• TABLE OF CONTENTS
HIDE
 Title Page
 Dedication
 Acknowledgement
 Table of Contents
 List of Tables
 List of Figures
 Abstract
 Introduction, literature, search,...
 Methodology
 Results
 Summary, conclusions, discussion,...
 Appendices
 Bibliography
 Biographical sketch














Group Title: effects of varying racial compositions upon attitudes and behaviors of supervisors and subordinates in simulated work groups
Title: The effects of varying racial compositions upon attitudes and behaviors of supervisors and subordinates in simulated work groups
CITATION PDF VIEWER THUMBNAILS PAGE IMAGE ZOOMABLE
Full Citation
STANDARD VIEW MARC VIEW
Permanent Link: http://ufdc.ufl.edu/UF00097638/00001
 Material Information
Title: The effects of varying racial compositions upon attitudes and behaviors of supervisors and subordinates in simulated work groups
Physical Description: xi, 127 leaves. : illus. ; 28 cm.
Language: English
Creator: Ruhe, John Alfred, 1932-
Publication Date: 1972
Copyright Date: 1972
 Subjects
Subject: Interpersonal relations   ( lcsh )
Management and Business Law thesis Ph. D   ( lcsh )
Dissertations, Academic -- Management and Business Law -- UF   ( lcsh )
Genre: bibliography   ( marcgt )
non-fiction   ( marcgt )
 Notes
Thesis: Thesis--University of Florida, 1972.
Bibliography: Bibliography: leaves 119-126.
Additional Physical Form: Also available on World Wide Web
Statement of Responsibility: by John A. Ruhe.
General Note: Typescript.
General Note: Vita.
 Record Information
Bibliographic ID: UF00097638
Volume ID: VID00001
Source Institution: University of Florida
Holding Location: University of Florida
Rights Management: All rights reserved by the source institution and holding location.
Resource Identifier: alephbibnum - 000871604
notis - AEG8827
oclc - 014277929

Downloads

This item has the following downloads:

effectsofvarying00ruherich ( PDF )


Table of Contents
    Title Page
        Page i
    Dedication
        Page ii
    Acknowledgement
        Page iii
    Table of Contents
        Page iv
        Page v
    List of Tables
        Page vi
        Page vii
    List of Figures
        Page viii
    Abstract
        Page ix
        Page x
        Page xi
    Introduction, literature, search, and hypotheses
        Page 1
        Page 2
        Page 3
        Page 4
        Page 5
        Page 6
        Page 7
        Page 8
        Page 9
        Page 10
        Page 11
        Page 12
        Page 13
        Page 14
        Page 15
        Page 16
    Methodology
        Page 17
        Page 18
        Page 19
        Page 20
        Page 21
        Page 22
        Page 23
        Page 24
        Page 25
        Page 26
        Page 27
        Page 28
        Page 29
        Page 30
        Page 31
        Page 32
        Page 33
    Results
        Page 34
        Page 35
        Page 36
        Page 37
        Page 38
        Page 39
        Page 40
        Page 41
        Page 42
        Page 43
        Page 44
        Page 45
        Page 46
        Page 47
        Page 48
        Page 49
        Page 50
        Page 51
        Page 52
        Page 53
        Page 54
        Page 55
        Page 56
        Page 57
        Page 58
        Page 59
        Page 60
        Page 61
        Page 62
        Page 63
        Page 64
        Page 65
    Summary, conclusions, discussion, and implications for future research
        Page 66
        Page 67
        Page 68
        Page 69
        Page 70
        Page 71
        Page 72
        Page 73
        Page 74
        Page 75
        Page 76
        Page 77
        Page 78
        Page 79
        Page 80
        Page 81
        Page 82
        Page 83
    Appendices
        Page 84
        Page 85
        Page 86
        Page 87
        Page 88
        Page 89
        Page 90
        Page 91
        Page 92
        Page 93
        Page 94
        Page 95
        Page 96
        Page 97
        Page 98
        Page 99
        Page 100
        Page 101
        Page 102
        Page 103
        Page 104
        Page 105
        Page 106
        Page 107
        Page 108
        Page 109
        Page 110
        Page 111
        Page 112
        Page 113
        Page 114
        Page 115
        Page 116
        Page 117
        Page 118
    Bibliography
        Page 119
        Page 120
        Page 121
        Page 122
        Page 123
        Page 124
        Page 125
        Page 126
    Biographical sketch
        Page 127
        Page 128
        Page 129
Full Text



















THE EFFECTS OF VARYING RACIAL COMPOSITIONS UPON
ATTITUDES AND BEHAVIORS OF SUPERVISORS AND
SUBORDINATES IN SIMULATED WORK GROUPS





by





JOHN A. RUHE


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


UNIVERSITY OF FLORIDA
1972


























To all the people regardless of race, creed, or sex

who have been denied jobs or better jobs because of

discriminatory employment practices.











ACKNOWLEDGEMENTS


The author wishes to express his appreciation to his Supervisory

Committee for their stimulation, interest, and support throughout the

doctoral program. The efforts of Dr. Hill, Dr. Blodgett, Dr. Keig,

and Dr. Shaw have proven invaluable in so many ways.

A special word of recognition and appreciation is given to Dr.

Walter Hill. His support, counsel, encouragement and time as teacher,

research supervisor, doctoral chairman, loyal friend, and mentor have

aided immeasurably.

The author wishes to acknowledge John Nagay, Burt King, and

Bill Gayman of the Office of Naval Research, Organizational Effective-

ness Program, who provided technical and financial assistance. Also,

the author expresses a word of thanks to Margie Kuenz who assisted in

the statistical analysis. Her support, time, and effort proved extreme-

ly helpful. In addition, others who have provided significant input

toward the development of this dissertation through the data collect-

ion, coding, recording, and typing are Fred Nutt, Corrine Brown,

Cathy Mathis, Martha Williams, and Susan Koch. Their efforts are

sincerely appreciated.

Finally, the author must recognize the prayers, works, and sacri-

fices made by his wife, Fran; sons, Michael and Chris; and daughters,

Laura, Lisa, and Judy. Their efforts and love have made this study

possible and worthwhile.












TABLE OF CONTENTS
Page

ACKNOWLEDGEMENTS . . . . . . . . . . . . iii

LIST OF TABLES . . . . . . . . ... . . . vi

LIST OF FIGURES . . . . . . . . . viii

ABSTRACT . . . . . . . . . . . . . . ix

CHAPTER

I. INTRODUCTION, LITERATURE SEARCH, AND HYPOTHESES

Introduction .......................................... 1
Statement of the Problem............................... 2
Review of the Literature .............................. 3
Attitude Studies ................................. 3
Behavior in Interaction Work Groups .............. 6
Performance Ratings .............................. 7
Racial Differences ............................... 7
Purpose of this Research Study ........................13
Hypotheses ............................................ 14
Notes .................................................16

II. METHODOLOGY

Subjects ..............................................17
Environment ........................................... 20
Tasks ............................................ 20
Observation ......................................25
Input Variables ..................................25
Mediating Structure ..............................26
Output Variables .................................27
Procedure ........................................ 29

III. RESULTS

Differences Between Supervisors .......................34
Differences Between Subordinates ......................41
Differences Between Subordinates in Dyads .............45
Differences Between Supervised Dyads .................. 51
Supplementary Findings ................................56
Differences in Heterogeneous Dyads ...............56
Differences Between Heterogeneous and
Homogeneous Dyads ......................... 62
Notes ................................................. 64









CHAPTER Page

IV. SUMMARY, CONCLUSIONS, DISCUSSION, AND IMPLICATIONS
FOR FUTURE RESEARCH

Summary ........................................ 66
Conclusions .................................... 68
Supplementary Findings ......................... 70
Discussion ..................................... 72
Limitations and Implications for Future
Research .............................. 81
Notes .......................................... 83

APPENDICES

A. Biographical Questionnaire ............................ 85
B. Ship Routing Information .............................. 90
C. Knot Tying Instructions ............................... 92
D. Job Description Index Supervisor .................... 95
Job Description Index Subordinate ................... 99
E. Social Orinetation Tasks ..............................103
F. Letter Rating Form ....................................111
G. Introduction and Instructions for Tasks ...............114

BIBLIOGRAPHY . . . . . . . . . . . . 119

BIOGRAPHICAL SKETCH . . . . . . . .... ..... 127











LIST OF TABLES


Table Page

1. Spearman Rank Correlations of White and Black Observers
on Interpersonal Behavior of White, Mixed, and Black
Dyads .......................................................33

2. U Values and Z Scores of IPA Behavior Differences
Between Black and White Supervisors Overall and in
Separate White, Mixed, and Black Dyads .......................36

3. U Values and Z Scores of Leadership Style Differences
Between Black and White Supervisors with All Subordinates
and with White, Black, or Mixed Racial Dyads of
Subordinates ................................................. 39

4. U Values and Z Scores of Duration of Speech, Self-
esteem, Satisfaction with Work and Subordinates Between
Black and White Supervisors ..................................40

5. U Values and Z Scores of IPA Behavior Differences
Between Black and White Subordinates .........................43

6. U Values and Z Scores of Duration of Speech, Self-
esteem, and Satisfaction with Fellow Subordinates,
Supervisors, and Work Between Black and White Subordinates ...44

7. U Values and Z Scores of IPA Behavior Differences
Between Dyads of White, Black, and Mixed Racial
Subordinates .................................................47

8. U Values and Z Scores Differences in Productivity,
Cohesiveness, Self-esteem, Duration of Speech, and
Satisfaction with Fellow Subordinate, Supervisor, and
Work Between Dyads of White, Black, and Mixed Racial
Subordinates .................................................49

9. Rank Sums and Resulting H Values of Six Different
Supervised Groups on Measures of Productivity, Cohesive-
ness, Duration of Speech, Satisfaction with Supervisors
and Work .................................... ............... 53

10. U Values of Cohesion Differences Between Individual
Pair Comparisons of Supervisors and Subordinate Dyads ........54










Table Page

11. Mean and Z Score Differences Between Black and White
Subordinates in Homogeneous and Heterogeneous Dyads . . 57

12. U Value and Z Score Differences in Interpersonal
Behavior Between Black and White Subordinates in
Heterogeneous Dyads . . . . . . . . . 60

13. U Values of IPA Behavior Differences Between Black
and White Subordinates in Heterogeneous Dyads with
Black and White Supervisors . . . . . . . .. 61

14. Mean Differences and Statistics Value Between Black
and White Subordinates in Homogeneous and Heterogeneous
Dyads in Measures of Self-esteem, Duration of Speech,
Satisfaction with Fellow Subordinate, and Satisfaction
with Work . . . . . . . . . . . . 63

15. Summary of Null Hypotheses Results . . . . .... 69

16. Summary of Supplementary Findings . . . . .... 71














LIST OF FIGURES


Figure Page

1. Selected Differences in Reported Economic and Social
Characteristics of Black and White Participants . . . 18

2. Photograph of Experimental Room as Seen Through the
Observation Room . . . . . . . .... ..... 21

3. Layout of Experimental and Observation Room . . ... 22













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





THE EFFECTS OF VARYING RACIAL COMPOSITIONS UPON
ATTITUDES AND BEHAVIORS OF SUPERVISORS AND
SUBORDINATES IN SIMULATED WORK GROUPS





By

John A. Ruhe

August, 1972



Chairman: Dr. Walter A. Hill
Major Department: Department of Management and Business Law








The purpose of this study was to investigate in a simulated job

setting differences in attitudes and interaction behavior between (a)

black and white supervisors, (b) black and white subordinates, (c)

black, white, and racially mixed dyads of subordinates, and (d) black

and white supervised dyads of black, white, and racially mixed subor-

dinates.










A sample of 96 male undergraduate students, 48 black and 48 white,

were hired for the study. Each of the 12 black and 12 white older

students who opted to be supervisors were randomly assigned to super-

vise three different racially mixed dyads of subordinates (one all-

black, one all-white, and one with a black and a white subordinate).

The remaining 72 students, 36 black and 36 white, were randomly

assigned to 1 of the 3 subordinate dyads that participated once with

a black supervisor and once with a white supervisor. All possible

orders of supervisor and dyad combinations were represented twice to

reduce the effect of order bias. Each group of a supervisor and a

subordinate dyad was given three similar problem-solving tasks, two

structured and one unstructured.

During the task performance, two trained graduate students (one

black and one white) observed and coded the group interaction through

a one-way mirror. Output measures were: (1) the Bales Interaction

Process Analysis (IPA), (2) the duration of speech for each group

member, (3) productivity of the three tasks, (4) cohesiveness, (5)

self-esteem, and (6) satisfaction ratings (JDI) of fellow subordinates,

supervisor, and work in the tasks.

Among the findings were the following major conclusions.

1. The only significant differences between black and white

subordinates originated within the heterogeneous (racially mixed)

dyads. As compared to white subordinates, black subordinates exhibited:

a. Less giving of suggestion

b. Less giving of information










c. Less giving of suggestions when with a white supervisor

d. Higher self-esteem

e. Less duration of speech

f. Greater satisfaction with work in the tasks.

In the aggregate comparison the black subordinate expressed greater

satisfaction of his fellow white subordinate.

2. The individual differences that occurred between black and

white subordinates in the heterogeneous dyads did not perceptibly

affect the group behavior measures of productivity of the three tasks

and of cohesiveness. There were no significant differences between

subordinates of black and white homogeneous dyads and heterogeneous

dyads in measures of interpersonal behavior (IPA), productivity of

the three tasks, cohesiveness, self-esteem, duration of speech, and

satisfaction with fellow subordinate, supervisor, and work in the

tasks.

3. The only significant differences between black and white

supervisors were that black supervisors exhibited less asking for

opinion and suggestion over all subordinates.

4. In differences between supervised dyads, white dyads super-

vised by white supervisors exhibited more cohesiveness than black

dyads supervised by white supervisors. No other significant differ-

ences in cohesiveness were found between any other supervised dyads.

There were no other significant differences in group measures of pro-

ductivity, duration of speech, and satisfaction with supervisor and

work between the six different supervised dyads.













CHAPTER I

INTRODUCTION, LITERATURE SEARCH, AND HYPOTHESES

Introduction

Today black Americans are striving for many things. High on their

priority list, if not at the top, are jobs. If they do not have jobs,

most want them. If they have jobs, most want better jobs as well as the

opportunity to prepare themselves for still better jobs in the future.

However, prejudicial attitudes and discriminating practices in our nation

have denied both jobs and better opportunities to many black Americans.

The psychological and social costs to the individual of this dis-

crimination is very difficult to measure, but it has been estimated that

the annual economic cost to our nation in wasted production and services

approximates $17 billion (U.S. Department of Labor, 1963.) Walter Heller

(1970) reported on studies by the Council of Economic Advisors that show:

the nation losing between three and four percent of its Gross
National Product annually through its failure to make full use
of existing skills and knowledge of nonwhites, coupled with its
failure to bring those skills and knowledge to a par with those
of whites.1

Despite recent employment advances by some members of the black com-

munity, little improvement in black employment compared to white employ-

ment has occurred. The Manpower Report of the President (1972) reported

almost one million jobless blacks in 1971. Their overall jobless rate of

8.2 in 1970 increased to 9.9 percent in 1971, while the teenage rate moved

up from 29.1 to 31.7 percent. The overall Negro/white ratio of jobless

rates is slightly less than 2 to 1 while the 31.7 percent jobless rate of










black teenagers was more than double the 15.1 percent rate for white teen-

agers.2 This discouraging profile is also reflected in the scarcity of

blacks in leadership positions. The Manpower Report of the President

(1972) reported that, in 1970, blacks represented only 1.6 percent of male

managers and officials in firms with 100 or more employees. Recently,

the Black Caucus in Congress claimed that business, education, and govern-

mental institutions have not integrated the Negro into all levels of their

organizations.

In defense, these organizations state a great concern about the

effectiveness of blacks as leaders -- particularly of integrated or all-

white groups. If blacks are placed in leadership positions, they are

usually limited to supervising other blacks. Unless significantly more

blacks reach leadership positions at all levels in these organizations,

the economic progress of other blacks may be stymied.


Statement of the Problem

Unfortunately, many qualified blacks are denied better jobs (or any

job at all) because of the assumption by white employers that blacks are

different from whites in a variety of work attitudes and behaviors. They

further assume that these differences eventually reduce group effective-

ness -- particularly in racially mixed work groups. Unfortunately, few

empirical studies have examined these assumptions. We have only an early

recognition by Ginzberg (1959) that race was one of the determinants of

group effectiveness in the military. This lack of social psychological

research concerning actual differences between blacks and whites and the

nature of supervisor-subordinate interaction in job setting has been









decried by Triandis (1971), King & Bass (1970), Dreger & Miller (1968),

and Moskos (1967).


Review of the Literature

Until recently, most social psychological research on interracial

relationships has been concerned with the narrow study of verbal prej-

udice. In a review of the literature on race and ethnic relations,

Harding, Proshansky, Kutner, & Chein (1969) found that approximately

eight times as much space was devoted to studies of attitudes as com-

pared with studies of overt behavior.

Attitude Studies

Recent literature reviews conducted by Katz (1970), Amir (1969), and

Pettigrew (1969) generally agree with Allport's (1954) Contact Hypothesis

regarding ethnic prejudice. This well-known hypothesis, implicitly

assumed in many desegregation plans, states that intergroup contacts in-

volving shared interests or goals, equal status of participants, and norma-

tive support tend to reduce prejudice. This hypothesis has received tenta-

tive support from studies in various environmental settings such as: the

military (Stouffer, Suchman, DeVinney, Star & Williams, 1949); department

stores (Harding & Hogrefe, 1952); housing projects (Deutsch & Collins,

1951; Wilner, Walkley & Cook, 1955); and experimental settings (Burnstein

& McRae, 1962). These studies generally suggest that cooperative, equal

status contact with Negroes may bring out favorable, but limited, change in

attitudes among whites. This statement must be qualified, however, as a few

studies have shown no change or no consistent change in attitudes among

children experiencing interracial contacts in summer camps (Mussen, 1950;










Yarrow, 1958) or in elementary schools (Shaw, 1970: Carithers, 1970).

Carithers (1970) concluded from his review of desegregated schools that

interracial contact, per se, will not bring about increased tolerance or

acceptance. In general, the equal status hypothesis has received impres-

sive, although not completely consistent, support as a means of bringing

about changes in white attitudes towards blacks.

One glaring omission in the above studies is their failure to deal

with the effects of contact on black attitudes. In view of the apparent

militancy of some blacks, we can no longer assume that black attitudes are

fixed or favorable. Prejudicial attitudes are not confined to whites

alone, but are now known to exist among members of both races. In addi-

tion to this omission, these studies have raised questions concerning the

reliability of the relationship between an expressed attitude of prejudice

and resulting overt behavior.

Relationship between attitude and behavior.-Katz (1970) reviewed the

literature concerning the predictive power of verbal attitudes on overt

behaviors such as: customer service (Kutner, Wilkins, & Yarrow, 1952;

Saenger & Gilbert, 1950); GSR rating (Rankin & Campbell, 1955; Porier &

Lott, 1967; Westie & DeFleur, 1959; Vidulich & Krevanick, 1966); verbal

conditioning (Smith & Dixon, 1968); and willingness to pose with a Negro

for a photograph (Linn, 1965; DeFleur & Westie, 1958). The field studies

involving behaviors such as customer service show a discrepancy between

verbal expression or intention and actual behavior. The four studies in-

volving galvanic skin responses to Negro stimulus photographs show fairly

consistently that prejudiced subjects give stronger autonomic reactions











than do relatively unprejudiced subjects. The implications of these studies

remain uncertain and will require studies of a wider range of biracial

situations (Katz, 1970).

The major findings of the studies asking whites to pose with a Negro

for a photograph indicated that subjects high on anti-Negro prejudice were

less willing than those low on anti-Negro prejudice to have their picture

widely publicized. Linn (1965), using a somewhat questionable procedure,

found no relationship between scores on anti-Negro prejudice scales and

willingness to pose with Negro subjects when told that such photographs

would be publicized widely. The implications of these studies are also

ambiguous since one of the seven graded uses of the photographs could be

considered a pro-Negro action. In addition, a refusal to pose with a Negro

may not have been seen by the subjects as harmful to Negroes because sub-

jects could assume that someone else would pose for the picture if they did

not. In general, the above-cited studies do not show a consistent relation-

ship between verbally expressed attitudes towards blacks and overt behavior

towards blacks. One logical reason for the observed lack of consistency

is that these studies missed the reality of most integration conditions

by not placing the white subject in a position where he actually interacted

with a Negro.

In a more recent study, Terry & Evans (1972) found that more class

and race discrimination was attributed to other persons, especially to

blacks, than to self and that class was a more important basis for dis-

crimination than race.









Influencing attitudes.-Several studies have investigated the relation-

ship of the sensitivity of attitudes towards minority groups to the indi-

vidual influence of group members. Particular attention has been paid to

the impact of the presence of a Negro peer or experimenter on the attitudes

of high-prejudiced subjects. Several studies (Himmelstein & Moore, 1963;

Bray, 1950, Berg, 1966, and Malof & Lott, 1962) indicated that behavior

was not obviously consistent with the racial attitudes measured. The in-

consistency between behaviors and attitudes may have been more apparent

than real since the measured attitudes and observed behaviors had no per-

ceived relationship to one another. No range of behavior alternatives

existed. The choice was not between behaving consistently or inconsistently

but only between engaging in specified types of behavior.

Behavior in Interaction Work Groups

Only a few studies have dealt with behavior in biracial work groups.

These studies were concerned with communication by whites to Negro mem-

bers (Burnstein & McRae, 1962; Katz, Goldston & Benjamin, 1958; Cohen

1969; Katz & Benjamin, 1960, Jaffee & Whitacre, 1969; Richards, 1970).

Generally whites initiated more communicative acts, tended to talk more,

and exercised more social influence than Negroes even though their success

was controlled. Katz & Cohen (1962) and Katz & Benjamin (1960) found that

the introduction of a shared goal (group bonus condition) did produce a

greater amount of cooperative behavior, but did not affect the tendency

of white subjects to avoid talking to Negroes on some tasks, or the ten-

dency of Negro subjects to prefer talking to whites. Burnstein & McRae

(1962) found that whites evaluated a Negro partner more favorably when

groups received failure feedback than they did after success feedback.









Katz & Cohen (1962) found further a greater degree of role equality in the

experimental condition, even though working toward a shared goal, seemed

to negatively affect attitudes of white subjects. On a final question-

naire, these whites downgraded the problem solving ability of Negro part-

ners even though both were correct an equal number of times on the task.

The whites were also less willing to continue the working relationships

than subjects in a control group.

Performance Ratings

Another area of difficulty in a working relationship is that attitudi-

nal biases of white raters might result in lower performance evaluations

of black supervisors. Cox & Krumboltz(1958) and DeJung & Kaplan (1962)

showed that racial bias existed in peer ratings among military personnel.

In both experiments, white and black soldiers received significantly high-

er performance ratings from members of their own race than from members of

the other race.

The results above generally agree with the cooperative goal attain-

ment hypothesis of Allport (1954). Deutsch & Collins (1951) and Sherif

(1958) suggest that an approach is needed that takes full account of the

cognitive predispositions that Negroes and whites bring to contact situa-

tions. This would include not only racial attitudes and stereotypes, but

also assumptions about the motives of Negroes as well as whites in given

racial encounters. A brief review of these and other black-white differ-

ences will aid our understanding of employment problems.

Racial Differences

General review.-Recent reviews of the literature by Dreger & Miller

(1960; 1968) have summarized black-white differences on a number of









environmental, social, personality, perceptual, and heredity variables.

They found that similarities continue to outweigh differences, but that

significant differences do exist and that blacks and whites are far from

having a clear understanding of the others' view of the world. Dreger &

Miller (1968) further indicate that recent research is less apt to lump

all blacks or all whites together such as the American Negro or the

American white.

A few examples of the differences as reviewed by Dreger & Miller

(1968) are:

1. Socioeconomic conditions affect both physical and psycho-
logical functioning, with higher status accorded light
skin; lower socioeconomic status correlates with prenatal
and perinatal experiences tending to induce neurological
damage.

2. When genetic factors are considered, some specific abili-
ties show up with differential racial patterns related to
class and age of the children. In addition, the recipro-
cal nature of language and thought processes have reveal-
ed contrasts between concretistic thinking of lower socio-
economic groups, both white or black, and the more abstract
processes of intellectual functioning of middle class child-
ren. Intellectual functioning has been the subject of a
continuing heated debate, particularly over the issue of
hereditary versus environmental influences. Bodmer &
Cavalli-Sforza (1970) reviewed the work of Shockley, Jensen
and others and concluded that the difference in I.Q. scores
between blacks and whites cannot be attributed to genetics
in the present circumstances.

3. Savoca (1965) correlated socioeconomic status with Origi-
nality (in terms of creative thinking) and Fluency. He
also found that white children scored higher on Semantic
Flexibility.

4. The presence or absence of the father in the home environ-
ment is a difference between black and white children. It,
with preschool or no-preschool experience, is related to
depressed I.Q. scores in the black group.

5. Negroes seem to continually express relatively high educa-
tional and occupational aspirations. Keig (1969) found that











many of these aspirations were not realistic with ex-
pectations from the environment.

6. A number of studies reveal comparatively negative self
attitudes in Negroes (Katz et al., 1958; Deutsch, 1965).
Gray (1964) found that the deprived Negro child, particu-
larly, has a problem of identity and self-esteem.

These are but a few of the differences reviewed by Dreger and Miller

(1968). In summary, they conclude that many variables (e.g., societal and

cultural goals, socioeconomic status, caste status, child rearing prac-

tices, speech and communication complexities, levels of aspiration, ex-

aminer variables, measurement instrument variables, and immediate situ-

ational stimulation) must be analyzed multivariately to better understand

the complexities of interaction of these variables on racial differences.

Dreger and Miller also suggest that recent changes resulting from "the

Second American Revolution" will have a significant effect on comparative

psychological research.

Racial differences relevant to employment.-In view of our nation's

pressing problems of employment integration, Triandis (1971) suggests

the study of four main kinds of difference:

a) motivational such as self-esteem. Long, Ziller, Ramana & Reddy
(1969) found self-esteem and socioeconomic status positively
associated among teenage students in India. Katz & Benjamin
(1960), Katz & Cohen (1962), Katz, Goldston & Benjamin (1958),
Katz & Greenbaum (1963), and Lefcourt & Ladwig (1965) indi-
cated that blacks tend to feel inadequate, and orient compli-
antly toward whites even when they are given objective evi-
dence of equal mental or achievement ability.

b) use of time such as socializing too much on the job. Kochman
(1969) found a great deal of "rapping" occurring among ghetto
members. This stereotype suggests that too much socializing
occurs among blacks on the job (APA Symposium, 1969). Jaffee
& Whitacre (1970) found that the number of votes received by
a leader was positively related to high verbal interaction
as measured by duration of speech, except for high black inter-
action.










c) interpersonal behavior such as cohesiveness, affiliation,
and suspicion. Guttentag (1970) reported greater group
cohesiveness when members receive more individual attention
and understanding from each other, communicate with one
another, and feel attracted to one another. Blacks appear
to be high on affiliation (Ferman, Kornbluk & Miller, 1968)
which should result in greater cohesion and satisfaction
scores. Huber (1970) found that supervisor-subordinate
similarity was an important factor in determining the sub-
ordinate's job satisfaction, but it bore no relationship
to performance evaluations. Katzell, Ewen, & Korman (1970)
found that black employees, in a specific industry, were
generally more satisfied with their jobs than white employees.

d) behavioral norms such as taking the initiative. King &
Bass (1970) indicate that black supervisors with white sub-
ordinates are more likely to exercise general rather than
close supervision in order to encourage initiation of in-
teraction by their subordinates.

Leadership studies.-Some recent writers (e.g., Blalock, 1967: Matthews

1969) have proposed a study of the differences between races in terms of

the power resources available to each party and the readiness of each to

employ these resources. Of particular importance is the power resource

of leadership and its significance to the economic success of black

individuals. King & Bass (1970) have examined the literature which in-

dicates the relative absence of Negroes in leadership positions in the

military, in business, in professional sport, etc.,and have concluded

that this absence amounts to an open indictment of the policies and prej-

udices of social institutions in this country. Further review of the

literature confirms their concern. Despite general satisfaction with

military integration from most quarters, the problem of rank and occu-

pational status of Negroes serving in the Armed Forces is a major con-

cern of Pentagon officials (Stillman, 1969). As of 1970, while 11 per

cent of the American population was Negro, 10 per cent of the military

was Negro but the officer corps consisted of only 2 per cent Negro (BLS,










1970). This distorted pyramid of Negro class structure with a very nar-

row vertex of people in leadership positions and a very broad base of

less-skilled people is also a problem among noncommissioned officers

(Stillman, 1969; Moskos, 1967).

In the business world the opportunities for black leadership are no

better. The largest division of one of our nation's giant corporations

in 1969 had only 17 black supervisors in its entire organization (King

& Bass, 1970). Higher level black managers are virtually non-existent.

Unterman (1971) concluded that it is virtually impossible to find

one successful minority management program in the entire nation. Faced

with a lack of understanding of the social, psychological, and economic

differences between blacks and whites, some companies have simply thrown

in the towel and are doing little or nothing in the way of minority-

group management programs (Unterman, 1971). The difficulties of black

supervisors fitting into the white man's business world were highlighted

in anecdotal form by Travaglio, Sloan & Walker (1971), Gooede (1970), and

National Industrial Conference Board (1966). However, Campbell (1971)

reported in a fifteen-city study of racial attitudes that 86 per cent

of the whites questioned stated that they would not mind having a black

supervisor on their job. Campbell (1971) suggested that the work situ-

ation apparently is not threatening to most whites because it involves

the more public "distant" aspect of personality. Furthermore, King &

Bass (1970) state that white supervisors are having trouble managing

black subordinates in mixed and all-black work groups.

One of the major questions raised by racially mixing work groups is

what will happen to the effectiveness of the group. This becomes an










especially debated dilemma when considerations are given to having blacks

supervise whites. Specific attention needs to be given to questions such

as: How do the individual, situational, and cultural differences of

supervisors and subordinates influence the productivity and satisfaction

of work groups? Do problems of interaction in racially mixed groups im-

pede group effectiveness? What problems do superiors have when they

supervise subordinates of a different race? Particularly, what problems

do blacks have when they supervise racially mixed or all white subordi-

nates. Unfortunately, few studies have examined these aspects of racial-

ly mixed groups.

Studies of blacks in leadership positions have generally been limit-

ed to such areas as temporary social settings (Winter, 1971; Fenelon,

1966); neighborhood opportunity centers (Delbecq & Kaplan, 1968):

experimenter or test administrators (Katz, Roberts, & Robinson, 1965;

Lefcourt & Ladwig, 1965; Battle & Rotter, 1963) and a business game

(Richards, 1970). These studies generally show that in black-white in-

teractions, the black has a higher expectancy of failure, shows less

self-assertion, lower self-esteem, is more anxious and less efficient

than when working among other blacks, talks to whites more than to other

blacks, performs more poorly than whites on the situational exercises

as a result of behavior of white subordinates, receives poorer perform-

ance ratings, and is influenced by evidence of equal ability with

whites and behaves in an external control fashion when requiring skill

but in an internal control manner when the attainment of rewards is

perceived as a matter of chance. Blacks reveal a greater expectancy











of control as being external if they see their goals as being unobtain-

able through their own efforts. Winter (1971) is the only one who found

an expression of confidence and aggressiveness among informal black male

leaders. This occurred in self-analytic groups of college students.

The lack of adequate research concerning differences in black-white

interactions in supervisor-subordinate relationships in job settings may

result from the scarcity of black supervisors in the natural world. Par-

ticularly in the less industrialized sections of the South, black super-

visors are a rarity,especially if they supervise whites. Even in the

laboratory studies, the scarcity of blacks in universities--particularly

in undergraduate psychology classes--has limited the research. Recently,

Katz (1970) concluded that this scarcity of research concerning black-

white interactions has seriously limited our knowledge.


Purpose of this Research Study

Based on the above literature review, it is obvious that little is

known about differences between black and white supervisors and subor-

dinates in actual job settings. The purpose of this study is to investi-

gate differences between:

a) black and white supervisors and their relationships with
black, white, and mixed racial groups of subordinates,

b) black and white subordinates and the nature of their in-
teraction and attitudes in a simulated job setting, and

c) black, white, and racially mixed subordinate dyads
supervised by black and white supervisors.

The laboratory setting of a simulated work group was chosen as the

site to test for differences between blacks and whites. In the laboratory











there is the greater opportunity to control the independent variables

of race and racial mixture and to assign subjects according to a sys-

tematic, unbiased plan.


Hypotheses

The specific null hypotheses tested are:

1. There are no differences between black and white supervisors

in measures of:

a) interpersonal behavior (IPA)
b) style of leadership (IPA)
c) duration of speech
d) self-esteem
e) satisfaction with subordinates
f) satisfaction with work in the tasks

2. There are no differences between black and white subordinates

in measures of:

a) interpersonal behavior (IPA)
b) duration of speech
c) self-esteem
d) satisfaction with their fellow subordinates
e) satisfaction with their supervisor
f) satisfaction with work in the tasks

3. There are no differences between dyads composed of black, white,

and racially mixed subordinates in group performance measures of:

a) interpersonal behavior (IPA)
b) productivity of the three tasks
c) cohesiveness
d) self-esteem
e) duration of speech
f) satisfaction with their fellow subordinates
g) satisfaction with their supervisors
h) satisfaction with work in the tasks

4. There are no differences between black and white supervised dy-

ads composed of black, white, and racially mixed subordinates

in measures of:











a) productivity of the three tasks
b) cohesiveness
c) duration of speech
d) satisfaction with their supervisors
e) satisfaction with work in the tasks

In Chapter I we have shown that racial discrimination in employment

can be a serious social and economic problem. The scarcity of black lead-

ers in the military, business, and professional sports suggests that our

nation's employment policies may be determined by assumptions relating

to behavioral and attitudinal differences between blacks and whites in

their delegated roles. It is often implied that these differences pro-

duce black-white interactions that impede group effectiveness. The ab-

sence of research in racial differences in work-roles prevents the deter-

mination of whether these assumptions are warranted or based on prejudicial

emotions.

Chapter II establishes a methodological framework of a simulated

work setting that manipulated the independent variables of race among

supervisory and subordinate roles. Output variables of attitude and

behavior measures are described in order to analyze the differences in

black-white interaction in the group process.

Chapter III describes the results of the analyses and Chapter IV

summarizes, concludes, and discusses possible reasons for the results

jointly with some directional hypotheses suggested by research evidence,

hunches, and stereotyped assumptions. Implications for future research

also are stated.











Chapter I

Notes


1Heller (1970) makes a compelling case for national interest

in eliminating discrimination by identifying groups that will benefit

from greater economic quality for the Negro. For example, employers

will have a larger and better prepared labor force; sellers will have

a more affluent Negro community; taxpayers will have a diminishing

welfare drain and a broader tax base to fund more productive government

expenditure; and city property owners will benefit from decreasing

crime, violence, and urban decay that accompany the economic gains of

the Negro.



2Ever since the Korean War, the Negro/white ratio of jobless rates

for teenagers (16-19 years of age) has steadily increased (i.e., 1.3 in

1954 to 2.1 in 1971). The jobless rate for black teenagers continued

to increase during the high employment period for the Vietnam conflict

(Manpower Report of the President, 1972, Table A-5).

















CHAPTER II

METHODOLOGY

This chapter presents the methodology used in the study in terms of

(a) subjects, (b) environment, (c) input variables, (d) mediating struc-

ture, (e) output variables, and (f) procedure.

Subjects


With a few exceptions, subjects were single male undergraduates from

the University of Florida. The investigator tried to obtain a matched

sample of white and black students of similar age, intellectual standing,

and socioeconomic class. Lists of students of comparable senior place-

ment scores and somewhat similar parental income were supplied by admin-

istrators of the Expanding Education Opportunity and Work Study Programs.

The investigator recruited participants through the use of personal let-

ters, phone calls, and counselor contacts. The study was introduced to

the students by a cover story that identified the need for research into

the way groups solve problems effectively. They were told they were cho-

sen because college students are among the groups that have the ability

and interest to participate in solving problems in business, in society,

and in our nation.

Students were asked to complete a biographical questionnaire prior

to scheduling their participation in the study (see Appendix A ). The





















W B


- B


W B


Idle Class Average and Living with White Collar
above above both parents and
Professional
;IAL CLASS FAMILY RESIDENCE OCCUPATION
INCOME WITH PARENTS OF FATHER

Figure I
Selected Comparisons in Reported Economic and Social

Characteristics of Black and White Participants


W B


2 or less

NUMBER OF
SIBLINGS


Mid
and
SOC


-U- B











purposes of the biographical questionnaire were: a) to establish cre-

dence for later announced group compatibility (Katzell et al., 1970b

and b) to check on similarity of the students. Despite a major attempt

to match the students, several differences existed between the black

and white students.(See Figure I.) The blacks reported themselves sub-

stantially different in social class, family income, father's occupation,

parental influence and number of siblings in their family. The difficul-

ty in matching the students in socioeconomic class appears to be charac-

teristic of the student populations in U.S. colleges and particularly

in recently integrated southern universities,(U.S. Bureau of the Census.

1970.) The black students did exceed whites in average age (19.5 vs.

18.9) and in a general satisfaction score toward a variety of activities

and friends at school and home. (32.2 vs. 30.7.)

A sample of 96 male undergraduate students, 48 black and 48 white,

were hired to role play as supervisors and subordinates in this study.

Each of 12 black and 12 white older students, were selected to be super-

visors, were randomly assigned to supervise three different racially

mixed dyads of subordinates (one all black, one all white, and one with

a black and a white subordinate ), The remaining 72 students, 36 black

and 36 white, were randomly assigned to one of the three different racial-

ly mixed subordinate dyads that participated once with a black supervisor

and once with a white supervisor. All possible orders of supervisor and

dyad combinations were represented twice to reduce the effect of order

bias. The participants were paid three dollars per scheduled hour.









If their group finished before the end of the scheduled time period,

they were permitted to leave. This increased the incentive to complete

the tasks quickly.


Environment

The studies were conducted in the NASA/Psychology Laboratory at

the University of Florida. Two adjacent rooms were used to conduct the

study: an experimental room and an observation room (see Figure 2). A

one-way mirror between the two rooms permitted observation of the parti-

cipants by two trained observers with a minimum of distraction to the

participants (see Figure 3). The experimental room contained:

a) three chairs around a circular table,

b) a rectangular table behind the circular table,

c) a small table and chair for the investigator,

d) two six-foot lengths of clothesline,

e) a microphone and tape recorder on the investigator's table,

f) and a microphone on the circular table connected to an ampli-
fier.

The observation room contained:

a) two chairs on a raised platform near the one-way mirror,

b) a speaker connected to the amplifier in the observation room,

c) and two large clipboards for the observers.


Tasks

Three tasks were chosen for study: two structured and one un-

structured. Two of the tasks, the ship routing task and the recruiting













* wgsj


Figure 2
Photograph of Experimental Room as Seen Through the Observation Room







































1 = Observer A (large circles
are chairs)

2 = Observer B

3 = Subordinate A

4 = Supervisor (chair on rollers)

5 = Subordinate B

6 = Investigator


= Work Table

= Microphone

= Amplifier

= Speaker

= One-way mirror

= Investigator's table

= Tape recorder

= Brief case with stop watches

= Evaluation table for
supervisor


Figure 3
Layout of Experimental and Observation Rooms











letter were similar to those used by Fiedler (1966 ). The recruiting let-

ter always was administered last to avoid contaminating behavior on the

other tasks with possible carryover from negative attitudes about the

military.

The structured tasks.-Two structured tasks were administered to

each group in alternating order: a ship routing problem and a knot-

tying problem. They were chosen because of shared goal and verbal in-

teraction orientations. The tasks differed largely in the degree to

which verbal skill was required. In the ship routing problem each group

was asked to work together quickly in order to find the shortest route

for a ship which had to touch five ports. Time for solution was the cri-

terion for productivity. This task was rated by Shaw (1963) to be high

on cooperation requirements, decision verifiability, and intellectual-

manipulative requirements. Each of the three group members had only

partial information sheets regarding the distances and availability of

the routes between the different ports.(See Appendix B.) To assure

group verbal interaction, the group members were told not to show each

other the information sheets randomly distributed to them. However,

they were encouraged to verbally communicate because the problem only

could be solved by thdr working together as a group. To reinforce his

leadership position, the supervisor had previously been trained in set-

ting up a scatter map to record the matrix of distances between the

ports. His function was to serve as the central communication point

in solving the problem. The sooner the group reported a solution, the

better was its productivity score measured by the total elapsed time to

solve the problem. The supervisor then reported the group's readiness











to report a solution by raising his hand. All groups tried to identify

several solutions before reporting a group solution. No group gambled

with the first solution found. When an incorrect solution was reported,

the investigator asked the group to work until it found the correct

solution.

The second structured task required the group members to perform

the physical task of tying one of two unusual knots: the French bowline

and the double becket bend. Only one of the participants admitted prior

knowledge of either of these knots. His individual and group performance

scores were not considered atypical and, therefore, not excluded.

The investigator spent approximately one hour instructing each

supervisor in a general familiarization of the three tasks. Knot tying

proficiency and job instructional techniques were emphasized. Each

supervisor received a personal five foot piece of clothesline and knot

diagram in order to practice the two unusual knots in his leisure time.

(See Appendix C.)

The supervisor instructed his two subordinates how to tie a desig-

nated knot. After the initial instruction period, the two subordinates

were asked to duplicate the knot in the shortest period of time. Two

clotheslines, one for each subordinate, were used to minimize time de-

lays during the knot tying. The investigator determined if the knot was

tied successfully. To maximize the group effort and verbal interaction,

the instructions preceding the task encouraged verbal but no physical

help from group members who had completed tying the knot successfully.

Productivity was the measure of the total elapsed time for the in-

struction by the supervisor and the duplication of the knot by the sub-











ordinates.

The unstructured task.-The unstructured task demanded a creative

product. Each group, after completing the two structured tasks, was

asked to develop a recruiting letter urging college students to join an

all-volunteer Navy. This letter, containing no more than 250 words, was

to be completed in 30 minutes. The groups were told that the letters

would be judged on style, form, persuasiveness and originality.

The Observation

During the participation of the three tasks, the group members were

observed by two observers (one black and one white) through the one-way

mirror. In addition, the investigator used four stop watches to measure

elapsed time of each task and total duration of speech of each of the

three group members. The investigator also used a tape recorder to re-

cord the speech interchange of the group members during the performance

of the three tasks so that group cohesiveness could be measured as de-

scribed later in this chapter.

The Post-task Questionnaire Forms

After the completion of the three tasks by the group, each of the

group members was asked to complete a three-page form evaluating the

tasks and fellow group participants.(See Appendix D.) In addition,

Ziller's self-esteem measure was completed by all participants after

their last scheduled task was completed. (See Appendix E.)

Input Variables

The study involved the independent variable of race in the roles

of subordinates and supervisors in problem solving groups. Race of










subordinate and supervisor was manipulated by the role playing of black

and white students assigned to the specific roles. Students were as-

signed to form an equal number of the three subordinate dyad varieties:

one all-black, one all-white, and one with a black and a white subor-

dinate. These subordinate dyads then joined a supervisor to form a three-

member problem solving group. Each subordinate dyad participated twice:

once with a black supervisor and once with a white supervisor. Older

students who opted to be supervisors were trained in the problem sol-

ving tasks. Each supervisor led three subordinate dyads: each of a

different racial mixture (two black, two white, and one black and one

white ).

Mediating Structure

The leader position of the supervisor was operationalized by intro-

ducing him as being selected because of his training, ability and inter-

est in group members and group activity. The investigator further strength-

ened the leader position by directing the supervisor to initiate and con-

clude all the tasks. For example, in the knot-tying task he was the in-

structor of the other two group members. In the ship routing task, he

served as a central communicator in solving the problem. He reported a

group solution by a signal of his hand. In the recruiting letter, his in-

structions were to pull together the group creative ideas and record them

on a note pad. Hollander and Julian (1970) used this approach in manipu-

lating subordinates' perceptions of the leader's authority, competence,

and motivation.

To further control the group relationship, perception of the com-

patibility of the group members was initially established as suggested by










Katzell et al. (1970b) At the introduction of each group, all partic-

ipants were advised that a biographical questionnaire they had completed

earlier was used to match their background and interests. This question-

naire called for the following data: age, place of birth, number of sib-

lings, community type when growing up, family income bracket, father's

occupation, social class, and attitudes toward a variety of activities

at school, home, etc. (See Appendix A.)

Output Variables

Differences in interaction, behavior, and attitudes were studied in

six separate areas: (1) group productivity on the tasks, (2) group in-

teraction processes, (3) group cohesiveness, (4) individual duration of

speech, (5) individual self-esteem, and (6) individual satisfaction to-

ward the tasks and fellow group members. Several instruments were used

to operationalize these output variables. First, the time required to

solve each of the two structured tasks was a measure of the productivity

of each group. A stop watch recorded the time of task solution. The pro-

ductivity of the unstructured task was measured by semantic differential

ratings of the recruiting letters by thirteen different judges. To con-

trol for order effects, each judge rated a randomly varied order of the

seventy-two letters. Since the product of this task was a recruiting let-

ter aimed at college students, six black and seven white undergraduate male

students did the rating. After a short training period to develop an under-

standing of the four dimensions to be judged, the students rated the letters

in two two-hour time periods.(See Appendix F for letter-rating form.) Fre-

quent breaks were allowed to reduce fatigue. The judges were also paid

three dollars per hour.










Second, the Bales Interaction Process Analysis (IPA) (1950,1953)

was used to operationalize the group processes (Bales and Gerbrands,

1948; Bales and Slater, 1955; Landsberger, 1955; Zdep, 1969; Richards,

1970; Katzell et al., 1970b). The twelve interaction variables were

measured by four trained observers (two blacks and two whites) by obser-

vation through a one-way mirror. Each group session was observed by one

black and one white observer to control for racial perception differences.

The scheduling of observers was based on time availability. No particu-

lar sequence was followed except that all four possible combinations of

observers served. An average interrater reliability coefficienct of .80

was obtained during the ten-hour training period from a set of video tapes

made during the pilot study of the three tasks. Initiations of interaction

in each of the 12 categories were coded and compared for supervisors and

subordinates of both races. Certain categories of the IPA that show the

supervisor initiating interactions on IPA categories 4 (giving suggestions),

5 (giving opinion), and 6 (giving information) were interpreted as direc-

tive leadership. Interaction by the supervisor on IPA categories 7 (asking

for information), 8 (asking for opinion), and 9 (asking for suggestions)

was interpreted as nondirective leadership as suggested by Katzell et al.

(197d .

Third, the proportionate usage of "we" or "I" was measured from re-

corded tapes of the group tasks to determine group cohesiveness. French

(1941) found that this unobtrusive measure related cohesiveness to quality

of interaction. Shaw (1971) concluded that no current measures take into

account all aspects of group spirit.

Fourth, individual group members were measured by three separate stop










watches activated by the experimenter. Bass (1960) and Jaffee (1970)

have shown this technique to be a reliable method of recording ab-

solute duration of speech.

Fifth, expressed self-esteem of each of three group members was

measured by the social self-esteem portion of the Social Orientation

Tasks of Ziller et al. (1969).(See Appendix E.) Ziller reports great-

er validity and utility of this linear ordering measurement of self in

the environment over the usual self-report measures because of limited

verbal demands and low visibility. Correlation of social self-esteem

with status identification by male college students was significant

(r = .46.) Results comparing social self-esteem reported negative or

very low correlations. Ziller suggests that social self-esteem is in

a different psychological domain that measures self-other perceptions

without the demand influence of social desirability.

Sixth, individual satisfaction toward work in the tasks and fel-

low group members (supervision and subordinate) was measured by having

the subjects complete the Job Description Index (JDI) scales of Smith

(1969 ). Only portions of the JDI scales evaluating work, supervision

and fellow subordinates were used in order to fit the particular work

and group environment (see Appendix D).

Procedure

Four graduate students, two black and two white, were hired and

trained as observers. Chapter 3 of Bales (1950) was used as background

training material. A pilot study of the three tasks with three paid

participants (one black and two whites) was run to familiarize the ob-

servers with the activities in each experimental condition. Video tapes










of the pilot study were further used to train the observers and to compute

interrater reliability. Observer training was concluded when reliability

reached .80. Two observers (one black and one white) sat in the obser-

vation room and measured the group interaction by IPA.

Each dyad of two subordinates was scheduled for two two-hour group

sessions. One supervisor was also scheduled to participate in the first

two-hour session, but, after one session, he was replaced with a super-

visor from the opposite race and another two-hour group session was held.

Each dyad of subordinates thus participated together in two two-hour ses-

sions, one with a black supervisor and one with a white supervisor. Each

supervisor returned at other times to lead two additional sessions of

different racially mixed groups. The group sessions were scheduled over

eight weeks in the spring of 1972 with a four-week intermission for final

exams, quarter break, and registration.

For each session, the participants gathered at the experimental room

at the appointed time with the investigator. The participants were in-

structed to sit at the round table with the supervisor in the middle on

a chair with rollers (See Figure 3.) Subjects were informed that the pur-

pose of the study was the analysis of group problem solving effectiveness,

but observation was a necessary part of this analysis. They were told

that the two observers were placed behind the one-way mirror in order not

to distract the participants. A standard set of introductory remarks (See

Appendix G) was read to each group. The group compatibility was estab-

lished by the following statement: The biographical questionnaire like

you filled out the first time you met with me has been successfully used

in determining the compatibility of groups in previous research. Your











group was selected to be compatible so you should enjoy the activity and

perform it efficiently. Leader position of the supervisor was also es-

tablished at this time by the investigator's introduction. Specific

task instructions were then explained and questions answered to clear

up misunderstandings before beginning each task. For each new subordi-

nate dyad a random selection of the two structured tasks was presented.

The second session of the dyad then performed the other task first. In

all group sessions the unstructured task of the recruiting letter was

presented last to avoid possible contamination of the other two tasks

if the participants held negative attitudes toward military recruiting.

After the investigator signaled the supervisor to begin the as-

signed task, the total participation time (duration of speech) of each

group member was recorded with stop watches. When a group member began

to speak, the investigator depressed the start button, allowed the stop

watch to run for the duration of the speech, and stopped the watch by

depressing the stop button when the group member ceased to speak. This

method accumulated total participation time for each group member. A

stop watch also recorded the total time for completion of the two struc-

tured tasks. The unstructured task had a thirty-minute time limit.

Time remaining was announced at the end of each ten-minute period to

ensure task completion.

At the completion of the three tasks, the group members were asked

to take a few minutes to complete a few forms. The investigator gave

the supervisor his forms and asked him to turn around and roll his

chair to the table behind him to reduce peer pressure to complete the

evaluation forms. The investigator also asked each group member to make











his comments on the forms as objectively as possible and related to the

activity involved in the tasks only. This was done to reduce the halo

effect that friends might bestow on each other. After completion of the

satisfaction and self-esteem forms, the investigator thanked the group

members for their participation in the study and asked them not to dis-

cuss details of the study with anyone until after April 30th, since ad-

vanced knowledge of the details would limit the value of the study and

all groups should be equally unprepared. They were also told that the

results of the study would be shared with them if interested. Payment

arrangements were then discussed and the group thanked again.

Interrater reliability.-After the first and second weeks of the

study, interrater reliability sample checks were run to test for dif-

ferences in ratings by the observers. No significant differences were

found at this point. The final reliability comparisons of the white

and black observers were checked with Spearman rank correlations for 12

categories over three dyads. Table 1 shows that white and black raters

were in greater agreement when the dyad of observation was white or mix-

ed rather than black. Whether this difference in correlation was at-

tributable to the race or the response set of the observer is open to

question since all dyads were rated by only one white and one black ob-

server. However, different response sets of raters have been reported

by Fiedler (1967 ). He explained that the differences in French-speaking

and Dutch-speaking judges resulted from different response sets inherent

in the culture.











TABLE 1

Spearman Rank Correlations of White and Black Observers on
Interpersonal Behavior of White, Mixed, and Black Dyads







Bales'IPA Categories White Mixed Black

1. Shows solidarity .969 .966 .940

2. Tension release .978 .679 .775

3. Agrees, accepts .828 .722 .596

4. Gives suggestion .840 .789 .929

5. Gives opinion .955 .871 .847

6. Gives information .934 .807 .863

7. Asks information .847 .927 .822

8. Asks opinion .848 .624 .507

9. Asks suggestion .878 .888 .586

10. Disagrees .901 .912 .873

11. Shows tension .850 .965 .812

12. Shows antagonism .888 .940 .595















CHAPTER III

RESULTS


A few of the hypotheses examined in this study have been tested in

other researches and the results have shown some evidence of direction-

ality. Other directional hypothesis have been suggested by hunches and

stereotyped assumptions held by blacks and whites alike. In order to

ferret out as many differences between blacks and whites as our measures

will test, we decided previously to state and test study results in the

null form and to present these results in this chapter. The discussion

in Chapter IV will bring out the relationships of the null hypotheses to

some directional hypotheses suggested by research evidence, hunches, and

stereotyped assumptions.

The results section is divided into five parts. Part one reports on

differences between black and white supervisors. Part two reports on dif-

ferences between black and white subordinates. Part three presents dif-

ferences between subordinates in black, white, and mixed dyads. Part

four presents differences between black and white supervised dyads. Part

five presents supplementary findings between black and white subordinates

in the mixed dyads.

Differences Between Supervisors

This part presents the results of tests of the null hypotheses that

there are no differences between black and white supervisors in measures

of (a) interpersonal behavior (IPA), (b) style of leadership (IPA), (c)










duration of speech, (d) self-esteem, (e) satisfaction with subordinates,

and (f) satisfaction with work in the tasks.

IPA behavior.-Hypothesis 1 (a) states there are no differences in

interpersonal behavior (IPA) between black and white supervisors. Inter-

personal behavior initiations by black and white supervisors were coded

by two observers into twelve different IPA categories. The total number

of behaviors initiated in all three tasks for each category was converted

to behavior frequencies per unit of time.1 The total frequencies per unit

of time for black and white supervisors then were ranked by the Mann-

Whitney U test (Siegel, 1956) with the assistance of the University of

Florida Statistical Program Library.2 Table 2 shows the U values and

z scores of IPA behavior and indicates that in the aggregate, white super-

visors asked for significantly more opinions (p< .05) and suggestions

(p< .05) than black supervisors. However, Table 2 also shows this dif-

ference was not significant within any one of the three types of dyads.

Thus, hypothesis 1 (a) was rejected because differences were found in

categories 8 and 9.

Styles of leadership.-Hypothesis 1 (b) states there are no differen-

ces between black and white supervisors in leadership style. Two types

of leadership styles were defined: directive and nondirective. Styles

of leadership were measured by summing the frequencies per unit of time

for three categories of interpersonal behavior. Directive leadership was

defined as the sum of behavior categorized by giving suggestions, opinion,

and information (categories 4, 5, and 6 ). Nondirective leadership was

defined as the sum of behaviors categorized by asking for information,






TABLE 2

U Values and Z Scores of IPA Behavior Differences Between Black and White
Supervisors Overall and in Separate White, Mixed, and Black Dyads


White Dyad


IPA Categories

1. Shows solidarity

2. Tension release

3. Agrees, accepts

4. Gives suggestion

5. Gives opinion

6. Gives information

7. Asks information

8. Asks opinion

9. Asks suggestion

10. Disagrees

11. Shows tension

12. Shows antagonism


Mixed Dyad


Supervisors Supervisors
White Black White Black


Black Dyad

Supervisors
White Black


Aggregate of all Dyads

Supervisors Observed
White Black Z

576 720 .8254

721 575 .8225

554 742 1.0643

652 644 .0450

618 678 .3322

702 594 .6082

607 689 .4618

432 864 2,4331 *

471 825 1.9936 *

623 673 .2816

645 651 .0377

662 634 .1573


1








TABLE 2 continued



NOTE: The value of Mann-Whitney U is determined by the number of times that a score of frequencies
per unit of time in the black supervisors with 12 cases precedes a score of frequencies per
unit of time in the white supervisors with 12 cases in the ranking.

A low score indicates high rank, hence, higher frequencies per unit of time
Critical values of U (nl = 12, n2 = 12), two-tailed test Critical value of Z, two-tailed test
p<.05 (U = 37) p<.05 (Z = 1.96)












opinion, and suggestions (categories 7, 8, and 9 ). The sums of the

categories showing a directive style of leadership with all subordi-

nates were ranked and compared for black and white supervisors. The

same procedure was followed for nondirective leadership styles. Table

3 shows U values and z scores which indicate no significant differences

in the aggregate between black and white supervisors in either direc-

tive or nondirective leadership style. Table 3 also shows no signifi-

cant differences between the leadership styles employed by black and

white supervisors when supervising dyads of black, white, or racially

mixed subordinates. Thus, hypothesis 1 (b) was accepted.

Duration of speech.- Hypothesis 1 (c) states that there are no

differences between black and white supervisors in duration of speech.

Duration of speech was measured by the total time talked by each super-

visor in the three tasks. The ranks of these speech durations were com-

pared for black and white supervisors. Table 4 shows no significant

differences between black and white supervisors in duration of speech.

Thus, hypothesis 1 (c) was accepted.

Self-esteem.-Hypothesis 1 (d) states that there are no differences

between black and white supervisors in self-esteem. Self-esteem was

measured at the end of the third group assignment by the social self-

esteem instrument of Ziller et al. (1969 ). Table 4 shows the U values

and z scores of self-esteem and indicates no significant differences be-

tween black and white supervisors in self-esteem. Thus, hypothesis 1

(d) was accepted.






TABLE 3

U Values and Z Scores of Leadership Style Differences Between Black and
White Supervisors with All Subordinates and with White,
Black, or Mixed Racial Dyads of Subordinates


Bales IPA

Directive Leadership Style
(Sum of Categories 4, 5, 6)





Nondirective Leadership Style
(Sum of Categories 7, 8, 9)


All Subordinates

Supervisor Observed
Black White Z


661 635


.1464


803 493 1.7457


White Dyad

Supervisor
Black White


Black Dyad

Supervisor
Black White


Mixed Dyad

Supervisor
Black White


57 87 70 74 89 55






92 52 76 68 100 55


NOTE: A low score indicates high rank, hence, higher frequency per unit of time

Critical value of observed z, two-tailed test Critical value of U (nl = 12, n2 = 12), two-tailed test

p<.05 (Z 1.96) p<.05 (U 37)














TABLE 4

U Values and Z Scores of Duration of Speech, Self-
esteem, Satisfaction with Work and Subordinates
Between Black and White Supervisors


Supervisors
Black White


Duration of speech


Self-esteem


Satisfaction with subordinates
White subordinates
Black subordinates


Satisfaction with work


Observed
Z

.3604


.1530


.1240
.1016
.0508


.0282


NOTE: A low score indicates high rank, hence, higher frequency per unit
of time
Critical values of observed Z (nl = 36, n2 = 36), two-tailed test
S<.05 (z=1.96)











Satisfaction with subordinates.-Hypothesis 1 (e) states that there

are no differences in satisfaction with subordinates between black and

white supervisors. Satisfaction with subordinates was one of the three

job satisfaction measures used from the Job Description Index (JDI)

scales of Smith (1969). Each supervisor was asked to complete a satis-

faction scale after supervising a dyad in the three tasks. Table 4 shows

the U values and z scores of satisfaction with subordinates and indicates

no significant differences between black and white supervisors. Thus,

hypothesis 1 (e) was accepted.

Satisfaction with work in the tasks.- Hypothesis 1 (f) states that

there are no differences in satisfaction with work between black and

white supervisors. Satisfaction with work was another of the JDI scales

completed by the supervisor after supervising each of the three dyads.

Table 4 shows the U values and z scores of satisfaction with work in the

tasks and indicates no significant differences between black and white

supervisors. Thus, hypothesis 1 (f) was accepted.

Differences Between Subordinates

This part presents the results of tests of the null hypotheses that

there are no differences between black and white subordinates in measures

of: (a) interpersonal behavior (IPA), (b) duration of speech, (c) self-

esteem, (d) satisfaction with their fellow subordinates, (e) satisfaction

with their supervisor, and (f) satisfaction with work in the tasks.

IPA behavior.-Hypothesis 2 (a) states there are no differences be-

tween black and white subordinates in interpersonal behavior. Initiations










of interpersonal behavior by black and white subordinates were observed

and coded by two observers in the same manner described in part one.

Behavior frequencies per unit of time were ranked for all black and

white subordinates disregarding racial mixture of the dyad. Table 5

shows the U values and z scores of IPA behavior and indicates that white

subordinates gave significantly more suggestions (p<.Ol) and information

(p<.001) than black subordinates. Thus, hypothesis 2 (a) was rejected

because differences were found in 4 and 6.

Duration of speech.-Hypothesis 2 (b) states there are no differences

between black and white subordinates in duration of speech. Duration of

speech was measured by the total time talked by each subordinate. Table

6 shows the U values and z scores of duration of speech and indicates no

significant difference between black and white subordinates in duration

of speech. Thus, hypothesis 2 (b) was accepted.

Self-esteem.-Hypothesis 2 (c) states there are no differences be-

tween black and white subordinates in self-esteem. Self-esteem was mea-

sured by Ziller's instrument at the end of the third task. The scores

were ranked and compared for all black subordinates and all white subor-

dinates irrespective of the racial mixture of the dyads. Table 6 shows

the U values and z scores of self-esteem and indicates that black subor-

dinates perceived themselves significantly higher (p< .05) in self-esteem

than white subordinates. Hypothesis 2 (b) was therefore rejected.

Satisfaction with fellow subordinates.-Hypothesis 2 (d) states there

are no differences between black and white subordinates in satisfaction













TABLE 5

U Values and Z Scores of IPA Behavior Differences
Between Black and White Subordinates


IPA Categories

1. Shows solidarity

2. Tension release

3. Agrees, accepts

4. Gives suggestion

5. Gives opinion

6. Gives information

7. Asks information

8. Asks opinion

9. Asks suggestion

10. Disagrees

11. Shows tension

12. Shows antagonism


Subordinates
Black White

2648 2536

2848 2336

2788 2395

3342 1842

2871 2313

3374 1810

2707 2477

2778 2406

2738 2446

2416 2768

2470 2713

2564 2619


n


NOTE: A low score indicates high rank, hence, higher frequency per unit
of time
Critical value of U (nl = 72, n2 = 72), two-tailed test
p< .05
** p< .01
*** p< .001


Observed
Z

.2399

1.0295

.7851

2.9947 **

1.1148

3.1265 ***

.4595

.7666

.7183

.7043

.6321

.1328


















TABLE 6

U Values and Z Scores of Duration of Speech, Self-esteem,
Satisfaction with Fellow Subordinates, Supervisors, and Work
Between Black and White Subordinates


Subordinates
Black White


Duration of speech


Self-esteem


Satisfaction with fellow
subordinates
White fellow subordinate
(n1 = 48, n2 = 24)
Black fellow subordinate
(nI = 24, n2 = 48)


Satisfaction with supervisor
White supervisor (nI = 36,
Black supervisor (nl = 36,


Satisfaction with work


2742


1982


1940

352

512


2387
n2=36)740
n2=36)606


2163


NOTE: A low U value indicates high rank, hence, a higher
critical values of observed Z (nl = 72, n2 = 72 or
p<.05, two-tailed test
** p< .01


value
n1 = 36, n2 = 36)


2442


3202



3244

800

640


2797
556
690


3021


Observed
Z

.5994


2.4448 *



2.6106 **

2.6759 **

.7683


.8210
1.0382
.4806


1.7164











with fellow subordinates. Satisfaction with fellow subordinate was another

JDI scale where each subordinate rated the other subordinate in his dyad

at the end of the third task. Table 6 indicates that black subordinates

rated all fellow subordinates significantly higher (p< .01) than white

subordinates did. This satisfaction with the fellow subordinate was

particularly significant (p< .01) when the fellow subordinate was white.

When the fellow subordinate was black, the difference between the black

and white ratings was not significant. Thus, hypothesis 2 (d) was reject-

ed.

Satisfaction with supervisor.-Hypothesis 2 (e) states there are no

differences between black and white subordinates in satisfaction with

their supervisor. Satisfaction with supervisor was another JDI measure

where each subordinate rated his supervisor. Table 6 indicates no signi-

ficant difference between black and white subordinates in satisfactions

with their supervisor whether the supervisor was black or white. Thus,

hypothesis 2 (e) was accepted.

Satisfaction with work in the tasks.-Hypothesis 2 (f) states there

are no differences between black and white subordinates in satisfaction

with work in the tasks. Satisfaction with work was also a JDI scale that

measured satisfaction with the three tasks. Table 6 indicates no signi-

ficant differences between black and white subordinates in satisfaction

with work in the tasks. Thus, hypothesis 2 (f) was accepted.

Differences Between Subordinates in Dyads

This part presents the results of tests of the null hypotheses that











there are no differences between dyads composed of black, white, and

racially mixed subordinates in measures of: (a) interpersonal behavior

(IPA), (b) productivity by type of tasks, (c) cohesiveness, (d) self-

esteem, (e) duration of speech, (f) satisfaction with fellow subordi-

nates, (g) satisfaction with their supervisors, and (h) satisfaction

with work in the tasks.

IPA behavior.-Hypothesis 3 (a) states there are no differences be-

tween dyads composed of black, white, and racially mixed subordinates

in interpersonal behavior. Interpersonal behavior initiations by sub-

ordinates in black, white, and mixed dyads were observed, coded, and

converted to frequencies per unit of time. The frequencies then were

ranked in pairs of the three dyads, since the Mann-Whitney U test can

compare two samples only. All possible comparisons of the three dyads

were ranked. As a result of this breakdown in comparing the three dy-

ads, the .05 significance level was divided by 3 to obtain a protection

level of .0167.4

Table 7 shows U values and observed z scores of IPA behavior and

indicates no significant differences between the three dyads for any of

the 12 IPA categories at the .0167 level. If a less stringent signifi-

cance level of .05 were used, category 9 would be significantly differ-

ent between white and mixed dyads. Subordinates in white dyads asked

for more suggestions than subordinates in mixed dyads. Because of the

number of Mann-Whitney U tests run, this difference may have resulted

from chance. On the basis of the more conservative significance level,

hypothesis 3 (a) was accepted.





TABLE 7

U Values and Z Scores of IPA Behavior Differences Between
Dyads of White, Black, and Mixed-Racial Subordinates


Bales IPA


DYADS
White Black


Shows solidarity

Tension release

Agrees, accepts

Gives suggestion

Gives opinion

Gives information

Asks information

Asks opinion

Asks suggestion

Disagrees

Shows tension

Shows antagonism


1114

1000

1040

968

1119

888

1062

1124

1024

1273

1234

1100


1190

1304

1264

1336

1185

1416

1242

1180

1280

1031

1070

1204


Observed
Z

.2990

1.1234

.8244

1.3446

.2418

1.9309

.6559

.2103

1.0930

.8870

.7786

.4499


DYADS
Mixed Black


1038

918

1099

928

1256

1094

1060

1252

1298

1387

1215

1170


1266

1386

1205

1376

1048

1210

1244

1052

1006

917

1089

1134


Observed
Z

.8971

1.7214

.3884

1.6377

.7657

.4250

.6742

.7570

1.4118

1.7260

.5899

.1669


DYADS
Mixed White


1099

1120

1202

1117

1306

1373

1152

1270

1412

1304

1126

1213


1205

1184

1102

1187

998

931

1152

1034

892

1000

1178

1091


NOTE: A low U value indicates high rank,
Protection level of observed Z
p< .0167 = 2.39


hence, a higher frequency per unit of time


Observed
Z

.4152

.2192

.3664

.2565

1.1248

1.6194

0

.9043

2.3775

1 1169

2496

.5383











Productivity by tasks.-Hypothesis 3 (b) states there are no dif-

ferences in productivity between dyads composed of black, white, and

racially mixed subordinates. Productivity of the two structured tasks

(knot tying and ship routing) was measured by the time required to com-

plete the tasks. Productivity of the unstructured task (letter writing)

was measured by the average ratings of the 13 judges (7 white and 6

black).5

The productivities on the three tasks were compared in pairs of

the three subordinate dyads with the aid of the Mann-Whitney U test.

Table 8 shows U values and z scores and indicates no significant dif-

ferences between the three dyads for measures of productivity in knot

tying, ship routing, and letter writing at the .0167 level. Thus, hy-

pothesis 3 (b) was accepted.

Cohesiveness.-Hypothesis 3 (c) states there are no differences be-

tween dyads composed of black, white and racially mixed subordinates in

cohesiveness. Cohesiveness was measured by the proportionate usage of

"we" to "I" tallied from recorded tapes of the dyad tasks. The result-

ing proportions were ranked with the Mann-Whitney U test. Table 8 shows

the U values and z scores of cohesiveness and indicates no significant

difference between the three dyads in cohesiveness at the .0167 level.

If a less stringent significance level of .05 were used, the mixed dyads

expressed more cohesion than the black dyads. However, at the stated

protection level of .0167, there were no differences and hypothesis 3 (c)

was accepted.





TABLE 8

U Values and Z Scores Differences in Productivity Cohesiveness,
Self-esteem, Duration of Speech, and Satisfaction With Fellow
Worker, Supervisor, and Work Between Dyads of White, Black, and
Mixed-Racial Subordinates


Productivity
Knot
Ship
Letter


DYADS
White Black

1422 882
1390 914
220 356


Observed DYADS
Z White Mixed

1.9796 1314 990
1.7447 1244 1060
1.3923 267 309


Observed
Z B

1.1876
.6744
.4333


DYADS Observed
Lack Mixed Z

964 1340 1.3782
988 1316 1.2022
344 232 1.1654


Cohesiveness

Self-esteem

Duration of speech

Satisfaction with
fellow subordinate

Satisfaction with
supervisor

Satisfaction with work


202 374 1.7750 294 258 .3834 374 178 2.0980


1268 1036

1264 1040


.8526 1314 990 1.1903 1120 1184

.8208 1260 1044 .7951 1060 1244


1462 842 2.2808 1448 856 2.1786 1106 1198


1235 1069

1266 1038


NOTE: A low U value indicates high rank,
Protection level of observed Z
p< .0167 = 2.39


.2357

.6742


.3422


.6099 1094 1210 .4222 992 1312 1.1799


.8364 1243 1061 .6679 1170 1134


.1285


hence, a higher score











Self-esteem.-Hypothesis 3 (d) states there are no differences be-

tween dyads composed of black, white, and racially mixed subordinates

in self-esteem. Self-esteem of individual subordinates was measured by

Ziller's instrument and grouped together in dyad rank comparisons with

other dyads. Table 8 shows U values and z scores of self-esteem and in-

dicates that there are no significant differences between the three dyads

at the .0167 level. Thus, hypothesis 3 (d) was accepted.

Duration of speech.-Hypothesis 3 (e) states there are no differences

between dyads composed of black, white, and racially mixed subordinates

in duration of speech. Duration of speech, as measured by the total

time spoken by each subordinate, was grouped together in dyad rank com-

parisons with other dyads. Table 8 shows U values and z scores and indi-

cates no significant difference in duration of speech between the three

dyads at the .0167 level. Thus, hypothesis 3 (e) was accepted.

Satisfaction with fellow subordinates.- Hypothesis 3 (f) states there

are no differences between dyads composed of black, white, and racially

mixed subordinates in satisfaction with their fellow subordinates. In-

dividual ratings of satisfaction with fellow subordinates were measured

by the JDI scale and grouped together in dyad rank comparisons with other

dyads. Table 8 shows U values and z scores which indicate no difference

between the three dyads at the .0167 level in satisfaction with fellow

subordinates. If the less stringent significance level of .05 were used,

subordinates in black and mixed dyads expressed greater satisfaction with

their fellow subordinates than subordinates in white dyads. However,










hypothesis 3 (f) was accepted at the .0167 level.

Satisfaction with supervisor.-Hypothesis 3 (g) states there are

no differences between dyads composed of black, white, and racially

mixed subordinates in satisfaction with their supervisors. The rat-

ings by individual subordinates of their supervisors were measured by

the JDI scale, grouped by type of dyad, and ranked with the Mann-

Whitney U test. Table 8 shows U values and z scores which indicate

no significant difference between the three dyads in satisfaction with

their supervisors. Thus, hypothesis 3 (g) was accepted.

Satisfaction with work in the tasks.-Hypothesis 3 (h) states there

are no differences between dyads composed of black, white, and racially

mixed subordinates in satisfaction with work in the tasks. Individual

ratings of satisfaction with work were measured by a JDI scale, grouped

by type of dyad, and ranked with the Mann-Whitney U test. Table 8

shows U values and z scores which indicate no significant differences

between the three dyads in satisfaction with work in the tasks. Thus,

hypothesis 3 (h) was accepted.

Differences Between Supervised Dyads

This part presents the results of tests of the null hypotheses that

there are no differences between dyads composed of black, white, and

racially mixed subordinates supervised by black or white supervisors

in measures of: (a) productivity by tasks, (b) cohesiveness, (c) du-

ration of speech, (d) satisfaction with their supervisors, and (e) sat-

isfaction with work in the tasks. The group measures of productivity and

cohesiveness, along with the combined individual measures of duration of

speech, satisfaction with supervisors, and satisfaction with work were











each ranked in a single series for each of the six different group com-

binations of supervisors and subordinate dyads (i.e., white supervisor

with white dyad, white supervisor with black dyad, white supervisor with

mixed dyad, black supervisor with white dyad, black supervisor with black

dyad, and black supervisor with mixed dyad).

Productivity by type of task.-Hypothesis 4 (a) states there are no

differences between black and white supervised dyad combinations in pro-

ductivity in the three tasks. Productivity in terms of completion time

of the two structured tasks (knot tying and ship routing) were ranked by

the Kruskal-Wallis (H) one-way analysis of variance to test whether the

six independent group combinations differed. Table 9 shows the rank sums

and observed H of the productivity measures of the two structured tasks.

No significant differences were found between the different groups. Pro-

ductivity in terms of average judge ratings of the recruiting letters al-

so was compared with the Kruskal-Wallis H test. Table 9 reports the sum-

med ranks and observed H which indicates no significant differences between

the six different groups. Thus, hypothesis 4 (a) was accepted.

Cohesiveness.-Hypothesis 4 (b) states there are no differences between

the black and white supervised dyad combinations in cohesiveness. Cohe-

siveness, as measured by the proportionate usage of "we" to "I" tallied

from recorded tapes of the different groups, were ranked with a Kruskal-

Wallis H test over the six different group combinations. Table 9 shows

the ranked sums and observed H which indicates that cohesiveness of the

groups varies significantly (p< .02). Further analysis of the groups, two

by two, with a Mann-Whitney U test reveals in Table 10 that white dyads







TABLE 9


Rank Sums and Resulting H Values of Six Different Supervised Groups
on Measures of Productivity, Cohesiveness, Duration of Speech,
Satisfaction with Supervisors, and Satisfaction with Work


Productivity


Supervisor
Subordinate Dyad


WHITE
White Black


Mixed


BLACK
White Black


369 473 432 400 518 438

388 570 510 346 386 386

398 281 364 562 459 538


526 355 531 431 340 480 14.17 **


Cohesiveness


Duration of
speech


Satisfaction
with supervisor


380 494 445 437 445 381



489 448 437 359 517 379



483 421 376 443 394 453


Satisfaction
with work


Knot


Letter


Mixed


Observed
H


4.61

4.90

8.28


p< .05
** p< .02


4.32



5.38



4.26








TABLE 10

U Values of Cohesion Differences Between Individual Pair
Comparisons of Supervisors and Subordinate Dyads


SUPERVISOR

White

White


White

White


White

White


Black

Black


Black

Black


Black

Black


DYAD

White

Black


White

Mixed


Black

Mixed


White

Black


White

Mixed


Black

Mixed


37 *

107


74

70


104

40


56

88


67

77


94

50


SUPERVISOR

White

Black


White

Black


White

Black


White

Black


White

Black


White

Black


DYAD

White

White


Mixed

Mixed


Black

White


Black

Mixed


Black

Black


White

Black


NOTE: A low score indicates high rank, hence, greater cohesion
Critical value of U (nl = 72, n2 = 72) two-tailed test

p<.05
p< .02
p< .01












supervised by white supervisors were significantly (p< .05) higher

in cohesiveness than black dyads supervised by white supervisors.6

Thus, hypothesis 4 (b) was rejected.

Duration of speech.-Hypothesis 4 (c) states there are no dif-

ferences between black and white supervised dyad combinations in

duration of speech. The duration of speech measures of the two

subordinates in the six different dyad combinations was totaled

and ranked with the Kruskal-Wallis H test. Table 9 shows the rank

sums and observed H which indicates no significant difference be-

tween the six different groups in duration of speech. Thus, hy-

pothesis 4 (c) was accepted.

Satisfaction with supervisors.-Hypothesis 4 (d) states there

are no differences between black and white supervised dyad combina-

tions in satisfaction with their supervisors. Satisfaction with

supervisor scores from the JDI rating scales of the subordinates

in the dyads were totaled and ranked with Kruskal-Wallis H test.

Table 9 shows ranked sums and observed H which indicate no signi-

ficant difference between the six different groups in satisfaction

with supervisors. Thus, hypothesis 4 (d) was accepted.

Satisfaction with work in the tasks.-Hypothesis 4 (e) states

there are no differences between the black and white supervised

dyad combinations in satisfaction with work in the tasks. Satis-

faction with work from the JDI rating scales of the two subordi-

nates in the dyads were totaled and ranked with the Kruskal-Wallis











H test. Table 9 shows ranked sums and observed H which indicates

no significant difference between the six different groups in

satisfaction with work in the tasks. Thus, hypothesis 4 (e) was

accepted.

Supplementary Findings

When relating the differences between black and white sub-

ordinates in the aggregate comparisons of Table 6 to the dyad

comparisons in Table 8, certain inconsistencies became apparent.

For example, Table 6 shows that the black subordinates in the ag-

gregate had significantly (p<.05) higher self-esteem than white

subordinates. However, Table 8 shows no difference in self-esteem

in the comparison of the homogeneous (all black and all white) dy-

ads of subordinates. An answer to this inconsistency was found in

a further analysis of black and white subordinates in the hetero-

geneous (mixed) dyads.

Difference in Heterogeneous Dyads

Statistical tests were made to determine if differences existed

within the heterogeneous dyads on measures of self-esteem, duration

of speech, satisfaction with fellow subordinate, satisfaction with

supervisor, satisfaction with work in the tasks, and IPA behavior.

Although Mann-Whitney U tests were used to test for differences,

mean comparisons were presented for illustration of the variance

between subordinates in homogeneous and heterogeneous dyads.

Self-esteem.-Table 11 indicates that although black and white

subordinates reported no difference in self-esteem when they worked












TABLE 11

Mean and Z Score Differences Between Black and White
Subordinates in Homogeneous and Heterogeneous Dyads


SELF-ESTEEM




DURATION OF
SPEECH




SATISFACTION
WITH FELLOW
SUBORDINATE




SATISFACTION
WITH SUPERVISOR




SATISFACTION
WITH WORK


Black

White


Black


White


Black



White


Black


White


Black


White


Critical value of observed
* p< .05
** p< .01


Homogeneous
Dyads
Means

26.42

25.54


9.66


9.39


39.27



36.06


35.71


34.60


30.17


29.38


Observed
Z


.8526





.8208







2.2808 *





.6099





.8364


Heterogeneous
Dyads
Means

29.17

23.50


7.89


10.27


41.33



37.25


34.79


33.50


33.29


29.33


Z, two-tailed test


Observed
Z


2.7361 **





2.2892 *







1.1087





.4654





1.9633 *












in all black or all white dyads, blacks exhibited a significantly

(p<.01) higher self-esteem than whites when grouped in hetero-

geneous dyads. Thus, the aggregate difference in self-esteem be-

tween blacks and whites reported in Table 6 is due to the tenden-

cy for blacks to show higher self-esteem scores and whites to ex-

hibit lower scores when placed in heterogeneous as compared to homo-

geneous dyads.

Duration of speech.-Although no difference in duration of

speech was found between black and white subordinates in the ag-

gregate comparison (Table 6) or in the homogeneous dyad comparison

(Table 8), Table 11 indicates that white subordinates illustrated

a significantly (p<.05) higher duration of speech than black sub-

ordinates in the heterogeneous dyads. Table 11 shows that the dif-

ference resulted because white subordinates had a tendency to talk

more and black subordinates had a tendency to talk less when placed

in heterogeneous as compared to homogeneous dyads.

Satisfaction with fellow subordinate.-Table 11 shows the sin-

gle pair comparison between black and white dyads where blacks

were significantly (p< .05) more satisfied with their fellow sub-

ordinate. When placed in heterogeneous as compared to homogeneous

dyads, the difference between blacks and whites was not significant.

Satisfaction with supervisor.-Table 11 shows z scores which

indicates no difference between black and white subordinates in

heterogeneous or homogeneous dyads. This finding confirms the












result found in Table 6 that race of the supervisor did not affect

the satisfaction expressed by black or white subordinates of their

supervisor.

Satisfaction with work.-Table 11 indicates that although black

and white subordinates reported no difference in satisfaction with

work when they worked in all black or all white dyads, blacks ex-

hibited significantly (p< .05) greater satisfaction with work than

whites when grouped in heterogeneous dyads. However, this differ-

ence was not great enough to affect the difference in the aggregate

comparison of black and white subordinates.

IPA behavior.-IPA behavior of the black and white subordinates

was also studied more thoroughly to determine if the type of inter-

personal behavior differed within the heterogeneous dyad. Table 12

reports that white subordinates gave significantly more suggestions

(p <.001) and information (p< .01) than their fellow black subordi-

nate. The increase in duration of speech reported in Table 11 for

white subordinates may have resulted from the increase in IPA be-

havior by white subordinates.

IPA with different supervisors.-Throughout the study we have

seen few, if any, differences traceable to the differences of race

of the supervisors. Table 13 reports that IPA behavior of black and

white subordinates differed on only one of the IPA categories in

heterogeneous dyads. White subordinates gave suggestions signi-

ficantly more (p< .01) with white supervisors than did black











TABLE 12

U Value and Z Score Differences in Interpersonal Behavior
Between Black and White Subordinates in Heterogeneous Dyads


Observed
Black White Z
Bales IPA Categories

1. Shows solidarity 286 290 .0441

2. Tension release 293 283 .1034

3. Agrees, accepts 295 281 .1443

4. Gives suggestion 458 118 3.5055 **

5. Gives opinion 374 202 1.7733

6. Gives information 414 162 2.5981 **

7. Asks information 286 290 .0309

8. Asks opinion 334 242 1.0164

9. Ask suggestion 274 302 .4088

10. Disagrees 281 295 .1455

11. Shows tension 290 286 .0678

12. Shows antagonism 253 323 .8988








NOTE: A lower U value indicates higher rank, hence, higher frequencies
per unit of time
Critical value of observed Z (nl = 48, n2 = 48), two-tailed test
p< .05
** p< .01
*** p <.001









TABLE 13

U Values of IPA Behavior Differences Between White and Black Subordinates in
Heterogeneous Dyads with White and Black Supervisors


White Supervisor Black Supervisor

Bales IPA White Black White Black

1. Shows solidarity 102 42 54 90

2. Tension release 60 84 74 70

3. Agrees, accepts 75 69 48 96

4. Gives suggestion 23 ** 121 62 82

5. Gives opinion 48 96 76 68

6. Gives information 62 82 56 88

7. Asks information 72 72 69 70

8. Asks opinion 50 94 68 76

9. Asks suggestion 70 74 76 68

10. Disagrees 54 90 63 81

11. Shows tension 74 70 70 74

12. Shows antagonism 61 83 82 62
0'
NOTE: A low score indicates high rank, hence, higher frequencies per unit of time
Critical value of U (nl = 12, n2 =12), two-tailed test p<.05 ** p< .01












subordinates. The interpersonal behavior (IPA) of black and

white subordinates did not significantly vary when supervised

by blacks.

Differences Between Heterogeneous and Homogeneous Dyads

A comparison of the differences between heterogeneous and

homogeneous dyads was also made. However, since the subordinates

in the homogeneous dyads were not the same as the subordinates in

the heterogeneous dyads, the comparison was limited. A contrast

test was made to determine if the differences between black and

white subordinates in the heterogeneous dyads were the same as the

differences between black and white subordinates in the two homo-

geneous dyads. This contrast test was made for self-esteem, dura-

tion of speech, satisfaction with fellow subordinate, and satis-

faction with work. Table 14 reports no significant differences

between subordinates in heterogeneous and homogeneous dyads.













TABLE 14

Mean Differences and Test Statistics Values Between Black and
White Subordinates in Homogeneous and Heterogeneous Dyads
in Measures of Self-esteem, Duration of Speech,
Satisfaction with Fellow Subordinate, and
Satisfaction with Work



DYADS


Homogeneous Heterogeneous
(w ) ( s )


Self-esteem


Duration of speech


Satisfaction with
fellow subordinate


Satisfaction with
work


3.50



.69


5.75


2.40



3.75



3.96


Test
Statistic

1.47


1.23


1.54


NOTE: Critical value of test statistic:
T.S. > 1.96 = p< .05


T.S. computed by Formula


-(G b)

2 2 2
+( + 0 + 0-
1 2 3
724 96


= T.S.











Chapter III

Notes



iThe adjustment for time was necessary because of unequal time

spent performing the three tasks by the different groups.



2The Mann-Whitney U test was chosen as the primary statistical

test because this study employs two or more independent samples with

measures that fit an ordinal scale. Siegel (1956) compares its power

efficiency to that of 95 percent of the T-test even for moderate-sized

samples.



3Although parametric tests are traditionally used to test for

differences in measures of duration of speech, self-esteem, satisfac-

tion, productivity, and cohesiveness, the Mann-Whitney U test was

used throughout the study because many of the populations were neither

equal in variance nor normally distributed. However, parametric F

tests were used to compare parametric results with nonparametric

results. Few differences were noted.



4Duncan (1955) recommends this procedure as protection against

finding wrong differences in multiple tests. This is thought to be a

conservative measure.





65




5The interjudge reliability of the judges' ratings was compared

with the Kendall coefficient of concordance. The resulting Kendall

W = .292 was tested for significance. We can conclude with consider-

able assurance (p<.001) that the agreement among the thirteen judges

is higher than it would be by chance.



6Champion (1970) states this procedure is necessary in order to

further identify where the difference lies.


















CHAPTER IV

SUMMARY, CONCLUSIONS, DISCUSSION, AND
IMPLICATIONS FOR FUTURE RESEARCH


This chapter is divided into four parts. Part one presents

a summary of the study design. Part two presents conclusions of

the test results of the null hypotheses. Part three discusses

possible reasons for the results in conjunction with some direc-

tional hypotheses suggested by research evidence, hunches, and

stereotyped assumptions. Part four presents implications for

future research.

Summary

The purpose of this study was to investigate in a simulated

job setting differences in attitudes and interaction behavior be-

tween (a) black and white supervisors, (b) black and white subor-

dinates, (c) black, white, and racially mixed dyads of subordinates,

and (d) black and white supervised dyads of black, white, and

racially mixed subordinates.

The specific null hypotheses tested were:

1. There are no differences between black and white super-
visors in measures of: (a) interpersonal behavior (IPA),
(b) style of leadership, (c) duration of speech, (d)
self-esteem, (e) satisfaction with subordinates, and (f)
satisfaction with work in the tasks.











2. There are no differences between black and white subor-
dinates in measures of: (a) interpersonal behavior (IPA),
(b) duration of speech, (c) self-esteem, (d) satisfaction
with fellow subordinate, (e) satisfaction with their su-
pervisor, and (f) satisfaction with work in the tasks.

3. There are no differences between dyads composed of black,
white, and racially mixed subordinates in measures of:
(a) interpersonal behavior (IPA), (b) productivity of
the three tasks, (c) cohesiveness, (d) self-esteem, (e)
duration of speech, (f) satisfaction with fellow subordi-
nates, (g) satisfaction with their supervisor, and (h)
satisfaction with work in the tasks.

4. There are no differences between black and white super-
vised dyads composed of black, white, and racially mixed
subordinates in measures of: (a) productivity of the
three tasks, (b) cohesiveness, (c) duration of speech,
(d) satisfaction with their supervisors, and (e) satis-
faction with work in the tasks.

A sample of 96 male undergraduate students, 48 black and 48

white, were hired for the study. Each of the 12 black and 12 white

older students selected to be supervisors was randomly assigned

to supervise three different racially mixed dyads of subordinates

(one all-black, one all-white, and one with a black and a white

subordinate.) The remaining 72 students, 36 black and 36 white,

was randomly assigned to 1 of the 3 subordinate dyads that partic-

ipated once with a black supervisor and once with a white super-

visor. All possible orders of supervisor and dyad combinations

was represented twice to reduce the effect of order bias.

Leader position of the supervisors and compatibility of the

supervisor and the subordinates in the dyads was established by

verbal instructions of the investigator.

Each group of a supervisor and a subordinate dyad was given













three similar problem solving tasks, two structured and one un-

structured. After performing the last of the three tasks, both

supervisor and subordinates completed satisfaction rating forms

(JDI) of each other and work in the tasks. At the end of their

group session, each supervisor and subordinate completed a self-

esteem form.

During the task performance, two trained graduate students

(one black and one white) observed and coded the group interaction

by Bales IPA through a one-way mirror. Other output measures re-

corded by the investigator were: (1) the duration of speech for

each group member, (2) productivity of the three tasks, and (3)

cohesiveness.

Conclusions

The results of the null hypotheses tested in this study are

summarized in Table 15 and discussed below.

Differences between supervisors.-As compared to white super-

visors, black supervisors exhibited:

a) Less asking for opinion
b) Less asking for suggestion

No significant differences were found in other IPA categories,

style of leadership, duration of speech, self-esteem, and satis-

faction with subordinates and work.

Difference between subordinates.-As compared to white subor-

dinates, black subordinates exhibited:








TABLE 15

Summary of Null Hypotheses Results*

Rejected
Hypotheses or Table
Number Hypotheses Accepted P No.

SUPERVISOR

la IPA Rejected 3
W >B #8 .05
W >B #9 .05
lb Style of leadership Accepted 4
Ic Duration of speech Accepted 5
Id Self-esteem Accepted 5
le Satisfaction with subordinates Accepted 5
If Satisfaction with work Accepted 5

SUBORDINATES

2a IPA Rejected 3
W >B #4 .01
W >B #6 .001
2b Duration of speech Accepted 5
2c Self-esteem Rejected 5
B >W .05
2d Satisfaction with fellow sub.
(if fellow subordinate is white) Rejected 5
B >W .01
(if fellow subordinate is black) Accepted 5
2e Satisfaction with supervisor Accepted 5
2f Satisfaction with work Accepted 5

SUBORDINATE DYADS

3a IPA Accepted 6
3b Productivity of tasks Accepted 7
3c Cohesiveness Accepted 7
3d Self-esteem Accepted 7
3e Duration of speech Accepted 7
3f Satisfaction with fellow subord. Accepted 7
3g Satisfaction with supervisor Accepted 7
3h Satisfaction with work Accepted 7

SUPERVISED DYADS

4a Productivity Accepted 8
4b Cohesiveness Rejected 9
(W dyad with W Supv.>B dyad with W Supv.) .05
4c Duration of speech Accepted 8
4d Satisfaction with supervisor Accepted 8
4e Satisfaction with work Accepted 8

*All hypotheses were tested at the .05 level or beyond.











a) Less giving of suggestions
b) Less giving of information
c) Higher self-esteem
d) Greater satisfaction with white fellow subordinate.

No significant differences were found in other IPA categories,

duration of speech, and satisfaction with fellow black subordi-

nate, supervisor, and work.

Differences between subordinate dyads.-No significant differ-

ences were found in group measures of interpersonal behavior (IPA),

productivity of the three tasks, cohesiveness, self-esteem, du-

ration of speech, satisfaction with fellow subordinate, supervisor,

and work between black, white, and racially mixed dyads of subor-

dinates.

Differences between supervised dyads.-As compared to white dyads

supervised by white supervisors, black dyads supervised by white

supervisors exhibited less cohesiveness. No significant differences

were found in group measures of productivity, duration of speech,

and satisfaction with supervisor and work between the six different

supervised dyads.

Supplementary Findings

Table 16 summarizes additional conclusions that were reached

from testing differences between black and white subordinates in

heterogeneous dyads.

Differences between subordinates in heterogeneous dyads.-As

compared to white subordinates in the heterogeneous dyads, black











TABLE 16

Summary of Supplementary Findings





Measures Differences

Between Subordinates in Heterogeneous Dyads


IPA
Category #8
Category #9

Duration of speech

Self-esteem

Satisfaction with fellow
subordinate

Satisfaction with supervisor

Satisfaction with work

IPA with black supervisor

IPA with white supervisor
Category #4

Contrast of Homogeneous

Self-esteem

Duration of speech

Satisfaction with work

Satisfaction with fellow
subordinate


Blacks lower than whites
Blacks lower than whites

Blacks lower than whites

Blacks higher than whites


No difference

No difference

Blacks higher than whites

No difference


Blacks lower than whites

Dyads with Heterogeneous Dyads

No difference

No difference

No difference


No difference












subordinates exhibited:

a) Less giving of suggestions
b) Less giving of information
c) Less giving of suggestions when with a white supervisor
d) Higher self-esteem
e) Less duration of speech
f) Greater satisfaction with work in the tasks.

No significant differences were found in other IPA categories, sat-

isfaction with fellow subordinate, or with supervisor between black

and white subordinates in heterogeneous dyads.

Differences between heterogeneous and homogeneous dyads.-When

comparing differences between subordinates in the homogeneous dyads

with differences between subordinates in the heterogeneous dyads, no

significant differences were found in individual measures of self-

esteem, duration of speech, and satisfaction with fellow subordinates

and work (See Table 16.)

Discussion

In considering the results of this study, it is important to

compare them with some directional hypotheses suggested by other

research evidence, hunches, and stereotyped assumptions. This com-

parison may be more meaningful in a framework that focuses on group

composition beginning at the lowest to the highest level of group

structure within the various groups in this study. Within the fol-

lowing framework we can examine the effect of composition on indi-

vidual and group attitudes and behavior:

a) subordinates in heterogeneous dyads











b) subordinate dyads
c) subordinates
d) supervisors, and
e) the supervised dyads.

Differences between subordinates in heterogeneous dyads.-The

major differences between blacks and whites found in this study

appear to have originated in the interactions of heterogeneous

dyads. The differences outlined in Table 16 also appear to be

interdependent.

A number of studies, Katz & Benjamin, 1960; Katz & Cohen, 1962;

Katz et al., 1958; Katz & Greenbaum, 1963; and Lefcourt & Ladwig,

1965, indicated that blacks tend to feel inadequate and low in

self-esteem compared to whites. However, a recent study by

Douglas (1969) of black and white junior high school students pre-

sents evidence that black youths express greater self-esteem than

whites particularly in the school self-esteem of the Coopersmith

Inventory. Although the findings of this study appear to agree with

the findings of Douglas, it is important to note that the difference

in self-esteem between black and white subordinates was a function

of an increase for blacks and a corresponding decrease for whites

compared to subordinates in homogeneous dyads. The apparent change

in self-esteem for both blacks and whites may have been associated

with the balanced heterogeneity. Blacks were in equal number as

well as equal status with whites. Both black and white students may

recognize that equal or balanced heterogeneity and an equal status












contact in a predominately white university is unusual. The bal-

anced heterogeneity with equal status may have caused the blacks

to feel less threatened and more satisfied and the whites to feel

more threatened and less satisfied. This possibility is supported

by the greater satisfaction with work expressed by the black subor-

dinates in the heterogeneous dyads. If this is true, and conversa-

tions with black participants after the study suggest it is, a bal-

anced heterogeneity and an equal status contact situation may be one

of the better ways to gain the satisfaction and cooperation of

blacks. The opportunityfor an equal status contact may have caused

a high sense of self-esteem because the reference group for the

black student appeared to be persons from a lower social class (See

Figure 1.) Few of his black peers from his described social class

would be expected to attend college -- particularly, the University

of Florida. Long et al. (1966) also found a similar anomaly among

Indian school children. Children from the lowest caste reported

higher self-esteem because few children from that caste attended

school. Long et al. (1966) reasoned that the lowest caste child

percieved higher status within his frame of reference. In Ziller's

construct of self-esteem, the field of comparison becomes crucial.

In addition, the white student who attends the university as

a matter of course could find the intergroup equal status contact

with a black student in a heterogeneous dyad incongruent and de-

meaning, thus, lowering his self-esteem compared to his reference











group of other white persons. Because the white subordinate may have

felt a reduced self-image, he may have reacted in a way that would per-

mit him to gain control of the situation. This possibility is support-

ted by his increased duration of speech (see Table 11). This signifi-

cant difference in speaking appears to be in the form of the cooperative

leadership acts of giving more suggestions and information. Richards &

Cuffee (1971) and Jaffee & Whitacre (1970) found these IPA categories

and higher duration of speech to be positively related to leadership

roles.

From the above list of differences between black and white subordi-

nates in heterogeneous dyads, we can begin to recognize the effect of

group composition on individual behavior.

Differences between subordinate dyads.-Table 15 presents a summary

that indicates no differences were found in group and combined indivi-

dual behaviors of interpersonal behavior, productivity, cohesiveness.

self-esteem, duration of speech, and satisfaction with fellow subordi-

nate, supervisor, and work between black, white, and mixed dyads of

subordinates. The reasons for the above and other lack of differences

in this study are not at all clear. At least five possibilities present

themselves: (1) There, indeed, may be no differences between the

subordinate dyads. (2) Differences in attitude and behavior did not

appear because the study time for groups was too short. (3) The

differences that may normally occur between heterogeneous and homo-

geneous dyads did not appear because of an exceptionally powerful

compatibility treatment. (4) Differences occurred, but faulty











instruments prevented their detection. (5) The individual differences

that did occur cancelled out in the combining of the dyadic scores and

were not dysfunctional for group behavior. Of these possibilities,

only the last one could be supported by available data.

Differences between subordinates.-The differences reported in the

aggregate comparison between black and white subordinates must be inter-

preted in light of the differences found between black and white subor-

dinates in heterogeneous dyads and the lack of differences found between

subordinate dyads. The differences reported in the heterogeneous dyad

of self-esteem and IPA categories of giving more information and

suggestion were potent enough (p< .01, .001 and .01 respectively) to

be manifested in the aggregate comparison. However, the other differ-

ences of duration of speech and satisfaction with work were not strong

enough (p< .05) to affect the aggregate.

The only other aggregate difference reported was that blacks ex-

pressed more satisfaction with their fellow white subordinate than

whites (see Table 15). Although Table 11 shows that the satisfaction

with fellow subordinate score of black subordinates tended to be higher

than whites at all comparison levels, it was not significant except in

the comparison of white subordinates. Further analysis of the difference

revealed that black subordinates were more satisfied with their fellow

white subordinate (heterogeneous dyads) than were white subordinates

(homogeneous dyads). From this we can again conclude that most of the

subordinate differences originated in the heterogeneous dyads.









Differences between supervisors.-The few differences found between

black and white supervisors in this study were contrary to some re-

search evidence and a few predictive hunches.

Richards (1970) found that black supervisors gave fewer suggestions

than white supervisors. However, in this study the blacks supervised

only whites in a business game. In addition, the unfamiliar situation-

al setting of supervising whites only in a business game may have caused

blacks to be reluctant to make suggestions. In contrast, the present

investigation emphasized equal status and training for supervisors.

It may be that this produced an equal status contact that provided

the black supervisor with a sense of confidence that was expressed by

an equally high self-esteem and a reduced need for opinion and suggestions

than white supervisors as presented in Table 15.

Another possible explanation for this racial difference in inter-

personal behavior could be that the white supervisors were especially

perceptive of the race variable and the leadership techniques that tend

to increase group participation. Because of a university administra-

tive confrontation with a small minority of black students the previous

year in which many of the white students were sympathetic to the black

cause, the white supervisors may have deliberately tried to involve

the black subordinates in the tasks.1

Another explanation for the lack of differences between black and

white supervisors is the small sample of black and white supervisors

participating. Although 72 supervisory situations were compared, only

24 different students 12 black and 12 white participated. Caution











must be exercised in generalizing from a sample of 12 black and 12

white college students who role-played the supervisory positions.

Differences between supervised dyads.-Contrary to several hunches

and directional hypotheses suggested by King & Bass (1970), Ferman

et al. (1968) and Kochman (1969), the composition of these six differ-

ent supervisor-subordinate groups did not appear to affect group be-

havior as measured. Only the cohesive difference between white

and black dyads supervised by white supervisors appearsto fit the

hypotheses of King & Bass (1970). They contend that all groups of one

race should be more cohesive. Nevertheless, Speroff & Kerr (1952) and

Hoffman & Maier (1961) assert that we should also expect less cohesive-

ness and lower productivity from groups with heterogeneous dyads. Per-

haps the tendency for higher satisfaction with fellow subordinates by

both black and white subordinates shown in Table 11 may explain the

lack of cohesiveness difference. This explanation receives some sup-

port from Guttentag (1970) who found that cohesiveness increases when

members feel attracted to one another.

However, this does not explain why black dyads did not appear

higher in cohesion. Ferman et al. (1968) found that blacks appear to

be high on affiliation which should result in greater cohesion and

satisfaction scores. The discussion about the subordinates in the

heterogeneous dyads raises some question about any study that does

not recognize that blacks are no more homogeneous than whites.

The conclusion that the racial composition of these groups did not

significantly affect group productivity on the three tasks finds some











support from the Belgian Navy study of Fiedler (1966) with French-

speaking and Dutch-speaking men. He found "strikingly small" differ-

ences in performance between homogeneous and heterogeneous groups.

Fiedler, Meuwese & Oonk (1961) also found no difference in Dutch groups

differing in religious and subcultural homogeneity. If we can equate

language and religious differences with racial differences, we may be-

gin to question the effect of racial composition on some aspects of

group behavior.

Another possible explanation was suggested by Ziller (1972). His

paradigm calls for future research in the relationship between group

composition (heterogeneity or homogeneity) and group productivity to

consider the intervening variables of affective group structure, group

processes, and group perception. According to Ziller (1972), these

intervening variables can be defined with variables already measured

in this study (i.e., affective group structure cohesiveness, group

processes amount of conflict expressed, and group perception indi-

vidual differences). An analysis of the lack of differences found in

the intervening variables in terms of our study helps us to understand

why the group composition (heterogeneity or homogeneity) did not ad-

versely affect group productivity. For example, cohesiveness (affec-

tive group structure) differed only between two groups; amount of

conflict expressed (group processes) did not differ between supervisor

or dyads from IPA observations (Tables 2 and 7); and individual differ-

ences (group perception) differed only in the heterogeneous dyad. From

this we might be able to conclude that racial composition did not











appear to adversely affect group productivity.

The lack of difference in leadership style is contrary to two

hypotheses by King & Bass (1970). They hypothesized that black super-

visors would exercise more nondirective leadership with white subordi-

nates than white supervisors because blacks would want to minimize

their leadership position and the status incongruity of their subordi-

nates by an open and participative leadership style. King & Bass (1970)

also theorized that white supervisors exercised more directive leader-

ship with black subordinates than black supervisors because racist

attitudes conscious or unconscious would legitimatize a directive

style.

The reasons that leadership style differences were not found in

this study (see Table 3) might be related to the peer association of

the students participating and to the temporary and artificial nature

of the supervisor-subordinate relationship. A student might be more

reluctant to be directive to a peer rather than a real subordinate. In

addition, it appears from Table 2 that both black and white supervisors

maintained their style of leadership regardless of racial character-

istics of the dyad being supervised. This lack of leadership style

difference also may be related to lack of difference in satisfaction

with supervisor expressed by their subordinates.

Another directional hypothesis not supported by this study was

Kochman's (1969) field study indicating that all-black groups "rapped"

more than other groups. A contrast in goal specificity and goal

orientation between this study and an industrial study may be able









to explain the absence of differences in "rapping" as measured by

duration of speech. In the uncertainty of the industrial world, blacks

may have difficulty in understanding from their white supervisors pre-

cisely the what, how, and why of the job. It does seem logical that

this uncertainty would lead to "rapping with their soul brothers" for

information and security. However, this study emphasized goal clarity

and goal specificity particularly in the repeated session by both

supervisors and subordinates.

Limitations and Implications for Future Research

Before discussing the implications it is necessary to point out

that the generalizability of this study is limited by its use of col-

lege students at the University of Florida as the subject population

and by its simulation of a supervisor-subordinate relationship in the

laboratory. Replication of this study at a non-southern university

or in the natural world might produce different results.

Another limitation of the study was the small sample of black

and white supervisors participating. Caution must be exercised in

generalizing from a sample of twelve black and twelve white students

to an entire race of people. In addition, the investigator recognizes

that the reported differences between black and white subordinates may

have been due to social class differences identified in Figure I.

The understanding of the impact of racial composition on group

attitudes and behavior represents a desirable goal in the literature

of group dynamics. The data in this dissertation suggest that few of

the individual differences between black and white supervisors and sub-

ordinates affected group productivity. Most of the reported individual








differences occurred between black and white subordinates in the hetero-

geneous dyads. Even these differences did not appear to perceptively

affect group behavior.

Although there are many important questions which need answering

before our knowledge of the impact of racial composition on groups is

adequate, some of the more pressing ones are: Will a replication of

this study yield similar results? Will a field study under hierarchi-

cally determined supervisor-subordinate relationships yield similar

results? Would the assignment of the same participants to both homo-

geneous and heterogeneous groups produce different results? Is the

compatibility treatment responsible for the lack of differences in

group behavior? If so, can compatibility treatments be helpful to in-

dustry in structuring group behavior? What is the significance of the

balanced heterogeneity in an equal status contact between blacks and

whites? Would an unbalanced heterogeneity produce different results

on individual and group behavior?






83




Chapter IV

Notes



1During the month of the study, a black male became the first

elected black president of the student body in this predominantly white

university.































APPENDICES































APPENDIX A


BIOGRAPHICAL QUESTIONNAIRE




















Biographical Information Instructions


Please complete this questionnaire if you are interested in par-

ticipating in this study. This information will be used in scheduling

you for your group assignment and in mailing your check to you. These

questions are meant to tell us more about you as an individual. The

information you give us will be strictly confidential. No one but the

researchers will see it.




Thank you.












BIOGRAPHICAL QUESTIONNAIRE

Part I



INSTRUCTIONS: It will help us in matching you with others in the study
to know something about your background. Please answer
the following questions.

NAME AGE

ADDRESS TELEPHONE
(for mailing of check)
SOCIAL SECURITY NUMBER

Please indicate your marital status by checking ( ) one of the following
categories.

Married? Divorced

Single? Separated?_

How many dependents do you have? (exclude yourself)

Where were you born?
city state


Where did you live during the time you were growing up? Please check one.


farm community medium-sized or large city

small town hard to say we moved often


When you were growing up, was your family's income: (Please check one.)

higher than average for the country?

about average for the country?

less than average for the country?

very much below average for the country?

so low we were mostly on welfare?














Whom did you live with most of the time while you were growing up?
(Please check one.)

your mother and father

your mother only

your father only

others (please specify relationship to you, if any)





How many brothers and sisters do you have?


You have seen in the paper stories about social class differences. In which
of these classes do you think you really belong? (Please check one.)


upper class working class


middle class poverty class


What was your father's occupation or major job duty? (Please check one.)


unskilled clerical or white collar


semi-skilled professional


skilled














Part II


INSTRUCTIONS: Now, please tell us
(Answer by checking



LIKE
VERY
MUCH


about how you feel about some things.
how much you like each item.)


NEITHER
LIKE NOR
DON'T DON'T VERY
LIKE LIKE LIKE MUCH


The academic life
at UF

What you do in your
spare time

Your family life

Friends at school

Taking part in social
activities

The way other students
treat you

Your life in general

A life in the Navy
(for others)




University of Florida Home Page
© 2004 - 2010 University of Florida George A. Smathers Libraries.
All rights reserved.

Acceptable Use, Copyright, and Disclaimer Statement
Last updated October 10, 2010 - - mvs