources, 1956, 39, 253-260.

Jones, E. & Daugherty, B. Political orientation and the
perceptual effects of an anticipated interaction.
Journal 9£_Abnormal g_Social Psychology, 1959,
s2, 340-349. ~r

Katz, D.,& Braly, K. Verbal stereotypes and racial
prejudice. In Eleanor E. Maccoby, T. M. Newcomb,
.& E. Hartley (Eds.), Readings is Social Psychology.
New York: Holt, 1958, 40-46.

Kirsch, A. Social distance and voting behavior related
to N variables, Dissertation Abstract, 1957,
AZJ 2673. .

 

Kleiner, R. The effects of threat reduction upon
interpersonal attractiveness. Journal g;
Personality, 1960, gs, 145-155.

 

Koch, Helen. The social distance between certain racial,
nationality, and skin-pigmentation groups in
selected populations of American school children.
Journal sf Genetic Psychology, 1946, gs, 63-95.

92

Krauss, R. Structural and attitudinal factors in
interpersonal bargaining. Journal 9£_Experimental
g_SOcial Psyghology, 1966, 2, 42-55.

Longmore, T. A matrix approach to the analysis of rank
and status in a community in Peru. Sociometry,
1948,.li, 192-206.

Loomis, C. Political and occupational cleavages in a
Hanoverian village, Germany: A sociometric study.
gociometry, 1946, 2, 316-333.

 

Lundberg, G. & Beazley, V. 'Consciousness of kind' in a
college population. ggciometgy, 1948,.11, 59-74.

Lundberg, G.,& Hertzler, V.,& Dickson, L. Attraction patterns
in a university. Sociometry, 1949, 13, 158-159.

Lundberg, G. & Steele, M. Special attraction-patterns
in a village. Sociometry, l938,_1, 375-419.

Maisonneure, J., Palmade, G. & Fourment, C. Selective
choices and propinquity. Sociometry, 1952, ig,
135-140.

Malof, M. & Lott, A. Ethnocentrism and the acceptance of
Negro support in a group pressure situation.
Journa1_g£_Abnorma1 g_Socia1 Psychology, 1962,
Igg, 254-258.

Mann, J. The influence of racial prejudice on sociometric
choices and perceptions. Sociometry, 1958, 2;,
150-158.

Mann, J. The effect of inter-racial contact on sociometric
choices and perceptions. Journal 9; Social Psychology,
1959, 50, 143-152. '

Melikian, L.,& Prothro, E. Social distance and social
change in the Near East. Sociolggical,§gcia1
Resources, 1952, gZJ 3-11.

93

Morris, C. Varieties 9; Human Value. Chicago:
University of Chicago Press, 1956.

Morris, C.,& Jones, L. Value scales and dimensions.
Journal 9§_Abnormal g_Social Psychology, 1955,
g;, 523-539.

McGrew, J. The cognitive consistency of left and
right authoritarians: A test of Rokeach's
“Belief Congruency“ Hypothesis. Journal 2;
Social Psychology, 1969, 12, 227-234.

Newcomb, T. An approach to the study of communicative
acts. Psychological Review, 1953, s9, 393-404.

Newcomb, T. The prediction of interpersonal attraction.
American Psychologist, 1956, 11, 575-586.

Osgood, C., Suci, G.,& Tannenbaum, P. The Measurement

2; Meaning. Urbana: University of Illinois
Press, 1957. . ~

Palmore, E. The introduction of Negroes into white
departments. Human Organization, 1955, 13,

Pepitone, A.,& Kleiner, R. The effect of threat and
frustration on group cohesiveness, Journal gg
Abnormal g.Socia1 Psychology, 1957, 54, 192-199.

Perlman, B. "Nonverbal Expressions of Prejudice."
Unpublished Master's thesis, Michigan State
University, 1969.

Pettigrew, T. Personality and sociocultural factors in
intergroup attitudes: A cross-national comparison.
Journal 9; Conflict Resolutiony 1958, g, 29-42.

Precker, J. Similarity of valuings as a factor in
selection. Journal of Abnormal & Social

Psychology, 1952, 31, 406-414.

94

Prothro, E.,& Miles, 0. Social distance in the deep
South as measured by a revised Borgardus scale.
Journal 2: Social Psychology, 1953, glJ 171-194.

Richardson, H. Community of values as a factor in
friendships of college and adult women. Journal
2g_Social Psychology, 1940, $1, 303-312.

Rickard, T., Triandis, H., & Patterson, C. Indices of
employer prejudice toward disabled applicants.
Journal gg_Applied Psychology, 1963, 51, 52-55.

Rokeach, M. Belief versus race as determinants of social
. distance: Comment on Triandis' paper. Journal
2; Abnormal g_Social Psychology, 1961, sg, 187-188.

ROkeach, M.,& Mezei, L. Race and shared belief as factors
in social choice. Science, 1966, 151, 167-172.

Rokeach, M.,& Rothman, G. The principle of belief
congruence and the congruity principle as models
of cognitive interaction. Psychological Review,
1965, 22, 128-142.

Rokeach, M., Smith, P.,& Evans, R. Two kinds of prejudice
or one? In M. R0keach, The Open and Closed Mind.
New York: Basic Books, 1960, 132-168.

Sartain, A.,& Bell, H. An evaluation of the Bogardus scale
' of social distance by the method of equal appearing
intervals. Journal 9; Social Psychology, 1949,
.22, 85—91.

Smith, A. Similarity of values and its relation to
acceptance and the projection of similarity.
Journal 9; Psychology, 1957, gs, 251-260.

Smith, F. An experiment in modifying attitudes toward
the Negro. Teachers' College Controlled Education.
1943, No. 887.

Smith, C., Williams, L.,& Willis, R.‘ Race, sex, and belief
as determinants of friendship acceptance. Journal
2£_Personalitypg_Social Psychology, 1967, s, 127-137.

95

Stein, D., Hardyck, J.,& Smith, M. Race and belief: An
open and shut case. Journal 9; Personality g
Social Psychology, 1965, ll 281-289.

 

Triandis, H. -A note on Rokeach's Theory of Prejudice.
Journal 9§_Abnormal g_Socia1 Psychology, 1961,
Q3, 184-186.

 

 

Triandis, H. Factors affecting employee selection in
two cultures. Journal g£_Applied Psychology,
1963, 31, 89—96.

Triandis, H. Cultural influences upon cognitive
processes. In L. Berkowitz (Ed.), Advances 13
Experimental Social Psychology. New York:
Academic Press, 1964, 1-48, (a).

 

 

Triandis, H. Exploratory factor analyses of the
behavioral component of social attitudes.
Journal 9£_Abnormal g_Social Psychology, 1964,
gs, 420-430, (b).

 

Triandis, H.,& Davis, E. Race and belief as determinants
of behavioral intentions. Journal gg Personality
g.Socia1 Psychology, 1965, gJ 715-725.

 

 

Triandis, H., Davis, E.,& Takezawa, S. Some determinants
of social distance among American, German, and
Japanese students. Journal 9: Personalitng
Social Psychology, 1965, 3, 540-551.

 

 

Triandis, H., Fishbein, M.,& Hall, E. Person perception
among American and Indian students. Technical
Report No. 15, 1964, University of Illinois,
Contract Nr 177-472, Nonr-1834 (36), Office of
Naval Research.

Triandis, H., Loh, W.,& Levin, L. A race, status,
quality of spoken English, and opinions about
civil rights as determinants of interpersonal
attitudes. Journal 9; Personality g gpcial
Psychology, 1966, s, 468—472.

 

 

96

Triandis, H.,& Triandis, L. Race, social class, religion,
and nationality as determinants of social distance.
Journal 9; Abnormal g_Socia1 Psychology, 1960, g;,
110-118.

Triandis, H.,& Triandis, L. A cross-cultural study of
social distance. Psychological Monographs, 1962,
lgj (21, Whole No. 540).

 

United States war Department, Information and Education
Division. Opinions about Negro infantry platoons
in white companies of seven divisions. In T. M.
Newcomb & E. L. Hartley (Eds.), Readings is Social
Psychology. New York: Holt, 1947, 542-546.

 

Whyte,‘W. The Organization Man. New York: Simon &
Schuster, 1956.

Wilkinson, F. Social distance between occupations.
Sociological Social Resources, 1929, lg, 234-244.

Willis, R., & Bulatao, R. Belief and ethnicity as
determinants of friendship and marriage acceptance
in the Phillippines. American Psychologist,

1967, gs, 539 (Abstract.)

Wilner, D., Walkley, R.,& Cook, S. Residential proximity
and intergroup relations in public housing
projects. Journal g£_Socia1 Issues, 1952,_§, 45-69.

Winslow, C. A study of the extent of agreement between
friends' opinions and their ability to estimate
the opinions of each other. Journal gg Social
Psychology, 1937, s, 433-442.

Wong, T. The effect of attitude similarity and prejudice
on interpersonal evaluation and attraction.
Unpublished master's thesis, University of Texas, 1961.

Zander, A, & Havelin, A. Social comparison and interpersonal
attraction. gpman Relations, 1960,_1g, 21-32.

 

Zimbardo, P., & Formica, R. Emotional comparison and self-
esteem as determinants of affiliation. Journa1_g§
Personality, 1963, s1, 141-162.

APPENDICES

APPENDIX A

Scripts

APPENDIX A

SCRIPTS

Conservative Speech

O.K. I think that one of the main causes of racial
disorders in the cities is a communist conspiracy.
Communistsorganized the riots so we would look bad to
other countries. That's why Stokely Carmichael is always
in Cuba and North Viet Nam. There wouldn't be any riots
if the policemen would just enforce the laws they have.
The problem is all the pointy-head intellectuals who have
raised such a big fuss that the officers are afraid to
enforce the law, because when they do, they get in
trouble themselves. Now, what kind of country is this
where cops get in trouble for enforcing the laws? Breaking
the law by anyone must not be tolerated, because When laws
are allowed to be broken, the victims are the peaceful,
law abiding citizens, and the Negro in the past, I think,
has not been a responsible citizen. I think he's rioted

and looted and burned and deserves much of the punishment

97

98

he's gotten. Now I know the white man in the past
hasn't been any angel, either, but times have changed and
the Negro shouldn't use past grievances as an excuse to
break the laws. Now maybe one of the reasons why
integration--at least pushy integration has not worked
is because there is such a cultural difference between
the races. Negroes have their own world, and I think
most Negroes would want to stay to themselves. They have
their own world and aren't interested in white values and
ideals. All they really want is to have a few middle-
class comforts—-a good car, a T.V., you know. And there
are plenty of jobs available for Negroes who want to work.
The middle class Negro works--he doesn't riot. He
stays home and is a good citizen. I think the people
who riot are the ones who haven't worked their way up.
They just want to get the same material things that the
hard—working members of the middle class have gotten for
themselves. The large and growing Negro middle class is
proof that all the Negro has to do is work as hard as
his white counterpart and he will get what he wants. The
trouble is coming from those who don't give a damn about

going through the usual channels to achieve economic and

99

personal success. The ghetto Negro is an easy-going
hedonist who wants to join the hard-working class, but
unless he changes his values, he's never going to get
in: and rioting is going to hurt his goals. In fact,
I'm not sure he wants to get in.

Another trouble is, everybody told the Negro that if
he'd just make a little trouble, the government would
give him everything he wants. Now, that's like giving
candy to a little boy for being bad rather than rewarding
him for being good. And the government, I think, has
encouraged riots, because every time the Negroes burn the
town down, the government gives them a nice new one. Now
the Negro thinks the solution is to destroy what he has
so that the government will give him better. Why can't
the lower class Negro work like everybody else to integrate
hbmself. Many minority groups have felt persecuted by
the main stream American culture. As soon as they worked
a little bit and got ahead, it's funny how they stopped
feeling victimized. And I think the Negro who really wants
to will do the same thing. ,The middle class Negro--he
doesn't feel victimized and go out in the street and riot

and loot, because he's worked hard, and he has accomplished

100

something--and he's integrated. And the solution to~
racial disorders is to enforce the laws, stop the rioting
and stop giving rewards to law breakers. Stop molly-
coddling people who don't appreciate this great country
of ours. And find the Commies who are stirring up all
this trouble, and punish them. Within the law, of course.
And we, we have to support the police, because if the
police are made powerless, who's going to defend us from
this lawlessness. We've got communists everyWhere. In
fact, a lot of Negroes wouldn't be rioting and looting if
it wasn't for Communists. Most of them are pretty
satisfied right where they are--on welfare or out in the
streets and they weren't complaining until the communists
came along.

The Negro must learn the white man's way of achieving
his aims. There just aren't any shortcuts. And trying to
take shortcuts results in the lawlessness which we are
witnessing today. All the Negro has to do is follow the
lead of his more industrious black brothers. It can be
done. It's the American way. And a lot of even the Black
leadership is doing it. Look at the way they publish books.

They're getting royalties from these books. So, you know

101

they're being industrious and earning their way. And
the people who promise the pot of gold to the Negro will
have to be silenced for they are adding fuel to the
flames. And I think we must be patient. Integration
takes time. The Negro must learn to postpone immediate
gratification for future goals. It will take time for
the Negro to fit in, and we have to wait and keep this

country in one piece while we wait.

102

Black Radical

First of all, Whitey's a racist. Now, not all whites
are Whiteys, but all Whiteys are racists. Whites kill
blacks. When they can't kill blacks directly, they do it
legally, economically, and socially. Racists put blacks
in the slums, charge higher prices for crummy goods, charge
ridiculous rent, rape our women, pay us less at work, hire
blacks last and fire us first. Racism, racism against the
working black ranges through unions, colleges, management,
welfare, government, and everywhere.

Now politically, blacks have about as many rights as
before the Civil War. Black communities are run by white
carpetbaggers, right now--today. Legislation is loOpholed
so all the laws sound just great on paper, but Whitey knows
they don't mean shit--he just has to work a little harder
to fuck the black man. And the police--he11 we all know
the police are pigs. Capital punishment is real handy
for eliminating blacks, and the Whitey generals in the
Armed Forces always try to use up the blood of the black

man so he doesn't lose any of his precious White boys. Of

course, the situation doesn't improve. You know, blacks

103

really can't vote--they can't even register down South and
everywhere else everybody screws up the districts or throws
the black vote out.

And then there's the social structure. The middle
class and its careful choice of friends, like, you know,
frats, country clubs, and all that crap. Of course, now,
every party has to have its black couple--the super-nigger
type--sort of like a badge saying, "I like Negroes." You
know, “I once had a friend who was Negro." Whitey's
getting real subtle, real subtle, but he's still a racist.
He knows it, we know it, so let's drop the act, man.

Whitey has seen to it that the blacks have no past.

We got lots of "White Knights“ but what about some black
heroes. You know, there have been lots of them. You
know--or didn't you know? Like Jimmy Brown, Stokely
Carmichael, Leroi Jones. You've heard of them? Blacks
have a history. Black children are tired of learning about
white Dick and Janes doing their thing in nice little white
houses when all they see is a world which is cruel, hostile
and filled with empty promises.

Now, what to do about this situation? As I see it,

get rid of Whitey. Just get him clear the hell out of here.

104

If it isn't done soon and peacefully, then someone is going
to get wiped off the face of the earth, and it ain't going
to be the black man. It's going to be good-bye Whitey--
George Wallace, Daley, cops, college presidents, the
establishment, and even some negroes. You know, some
Negroes look black on the outside but they're really Whiteys
on the inside. Blacks must be given power--power to run
their own businesses, schools, apartments, towns, and
governments. And some laws with guts have to be assessed--
not this messing around with laws aimed at Tokenism but

at real, relevant problems. Laws are needed to protect the
black man from Whitey and to protect Whitey from himself.
And then there's all this shit about law and order--like
old J. Edgar saying justice is only incidental to law and
order and Nixon wanting to jail people before they commit

a crime--we know WhO is going to get screwed by that kind
of a deal. Economically, all the government has to do is
to spend a little less money on wars and stop screwing the
black man and give us some credits for a JJTt of blood,
sweat, and grief. The white man owes us more than he

could ever pay for in a million years. ‘Whitey owes it and

we're going to collect from those who wrenched us from our

105

homeland, enslaved us, lynched us, beat and tortured us,
humiliated us, stripped away our heritage, our pride, our
dignity, our humanness, and he's trying to keep us slaves,
even today--trying to coop us up into little stinking
ghettos sitting with the rats underneath the freeway. The
unions have to be opened up, equal hiring laws have to be
enforced, the welfare system has to be reorganized. There's
got to be some black controlled businesses and government
for black communities and schools. We blacks must have
real power over our own real problems, not this piddling
shit we've been getting. And finally, Whitey's got to
learn, one way or the other, that the black man is going
to change this country, and if Whitey resists, he's going

to have one hell or a fight on his hands.

106

White Radical

 

First of all, Whitey's a racist. Now, not all whites
are‘Whiteys, but all Whiteys are racists. Whites kill
blacks. When they can't kill blacks directly, they do it
legally, economically, and socially. Racists put blacks
in the slums, charge higher prices for crummy goods,
charge ridiculous rent, rape black women, pay less at work,
hire blacks last and fire them first. Racism, racism
against the working black ranges through unions, colleges,
management, welfare, government, and everywhere.

Now politically, blacks have about as many rights as
before the Civil War, Black communities are run by white
carpetbaggers, right now--today. Legislation is loopholed
so all the laws sound just great on paper, but Whitey knows
they don't mean shit--he just has to work a little harder
to fuck the black man. And the police--hell, we all know
the police are pigs. Capital punishment is real handy
for eliminating blacks, and the Whitey generals in the
Armed Forces always try to use up the blood of the black
man so he doesn't lose any of his precious white boys.

Of course, the situation doesn't improve. You know, blacks

107

really can't vote-—they can't even register down South
and everywhere else everybody screws up the districts
or throws the black vote out.

And then there's the social structure. The middle
class and its careful choice of friends, like, you know,
frats, country clubs, and all that crap. Of course, now,
every party has to have its black couple-—the super-nigger
type--sort of like a badge saying, “I like Negroes." .You
know, "I once had a friend who was Negro." Whitey's
getting real subtle, real subtle--but he's still a racist.
He knows it, we know it, so let's drop the act, man.

Whitey has seen to it that the blacks have no past.
There are lots of 'White Knights" but what about some
black heroes. You know, there have been lots of them. You
know--or didn't you know? Like Jimmy Brown, Stokely
Carmichael, Leroi Jones. You've heard of them? Blacks
have a history. Black children are tired of learning
about white Dick and Janes doing their thing in nice little
white houses when all they see is a world which is cruel,
hostile and filled with empty promises.

Now, what to do about this situation? As I see it,

get rid of Whitey. Just get him clear the hell out of here.

108

If it isn't done soon and peacefully, then someone is
going to get wiped off the face of the earth, and it ain't
going to be the black man. It's going to be good-bye
Whitey--George Wallace, Daley, c0ps, college presidents,
the establishment, and even some blacks. You know, some
Negroes look black on the outside but they're really
Whiteys on the inside. Blacks must be given power--

power to run their own businesses, schools, apartments,
towns, and governments. And some laws with guts have to
be passed--not this messing around with laws aimed at
tokenism but at real relevant problems. Laws are needed
to protect the black man from Whitey and to protect whitey
from himself. And then there's all this shit about law
and order--like old J. Edgar saying justice is only
incidental to law and order and Nixon wanting to jail
people before they commit a crime--we know who is going to
get screwed by that kind of a deal. Economically, all the
government has to do is to spend a little less money on
wars and Stop screwing the black man and give him the due
share of money he has paid in--and give some credit for a
lot of blood, sweat, and grief. The white man owes more

than he could ever pay for in a million years. Whitey owes

109

it and they're going to collect from those who wrenched
them from their homeland, enslaved them, lynched them,

beat and tortured them, humiliated them, stripped away
their heritage, their pride, their dignity, their humanness,
and he's trying to keep them slaves, even today--trying to
coop them up into little stinking ghettos sitting with the
rats underneath the freeway. The unions have to be opened
up, equal hiring laws have to be enforced, and the welfare
system has to be reorganized. There's got to be some black
controlled business and government for black communities
and schools. Blacks must have real power over their own
real problems, not this piddling shit they've been getting.
And finally, Whitey's got to learn, one way or the other,
that the black man is going to change this country, and if
Whitey resists, he's going to have one hell of a fight on

his hands.

APPENDIX B

Instructions

APPENDIX B

INSTRUCTIONS

The purpose of this study is to obtain undergraduate
opinions concerning certain social issues in the United
States. To determine which social issues in the United
States you will discuss, will one of you pick a piece of
paper from this box. According to the letter picked, you
two will discuss the causes and solutions to racial
disorders in the United States, specifically in the big
city. You will speak first (point to the accomplice).
You will speak for up to ten minutes concerning your
viewpoints as to the causes of racial disorders and
solutions to the problem. After you speak, the other
subject will speak for up to ten minutes. While one of
you is speaking we would appreciate no interruptions from
the other person. Immediately after the second person is
done speaking, you will have ten minutes to discuss your

opinions between yourselves.

110

111

Feel free to offer your Opinion no matter what it
is. The purpose of this study is to gain YOUR Opinions
to certain social issues. You can use any language,
comparisons, or examples you like if it helps you in
expressing your view. This session is being tape recorded.
All that is said here is confidential and will be known
only to the experimenters.

To give you a chance to organize your thoughts, you
will have about 3 minutes to think about what you will
say. When you are done, please leave the room. One of
the experimenters will meet you in the hallway.

If you have any questions, please ask them now.

APPENDIX C

Analyses of Variance

APPENDIX C

TABLE 10

Analyses‘g§_Variance

 

 

‘Discrepancy Score (4&6)l

 

 

 

'Source df MS F
Race (A) 1 2668.050 12.4122
Speech (B) 1 312.050 1.451
A X B l ' 7.200 0.033
Error (w) 76 214.956

1

Discrepancy scores of the g_rating how the actor viewed
himself compared to the gfs rating how the actor viewed the s,

2p (.01.

112

113

Table 10—continued

Discrepancy Score (2&3)3

 

 

 

Source df
Race (A) 1
Speech (B) l
A X B ' 1
Error (w) 76

MS

0.000

0.050

684.450

150.306

0.000
0.000

4.5534

 

3Discrepancy scores of the ideal—self compared to how

the g viewed the actor.

4p < .05.

Discrepancy score (1&2)5

 

 

Source df
Race (A) 1
Speech (B) 1
A X B 1
Error (w) 76

MS

74.113

1.013

262.812

228.224

0.324

0.004

1.151

 

5Discrepancy scores of the s comparing himself to the

ideal-self.

114

Table lO-continued

 

Discrepancy Score (1&3)6

 

 

Source df MS F
Race (A) 1 40.613 . 0.241
Speedh (B) 1 43.513 0.259
A X B 1 546.012 3.250
Error (w) - 76 _ 167.958

 

6Discrepancy scores of the g_rating himself compared to
the g_rating the actor.

Discrepancy Score (3&5)7

 

Source . df MS _ F
Race (A) 1 391.613 1.761
Speech (B) 1 40.613 0.182
.A X B 1 5.512 0.024
Error (w) 76 222.371

 

7Discrepancy scores of the g_rating the actor compared
to the g_rating the average college student.

115

Table 10-continued

 

Discrepancy Score (1&6)8

Source df MS F
Race (A) 1 525.313 3.161
Speech (B) 1 19.013 7 0.114
A X B 1 382.812 2.303
Error (w) 76 166.178

 

8Discrepancy scores of the g rating himself compared to
the g rating how the actor viewed the g,

Discrepancy Score (pooled)9

 

 

Source df MS F
Race (A) 1 8487 .200 ' 2 .101
Speech (B) 1 ' 151.250 0.037
A X B 1 7220.000 1.787
Error (w) 76 4039.372

 

9The results of the summation discrepancy scores from
columns 1, 2, 3, 4, 5, and 6.