Citation
Conformity, majority effect, and perceived competency

Material Information

Title:
Conformity, majority effect, and perceived competency
Creator:
Reitan, Harold Theodore, 1928-
Publication Date:
Language:
English
Physical Description:
viii, 97 leaves : illus. ; 28 cm.

Subjects

Subjects / Keywords:
Analysis of variance ( jstor )
Conformity ( jstor )
Error rates ( jstor )
Experimentation ( jstor )
Judgment ( jstor )
Mental stimulation ( jstor )
Pressure control ( jstor )
Questionnaires ( jstor )
Self ( jstor )
Squares ( jstor )
Conformity ( lcsh )
Dissertations, Academic -- Psychology -- UF ( lcsh )
Psychology thesis Ph. D ( lcsh )
Genre:
bibliography ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis - University of Florida.
Bibliography:
Bibliography: leaves 93-96.
General Note:
Manuscript copy.
General Note:
Vita.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
This item is presumed in the public domain according to the terms of the Retrospective Dissertation Scanning (RDS) policy, which may be viewed at http://ufdc.ufl.edu/AA00007596/00001. The University of Florida George A. Smathers Libraries respect the intellectual property rights of others and do not claim any copyright interest in this item. Users of this work have responsibility for determining copyright status prior to reusing, publishing or reproducing this item for purposes other than what is allowed by fair use or other copyright exemptions. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder. The Smathers Libraries would like to learn more about this item and invite individuals or organizations to contact the RDS coordinator (ufdissertations@uflib.ufl.edu) with any additional information they can provide.
Resource Identifier:
021895627 ( ALEPH )
13421941 ( OCLC )

Downloads

This item has the following downloads:


Full Text















CONFORMITY, MAJORITY EFFECT,
AND PERCEIVED COMPETENCY












By
HAROLD T. REITAN













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
December, 1967






















ACKNOW~LEDG 4ENTS


The writer wishes to express his sincere appreciation to Dr. Marvin E. Shaw, chairman of the supervisory committee, for his generous assistance, his welcomed encouragement, and his valuable guidance throughout the graduate career of the writer, as well as during the preparation of this dissertation. Sincere thanks are also extended to Dr. Jack m. Wright, Dr. Stephen T. Margrulis, Dr. Henry S. Pennypacker, Dr. Sidney M. Jourard, and Dr. E. Wilbur Bock for their advice and helpful cooperation.


ii


















TABLE OF CONTENTS


ACKNOWLEDGMENTS......... .. .. ..

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

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

CHAPTER
T Tmr Clnhlr'1TnVfl


II EXT


r, J... .. .. .. .. .

Situational Variables.... .. .. Purpose. .......... .. .

~ERIENT 1. ......... .. .

Hypotheses ...........
Method .............

Experimental Design ....
Subjects......... .. .. ..
Apparatus .........
Stimulus Materials... ....
Procedure .........

Results. .......... .. .

Scoring ..........
Analysis of Conformity Scores Analysis of Questionnaire Res
Summary of Results... ....

Discussion ...........

Conformity Behavior ...
Questionnaire Responses ..
Perceived Difficulty ...


Confidence in the Judgments of Others


Confidence in own Judgments


Conformity Behavior as a Rational Process.


iii


page

ii

V

Vii


1

2
6

8


16 16

16 17 18 18

21 21 23 27
40 142 42
45 45
46


ponses


. . . . . 47


47












TABLE OF CONTENTS (Continued) CHAPTER

III EXPERIMENT 2. ............. ..

Hypotheses . . . . . . .
Method . . . . . . . .

Experimental Design.......
Subjects. .......... .. .
Procedure ..... .... ...

Results . . . . . . .

Conformity Behavior.......
Questionnaire Responses.....
Summary of Results .......

Discuss ion . . . . . . .

IV SUMMARY . .. . . . . . .


APPENDICES ................

A EXPERIMENTAL MATERIALS.... .. ....

B RAW DATA AND SUPPLEMENTARY ANALYSIS, C RAW DATA AND SUPPLEMENTARY ANALYSIS$


EXPERIMENT 1 EXPERIMENT 2


BIBLIOGRAPHY . . . . . . . . . . . . .


BIOGRAPHICAL SKETCH ......................


iv


Page

149 51 52 52 53 53

514 514
58 63

64 69


73

714

78

87 93 97
















LIST OF TABLES


Table Page

1. Summary of the Analysis of Variance of a Portion of
Data from Shaw et al. (1966).................13

2. Predicted Relationships of Combined Levels of Variables 15

3. Summary of the Analysis of Variance of Conformity Scores,
Experiment 1.........................25

4. Summary of the Analysis of Variance of Difficulty Ratings,
Experiment 1.........................30

5. Summary of the Analysis of Variance of Ratings of Confidence in Judg-ments of Others, Experiment 1 . . . 32

6. Summary of the Analysis of Variance of Ratings of Confidence in own Judgments, Experiment 1 ............35

7. Reported Levels of Homogeneity of Majority-Competency .. 54

8. Summary of the Analysis of Variance of Conformity Scores,
Experiment 2.........................56

9. Summary of the Analysis of Variance of Difficulty Ratings,
Experiment 2.........................60

10. Summary of the Analysis of Variance of Ratings of Confidence in Own Judgments, Experiment 2 ............61

11. Summary of the Analysis of Variance of Ratings of Confidence in the Judgments of Others, Experiment 2 ........62

12. Description of Stimuli ....................75

13. The Postexperimental Questionnaire. .............77

14. Individual Ccnformity Scores and Questionnaire Responses
by Experimental Conditions, Experiment 1 ..........79

15. Means and Standard Deviations of Corrected Conformity
Scores by Experimental Conditions, Experiment 1
(N = 12 in Each Condition Except in the No Pressure
Control Group)........................85


v












LIST OF TABLES (Continued)


Table Page

16. Frequency Analysis of Ratings of Confidence D Values
by Levels of Conformity..................86

17. Individual Conformity Scores and Questionnaire Responses
by Experimental Conditions, Experiment 2 ............88

18. Means and Standard Deviations of Corrected Conformity
Scores by Experimental Conditions, Experiment 2,
(N =12 in Each Condition Except in the No Pressure
Control Group) .........................92


vi


















LIST OF FIGURES


Figure Page

1. Mean conformity scores for the three levels of majority
size, and for the Pressure Control (PC) and No Pressure
Control (NPC) groups, Experiment 1. ............26

2. Mean conformity scores for high and low levels of
perceived self-competency and perceived majoritycompetency, Experiment 1..................26

3. Obtained mean conformity scores by experimental
conditions compared with the predicted relationships
of the experimental groups, Experiment 1 ..........28

4i. Mean difficulty ratings for high and low levels of
perceived self-competency, and for the Pressure Control
(PC) and No Pressure Control (NPC) groups,
Experiment 1.........................31

5. Mean ratings of confidence in the judgments of others
for high and low levels of perceived self-competency
and perceived majority-competency, and for the
Pressure Control (PC) group, Experiment 1.........33

6. Mean ratings of confidence in the judgments of others
by the three levels of majority size for high and low
levels of perceived self-competency, Experiment 1 . 33

7. Mean rating's of conf idence in own judgments for high and
low levels of perceived self-competency and perceived majority-competency, and for the Pressure Control (PC)
and No Pressure Control (NPC) groups, Experiment 1 .. 36

8. Mean ratings of confidence in own judgments for the
three levels of majority size, Experiment 1.........36

9. The relationship between ratings of confidence in own
judgments and confidence in the judgments of others
by levels of conformity, Experiment 1...........38


v ii













LIST OF FIGURES (Continued)


Figure Page

10. Mean conformity scores for high and low levels of
perceived self-competency and perceived majoritycompetency, and for the Pressure Control (PC) and
No Pressure Control (NPC) groups, Experiment 2........57

11. Me~an conformity scores for the triple interaction
effects, Experiment 2...................59

12. A sample stimulus card...................76


viii

















CHAP'T 'P I


INTRODUCTION


A great deal of research in recent years has been devoted to

the investigation of conformity behavior under a variety of conditions. Two studies which have become classics in the area are those of Sherif (1935) and Asch (1951). Sherif demonstrated the effect of a totally unstructured stimulus on norm formation, and Asch showed

that the group has considerable influence even with regard to stated opinions about undisputed facts. With the impetus provided by these studies, the research that followed has focused on three general classes of variables: stimulus, personality, and situational factors.

It has been generalIly found that objective stimauli elicit less conformity than subjective stimuli and that the more ambiguous the stimuli the greater the conformity behavior (Crutchfield, 1955; Sherif and Sherif, 1956; Luchins and Luchins, 1955a; Nichols, 1964). The studies devoted to identifying the personality correlates of

conformity behavior have provided soma evidence to indicate that the type of individual who is least able to resist conformity pressure is: "submissive, low in self-confidence, less intelligent, lacking in originality, authoritarian minded, lacking in achievement motivation, conventional, and desires social approval" (Blake and Mouton, 1961, p. 23). The attempts to identify the personality correlates of


I






2


conformity behavior have, however, encountered problems similar to those which have stymied investigators who have searched for leadership "traits ." In the samei- sense that leadership is not unique to a given "type" of indi-vidual, it wouldJ be inappropriate to label a person demonstrating the above array of personality characteristics as a conformistst" or one who demonstrates antithetical characteristics as a "nonconformist." As indicated by virtually every study

in the area, conformity behavior i~s a complex process which is highly dependent upon situational variables. Individuals who conform in one social situation may maintain total independence in another situation. It may well be that pe2rsonality characteristics determine, to a large extent, the conditions undecr which an individual will tend to conform, and the situational and stimulus variables, in interaction with the

personality variables, determine the degree to which one conforms.




Situational Variables


From the research that has focused on conformity behavior, it appears that the investigation of situational, in contrast to

personality, variables is the more fruitful approach to gaining a better understanding of this aspect of social behavior. Allen (1965) has indicated the considerable attention situational variables have received and the extent to which conformity behavior is modified under a variety of social stimulus situations. Some situational variables, which have been shown to significantly influence conformity behavior are:






3


The attractiveness of the group. Generally, the more attractive the group the more influence is exerted on deviant members and the greater the conformity behavior of the deviant (Festinger et al., 1950; Schachter, 1951; Gerard, 1954; Lott and Lott, 1961; Walker and Heyns, 1962; Newcomb, 1961).

Status in the group. Hotnans' (1950) and Hollander's (1958) opposing hypotheses serve well to reflect the conflicting results which have been obtained with respect to this variable. 1{omans suggests that the person with the highest status will more closely

follow the group norms, and the more certain a person is about his status, the less he will conform to the norms of the group. Hollander hypothesizes that the high status person will be free to deviate from the norms of the group because of "idiosyncratic credits" he has built up in the group over a period of time. Sundby, as cited by Harvey and Consalvi (1960), found status to be positively related to conformity when the task was relevant to the group.

Interdependence MLIore conformity behavior is evidenced in

interdependent conditions than in independent conditions. When group pressure is increased, conformity also increases in interdependent groups and decreases slightly in independ ent groups (Thibaut and Strickland, 1956).

Size of the majority. Asch (1956) systematically investigated this variable and concluded that conformity increased significantly when the unanimous majority was increased from 1 to 3 members. Ttere was no appreciable increase in inffiuc ncpc for majorities of greater











size. Rosenberg (1963) found a curvilinear relationship between group size and conformity behavior. Using 1-, 2-, 3-., and 4-member majoriti'-s, the latter elicited significantly less conformity than did 3-member majorities which evokedI the greatest amount of conformity. Luchins and Luchins (1955b) reported that 3-membar majorities demonstrated sigrnificantly more conformity behavior than 1-member majorities. Nichols (1964) reported similar finding-s. K idd (1958) and Goldberg (1954) failed to find significant relationships between majority size and

conformity. In both of these studies the majority judgments were conveyed to the naive subject by the experimenter. Goldberg (1954) suggested that the subjects may have been conforming to the experimenter and the size of the majority may have made little difference. In a related study, Shaw et al. (1957) concluded that direct conformity pressure is more. effective than indirect pressure. The, indirect pressure exerted by the experimenters in tl-.: Kidd (1958) and Goldberg (1954) studies may have been insufficient to elicit differential conformity behavior. In any event, the bulk of the evidence suggests that the size of the majority is a highly relevant variable with respect to conformity behavior.

Composition of the group. Most of the research in this area

has been devoted to comparisons of homogeneous and heterogeneous groups along a variety of dimensions. Considering_ the diversity of the dimensions used--disability, skill, knowledge, sex, race--it is noteworthy that sotme uniformity of results has been achieved. Although there are conflicting results, the bulk of the evidence indicates






5


that individuals in homogeneous groups demonstrate more conformity behavior than do those in heterogeneous groups (Festinger and Thibaut, 1951; Gerard, 1953; Tuddenham,MacBride, and Zahn, 1958; Linde and Patterson, 196'4). Luchins and Luchins (1955b) and Reitan and Shaw (1964) found mixed-sex groups to demonstrate more conformity behavior than same-sex groups. These results are not necessarily contradictory in the sense that the subjects may have considered the mixedsex groups to be more homogeneous on some other dimension, and it is highly probable that the mixed-sex groups were seen as more attractive than the same-sex groups. Newcomb (1961) suggests that perceived

similarity leads to increased attraction which in turn is seen to produce a high level of conformity.

Competency on the assigned task. An individual who perceives himself as less competent than others in performing a task will conform to a greater extent than an individual who has reason to believe that he is more competent than the majority members (Samelson, 1957; Hochbaum, 1954; Mausner, 1954b; Di Vesta, 1959; Fagen, 1963). By manipulating the competency of both the naive subject and the majority, Fagen (1963) demonstrated that conformity behavior was more dependent on relative competency than absolute levels of competency. Shaw et al. (1966) found similar results.






6


Purpose


Two of the mentioned situational variables were selected for further investigation--group size and competency. As Allen (1965) stated, "Only slight attention has beeit given to the effect on conformity of the size of the group opposing a person" (P. 160). Several studies, as indicated, have examined this varicble, but the nature of the effect and the relationship of this situational variable to other situational variables warrants additional attention. It is conceivable, for example, that group size and perceived relative competency function in much the same manner and are additive in their effects on conformity behavior.

Perceived relative competency has received some attention, but little is known about the possible differential effects on conformity by majority members who are perceived as having homogeneous or heterogeneous levels of competency. Thus, the three primary interrelated purposes of the present study are to:

1. Examine the relationships among majority size, perceived
self-competency and perceived majority-competency.

2. Investigate the possible differential effects of homogeneous and heterogeneous majority-competency levels.

3. Determine if order of response of majority members
attributed with heterogeneous competencies influences
the degree to which the naive subject conforms.

Two experiments were performed to accomplish the above purposes.

The first experiment was devoted to examining the relationships among majority size and perceived self- and majority-competency levels.





7





The second experiment was designed to investigate the differential effects on conformity behavior of homogeneous and heterogeneous majority competencies and the order effects associated with a majority

perceived as having heterogeneous competency levels.



















CHAPTER I I


EXPERIMENT 1


Deutsch and Gerard (1955) offered the following hypothesis: "The more uncertain the individual is about the correctness of the judgment of others, the less likely he is to be susceptible to informational social influences in making his judgments" (P. 630). They

suggest that for the judgments of others to represent conformity pressure, the naive subject must perceive the others as being more accurate than himself.

Thibaut and Strickland (1956) conceptualized the reliance or nonreliance upon the judgments of others in terms of "group set" and

"task set." These investigators suggest that in "task set" the person is disposed to view the other individuals in the group as "mediators of fact" and, in responding to the perceptions and attitudes of others, the person is concerned not with achieving or maintaining a social relationship but with achieving or maintaining a cognitive clarity about his environment. Thibaut and Strickland (1956) provide a hypothesis similar to that of Deutsch and Gerard (1955): ". .to the extent that he . lacks confidence in the validity of his judgments, he will tend to depend on other observers for his attitudes and assessments" (P. 116).


8






9


There is a great deal of empirical evidence to support these hypotheses. Keiman (1950) found prior success reduced suggestibility and prior failure increased suggestibility. Mausner and Bloch (1957) examined the interaction of success and failure on conformity behavior. In 2-person situations, the past success of the subject (S) coupled with the past failure of his partner resulted in significantly less conformity than when S had previously experienced failure and his partner had experienced success. Mausner (1954ta) had earlier reported that when a person experiences a situation which causes him to lower his "confidence in self," conformity increased, and (Mausner, 1954b) if one's partner is perceived as more accurate, conformity is increased. Fisher, Williams, and Lubin (1957) found that the more confident one is in his judgments, the less he will conform to the judgments of others. Fagen (1963) manipulated real and perceived subject-competency and perceived majority-competency. He found conformity to be a function of all three variables. Shaw et al. (1966), in an unpublished study, obtained similar results with respect to perceived self- and perceived majority-competency.

Rosenberg (1963) examined the effects of perceived low competency on conformity behavior and the resultant confidence in self and confidence in partner(s). Ss were told that they had been wrong on 8 of 10 initial trials and partner(s) had been right on 7 of the 10 trials. Subjective measures of confidence in self and confidence in partner(s) were used and it was found that conformity had (1) a negative relationship with confidence in self, (2) a positive relationship






10


with confidence in partner(s), and (3) the difference between confidence in self and confidence in partner(s) was negatively correlated with conformity. Rosenberg varied the size of the majority but limited competency to a single level. He reported a curvilinear relationship between majority size and conformitywith maximum conformity elicited by majorities of three and significantly less by majorities of four. These results support Asch's (1956) findings although Asch did not find the significant curvilinear effect reported by Rosenberg. As previously mentioned, Luchins and Luchins (1955a) and Nichols (1964) also found majorities of three to evoke significantly more conformity than majorities of one.

The empirical results show that conformity varies directly

with perceived majority-competency and majority size (at least to a maximum of three majority members), and inversely with perceived selfcompetency. As suggested, it is conceivable that majority size and perceived majority-competency may function similarly in eliciting conformity behavior and may be additive in their effects. When confronted with the unanimously wrong judgments of the majority, S quite likely attributes some level of relative competency to himself with respect to the group. The degree to which S evidences ag-reement with the majority judgment depends to a great extent on his perceived relative competency. If S is opposed by a 3-member majority and suspects that all of the others are more capable than he is in the performance of the assigned task, he is more likely to agree with the unanimous majority than if he has reason to believe that the others are less






11


competent than he is. A similar appraisal would be made by S opposed by 2- and 1-member majorities; however, less conformity would be expected in the smaller groups.

If S is told that his performance on a task was better or worse than the performance of the other members of the group, S has a more objective evaluation of his relative competency. The empirical evidence indicates that such information has a highly significant effect upon conformity behavior.

Samelson (1957) suggested that the cognitive process associated with conformity possibly takes the following form: faced with highly discrepant information from two normally veridical sources-his own perceptions and the reported judgments of others--S must determine which source to reject and which to accept. In such a situation,

the so-called "conformist" tends to rely most heavily upon social reality, whereas the "independent" and the "anticonformist"' would be more disposed to reject the judgments of others and rely upon physical reality. Any information which will serve to support social reality or physical reality will tend to enhance conformity or independence, respectively. Information concerning performance on a task could serve to influence both types of behavior as would variations in majority size.


Hypotheses


It was anticipated that the present study would support previous findings; therefore the hypotheses with regard to main effects are as follows:






12


1. A majority of three evokes more conformity than a majority of two, which in turn evokes more conformity than
a majority of one.

2. A majority perceived as having high-competency elicits
more conformity than a majority perceived as possessing
low-competency.

3. Individuals who perceive themselves as having low-competency conform more than those who perceive themselves
as having high-competency.


If, as suggested, majority size and perceived relative competency function in much the same manner in the elicitation of conformity behavior, predictions concerning the effects of the various combinations of levels of the variables were deemed feasible. Previous research provides an indication of the relative influences these variables have on conformity.

An examination of Nichols' (1964) data concerning majority size reflects that female Ss who were required to select the largest

of three g-eometric figures conformed 43 percent of the time when faced with the unanimously wrong judgment of a majority of three. When a majority of one was used, conformity dropped to 17 percent. This difference yields a t value of 4.205 which is significant beyond the .001 level of confidence. The proportion of the variance accounted for by the majority size variable under these conditions is 27.5 percent. The proportion of the variance ('u) ) was computed in accordance with the procedure outlined by Hays (1963, p. 327). 1



1 Statisticians are not in complete accord as to the appropriate use of this statistic. In particular, it is questioned as a population estimate of strength of association.






13


Shaw et al. (1966) used 4-member female groups and virtually the same stimulus materials as those used by Nichols (1964). The results of this study with regard to high and low levels of perceived self- and majority-competency are indicated in Table 1.




TABLE 1

SUMMARY OF THE ANALYSIS OF VARIANCE OF A PORTION OF DATA FROM SH-AW ET AL. (1966) Suc f Me an Fpw2
Source df squareF


Perceived self -competency
(5) 1 147.00 22.07 .001 30.4

Perceived majoritycompetency (M) 1 102.08 15.33 .001 23.0 S xM 1 20.34 3.05 -Error 44 6.66





Shaw's Ss who perceived themselves as having high-competency conformed only 12.53 percent of the time, while those who had been informed that they possessed low-competency conformed 35.87 percent of the time. Perceived high majority-competency evoked conformity behavior on 33.87 percent of the occasions as compared with 18 percent for the perceived low-competent majorities.

Perceived high and low self-competency was produced by telling the Ss that they had been correct on either 8 or 4 of 10 initial trials. Majority-competency was manipulated in a like manner






14


by stating that the majority members had each been correct on 8 of the 10 initial trials (high competency) or 4 of the 10 trials (low competency).

The obtained w uswol indicate that if the variables were

to be ordered from greatest to least effect, perceived self-competency would be ranked first; majority size, second; and perceived majoritycompetency, third. The differences, however, do not appear to be significant and a best guess would be that, at the levels tested, the three variables are equivalent in their effects upon conformity behavior. If weights are assigned to levels accordingly and if the effects of these three variables are additive, Table 2 shows the predicted relationships of various combinations of levels. The greater the value of the combined weights, the greater the level of expected conformity behavior. Also included in the table are means for 6 of the 12 conditions which were obtained in a pilot study. Each mean is based on an n of 8. As can be seen, the pilot data generally support the predicted relationships.

Hypothesis 4 deals with the bases of Table 2 and the resultant predictions.

4a. The effects of the three variables on conformity
behavior are approximately equivalent to one another.

4b. The variables are additive in their effects on conformity behavior to the extent that they will produce
the relationships indicated by the combined weights.

The combined weights are cons idered to represent an ordinal scale but it is conceivable that the values of the lower end of the scale are below threshold; therefore there may be no differences in conformity under conditions represented by combined weig-hts of 5, 4,

and 3.





15


TABLE 2

PREDICTED RELATIONSHIPS OF COMBINED
LEVELS OF VARIABLES


Self- Majority- Majority Pilot Competency Competency Size Combined Study Level Weight Level Weight Level Weigrht Weights Means


low 3 high 3 high 3 9 8.0 low 3 high 3 inter 2 8 5.0

low 3 high 3 low 1 7

low 3 low 1 high 3 7 3.75

low 3 low 1 inter 2 6 low 3 low 1 low 1 5

high 1 high 3 high 3 7 2.40 high 1 high 3 inter 2 6 1.78 high 1 high 3 low 1 5 high 1 low 1 high 3 5 1.40 high 1 low 1 inter 2 4 high 1 low 1 low 1 3






16


Method


Experimental Design

A 2 x 3 x 2 factorial design was employed testing perceived self-competency (high vs low), majority size (3-member vs 2-member vs 1-member), and perceived majority-competency (high vs low). Twelve Ss were tested in each of the conditions, representing a total of 144 experimental subjects. Fifty-six control Ss were also included; 12 Ss under each level of majority size were exposed to conformity pressure but did not receive feedback information relative to self- or majority-competency, and 20 additional control Ss made all judgments with neither competency information nor conformity pressure.


Subjects

Due to significant differences in conformity behavior of males and females (Beloff, 1958; Reitan and Shaw, 1964), only females were used as Ss. The Ss were obtained on a volunteer basis from introductory psychology, sociologyand speech courses at the University of Florida. Those in introductory psychology courses received experimental credit for their participation. Ss were requested to volunteer for experimental sessions in groups of four, three or two. Sampling

bias was minimized by randomizing the sequence of conditions and by assigning Ss to conditions in the order in which they volunteered for participation in the experiment.






17


A total of 251 Ss participated in the experiment. Data from 32 Ss were discarded due to prior knowledge of the experimental treatment and an additional 19 Ss were randomly eliminated from the sample to achieve equal n's.


Appara tus

The apparatus was similar to that employed by Crutchfield

(1955). Five adjacent cubicles were located at one end of a 12' x 20' room. The experimenter's cubicle, which was equipped with a Beseler opaque projector and a master panel of lights and switches, was in the middle position. Ss occupied the four other cubicles, each of which contained a panel of twelve lights arranged in four rows of three lights (one row of lights for each of four Ss and one light for each of three possible responses). Mercury switches corresponding to the three columns of lights were mounted on each panel to permit Ss to report their judgments. Each switch activated one light in the fourth row of lights on a given S's panel, and a corresponding light on Ets master response panel. Although each S was led to believe that the first three rows of lights reflected the judgments of the other Ss, the lights were actually controlled by E by means of his master switch panel, which was wired to permit lights in identical positions to be illuminated in all cubicles simultaneously. Each S was led to believe that she was Subject 14 and the light responses she saw on her panel for Subjects 1, 2, and 3 represented the judgments of the other Ss.






18


Stimulus Materials

Two sets of stimulus cards were used with each card depicting three geometric figures (squares, rectangles, triangles, parallelograms) of different sizes. The first set of ten cards was used for "Practice" trials and the second set of twenty cards was used under conformity pressure conditions. The two sets were matched with respect to areas represented but differed with regard to the

shape of the figures, i.e., the figures on cards 1, 11, and 21 were identical in area, as were the figures on cards 2, 12, and 22, etc. The cards were projected on a screen at the opposite end of the room and Ss were instructed, in all cases, to select the figure representing the largest area. Table 12 in Appendix A provides the area sizes, order of presentation, and majority selections. A sample stimulus card is shown in Figure 12, Appendix A.


Procedure

Upon reporting to the laboratory to participate in a "perception" experiment, Ss were requested to select a cubicle position and be seated. The following instructions were read to the Subject:

I request that there be no discussion or communication
among you while you are seated in the cubicles. If you have any questions during the experiment I will discuss them with
you individually.

A number of cards will be projected on the screen you
see in front of you. On each card will be three figures of differing areas. In each case you are requested to determine which of the three figures represents the largest area.
Each card will be shown on the screen for a period of three
seconds. Following removal of the card you are requested to
make your judg-ment. (Response sheets were distributed at
this point.)






19


Please note the subject number on your response sheet.
This is your subject number and it also represents the
cubicle you have selected. Please indicate your judgments as to which figure represents the largest area by checking
the appropriate column on your response sheet. Let's try a
sample card. In this case, if you determine that figure
11C11 represents the largest area, you would check "C" on your
response sheet for "sample" card. Indicate your judgments in
the same manner for cards 1-10. Please be as accurate as
possible. Are there any questions?


The response sheets were collected by E following the presentation of the tenth stimulus card and E purportedly scored them.

The instructions for the second phase of the experiment were then

read with appropriate feedback information for each experimental

condition:

We have now completed the first phase of the experiment.
I will tell you how you have performed so far; however, in order
that you remain anonymous, I will give you results by subject
number, the number which you noted on your response sheet. It should be noted that there is no logical order of subject numbers according to cubicle location. (Competency feedback was
provided at this point.)

During, the second phase of the experiment you will indicate
your judgments relative to the figure representing the largest
area by means of the switches in each of your cubicles. You
will notice that your panel has four rows of lights numbered 1, 2, 3, and 4. These numbers represent subjects--each of you--l,
2, 3, and 4. The columns of lights are lettered A, B, and C.
Corresponding to the columns of lights are three switches,
A, B, and C. If, for example, you are Subject 2 and you
decide that figure "C" represents the largest area, you would
turn "on" the "C" switch and the "2C" light would be illuminated on your panel, on the panel(s) of the other subject(s)
and on my panel as well.

There are 20 cards in this sequence and they will be presented
in the same manner as were the first 10 cards. Do not make your
responses until I request them by calling out your subject number.
Each card will be presented for three seconds; I will remove the
card and will then ask for your judgments by subject number.
Please leave your lights on until I have had the opportunity to






20


record your judgments. After I have done so, I will request
you to turn your lights off. Again, please be as accurate
as possible. Are there any questions?


The two levels of perceived self-competency were manipulated by stating that Subject 14 was correct on either 3 or 7 of the initial 10 trials. Perceived majority-competency was manipulated by stating that Subject 1 (majority of one), Subject 1 and 2 (majority of two), or Subjects 1, 2, and 3 (majority of three) were each correct on

3 or 7 of the initial 10 trials.

In all cases Subject 4 was the last to respond. On 15 of

the 20 trials in the second phase of the experiment, all experimental Ss, and the control Ss exposed to conformity pressure, were required to respond following the judgments of the unanimously wrong majority.

On these trials the majority "selected" the smallest of the three figures in each case. On trials 11, 12, 18, 21, and 28, the majority unanimously "selected" the largest figure. Agreement with the unanimously wrong majority constituted conformity behavior.

Following presentation of the stimuli, Ss were requested to

complete a questionnaire designed to obtain additional data concerning the effects of conformity behavior. Ss were asked to indicate on a seven-point scale how difficult the task was, how confident they were in the correctness of their judgments, and how confident they were in the correctness of the judgments made by the others. Additionally, they indicated whether they were concerned about the disagreement; were doubtful of their accuracy; were tempted to answer as the others

did; and did answer as the others did against their own first choice.






21


The questionnaires were also used to identify those Ss who had previous information about the experimental condition. The postexperimental questionnaire is reproduced in Table 13, Appendix A.

The instructions read to the control Ss, who were not exposed to conformity pressure, were modified to omit any reference to the judgments of Subjects 1, 2, and 3. Naturally, switches representing the judgments of Subjects 1, 2, and 3 were not used while these control Ss made their judgments.

All Ss were debriefed and requested not to divulge any information about the experiment to potential participants.




Results


Scoring

Raw conformity scores were the number of times S agreed with the unanimously wrong majority. Raw conformity scores, therefore, had a possible range of 0 to 15. During the initial ten trials the Ss who were subsequently exposed to conformity pressure (all experimental Ss and the Pressure Control Ss) selected the smallest of the three figures a total of 149 times. Although there was no systematic relationship between these errors and experimental conditions, the conformity scores of the 42 Ss who were responsible for these judgments were proportionally reduced to correct for errors of this nature. In a few cases negative conformity scores resulted, so a value of 1131' was added to all conformity scores. The scores of the No Pressure






22


Control Ss were corrected in the same manner. The corrected scores were used for statistical analysis.

The postexperimental questionnaire responses were scored as follows:

Question 1. The open-ended question provided only qualitative data. Responses indicating the S was aware of the purpose of the experiment served as one basis for the elimination of S from the experiment. The responses to this question by all retained Ss were examined by three judges to determine degrees of suspicion as to the nature of the experiment. Ten Ss were identified by one or more of the judges as expressing some degree of suspicion, including two Ss who were identified by all three judges as indicating a high level of suspicion. No experimental or control condition included

more than two of the ten "suspicious" Ss.

.Questions 2, 3, and 4. The scoring of perceived difficulty, confidence in the j udgrments of others, and confidence in own judgments was identical. Ss responded to these questions on a sevenpoint scale with the highest score indicating greater difficulty and greater confidence. Control Ss who were not exposed to conformity pressure (No Pressure Control) responded only to Question 2 (level of difficulty) and Question 4 (confidence in own judgments).

Questions 5, 6, 7, and 8. Responses to these questions were "lYes" or "No" which were coded "1"l and "10," respectively.

Question 9. Affirmative responses to this question, about prior knowledge of the experiment, with an appropriate explanation, served to eliminate 28 Ss from the experiment.


A






23


All conformity scores and questionnaire responses are given in Appendix B, Table 14.


Analysis of Conformity Scores

The uncorrected conformity responses showed that the experimental Ss and the Pressure Control (PC) Ss agreed with the unanimously wrong majority judgments on 566 occasions, or 20.96 percent of the time. These conforming responses were made by 130 of the 180 Ss (72.22 percent). On the initial ten trials, under no conformity pressure, these Ss selected the smallest figure on 49 occasions, or on 2.72 percent of the opportunities. The errors committed under no pressure conditions were random across conditions and were not related to level of conformity behavior. The difference in errors under conformity pressure and under no pressure to conform is significant (p <.001).

A comparison of the experimental and PC Ss with the 20 Ss

who responded to all 30 stimulus cards under no conformity pressure provides similar results. Four of the No Pressure Control (NPC) Ss made one or more of the erroneous majority selections on a total of six occasions (2 percent). These results are not significantly different from those of the experimental and PC Ss under no pressure conditions, but are significant (p <.001) when compared with the responses of the experimental and PC Ss under conformity pressure conditions.

A 2 x 3 x 2 factorial design with a single control group (PC) was employed to examine the effects on conformity behavior of perceived majority-competency (high vs lo, ), majority size (3-member vs 2-member vs 1-member), and perceived self-competency (high vs low).






24


Due to the relatively simple nature of the task, it was anticipated that there would be no systematic relationship between real competency and conformity behavior. Therefore, real competency was not included as an independent variable. An examination of the data revealed no discernible relationship between real competency and conformity behavior.

Table 3 summarizes the analysis of variance results and indicates the percentage of the variance accounted for by each main effect. Means and standard deviations of the conformity scores by experimental conditions are provided in Table 15, Appendix B.

The PC group did not differ from the experimental grroups; however, a comparison between the PC and NPC groups yielded a t of 21.64 which is significant (p <.001) and indicates the general effectiveness of the conformity pressure.

All three independent variables significantly influenced conformity behavior, as hypothesized. Of the three, group size, which accounted for 13.04 percent of the variance, proved to be the the strongest. A comparison of means indicated that the 1-member majorities elicited significantly less conformity behavior than did the 2-member majorities (p <.001), but the difference between the 2- and 3-member majorities was not significant, although it was in the



1 The control grroup was not included in the computation of w 2
since the primary focus of interest was upon the relative contribution of the three experimental variables occurring in combination. This holds true for the computation of w2 throughout the remainder of Experiment 1 and for Experimecnt 2.-






25


anticipated direction. Hypothesis 1, therefore, gained partial support. Figure 1 shows the differential effects of the three levels of majority size. Also included are the mean conformity scores for the PC and NPC groups.




TABLE 3

SMMIARY OF THE ANALYSIS OF VARIANCE OF CONFORMITY SCORES, EXPERIMENT 1



Source Sums of df Mean F 2* Squares square FM%


Control vs all others 3.82 1 3.82 Majority-competency (M) 170.08 1 170.08 15.62 .001 7.01 Majority size (G) 318.14 2 159.07 14.61 .001 13.04 Self-competency (S) 102.52 1 102.52 9.141 .01 4.02 M x G 52.23 2 26.12 2.40 M x S 2.92 1 2.92 G x S 30.00 2 15.00 1.38 M xG xS 21.07 2 10.53 Error 1819.37 167 10.89


The relevant values for the computation Squares, 22148.11; MS Error, 11.75.


of IX are: Total Sums of


Hypotheses 2 and 3 were also supported. Majorities depicted as possessing high competency elicited significantly more conformity behavior than did majorities having low competency (p <.001), and Ss





















U' - I -


Majority


2- me m' be r
S i Ze


3-rnembe r


Figure l.--Mean conformity, scores for the size, and for the Pressure Control (PC) and No groups, Experiment 1.


7


6

S.

4, 3,


three. levels of majority Pressure Control (NPC)


-Self
- -Majority


H I gh
Perceived Compete nc y


[6w


Figure 2.--Mean conformity scores for high and low levels of perceived self-competency and perceived majority-competency, Experiment 1.


26


8' 7, 6,

5

4

3


E
C~-)


n1


1-me mnber


NPC-


Control Groups


CU


'., t


PC -


8,






27


who perceived themselves as having low competency conformed less than did Ss who perceived that they had performed at a high level (p <.01). Figure 2 indicates the mean conformity scores at the two levels of these variables.

It was predicted that the three independent variables would have equivalent (Hypothesis 4a) and additive (Hypothesis 4b) effects on conformity behavior. To test this two-part hypothesis a comparison was made between the obtained experimental group means and the predicted relationships of these groups (see Table 2). A Spearman rank order correlation of .83 was obtained (p <.01). This would indicate some support for the hypothesis. Figure 3 depicts the predicted and obtained ordering of means by experimental condition. The differential strength of the majority size and perceived selfcompetency variables, the relatively high level of conformity behavior that occurred in the 3-member homogeneous groups, and the consistently low level of conformity in the 2-member groups accounted for most of the discrepancies between the predicted and obtained results.


Analysis of Questionnaire Responses

Responses to Questions 2, 3, and 4 yielded information relative to how difficult S perceived the task, how confident S was in the judgments of the others, and how confident S was in her own judgments. These data were examined by analysis of variance.





28


10

9

8


7

6

5

4


31*
nT


-C


LI I


I
1-member


I
2-memnber


-Obtained Conformity
--Predicted Conformity M-Majo rity S self




Low M Low S


II I
3-m~ebr 1-miemrber 2-member 3-memnber


Majority Size

Figure 3.--Obtained mean conformity scores by experimental
conditions compared with the predicted relationships of the experimental groups, Experiment 1.


E

.W
2t






29


Question 2. Please circle on the. line below, the number
which represents how easy or difficult you feel it is to select
the correct choice for this perceptual task.

A summary of the 2 x 3 x 2 analysis of variance carried out on the perceived difficulty ratings is shown in Table 4. The only variable that achieved significance was perceived self-competency (p <.001), and there were no significant interactions. Figure 4 indicates the mean levels of difficulty ratings for the two levels of perceived selfcompetency as well as the mean ratings of difficulty for the PC and NPG groups. A comparison between the two control groups means yielded a t of 8.86, which is significant beyond the .001 level of confidence.

Question 3. How confident were you in the judgments of
the others?

The analysis of variance summary table for the ratings of confidence in the judgments of others (Table 5) reflects that such confidence was a function of perceived majority-competency (p <.001), perceived self-competency (p< .001), and an interaction effect between the size of the majority and perceived self-competency (p <.01). A plot of mean confidence in the judgments of others for high and low levels of perceived self-competency and perceived majority-competency (Figure 5) indicates that confidence in the judgments of others varied directly with perceived majority-competency and inversely with perceived self-competency. Figure 5 also indicates the PC group mean rating of confidence in the judgments of others.

Figure 6 indicates the interaction between majority size and perceived self-competency. Tests for simple main effects indicated





30


TABLE 4

SUMMAkRY OF THE ANALYSIS OF VARIANCE
DIFFICULTY RATINGS, EXPERIMENT


OF
1


Source ~Sums of d en F P w2
SourceSquares df Square FM%



Control vs all others .52 1 .52 Majority-competency (M) .18 1 .18 Majority size (G) 2.73 2 1.36 Self-competency (3) 47.84 1 47.84 27.81 .001 15.12

M x G 4.04 2 2.02 1.17 M x S 1.17 1 1.17 G x S 9.55 2 4.78 2.78 M x G x S 4.07 2 2.04 1.19 Error 286.89 167 1.72 The relevant values for the computation of w 2are: Total Sums of Squares, 302.83; MS Error, 1.77.






31


6






5 P

C

4



NPC *

C





HihLow Control Groups

Perceived Self-Competency

Figure 4.--Mean difficulty ratings f or high and low levels of perceived self-competency, and for Pressure Control (PC) and No Pressure Control (NPC) groups, Experiment 1.






32


TABLE 5

SUMMARY OF THE ANALYSIS OF VARIANCE OF RATINGS CONFIDENCE IN JUDGMENTS OF OTHERS) EXPERIMENT


OF
1


Sums of Ma 2
SoreSquares df square F


Control vs all others .19 1 .19 Majority-competency (M) 19.51 1 19.51 14.114 .001 7.85 Majority size (G) 3.60 2 1.80 1.30 Self-competency (S) 21.01 1 21.01 15.22 .001 8.50 M x G .68 2 .341 M x S 2.00 1 2.00 1.45 G x S 13.51 2 6.75 4.89 .01 4.71 M x G x S .60 2 .30 Error 230.92 167 1.38


The relevant values for the computation Squares, 230.84; MS Error, 1.29.


of w are: Total Sums of






33




4-1 S af
4. Majority




oo



-6-- PC.

COc -= CO)




1-14 Lo'wControl 'Group

Perceived Competency

Figure 5.--Mean ratings of confidence in the judg,,ments of others for high and low levels of perceived self-competency and perceived majority-competency, and for the Pressure Control (PC) group, Experiment 1.


-High Self-Competency

4, Low Self -Competency






OC 3




CD
CD- V)
1-eme 2-ebe -mme MaoiySz
Fiur 6.-enrtnso cniecMntejuget fohr




by te thee lvelsofMajority iefoSighadlweeeso perceived self-competency, Experiment 1.






34


that when perceived self-competency was high the confidence in judgments of others was uniformly low regardless of majority size, but when perceived self-competency was low, the confidence in the judgments of others increased as the size of the majority increased (p <.01). The interaction also resulted in a significant difference (p< .001) between the confidence in judgments of others by high-and low-competent Ss opposed by 3-member majorities.

Question 4. How confident were you of the correctness of
the judgments you made?

A 2 x 3 x 2 analysis of variance carried out on the ratings

of confidence in own judgments indicated that all three main variables influenced the confidence S had in her own judgments. The strongest variable, in this case, was perceived self-competency which accounted for 13.86 percent of the variance. Table 6 shows the summary of the analysis of variance carried out on confidence in own judgments. Figure 7 indicates graphically the mean ratings of confidence in own

judgments for the levels of perceived self- and major ity-competency. Also included are the mean ratings of confidence in own judgments for the PC and NPC Ss. A t test comparing the mean ratings of the control groups indicated that they were not significantly different. Figure 8 portrays the mean ratings of confidence in own judgments

for the three levels of majority size.

A further examination of the responses to Questions 3 and 4 was made in terms of the frequency with which Ss rated confidence in the judgments of others higher than confidence in own judgments.






35


TABLE 6

SUMMARY OF THE ANALYSIS OF VARIANCE OF RATINGS OF
CONFIDENCE IN OWN JUDGMENTS, EXPERIMENT 1


SourceSums of df Mean Fp 2
SoreSquares square M



Control vs all others .73 1 .73Majority-competency (M) 7.56 1 7.56 4.47 .05 1.97 Majority size (G) 11.93 2 5.96 3.53 .05 2.88 Self-competency (S) 43.34 1 43.34 25.64 .001 13.86 M x G 4.55 2 2.27 1.34 M x S 2.51 1 2.51 1.49 GCx S 5.68 2 2.84 1.68 M x GxS 6.67 2 3.34 1.98 Error 282.81 167 1.69


The relevant values for the computation Squares, 299.16; MS Error, 1.64.


of w 2 are: Total Sums of









-SElf
- -- Majority


CD, CD
S E 04-.

0U


C-


High


L ow


36


5




4


Perceived Competency
Figure 7.--Mean ratings of confidence in own judgments for high and low levels of perceived self-competency and perceived majoritycompetency, and for the Pressure Control (PC) and No Pressure Control (NPC) groups, Experiment 1.


2-member Majority Size


3-mem ber


Figure 8.--Mean ratings of confidence in own judgments f or the three levels of majority size, Experiment 1.


NPC









Control GroLups


3J,


C-)



CV

CD CU WC :E)


5.




4




3


1-member


K__ -






37


Chi-square tests revealed that more Ss opposed by high-competency majorities reported a higher level of confidence in majority judgments
2
than in their own (23 vs 9; X = 4.87; p <.05). Similarly, more Ss who perceived themselves as demonstrating low- versus high-competency reported that they had greater confidence in the judgments of others than in their own judgments (27 vs 5; X2 = 19.4I4; p <.O0l).

To determine the relationship between conformity behavior and the tendency to assign greater confidence to the judgments of others, the difference between confidence in own judgments and judgments of

others was determined for each of the experimental Ss. Three categories were used to classify the relationships of responses: confidence in self greater than confidence in others, equal ratings of confidence in self and others, and confidence in self less than confidence in others. Figure 9 portrays these relationships by five levels of conformity behavior. A chi-square test on these data, with the three cells of greatest conformity pooled to permit sufficiently large expected cell frequencies, indicated that as conformity behavior increased, proportionally fewer Ss reported greater confidence in own judgments and proportionally more Ss reported equal levels of confi2
dence and grreater confidence in the judgments of others (X = 16.68; p <.005). Table 16 in Appendix B contains the frequency data which provided these results.

Chi-square tests were also carried out on the frequencies of "lYes" and "No" responses to Questions 5 through 8. Although more Ss who perceived themselves as having low-rather than high-competency





38


8E3 Confidence in Self > Confidence in Others
U80fdneinSl ofdec nOhr
70 N48 E Confidence in Self

60N5
N::12
50- N=z67
N=12


c / 3 0 1

~20'
CII!


0-3 4-7 8-11 12-15 15-18
Conformity Scores

Figure 9 .-The relationship between ratings of confidence in own judgments and confidence in the judgments of others by levels of conformity, Experiment 1.





39


indicated that they were concerned about disagreements (28 vs 21), were doubtful of their accuracy (49 vs 42), were tempted to answer as the others did (46 vs 35), and answered as the others did against their own first choice (20 vs 114), none of these differences achieved an acceptable level of significance.

The perceived majority-competency variables provided significant results with regard to all four questions. More Ss who perceived themselves opposed by high rather than low majority-competency reported that they were concerned about disagreements (32 vs 17; x 2=6.80; p <.01), were doubtful of their accuracy (52 vs 39; y, 2= 5.04;
2
p <.05), were tempted to answer as the others did (48 vs 33; X = 6.35; p <.02), and answered as the others did against their own first choice

2
(22 vs 12; _X = 3.85; p <.05).

The majority size variable also influenced responses to Questions 5, 7, and 8. As majority size increased,more Ss stated that they were tempted to answer as the others did (14 vs 33 vs 34;
2
x=21.52; p <.00l), and more reported that they answered as the
2
others did against their own first choice (3 vs 11 vs 20; x = 16.69; p <.001). With regard to concern over disagreements, Ss opposed by 1-member majorities made the fewest number of affirmative responses, and Ss opposed by 2-member majorities made the most (7 vs 24 vs 18;
2
x = 13.80; p <.01).





40


Summary of Results

All three independent variables (majority size, perceived majority-competency, and perceived self-competency) were found to significantly influence conformity behavior with majority size being the strongest of the three variables by accounting for 13.04~ percent of the variance. Hypothesis 1 received partial support in that 1-member majorities evoked significantly less conformity behavior than did 2-member majorities, but 3-member majorities did not elicit significantly more conformity behavior than did the 2-member majorities. Hypotheses 2 and 3 received strong support; majorities perceived as having high competency elicited significantly more conformity than did majorities perceived as possessing low-competency, and Ss who perceived themselves as having low-competency conformed more than did Ss who perceived themselves as having high-competency. Hypothesis 4, which predicted that the three variables were additive and equivalent, received some support by virtue of the fact that the predicted relationships among, experimental groups correlated favorably with the obtained experimental group means (p = .83; p <.Ol). The differences between the predicted relationships and the obtained experimental group means were primarily due to (1) the relatively high level of conformity elicited by 2-member majorities when perceived majority-competency and perceived self-competency were equivalent, and (2) the unitormly low level of conformity behavior in response to 1-member majority judgments.

The questionnaire results were consistent with the more objectively measured conformity behavior. Perceived self-competency was the






41


only variable that significantly influenced ratings of difficulty; however, all three variables significantly influenced the ratings of confidence in own judgments. Confidence in the judgments of others was primarily a function of perceived self-competency and perceived competency of the majority; the confidence in the judgments of others varied directly with perceived majority-competency and inversely with perceived self-competency. It was also found that a significant interaction occurred between majority size and perceived self-competency; if perceived self-competency was low, confidence in the judgments of others varied directly with the size of the majority. This relationship resulted in a significant difference between perceived high and low self-competency Ss exposed to 3-member majorities, but not at the other two levels of majority size.

It was also found that perceived majority-competency and perceived self-competency significantly influenced the relationship of confidence in the judgments of others and confidence in own judgments. More Ss opposed by high-competent majorities and more Ss who perceived themselves as having low-competency reported less confidence in their own judgments than they did in the judgments of others. Furthermore, the difference between confidence in own judgments and confidence in the judgments of others was negatively correlated with conformity behavior.

The responses to the questionnaire items dealing with concern over disagreements, doubtfulness of accuracy, temptation to answer as the others did, and whether they answered as the others did against






42


their own first choice, supported the indication of the relative strength of the variables. The levels of perceived self-competency failed to yield any significant differences in responses to these questions; whereas perceived majority-competency reflected significant differences for all four questions. Majority size provided significantly differing frequencies with regard to concern over disagreement and the temptation to respond as the others did.



Discuss ion


Conformity Behavior

Perceived self-competency, perceived majority-competency, and majority size all significantly influence the level of conformity behavior. As was hypothesized, perceived high self-competency resulted in less conformity behavior than did perceived low self-competency; perceived high majority-competency elicited more conforming responses

than did perceived low majority-competency; and, conformity behavior varied directly with majority size. These results serve to extend and clarify results from other studies dealing with these variables in singular and paired relationships.

The prediction that the three variables are equivalent and additive was basically supported; however, relationships among these variables were not consistent. Regardless of majority size, virtually no conformity behavior was demonstrated when S perceived herself as possessing greater competency than the other member(s) of the group.






'43


It is suggested that under such conditions the combined effect of the variables was below threshold.

When all group members are perceived as demonstrating the

same level of competency, the 1-member majority again failed to elicit conformity behavior. Under these conditions, 2- and 3-member majorities elicited significant levels of conformity; however, there was no significant difference in conformity elicited by these larger majorities.

only under conditions when S perceived herself as less competent than the other member(s) of the group did the three variables approach additivity and equivalence of effect. These results, and those mentioned above, indicate that the majority size effect on conformity behavior is, to a great extent, a function of how competent the other members of the group are perceived by S. Quite logically, if S has reason to believe that she is more competent than all the other members of the group, regardless of majority size, she is not inclined to rely on the majority judgments under informational influence conditions. If, however, the majority members are each perceived as having levels of competency equal to or greater than S the size of the majority becomes salient. The subjec t apparently perceives that there is greater probability that 2- and 3-member majorities would be correct in unanimous judgments than would be a single member of a group. When the majority members are portrayed as more competent than the naive subject, the majority effect is highly differentiated as a function of majority size.






44


Although the correlation between the obtained experimental

group means and the predicted relationships among them was significant (p <.01), the variables did not account for equivalent proportions of the variance. This was in part due to the minimal conformity in response to the 1-member majority. The limited effect that a majority of one has on conformity behavior is well documented (Asch, 1956; Luchins and Luchins, 1955a; Nichols, 1964; Rosenberg, 1963); however, it was anticipated that a high competent majority of one would elicit more conformity behavior than it did. More ambiguous stimulus materials would possibly have increased the conformity influence of the 1-member majority without appreciably increasing the influence of the 3-member majority (Bass, 1961). In any event, the relative contribution any relevant situational variable has on conformity behavior depends on the levels employed and the context in which they are tested.

MacBride and Tuddenham (1965) concluded that a demonstration of ability to perform a task can prepare an individual better to withstand group pressure and a demonstration of inaccuracy can weaken selfconfidence and reduce ability to withstand pressure. A similar conclusion could be made from the results of the present study with regard to the influence of the perceived self-competency variable on conformity behavior. When the effects of the other variables are considered, however, the conclusion must be qualified. At the levels tested, the absolute level of competency was largely irrelevant. If one is unsuccessful at a task, and everyone else in the group is also unsuccessful, conformity behavior is little different from that in a situation in






4~5


which all group members perceive themselves as having high levels of competency. Inaccuracy per se does not reduce one's ability to withstand pressure, nor does a demonstration of ability necessarily dispose: one to independent behavior. However, if one is aware of one's own level of competency and the competency of the other members of the g roup, conformity behavior becomes a function of the individual's perceived relative competency and the size of the group.


Questionnaire Responses

Responses to the questionnaire items were consistent with the behavior exhibited under conformity pressure. This is to be expected since the conformity behavior and the questionnaire responses were not independent responses. It is reasonable to expect that conformity behavior varies directly with perceived difficulty and confidence in the judgments of others, and inversely with confidence in one's own judgments. Such results were obtained in the present study and provide support for similar results reported by Rosenberg (1963).


Perceived Difficulty

Ratings of the task difficulty were found to be a function of perceived self-competency but were not related to the other independent variables. The significant difference between the difficulty ratings of the PC and NpC Ss would indicate that ratings of task difficulty were also influenced by the experimental treatment. For the experimental Ss, the effect of the disagreeing majority was modified by the individual's perceived self-competency. All of this points to





46


the consideration that perceived difficulty ratings are highly susceptible to influence.


Confidence in the Judgments of Others

Confidence in the judgments of others was found to be a function of perceived self-competency, perceived majority-competency and an interaction between majority size and perceived self-competency. The findings relative to the relationship of the two competency variables indicate that the Ss were attendant to the reported results on the initial ten "practice" trials. The interaction effect is interesting in that it may reflect the perceived credibility of the majority judgments when the perceived self-competency is low. Under conditions of perceived high self-competency, majority size had a very slight negative relationship with reported confidence in the judgments of others, i.e., as majority size increased, confidence in the majority judgments decreased. On the other hand, when Ss perceived themselves as having low-competency,, confidence in the judgments of the majority increased as majority size increased. The difference between high and low perceived self-competency when Ss were opposed by 3-member majorities, is significant at beyond the .001 level of confidence.

Unfortunately, it is not possible to determine to what degree this interaction effect is attributable to the perceived levels of relative competency, the experimental treatment and/or the conformity behavior itself. It can be concluded, however, that reported confidence in the judgments of others is modified to some degree by one's

own behavior.






47


Confidence in Own Judgments

Confidence in S's judgments varied directly with perceived

self-competency and inversely with perceived majority-competency which would be expected. However, the finding that confidence in own judgments increased as majority size increased was not expected and this, finding is incongruent with all other findings in the study. It is suggested that this is a spurious finding and should be disregarded. This is supported by the fact that the mean ratings of confidence in own judgments by Ss opposed by 3-member majorities were not significently different from those made by Ss opposed by 1-member majorities. The only significant difference existed between the ratings made by Ss opposed by 2- and 3-member majorities (p <.05).


Conformity Behavior as a Rational Process

The examination of the relationship of the difference between the two confidence ratings and conformity behavior provides strong support for Rosenberg's (1963) results in this regard. At the lower levels of conformity behavior, proportionally more Ss indicated greater confidence in their own judgments than in the judgments of others. At the higher levels of conformity, however, proportionally more Ss reported equal levels of confidence in own and other judgments and more confidence in the judgments of others than in their own.

These results and the relationship of conformity behavior to

perceived self-competency, perceived majority-competency, and majority size, support Samelson's (1957) supposition concerning the cognitive process associated with conformity behavior. In the experimental






'48


session, the "conformist," the "independent," and the "anticonformist" all may be considered as demonstrating rational and logical behavior. Each attends to the cues she had determined to be relevant in that particular situation and, depending on the direction and perceived strength of these selected cues, demonstrates conforming or independent behavior. The present study found perceived relative competency and majority size to represent salient cues. The naive subject demonstrated conformity behavior when she perceived herself to be less competent than the other member(s) of the group and when she was outnumbered by a unanimous majority. Independent behavior occurred, regardless of majority size, when the majority members were perceived as less competent than the naive subject perceived herself to be. Furthermore, the individual who perceived herself to have low-competency perceived the task to be relatively difficult, had relatively little confidence in the judgments she made, and relatively high confidence in the judgments made by others. Conversely, the individual who perceived herself as possessing high-competency perceived the task to be relatively simple, was relatively confident in the judgments she made and had relatively little confidence in the judgments made by the majority-

















CHAPTER III


EXPERIMENT 2


Experiment 1 was designed to investigate the effects oh conformity of perceived self-competency, perceived majority-competency, and majority size. In all cases the majority members were portrayed as having competency levels equivalent to one another. The purpose of this experiment is to investigate the nature of conformity behavior when majority members are portrayed as possessing differing levels of competency.

If majority members are perceived as having homogeneous competency levels, the order in which the majority members report their judgments is immaterial. However, if the majority members are perceived as possessing differing levels of competency, the degree to which the naive S is influenced by the majority may depend to some extent on the order in which the majority members report their judgments and the deg-ree to which the naive S perceives the judgment of one majority member influencing the other(s).

As previously mentioned, the studies which have dealt with

the composition of the group as an independent variable have provided no conclusive evidence concerning the effects of homogeneous and heterogeneous groups. Gerard (1953) examined the effects on opinion change of small groups (8 14 members) perceived as having homogeneous


49







50


and heterogeneous levels of competency. He found limited evidence to indicate that groups having homogeneous levels of competency exerted greater pressure toward uniformity and that change toward uniformity occurred only in a homogeneous high-pressure condition. The studies dealing with the same-sex versus mixed-sex variable (Luchins and Luchins, 1955b; Tuddenham et al., 1958; Reitan and Shaw, 1964) may be interpreted in terms of perceived competency, but the conflicting results indicate that factors in addition to perceived competency are influencing conformity behavior when the sex composition of the group is varied.

It is suspected that the heterogeneity of majority-competency, in and of itself, is not a critical factor but that the order of majority member responses is probably a significant variable. Thibaut and Riecken (1965) have demonstrated that the degree to which a person perceives himself as influencing others depends to a significant degrree on the perceived power relationships. Ss perceived the causal locus for compliance as "internal" to the high-power stimulus person and "external" to the low-power stimulus person. Although Thibaut and Riecken were concerned with power relationships and the effects of perceived conformity of majority members, their results are, in a sense, applicable to the proposed investigation of order effects of heterogeneous majorities. If, in a 2-member major ity, the high-competent majority member is the first to indicate her judgment, the low-competent majority member may be seen as merely






51


conforming to the judgment of the high-competent member when she agrees with the initial judgment. If S perceives such to be the case, the majority-size effect may be reduced, since S would possibly be influenced only by the high-competent member's response. on the other hand, when the low-competent majority member is the first to respond, the high-competent majority member would not likely be seen as conforming to the initial judgment. Instead, she may be viewed as confirming it. The judgments of the majority members, in this case, would quite likely evoke more conformity behavior than the reverse order of majority member responses.



Hypotheses


The hypotheses offered in Experiment 1 with respect to perceived self-competency and perceived majority-competency were not expected to be altered by the inclusion of the homogeneity of majoritycompetency variable. Thus, again, it was expected that:

1. Ss who perceive themselves as possessing low-competency
conform more- than Ss who perceive themselves as having
high-competency.

2. Majorities perceived as having high-competency evoke
more conformity than majorities perceived as having
low -competency.

No directional hypothesis is offered with regard to homogeneous majorities versus heterogeneous majorities. Although the bulk of evidence suggests that more conformity occurs in homogeneous groups, this evidence is relatively weak. It is expected, however, that conformity under the two conditions may differ and the hypothesis becomes:






52


3. Majorities perceived as having homogeneous and heterogeneous competencies differ in their elicitation
of conformity.

With regard to Hypothesis 3 an assumption must be made as to what represents equivalent levels of homogeneous and heterogeneous competencies. For purposes of this experiment, it was assumed that if the sums of majority member competency levels are equal, regardless of the individual values, the majorities would be equally competent. It is recog-nized that this is a rather gross assumption but is considered appropriate at this point.

As has been suggested, order effects associated with heterogeneous majorities may influence conformity as a function of perceived locus of causality. The hypothesis is:

4. More conformity is elicited when the low-competent member
of a 2-member majority is the first to report her judgment than when the high-competent member reports her
judgment initially.


Method


Experimental Design

A 2 x 3 x 2 factorial design was used to test perceived

majority-competency (high vs low), homogeneity of perceived majoritycompetency (homogeneous vs heterogeneous vs heterogeneous2), and perceived self-competency (high vs low). The two levels of heterogeneity refer to the two possible orders in which 2-member majorities can report their judgments. Three-member groups were used and, as in Experiment 1, 12 Ss were tested in each condition.






53


The data representing the judgments of homogeneous majorities were those which were obtained for 3-member groups in Experiment 1. The 3-member control Ss from Experiment 1 were also used for Experiment 2. Data for the two experiments were collected simultaneously and Ss were randomly assigned to the experimental conditions.


Subjects

Subjects for this experiment came from the same subject pooi as did *those who served in Experiment 1. In addition to the 3-member groups and the NPC subjects from Experiment 1, 114 more female undergraduates served in the present experiment. Data from 18 Ss were discarded as follows: 6 Ss were familiar with the experimental treatment, experimenter error invalidated data from three Ss, and 9 Ss were randomly discarded from within conditions to achieve equal n's. Procedure

The same apparatus and stimulus materials that were used in Experiment 1 were used in this experiment and all procedures in Experiment 1 pertaining to instructions, presentation of stimuli, manner of making responses, administration of questionnaire, and debriefing, also apply to Experiment 2.

Table 7 indicates the levels of homogeneous and heterogeneous majority-competency.

The two levels of perceived self-competency were again manipulated by stating that Subject 4 was correct on either 3 or 7 of the 10 initial trials. The high- and low-homogeneous majority-competency





54


data were the 3-member group data from Experiment 1: high-homogeneous competency was indicated by reporting that Subjects 1 and 2 had both been correct on 7 of the 10 initial trials; low-homogeneous majoritycompetency was indicated by stating that Subjects 1 and 2 were each correct on 3 of the 10 initial trials.

The high-competent heterogeneous majority was perceived by

S as having competency levels of 4 and 10. Low-competent heterogeneous majorities were attributed with competency levels of 0 and 6.




TABLE 7

REPORTED LEVELS OF HOMOGENEITY OF MAJORITY-COMPETENCY



Homogeneous Heterogeneous Heteonou2


High 7, 7 103% 4 4, 10 Low 3S 3 6s 0 0,% 6


Results


Conformity Behavior

Scoring of conformity and questionnaire responses for Experiment 2 were identical to those used for Experiment 1.

The experimental and PC Ss conformed 536 times out of 2,340 opportunities to do so, or 21.52 percent of the time. Compared with the NPC group, this represented a significant level of conformity behavior (p <.001). Table 8 summarizes the analysis of variance that





55


was carried out on the corrected conformity scores of the experimental and PC Ss. The table also provides the percentage of variance (w2 ) accounted for by the significant main effects and the one significant interaction.

Hypotheses 1 and 2 predicted that, as in Experiment 1, the two levels of perceived majority-competency and perceived selfcompetency would again have significant and opposite effects on conformity behavior. These hypotheses were supported (p <.001). The means for the levels of these variables and for the two control groups are shown in Figure 10.

Hypothesis 3 stated that majorities perceived as having homogeneous and heterogeneous levels of competency would differ in their elicitation of conformity. The main effect was not considered significant (p <.10); however, a significant triple interaction occurred (p <.05) which indicated that when both self-competency and majoritycompetency were perceived to be high, the homogeneous majority elicited significantly more conformity than did the heterogeneous majorities (p < 01).

Based on an extension of Thibaut and Rieckents (1965) results, Hypothesis 4 predicted that when the low-competent member of a 2-member heterogeneous majority was the first to report her judgment, conformity behavior on the part of a naive subject would be greater than when the high-competent member was the first to respond. This hypothesis was not supported. In fact, when perceived self-competency and perceived majority-competency were both high, the opposite results were obtained

(p < .05).







56


TABLE 8

SUMMARY OF THE ANALYSIS OF VARIANCE OF
CONFORMITY SCORES, EXPERIMENT 2


Source Sums of df Mean F 2 Squares Square M%



Control vs all others .70 1 .70 Majority-competency CM) 178.78 1 178.89 19.51 .001 9.43 Majority-homogeneity (H) 49.15 2 24.58 2.68 .10 1.72 Self-competency (S) 276.39 1 276.39 30.14 .001 14.86 M x H 2.70 2 1.35 M x S 5.64 1 5.64 H x S 26.33 2 13.17 1.14 M x H x S 60.82 2 30.41 3.32 .05 2.38 Error 1311.73 143 9.17


The relevant values for the computation Squares, 1790.23; MS Error, 9.02.


of w 2 are: Total Sums of





57


-Self

8. -- -Majority


7





E
L..
C.


co NPC*


2




High Low Control Groups

Perceived Competency

Figure lO.--Mean conformity scores for high and low levels of perceived self-competency and perceived majority-competency, and for the Pressure Control (PC) and No Pressure Control (NPC) groups, Experiment 2.






58


A graphic representation of the triple interaction effects is

provided in Figure 11. Results of the triple interaction, in addition

to those mentioned above, are as follows:

1. A significant two-way interaction between perceived selfcompetency and perceived majority-competency occurred
when the low-competent member of the heterogeneous majority responded first (p <.05).

2. A significant two-way interaction between homogeneity of
majority-competency and perceived self-competency occurred
when perceived majority-competency was high (p< .05).

3. When the low-competent member of a heterogeneous majority
was indicated as the first to respond, significantly more conformity occurred when perceived majority-competency was high and perceived self-competency was low, than under any
other conditions at this level of majority-homogeneity
(p < 001).

4. When the high-competent member of a high-competent heterogeneous majority was portrayed as the first to report
her judgment, Ss who perceived themselves as having lowcompetency confTormed significantly more than did Ss who
perceived themselves as possessing high-competency (p <.05).


Questionnaire Responses

The responses to the questionnaire items, as in Excperiment 1,

provided strong support for the demonstrated conformity behavior.

Analysis of variance summary tables for ratings of difficulty, confidence in own judgments, and confidence in the judgments of others are

provided in Tables 9, 10, and 11, respectively. The significant findings from these analyses are:

1. Ss who perceived themselves as having low-competency
rated the task as more difficult than did Ss who perceived themselves as possessing high-competency
(p < 001).





59


-- High Majority -Competency and Low Self-Competency


High Majority-Conipetency and .Low Majority-Competency and
Low Majority-Competency and


High Self-Competency Low Self-Competency High- Self-Competency


S.-


S..
-s
-5
S..
S.
-S.
5
5-


* 5-.5.
-. -. -. 5-.


Homogeneous Homogeneity of


Heterogerieou (High, Low) Perceived


sl Heterogeneous2
(Low, High)
Majority Competency


Figure ll.--Mean conformity scores for the triple interaction effects, Experiment 2.


10

9

8

7


E



C


6' 5,

4.

3-


II






60


TABLE 9

SUMMARY OF THE ANALYSIS OF VARIANCE OF
DIFFICULTY RATINGS, EXPERIMENT 2


Source Sums of df MenF p .2
Squares Square M% Control vs all others .36 1 .36 Majority-competency (M) .70 1 .70 majority-homogeneity (H) 1.27 2 .63 Self-competency (S) 34.03 1 34.03 20.50 .001 12.12 M x H- 2.09 2 1.05 M x S 6.25 1 6.25 3.76 .10 H x S .18 2 .09 M x HxS 1.29 2 .65 Error 237.75 143 1.66


The relevant values for the computation Squares, 265.31; MS Error, 1.66.


of w are: Total Sums of






61


TABLE 10

SUMMARY OF THE ANALYSIS OF VARIANCE OF RATINGS OF
CONFIDENCE IN OWN JUDGMENTS, EXPERIMENT 2


SoreSums of df Mean F P w2
SouceSquares square W%



Control vs all others .21 1 .21 Majority-competency CM) 3.36 1 3.36 2.35 Majority-homogeneity (H) 9.39 2 4.70 3.29 .05 2.66 Self-competency (S) 34.02 1 34.02 23.84 .001 13.22 M x H 3.72 2 1.86 1.30 M x S 1.01 1 1.01 H x S 5.06 2 2.53 1.77 M x HxS .66 2 .33 Error 204.00 143 1.43


The relevant values f or the computation Squares, 245.22; MS Error 1.42.


of w2 are: Total Sums of






62


TABLE 11

SUMMARY OF THE ANALYSIS OF VARIANCE OF RATINGS OF
CONFIDENCE IN THE JUDGMENTS OF OTHERS, EXPERIMENT 2


Source Suasof d Mean F E2 SursS quare M%


Control vs all others .09 1 .09Majority-competency (M) 15.34 1 15.34 11.12 .01 5.77 Majority-homogeneity (H) .52 2 .26 Self-competency (S) 22.57 1 22.57 16.36 .001 8.77 M x H 3.43 2 1.72 1.25 M x S .84 1 .84 H x S 7.03 2 3.52 2.55 .10 M x H x S 4.18 2 2.09 1.51 Error 197.92 143 1.38


The relevant values for the computation
Squares, 239.83; MS Error 1.41.


of U) are: Total Sums of






63


2. Confidence in own judgments varied directly with perceived self-competency (p <.001) and Ss opposed by
majorities having heterogeneous levels of competency
were more confident in their judgments than were Ss opposed by majorities having homogeneous competency
(p < .05).

3. Confidence in the judgments of others varied directly
with perceived majority-competency (p <.Ol) and inversely
with perceived self-competency (p< .001).

The only variable that significantly influenced responses to

the questionnaire items requiring "rYes"t or "No" responses was perceived self-competency. More Ss who perceived themselves as having low-compe2
tency were doubtful of their accuracy (55 vs 44; X = 3.91; p <.05),

2
were tempted to answer as the others did (517 vs 36; X = 13.39; p <.001), and answered as the others did against their own first choice (32 vs 12;

2
X = 13.09; p <.001).


Summary of Results

Perceived majority-competency had a direct relationship and

perceived self-competency had an inverse relationship with conformity behavior. These results provided strong support for Hypotheses 1 and 2.

Hypothesis 3, which predicted that majorities perceived as

having homogeneous levels of competency would differ in their effect on conformity behavior from majorities perceived as having heterogeneous levels of competency, gained partial support as a result of a triple interaction effect. At high levels of perceived self-competency and perceived majority-competency, majorities portrayed as having homogeneous levels of competency evoked significantly more conformity behavior than did majorities having heterogeneous competency levels.






614


Hypothesis 14 suggested that the influence of 2-member majorities having heterogeneous competency levels would be greater when the low-competent member was the first to respond. This hypothesis not only was not supported but the opposite occurred when perceived self- and majority-competency were both high. This was also a function of the triple interaction.

The questionnaire results were in accord with the conformity results. Perceived difficulty was inversely related to perceived self-competency. Confidence in the judgments of others was directly related to perceived majority-competency and inversely related to perceived self-competency. Confidence in own judgments was directly related to perceived self-competency and was lower in groups having homogeneous versus heterogeneous levels of majority-competency. More Ss with perceived low versus high self-competency reported that they were doubtful of their accuracy, were tempted to answer as the others did, and answered as the others did against their own first choice.




Discuss ion


The findings concerning the effects of perceived self-competency and perceived majority-competency on conformity behavior are similar to the results obtained in Experiment 1 with regard to these variables. These results further indicate the strength of these variables and the relevance they have to conformity behavior. Under conditions of disagreement, individuals must determine to what degree they will rely on their own judgments and to what extent-they will depend on the judgments






65


of others. If information is available to establish relative levels of competency the reliance upon the judgments of the majority tends to be modified accordingly. In the present experiment, when the naive subject had evidence to indicate that she was better able than the majority members to make the type of judgments called for, she demonstrated virtually no tendency to rely on the majority judgment. However, when she perceived herself to have the same level of competency as each of the other members of the group she was inclined to rely on the majority judgment. Such reliance became more pronounced when she viewed herself as being less competent than the majority members.

The results found with regard to the homogeneity of majoritycompetency are of primary interest as the main focus of the experiment was upon this variable and its interaction with perceived selfand majority-competency. The main effect of homogeneity of majoritycompetency was not statistically significant (p <.10); however, the trend was in the same direction as that reported by Gerard (1953) in another context. En both cases majorities perceived as having homogeneous-competency levels tended to elicit somewhat more conformity behavior than did heterogeneous majoritie s. Gerard speculated that when others in the group are perceived as having equal ability, the subjects depend upon all equally and, therefore, upon the group as a whole. In the heterogeneous condition, dependency is not upon the group as a whole but upon the expert within the majority.

This may represent a partial explanation of what occurs but implicitly it states that a 2-member heterogeneous majority would,






66


in effect, be reduced to a majority of one. The results of the present experiment bear this out to a degree. However, the order effects which were associated with heterogeneous majority responses cannot be fully accounted for by suggesting that the dependency is upon the expert within the majority and not upon the majority as a whole. The results of this study indicate that the tendency to perceive a heterogeneous majority as a monolithic whole depends in part upon the naive subject's perceived level of relative competency.

It was predicted that when the low-competent member of a

heterogeneous majority was perceived to be the first to respond, and the most competent majority member's response reflected agreement, the confirmation would serve to provide more informational influence than when the response order was reversed. This was not the case, particularly when the perceived self- and majority-competencies were both at high levels. Under these conditions, significant results were obtained which were opposite from what was predicted. Similar, but nonsig-nificant, results occurred when perceived self- and majoritycompetencies were both at low levels.

In these conditions the naive subject was, in effect, led to

believe that her level of competency was intermediate to the competency levels of the two majority members. When the most competent majority member was seen as responding first and the least competent member followed, the agreement of the majority members was in accord with what would possibly be expected by the naive subject. Gerard and Greenbauma (1962) report that interviewed naive subjects indicated that they tended to perceive Subject "1ll as the "leader" and Subject "2"1 as






67


a "spineless sheep.- Such an observation would be congruent with

the reported levels of competency attributed to Subjects "1" and '12.''

When the least competent member of the majority is perceived to be the first to respond, the situation is considerably different. In this case, the "leader" is perceived as relatively incompetent and the "spineless sheep" is seen as the most competent member of the group. This incongruence apparently reduces the credibility of the majority

judgment and encourages the naive subject to demonstrate independent behavior. When the majority members are perceived as possessing levels of competency equal to that of the naive subject, the credibility of the majority judgment does not suffer and the influence of the majority is maintained.

The order effects of heterogeneous majority responses did not differentially influence conformity behavior when both members of the majority were perceived as having higher levels of competency than the naive subject. That is, when Subjects "1," "2," and "14" were perceived as having competency levels of 4, 10, and 3, respectively, conformity behavior was not different fromn when the respective levels of perceived competency were 10, 4, and 3. In this case the naive subject possibly sees the majority as a whole and is apparently not inclined to view the differing levels of majority member competencies as incongruent with their agreement when the low-competent member is perceived as the first to respond.

The extension of Thibaut and Riecken's (1965) results to thirdperson behavior did not hold up. Apparently when a perceived low-






68




competent third person is viewed as "influencing" a high-competent individual, the perception with regard to locus of causality is considerably different from that which occurs when the subject perceives herself as the influencing agent. The overriding factor seems to be the perceived credibility of the situation which is influenced by the subject's perceived relative competency and the response order of the majority members.


















CHAPTER IV


SUMMARY


The purpose of the research reported here was to investigate the relationships to conformity behavior of majority size, perceived self-competency, perceived majority-competency, and homogeneity of majority-competency. Two experiments were performed to evaluate these situational variables.

Experiment 1 was designed to investigate the effects of majority size, perceived self-competency and perceived majoritycompetency. Previous research has indicated that conformity varies directly with perceived majority-competency and majority size (to a limit of three members), and inversely with perceived self-competency. Similar results were predicted in the present research. Additionally, it was suggested that the three variables are additive in their effects on conformity and, as employed, approximately equivalent.

A 2 x 3 x 2 factorial design with a single control group was employed to test the relationships of the variables. Two levels (high vs low) of the competency variables were used and three levels (3-members vs 2-members vs 1-member) of majority size were employed. Levels of self- and majority-competency were conveyed to the subjects by false feedback indicating performance on a judgmental task concerning relative area represented by geometric figures. A Crutchfield-(1955)


69






70


type apparatus was used and 12 Ss served in each condition. Following the experimental session Ss completed a questionnaire indicating their ratings of task difficulty, levels of confidence in judgments,

and the like.

It was found that all three variables significantly influenced conformity behavior in the predicted directions. The variables were found to be additive in their effects; however, they could not be considered equivalent due primarily to the disproportionate conformity that occurred in response to 2-member majority judgments and the uniformly low level of conformity elicited by majorities of one.

The questionnaire responses indicate the following:

1. Perceived difficulty was inversely related to perceived
self-competency.

2. Confidence in own judgments varied directly with perceived
self-competency and inversely with perceived majoritycompetency.

3. Confidence in the judgments of others varied directly with
perceived majority-competency and inversely with perceived
self-competency.

4. Confidence in the judgments of others was also found to be
a function of an interaction between majority size and
perceived self-competency. When perceived self-competency
was high, majority size had no influence on confidence in
the judgments of others; however, when perceived selfcompetency was low, confidence in the judgments of others
increased in proportion to the size of the majority.

Experiment 2 investigated the interrelated influence of perceived majority-competency (high vs low), homogene ity of majority-competency (homogeneous vs heterogeneous vs hetrgnos) and perceived selfcompetency (high vs low). The two heterogeneous levels refer to the two






71


possible orders in which 2-member majorities can report their judgments. The data for the two experiments were collected simultaneously and the data for the 3-member groups having homogeneous majorities were used in both analyses.

It was hypothesized that perceived majority-competency and

perceived self-competency would have the same relationship to conformity behavior in the context of Experiment 2 as these variables had in the first experiment. Results supporting these hypotheses were obtained.

It was also hypothesized that the level of conformity in groups having majorities with homogeneous levels of competency would differ from that elicited by heterogeneous majorities. Results indicated that only when both perceived self- and majority-competency were high was there a difference. In that case, homogeneous majorities elicited more conformity.

Based on an extension of Thibaut and Riecken's (1965) results

concerning perceived locus of causality, it was predicted that when the least competent member of a heterogeneous majority was perceived as the first to report her judgment, conformity behavior would be greater than when the response order was reversed. Opposite results were obtained when perceived self- and majority-competency were at high levels. Similar findings occurred when both perceived selfand majority-competency were at low levels; however, these results were not statistically significant.






72




It was concluded that perceived relative competency does

indeed influence conformity behavior and serves to modify the influence of majority size. The order effects associated with responses of heterogeneous majorities were interpreted as being a function of the perceived credibility of the situation.









































APPENDICES







































APPENDIX A

EXPERIMENTrAL MATERIALS







75


TABLE 12

DESCRIPTION OF STIMULI


Area Size


R 8 T 6 T 9
R 12 P 12 T 9
T 13.75 T 16 T 6
T 10 T 10 S 9 T T T 13.50 T 15.75 P 12 R 10.5 S 16 T 5 T ff T 12. 50
T 6.75 T 9 p 10 R 70 R 10 T 11I
P 20 T 5 R 17. 5


T 10 T 6.75 R 6
T 10 R 10 R 10 T 11 T 20 R 5 R 17.50 R 12.50 T 6.75 R 6 R IT T T T 9 T 13. 75 R 18 T 7.50 S 12.25 T 8 S 9
T 7 T 12 T 15.75 P 12 P 10.50 R 18 R 6
T 10


T 12.50 S 9 T 7 T 13.50 T 15.75 T 12 R 10.50 P 18 T 7.50 S 12.25 T 8 T 6 T 9
T 12 R 10 R f T 11 R 20

R 17.50 R 10 T 6 R T T 13. 50 P 12 T 9 T 13. 75
P 16 T 7.50 P 12.25


Critical stimuli.

Key: The shape of geometric figures is indicated by P (parallelogram), R (rectangle), S (square), or T (triangle). Underlined values represent the majority response. The values given are area sizes in square units.


Stimulus Number


1
2
3
4
5
6
7
8
9
10 11
12
13*
14* 15* 16* 17* 18
19*
20* 21
22* 23*
24* 25* 26* 2 7* 28
2 9*
30*






76


Figure 12.-A sample stimulus card.


C


A B-






77


TABLE 13

THE POSTEXPERIL4ENTAL QUESTIONNAIRE


1. Please describe in your own words your experience during this
experiment.




2. Please circle on the line below, the number which represents how
easy or difficult you feel it is to select the correct choice for
this perceptual task

EASYl1 2 3 4 5 6 7 DIFFICULT

3. How confident were you of the correctness of the judgments of the
others?

CONFIDTL 1 2 3 4 5 6 7 VERY
CONFDENTCONF IDEN4T ~4. How confident were you of the correctness of the judgments
you made?

NOT AT ALL 1 2 3 4 5 6 7 VERY
CONFIDENT CONFIDENT

5. Would you say that you were quite concerned about those times
that you disagreed with the others? YES NO

6. Did the others make you doubtful of your accuracy? YES NO

7. Would you say that you were tempted to answer as the others did on
some trials? YES NO

8. Did you ever answer as the others did against your own first choice?
YES NO

9. The results of this study depend to a large extent upon identifying
those individuals who have prior knowledge about this particular
experiment or other similar experiments. Please indicate below
whether you had prior information of any type. You will, of course,
receive experimental credit in either case.

YES, I did have prior knowledge of this type of experiment.

NO, I did not have prior knowledge of this type of experiment.

If you answered yes, please indicate below what you knew or had
heard, and where or how you received the information, e.g., from
previously serving in a similar experiment, from class discussion,
a textbook, another student, etc.









































APPENDIX B

RAW DATA AND SUPPLEMENTARY ANALYS IS, EXPERIMENT 1






79


TABLE 14l

INDIVIDUAL CONFORMITY SCORES AND QUESTIONNAIRE RESPONSES BY EXPERIMENTAL CONDITIONS, EXPERIMENT 1

HIGH SELF-COMPETENC Y, LOW MAJOR IT Y-COM{PETENCY

Raw Corrected Postexperimental Questionnaire Items Conformity Conformity 2 3 4 5 6 7 8

1-Member Majority
0 3 4 4 5 0 0 0 0 0 3 5 4 3 0 0 0 0 0 3 5 1 2 0 0 0 0 0 3 6 2 5 0 1 0 0 0 3 2 2 6 0 0 0 0 0 3 5 2 4 0 0 0 0 0 3 6 2 3 0 0 0 0 1 4 2 2 6 0 1 1 0 1 2.5 7 3 2 0 1 0 0
3 6 5 2 4 0 1 0 0 3 4.5 3 4 5 0 0 0 0 3 4.5 6 2 2 0 1 0 0

2-Member Majority
0 3 4 2 3 0 1 1 0 0 3 3 3 5 0 0 0 0 0 3 3 2 5 0 0 0 0 1 2.5 5 3 3 0 1 1 1
1 4 4 2 5 0 1 1 0 1 4 4 2 4 0 0 1 0 1 2.5 5 2 2 1 1 1 0
1 4 3 1 4 0 0 0 0 1 4 6 2 2 0 1 0 0 3 6 3 3 5 1 1 1 0 3 6 4 3 3 0 1 0 0 7 10 4 3 5 0 0 0 0

3-Member Majority
0 3 6 3 4 0 0 1 0 0 3 2 3 5 0 0 0 0 0 3 2 1 7 0 0 0 0 0 3 1 1 7 0 0 0 0 1 4 4 3 3 1 1 1 0 1 1 6 2 6 0 0 1 0 1 4 1 2 6 0 1 0 0 1 4 1 1 7 0 0 0 0 1 4 3 2 5 1 1 1 1 1 4 2 2 6 0 0 0 0 1 4 3 2 5 0 0 0 0 2 5 5 4 6 1 1 1 1






80


TABLE 14 (Continued)

HIGH1 SELF -COMPETENCY, IGH MAJORITfY-COMPETENCY

Raw Corrected Postexperimental Questionnaire Items Conformity Conformity 2 3 4 5 6 7 8

1-Member Majority
O 3 4 2 5 0 1 0 0 0 3 4 2 5 1 1 1 0 0. 3 3 4 6 0 1 0 0
0 3 4 2 5 0 0 0 0 0 3 4 4 4 0 0 0 0 0 0 5 3 3 0 1 0 0 1 1 4 3 3 1 0 1 1 1 4 5 4 6 0 1 1 1 1 4 3 1 6 1 0 0 0 1 4 5 4 3 0 1 0 0 2 2 5 3 4 0 1 1 0 9 10.5 4 4 4 0 0 1 0

2-Member Majority
O 3 5 2 4 1 1 1 0 1 4 3 2 5 0 1 0 0 1 4 6 2 2 1 1 0 0 1 4 5 1 6 1 0 0 0 1 4 1 2 3 1 1 1 1 3 6 4 4 4 1 1 1 0 4 5.5 4 4 4 1 1 1 1 5 6.5 4 4 4 1 1 1 0 7 10 6 2 2 0 1 1 1 9 12 3 4 4 1 0 0 0
11 14 5 3 3 0 1 1 1 15 18 5 5 5 1 0 1 0

3-Member majority
0 3 6 3 4 1 1 1 1 0 0 2 1 6 0 0 0 0 2 5 4 1 6 0 1 0 0 3 6 6 5 4 0 1 1 1 3 6 4 3 4 0 1 1 0 4 7 3 3 6 0 1 1 0 4 7 3 1 5 0 0 1 0 6 9 5 1 5 0 1 1 1 6 9 3 2 3 1 1 1 0 8 9.5 6 3 3 0 1 1 0 11 14 3 4 4 1 1 1 1 15 18 3 4 3 1 1 1 1






81


TABLE 14 (Continued)

LOW SELF-COM'PETENCY, LOW WJORITY-COHPETENCY

Raw Corrected Postexperimental Questionnaire Items Conformity Conformity 2 3 4 5 6 7 8

1-Member Majority
0 3 6 1 5 0 0 0 0 0 3 5 2 5 0 0 0 0 0 3 6 1 1 0 0 0 0 O 3 4 1 5 0 1 0 0 0 1.5 3 3 3 1 1 0 0 0 1.5 4 1 5 0 0 0 0 1 2.5 7 1 1 0 0 0 0
1 1 6 3 5 0 0 0 0 1 4 4 1 4 0 0 0 0 3 4.5 6 4 4 0 0 0 0
4 7 6 3 5 0 1 1 0 5 6.5 4 4 2 0 1 1 0

2-Member Majority
O 3 3 2 4 1 1 1 0 1 4 5 3 4 1 1 1 0 3 6 5 4 4 0 1 1 0 3 6 6 4 5 0 0 1 0 3 6 6 2 5 0 1 1 0 3 4.5 7 2 2 0 0 0 0
4 7 3 3 3 1 1 1 1 4 7 7 2 2 0 0 1 0 5 6.5 6 2 3 1 1 1 0 7 8.5 6 3 2 1 1 1 0 8 9.5 4 5 3 1 1 1 1 11 14 5 4 3 1 1 1 0

2-Member Majority
0 3 6 4 4 0 1 1 0 0 3 6 4 3 1 1 0 0 0 3 6 1 5 0 0 0 0 0 3 1 4 7 0 0 0 0 1 4 6 4 3 1 1 1 1 4 7 6 5 3 1 1 1 1 4 7 5 4 4 0 1 0 0 4 5.5 6 3 3 0 1 1 1
6 9 6 3 3 1 1 1 1 9 12 4 3 3 0 1 1 1 10 11.5 6 3 3 0 1 1 1 14 17 6 4 4 0 1 1 1






82


TABLE 114 (Continued)

LOW SELF-COMPETENCY, HIGH MAJOR ITY-COMPETENCY

Raw Corrected Postexperimental Questionnaire Items Conformity Conformity 2 3 4 5 6 7 8

1-Member Majority
0 3 6 2 2 0 0 1 0 0 3 6 5 3 0 1 0 0 0 1.5 5 5 2 0 1 0 0
1 4 6 5 2 1 1 0 0 2 5 5 14 2 0 0 0 0 2 5 7 1 1 0 1 1 0 2 5 3 4 14 1 0 1 0 3 4.5 5 2 2 0 1 0 0
4 7 4 4 4 0 1 1 0 4 7 5 3 5 0 1 0 0 5 6.5 3 14 3 0 0 1 0
6 6 7 2 2 1 1 1 1

2-Member Majority
0 3 3 .6 1 1 1 1 0
0 3 5 2 5 0 0 0 0 1 4 3 2 2 0 0 0 03 6 5 3 14 0 0 1 0 3 6 5 5 4 0 0 1 1 4 7 6 4 2 1 1 1 0 6 7.5 3 4 4 1 1 1 1 9 12 4 4 3 0 1 1 1 9 12 7 2 1 1 1 1 0 10 13 5 5 14 1 1 0 0 10 11.5 5 4 2 1 1 1 0 11 14 6 5 4 0 1 0 1

3-Member Majority
0 3 5 4 2 0 1 1 0 2 5 6 5 3 0 1 1 0 3 4.5 7 4 1 1 1 1 0
3 6 5 3 5 0 0 0 0 4 7 5 7. 1 1 1 1 0 4 5.5 5 4 2 1 1 1 1
6 9 5 4 5 1 1 1 1 10 13 6 3 4 1 1 1 1 12 15 5 3 3 1 1 1 1 12 15 6 5 4 0 1 1 0 15 18 4 4 4 1 1 1 1 15 18 6 5 2 0 0 1 1






83


TABLE 14 (Continued)

PRESSURE CONTROL

Raw Corrected Postexperimental Questionnaire Items Conformity Conformity 2 3 4 5 6 7 8


1-Member Majority
0 3 O 3
0 3 0 3


1
1


4
4
2.5
4
4
2.5
5
6


2-Member Majority O 3 0 3 0 3 1 4


1
2
2
4
5
5
8 10


4
5
3.5
7
8
6.5 11 13


3-Member Majority O 3 O 3 O 3
1 4
2 3.5 4 5.5
6 9 6 9
6 7.5
6 9 6 9
8 9.5


5
5
4
4
3
6
6
5
6
4
4
6


6
4
5
3
6
6
4
5
2
5
5
6


2
7
6
5
5
2
5
5
6
5
5
5






814


TABLE 14 (Continued)


NO PRESSURE CONTROL

Raw Corrected Postexperimenta 1 Questionnaire Items Conformity Conformity 2 4

0 3 5 5 o 1.5 6 2 0 1.5 2 5
0 3 4 5 1 4 4 4 o 3 3 5 2 5 5 5 o 3 6 1 1 4 2 5 o 3 2 5 o 3 2 6 o 3 1 5 o 3 3 5 o 3 4 5 o 3 2 6 0 3 3 4 2 3.5 4 3
0 3 2 5 0 1.5 6 2 0 1.5 5 2






85


TABLE 15

MEANS AND STANDARD DEVIATIONS OF CORRECTED CONFORMITY SCORES BY
EXPERIMENTAL CONDITIONS, EXPERIMENT 1 (N = 12 IN EACH
CONDITION EXCEPT IN THE NO PRESSURE CONTROL, GROUP)


Majority Size (1-Member)


Self -Competency Majority-Competency


S.D.


HIGH High

3.33 2.50


LOW


Low 3.54 .97


High 4.79 1.65


Low 3.33 1.87


Majority Size (2-Members)


Self-Competency HIGH Majority-Competency High Low

y7.58 4.33
S.D. 4.59 2.05




Majority Size (3-Members)


Selif-Competency Majority-Competency


S.D.


HIGH


High 7.79
4.55


LOW


High 8.25 3.87


Low 6.83 2.77


LOW


Low


3 .50 .96


High 9.91 5.31


Low 7.08
4.33


Control Groups


No Pressure

(n = 20)


S.D.


Pressure

2-Member 3-Member 4-Member


2.92
.04


3.67 .99


5.92 3.17


6.25 2.70






86


TABLE 16

FREQUENCY ANALYS IS OF CONFIDENCE D VALUES
BY LEVELS OF CONFORMITY


Confidence d Value *


+


0


Conformity Level


0-3 4 7


31 31

5


8 +


10

22 13


7

14

11


Total 48 67 29


Total 67


y,2 16.68)


45


32


144


p < .005


Rating of confidence in own judgments minus rating of confidence in the judgments of others.







































APPENDIX C

RAW DATA AND SUPPLEMENTARY ANALYSIS, EXPERIMENT 2






88








TABLE 17

INDIVIDUAL CONFORMITY SCORES AND QUESTIONNAIRE RESPONSES BY EXPERIMENTAL CONDITIONS, EXPERIMENT 2

HIGH SELF-COMPETENCY, LOW MAJORITY-COMPETENCY

Raw Corrected Postexperimental Questionnaire Items Conformity Conformity 2 3 14 5 6 7 8

Heterogeneous1I


0
0
0
0
0
0
0
0
1


3
3
3
3
3
3
1.5
3


1 '4
4 5.5
5 8


He terogeneous
2
0 3 0 3 0 3 o 3


0
0
0




4
5


3
1.5 1.5
4 '4
2.5
7
8






89


TABLE 17 (Continued)


HIGH SELF-COMPETENCY, HIGH MAJORITY-COMPETENCY

Raw Corrected Postexperimental Questionnaire Items Conformity Conformity 2 3 4 5 6 7 8

Heterogeneous1I


0
0
0




3
3
4
5
6 10


3
3
3
4
4
2.5
6
6
7
6.5
9 13


2
4
4
3
5
5
5
6
4
4
6
4


6
6
6
4
4
5
4
4
4
2
6
4


1
1

0


0
0
0


0
0




1


0
0
0
0
0
1
0
0
0
0
0
1


Heterogreneous
02


0
0
0
0





2
2
3
5


3
3
1.5
3
4
4
4
2.5
5
5
4.5
8


2
5
3
5
5
7
5
2
5
4
6
5


3
3
2
4
2
1
5
1
5
3
6
3


5
5
5
4
4
3
4
5
6
5
2
5


1
0
0
0
0
0
0
0
0
0
1
0


1

1
0

1
0
1
0


1
0
0
0
1
0
0
0
1
0


0
0
0
0
0
0
0
0
1
0






90


TABLE 17 (Continued)

LOW SELF -COMPETENCY, LOW MAJORITY-COMPETENCY

Raw Corrected Postexperimental Questionnaire Items Conformity Conformity 2 3 4 5 6 7 8


Heterogeneous


0


3
5
3.5
6
6
8
8
8
9
9 10
12


7
6
6
6
3
4
5
6
6
5
5
4


1
3
4
5
3
5
4
4
6
4
3
4


4
3
5
4
5
4
4
5
2
4
2
4


0
0

1
1
0

1
0

1
0


0
1
1
1
0
1
1
1
1
1
1
1


0
1
1
1
1
1
1
0
1
1
1
1


0


1
1
1
0


Heterogeneous 2


0
0

1

2
2
2
2
3
3

5


3
3
4
4
5
5
3.5 3.5
6
6
6
6.5


0
0
0
1
0




0
0
1
0


0
0
0
0




0


I
0
0
0




1


0
1
0
0
1
0
1
0
0
1
0
0


5
2
6
4
5
6
7
4
6
4
6
6






91


TABLE 17 (Continued)

LOW SELF-COMPETENCY, HIVGH MAJORITY-COMPETENCY

Raw Corrected Postexperimental Questionnaire Items Conformity Conformity 2 3 4 5 6 7 8


Heterogeneous1I


1
2
2
4
4
4
5
6
8


15


4
5
5
7
7
7
8
9 11
14 14 18


4
1
5
6
6
6
5
4
5
6
5
3


2
2
2
4
5
4
5
5
4
4
4
5


4
7
2
2
2
3
2
5
3
3
4
5


0
0
0
0
1
0





0
0


0
0
1
0


0





1
1


0
0
1
0
0
0

1
1
0
1


Heterogeneous 2


2
2
4
5
6
7
8
8
9
9 10 10


5
3.5
7
6.5
9
10 11 11
12 12 13
11.5


5
6
6
4
5
7
7
4
4
6
5
5


3
4
4
6
2
4
4
4
3
5
4
5


5
3
2
2
2
2
1
3
5
3
3
4


0
0


1


1
0
0


1
1
1
1
1
1
1
1
1
1
1
1


0
1
1
1
1
0
1
1
1
1
1
1


0
1
0


0






92


TABLE 18

MEANS AND S.TANDARD DEVIATIONS OF CORRECTED CONFORMITY SCORES BY
EXPERIMENTAL CONDITIONS, EXPERIMENT 2 (N = 12 IN EACH
CONDITION EXCEPT IN THE NO PRESSURE C ONTROL GROUP)



Majority-Homogeneity (Homogeneous)


Self-Competency Majority-Competency


S.D.


High 7.58
4.59


HIGH
Low


4.33 2.05


LOW


High 8.25 3.87


Low 6.83 2.77


Majority-Homogeneity (Heterogeneous,)


Self-Competency HIGH Majority-Competency High Low Hig

y 5.58 3.67 9.0~ S.D. 2.96 1.58 4.1~


Majority-Homogeneity (Heterogeneous2)


Self-Competency Major ity-Competency


S.D.


HIGH


High 3.96 1.57


8
4


LOW


Low 7.29 2.56


LOW


Low


3.62 1.91


High 9.29 3.05


Low 4.62 1.25


Control Groups


S. D.


No Pressure
(n = 20)

2.92
.04


Pressure


5.92 3.17


The homogeneous to Experiment I


majority data and control group data are common and Experiment 2.


h




Full Text

PAGE 1

CONFORMITY, MAJORITY EFFECT, AND PERCEIVED COMPETENCY By HAROLD T. REITAN A DISSERTATION PRESENTED TO THE GRADUATE COUNCIL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FXJfLFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA December, 1967

PAGE 2

ACKNOWLEDGMENTS The writer v/ishes to express his sincere appreciation to Dr. Marvin E. Shaw, chairman of the supervisory committee, for his generous assistance, his welcomed encouragement, and his valuable guidance throughout the graduate career of the writer, as well as during the preparation of this dissertation. Sincere thanks are also extended to Dr. Jack M. V7right, Dr. Stephen T. Margulis, Dr. Henry S. Pennypacker, Dr. Sidney M. Jourard, and Dr. E. Wilbur Bock for their advice and helpful cooperation. ii

PAGE 3

TABLE OF CONTENTS Page ACKNOWLEDGMENTS iiLIST OF TABLES v LIST OF FIGURES vii CHAPTER I INTRODUCTION 1 Situational Variables 2 Purpose 6 II EXPERIMENT 1 8 Hypotheses 11 Method 16 Experimental Design 16 Subjects 16 Apparatus 17 Stimulus Materials 18 Procedure 18 Results 21 Scoring 21 Analysis of Conformity Scores 23 Analysis of Questionnaire Responses 27 Summary of Results 40 Discussion 42 Conformity Behavior 42 Questionnaire Responses 45 Perceived Difficulty 45 Confidence in the Judgments of Others 46 Confidence in own Judgments 47 Conformity Behavior as a Rational Process .... 47 iii

PAGE 4

TABLE OF CONTENTS (Continued) CHAPTER Page III EXPERIMENT 2 ^9 Hypotheses 51 Method 52 Experimental Design 52 Subjects 53 Procedure 53 Results 54 Conformity Behavior SyQuest ionna ire Responses 58 Summary of Results 63 Discussion 64 IV SUMMARY 69 APPENDICES 73 A EXPERIMENTAL MATERIALS 74 B RAW DATA AND SUPPLEMENTARY ANALYSIS, EXPERIMENT 1 . 78 C RAW DATA AND SUPPLEMENTARY ANALYSIS, EXPERIMENT 2 . 87 BIBLIOGRAPHY 93 BIOGRAPHICAL SKETCH 97 iv

PAGE 5

LIST OF TABLES Table P^g^ 1. Summary o£ the Analysis of Variance of a Portion of Data from Shaw et al (1966) 13 2. Predicted Relationships of Combined Levels of Variables 15 3. Summary of the Analysis of Variance of Conformity Scores, Experiment 1 4. Summary of the Analysis of Variance of Difficulty Ratings, Experiment 1 5. Summary of the Analysis of Variance of Ratings of Confidence in Judgments of Others, Experiment 1 32 6. Summary of the Analysis of Variance of Ratings of Confidence in Own Judgments, Experiment 1 35 7. Reported Levels of Homogeneity of Majority-Competency 54 8. Summary of the Analysis of Variance of Conformity Scores, Experiment 2 56 9. Svimmary of the Analysis of Variance of Difficulty Ratings, Experiment 2 60 10. Summary of the Analysis of Variance of Ratings of Confidence in Own Judgments, Experiment 2 61 11. Summary of the Analysis of Variance of Ratings of Confidence in the Judgments of Others, Experiment 2 62 12. Description of Stimuli 75 13. The Postexperimental Questionnaire 77 14. Individual Conformity Scores and Questionnaire Responses by Experimental Conditions, Experiment 1 79 15. Means and Standard Deviations of Corrected Conformity Scores by Experimental Conditions, Experiment 1 (N = 12 in Each Condition Except in the No Pressure Co^ntrol Group) 85 V

PAGE 6

LIST OF TABLES (Continued) Table P^'^g^ 16. Frequency Analysis of Ratings of Confidence D Values by Levels of Conformity 86 17. Individual Conformity Scores and Questionnaire Responses by Experimental Conditions, Experiment 2 88 18. Means and Standard Deviations of Corrected Conformity Scores by Experimental Conditions, Experiment 2, (N 12 in Each Condition Except in the No Pressure Control Group) 92 vi

PAGE 7

LIST OF FIGURES Figure Page 1. Mean conformity scores for the three levels of majoritysize, and for the Pressure Control (PC) and No Pressure Control (NPC) groups, Experiment 1 26 2. Mean conformity scores for high and low levels of perceived self -competency and perceived majoritycompetency, Experiment 1 26 3. Obtained mean conformity scores by experimental conditions compared with the predicted relationships of the experimental groups. Experiment 1 28 4. Mean difficulty ratings for high and low levels of perceived self -competency, and for the Pressure Control (PC) and No Pressure Control (NPC) groups. Experiment 1 31 5. Mean ratings of confidence in the judgments of others for high and low levels of perceived self -competency and perceived majority-competency, and for the Pressure Control (PC) group. Experiment 1 33 6. Ifean ratings of confidence in the judgments of others by the three levels of majority size for high and low levels of perceived self -competency. Experiment 1 33 7. Mean ratings of confidence in own judgments for high and low levels of perceived self -competency and perceived majority-competency, and for the Pressure Control (PC) and No Pressure Control (NPC) groups, Experiment 1 36 8. Mean ratings of confidence in own judgments for the three levels of majority size, Experiment 1 36 9. The relationship between ratings of confidence in own judgments and confidence in the judgments of others by levels of conformity. Experiment 1 38 vii

PAGE 8

LIST OF FIGURES (Continued) Figure Page 10. Mean conformity scores for high and low levels of perceived self -competency and perceived majoritycompetency, and for the Pressure Control (PC) and No Pressure Control (NPC) groups, Experiment 2 57 11. Mean conformity scores for the triple interaction effects, Experiment 2 59 12. A sample stimulus card 76 viil

PAGE 9

CHAPTER I INTRODUCTION A great deal of research in recent years has been devoted to the investigation of conformity behavior under a variety of conditions. Two studies which have become classics in the area are those of Sherif (1935) and Asch (1951). Sherif demonstrated the effect of a totally unstructured stimulus on norm formation, and Asch showed that the group has considerable influence even with regard to stated opinions about undisputed facts. With the impetus provided by these studies, the research that followed has focused on three general classes of variables: stimulus, personality, and situational factors It has been generally found that objective stimuli elicit les conformity than subjective stimuli and that the more ambiguous the stimuli the greater the conformity behavior (Crutchf ie Id 1955; Sherif and Sherif, 1956; Luchins and Luchins, 1955a; Nichols, 1964). The studies devoted to identifying the personality correlates of conformity behavior have provided some evidence to indicate that the type of individual who is least able to resist conformity pressure is "submissive, low in self-confidence, less intelligent, lacking in originality, authoritarian minded, lacking in achievement motivation, conventional, and desires social approval" (Blake and Mouton, 1961, p. 23). The attempts to identify the personality correlates of 1

PAGE 10

2 conformity behavior have, however, encountered probleras similar to those which have stymied investigators who have searched for leadership "traits." In the same sense that leadership is not unique to a given "type" of individual, it would be inappropriate to label a person demonstrating the above array of personality characteristics as a "conformist," or one who demonstrates antithetical characteristics as a "nonconformist." As indicated by virtually every study in the area, conformity behavior is a complex process which is highly dependent upon situational variables. Individuals who conform in one social situation may maintain total independence in another situation It may well be that personality characteristics determine, to a large extent, the conditions under which an individual will tend to conform and the situational and stimulus variables, in interaction with the personality variables, determine the degree to which one conforms. Situational Variables From the research that has focused on conformity behavior, it appears that the investigation of situational, in contrast to personality, variables is the more fruitful approach to gaining a better understanding of this aspect of social behavior. Allen (1965) has indicated the considerable attention situational variables have received and the extent to which conformity behavior is modified under a variety of social stimulus situations. Some situational variables, which have been shown to significantly influence conformity behavior are:

PAGE 11

The attractiveness of the group Generally, the more attractive the group the more influence is exerted on deviant members and the greater the conformity behavior of the deviant (Festinger et a 1 1950; Schachter, 1951; Gerard, 1954; Lott and Lott, 1961; Walker and Heyns, 1962; Newcomb, 1961). Status in the group Romans* (1950) and Hollander's (1958) opposing hypotheses serve well to reflect the conflicting results which have been obtained with respect to this variable. Homans suggests that the person with the highest status will more closely follow the group norras and the more certain a person is about his status, the less he will conform to the norms of the group. Hollander hypothesizes that the high status person will be free to deviate from the norms of the group because of "idiosyncratic credits" he has built up in the group over a period of time. Sundby, as cited by Harvey and Consalvi (1960), found status to be positively related to conformity when the task was relevant to the group. Interdependence More conformity behavior is evidenced in interdependent conditions than in independent conditions. When group pressure is increased, conformity also increases in interdependent groups and decreases slightly in independent groups (Thibaut and Strickland, 1956). Size of the majority Asch (1956) systematically investigated this variable and concluded that conformity increased significantly when the unanimous majority was increased from 1 to 3 members. There was no appreciable increase in influence for majorities of greater

PAGE 12

size. Rosenberg (1963) found a curvilinear relationship between group size and conformity behavior. Using 1-, 2-, 3-, and 4-member majorities, the latter elicited significantly less conformity than did 3-member majorities which evoked the greatest amount of conformity. Luchins and Luchins (1955b) reported that 3-raember majorities demonstrated significantly more conformity behavior than 1-member majorities. Nichols (1964) reported similar findings. Kidd (1958) and Goldberg (1954) failed to find significant relationships between majority size and conformity. In both of these studies the majority judgments were conveyed to the naive subject by the experimenter. Goldberg (1954) suggested that the subjects may have been conforming to the experimenter and the size of the majority may have made little difference. In a related study, Shaw et al (1957) concluded that direct conformity pressure is more effective than indirect pressure. The indirect pressure exerted by the experimenters in the Kidd (1958) and Goldberg (1954) studies may have been insufficient to elicit differential conformity behavior. In any event, the bulk of the evidence suggests that the size of the majority is a highly relevant variable with respect to conformity behavior. Composition of the group Most of the research in this area has been devoted to comparisons of homogeneous and heterogeneous groups along a variety of dimensions. Considering the diversity of the dimensions used — disability, skill, knowledge, sex, race — it is noteworthy that some uniformity of results has been achieved. Although there are conflicting results, the bulk of the evidence indicates

PAGE 13

5 that individuals in homogeneous groups demonstrate more conformity behavior than do those in heterogeneous groups (Festinger and Thibaut, 1951; Gerard, 1953; Tuddenham,MacBride, and Zahn, 1958; Linde and Patterson, 1964). Luchins and Luchins (1955b) and Reitan and Shaw (1964) found mixed-sex groups to demonstrate more conformity behavior than same-sex groups. These results are not necessarily contradictory in the sense that the subjects may have considered the mixedsex groups to be more homogeneous on some other dimension, and it is highly probable that the mixed-sex groups were seen as more attractive than the same-sex groups. Newcomb (1961) suggests that perceived similarity leads to increased attraction which in turn is seen to produce a high level of conformity. Competency on the assigned task An individual who perceives himself as less competent than others in performing a task will conform to a greater extent than an individual who has reason to believe that he is more competent than the majority members (Samelson, 1957; Hochbaum, 1954; Mausner, 1954b; Di Vesta, 1959; Fagen, 1963). By manipu lating the competency of both the naive subject and the majority, Fagen (1963) demonstrated that conformity behavior was more dependent on relative competency than absolute levels of competency. Shaw et al. (1966) found similar results.

PAGE 14

6 Purpo sg Tv7o of the mentioned situational variables were selected for further investigation — group size and competency. As Allen (1965) stated, "Only slight attention has been given to the effect on conformity of the size of the group opposing a person" (P. 160). Several studies, as indicated, have examined this variable, but the nature of the effect and the relationship of this situational variable to other situational variables warrants additional attention. It is conceivable, for example, that group size and perceived relative competency function in much the same manner and are additive in their effects on conformity behavior. Perceived relative competency has received som.e attention, but little is known about the possible differential effects on conformity by majority members who are perceived as having homogeneous or heterogeneous levels of competency. Thus, the three primary interrelated purposes of the present study are to: 1. Examine the relationships among majority size, perceived self -competency and perceived majority-competency. 2. Investigate the possible differential effects of homogeneous and heterogeneous majority-competency levels. 3. Determine if order of response of majority members attributed with heterogeneous competencies influences the degree to which the naive subject conforms. Two experiments were performed to accomplish the above purposes. The first experiment was devoted to examining the relationships among majority size and perceived selfand majority-competency levels.

PAGE 15

7 The second experiment was designed to investigate the differential effects on conformity behavior of homogeneous and heterogeneous majority competencies and the order effects associated with a majority perceived as having heterogeneous competency levels.

PAGE 16

CHAPTER II EXPERIMENT 1 Deutsch and Gerard (1955) offered the following hypothesis: "The more uncertain the individual is about the correctness of the judgment of others, the less likely he is to be susceptible to informational social influences in making his judgments" (P. 630). They suggest that for the judgments of others to represent conformity pressure, the naive subject must perceive the others as being more accurate than himself. Thibaut and Strickland (1956) conceptualized the reliance or nonreliance upon the judgments of others in terms of "group set" and "task set." These investigators suggest that in "task set" the person is disposed to view the other individuals in the group as "mediators of fact" and, in responding to the perceptions and attitudes of others, the person is concerned not with achieving or maintaining a social relationship but with achieving or maintaining a cognitive clarity about his environment. Thibaut and Strickland (1956) provide a hypothesis similar to that of Deutsch and Gerard (1955): ". to the extent that he lacks confidence in the validity of his judgments, he will tend to depend on other observers for his attitudes and assessments" (P. 116). 8

PAGE 17

9 There is a great deal of empirical evidence to support these hypotheses. Kelman (1950) found prior success reduced suggestibility and prior failure increased suggestibility. Mausner and Bloch (1957) examined the interaction of success and failure on conformity behavior. In 2-person situations, the past success of the subject (S_) coupled with the past failure of his partner resulted in significantly less conformity than when S had previously experienced failure and his partner had experienced success. Mausner (1954a) had earlier reported that when a person experiences a situation vjhich causes hira to lower his "confidence in self," conformity increased, and (Mausner, 1954b) if one's partner is perceived as more accurate, conformity is increased. Fisher, Williams, and Lubin (1957) found that the more confident one is in his judgments, the less he will conform to the judgments of others. Fagen (1963) manipulated real and perceived subject-competency and perceived majority-competency. He found conformity to be a function of all three variables. Shaw etal. (1966), in an unpublished study, obtained similar results with respect to perceived selfand perceived majority-competency. Rosenberg (1963) examined the effects of perceived low competency on conformity behavior and the resultant confidence in self and confidence in partner(s). Ss were told that they had been wrong on 8 of 10 initial trials and partner(s) had been right on 7 of the 10 trials. Subjective measures of confidence in self and confidence in partner(s) were used and it was found that conformity had (1) a negative relationship with confidence in self, (2) a positive relationship

PAGE 18

10 with confidence in partner(s), and (3) the difference between confidence in self and confidence in partner(s) was negatively correlated with conformity. Rosenberg varied the size of the majority but limited competency to a single level. He reported a curvilinear relationship between majority size and conformity, with maximum conformity elicited by majorities of three and significantly less by majorities of four. These results support Asch's (1956) findings although Asch did not find the significant curvilinear effect reported by Rosenberg. As previously mentioned, Luchins and Luchins (1955a) and Nichols (1964) also found majorities of three to evoke significantly more conformity than majorities of one. The empirical results show that conformity varies directly with perceived majority-competency and majority size (at least to a maximum of three majority members), and inversely with perceived selfcompetency. As suggested, it is conceivable that majority size and perceived majority-competency may function similarly in eliciting conformity behavior and may be additive in their effects. When confronted with the unanimously wrong judgments of the majority, S quite likely attributes some level of relative competency to himself with respect to the group. The degree to which S evidences agreement with the majority judgment depends to a great extent on his perceived relative competency. If S_ is opposed by a 3-member majority and suspects that all of the others are more capable than he is in the performance of the assigned task, he is more likely to agree with the unanimous majority than if he has reason to believe that the others are less

PAGE 19

11 competent than he is A similar appraisal would be made by S_ opposed by 2and 1-member majorities; however, less conformity would be expected in the smaller groups. If S is told that his performance on a task was better or worse than the performance of the other members of the group, S_ has a more objective evaluation of his relative competency. The empirical evidence indicates that such information has a highly significant effect upon conformity behavior. Samelson (1957) suggested that the cognitive process associated with conformity possibly takes the following form: faced with highly discrepant information from two normally veridical sources — his own perceptions and the reported judgments of others--S must determine which source to reject and which to accept. In such a situation, the so-called "conformist" tends to rely most heavily upon social reality, whereas the "independent" and the "anticonf ormist" would be more disposed to reject the judgments of others and rely upon physical reality. Any information which will serve to support social reality or physical reality will tend to enhance conformity or independence, respectively. Information concerning performance on a task could serve to influence both types of behavior as would variations in majority size Hypotheses It was anticipated that the present study would support previous findings; therefore the hypotheses with regard to main effects are as follows:

PAGE 20

12 :^|. ; / 1. A majority of three evokes more conformity than a majority of two, which in turn evokes more conformity than a majority of one. 2. A majority perceived as having high-competency elicits more conformity than a majority perceived as possessing 1 owc ompe ten c y 3. Individuals who perceive themselves as having low-jcompetency conform more than those who perceive themselves as having high-competency. If, as suggested, majority size and perceived relative competency function in much the same manner in the elicitation of conformity behavior, predictions concerning the effects of the various combinations of levels of the variables were deemed feasible. Previous research provides an indication of the relative influences these variables have on conformity. An examination of Nichols' (1964) data concerning majority size reflects that female S_s who were required to select the largest of three geometric figures conformed 43 percent of the time when faced with the unanimously wrong judgment of a majority of three. When a majority of one was used, conformity dropped to 17 percent. This difference yields a t value of 4.205 which is significant beyond the .001 level of confidence. The proportion of the variance accounted for by the majority size variable under these conditions is 27.5 per2 cent. The proportion of the variance (uo ) was computed in accordance with the procedure outlined by Hays (1963, p. 327). Statisticians are not in complete accord as to the appropriate use of this statistic. In particular, it is questioned as a population estimate of strength of association.

PAGE 21

13 Shaw et al (1966) used 4-member female groups and virtually the same stimulus materials as those used by Nichols (1964). The results of this study with regard to high and low levels of perceived selfand majority-competency are indicated in Table 1. TABLE 1 SUMMARY OF THE ANALYSIS OF VARIANCE OF A PORTION OF DATA FROM SHAW ET AL. (1966) Source df Mean „ uj F p Square *^ 2 (%) Perceived self -competency (S) 1 147.00 22.07 .001 30.4 Perceived majoritycompetency (M) 1 102.08 15.33 .001 23.0 S X M 1 20.34 3.05 Error 44 6.66 Shaw's Ss who perceived themselves as having highcompetency conformed only 12.53 percent of the time, while those who had been informed that they possessed lowcompetency conformed 35.87 percent of the time. Perceived high majority-competency evoked conformity behavior on 33.87 percent of the occasions as compared with 18 percent for the perceived low-competent majorities. Perceived high and low se If -competency was produced by telling the S^s that they had been correct on either 8 or 4 of 10 initial trials. Majority-competency was manipulated in a like manner

PAGE 22

14 by stating that the majority members had each been correct on 8 of the 10 initial trials (high competency) or 4 of the 10 trials (low competency) 2 The obtained cu 's would indicate that if the variables were to be ordered from greatest to least effect, perceived se If -competency would be ranked first; majority size, second; and perceived majoritycompetency, third. The differences, however, do not appear to be significant and a best guess would be that, at the levels tested, the three variables are equivalent in their effects upon conformity behavior. If weights are assigned to levels accordingly and if the effects of these three variables are additive, Table 2 shows the predicted relationships of various combinations of levels. The greater the value of the combined weights, the greater the level of expected conformity behavior. Also included in the table are means for 6 of the 12 conditions which were obtained in a pilot study. Each mean is based on an n of 8. As can be seen, the pilot data generally support the predicted relationships. Hypothesis 4 deals with the bases of Table 2 and the resultant predictions. 4a. The effects of the three variables on conformity behavior are approximately equivalent to one another. 4b. The variables are additive in their effects on conformity behavior to the extent that they will produce the relationships indicated by the combined weights. The combined weights are considered to represent an ordinal scale but it is conceivable that the values of the lower end of the scale are below threshold; therefore there may be no differences in conformity under conditions represented by combined weights of 5, 4, and 3

PAGE 23

15 TABLE 2 PREDICTED RELATIONSHIPS OF COMBINED LEVELS OF VARIABLES SelfCompetency MajorityCompetency Majority Size Combined Weights Pilot Study Means Le ve 1 Wc IlL 1 ^ \Tl> 1 Lit. V t, J. LTo 1 cf h -iWti J-^l'L L T T ro 1 Ljti Vc. J. low 3 high 3 high 3 9 8.0 low 3 high 3 inter 2 8 5.0 low 3 high 3 low 1 7 low 3 low 1 high 3 •7 / 7 C J / D low 3 low 1 inter 2 6 low 3 low 1 low 1 5 high high 3 high 3 7 2.40 high high 3 inter 2 6 1.78 high high 3 low 1 5 high low 1 high 3 5 1.40 high low 1 inter 2 4 high low 1 low 1 3

PAGE 24

16 Method Experimental Design A 2 X 3 X 2 factorial design was employed testing perceived self-competency (high vs low), majority size (3-member vs 2-member vs 1-member), and perceived majority-competency (high vs low). Twelve S_s were tested in each of the conditions, representing a total of 144 experimental subjects. Fifty-six control £s were also included 12 S^s under each level of majority size were exposed to conformity pressure but did not receive feedback information relative to selfor majority-competency, and 20 additional control Ss made all judgments with neither ccmpetency information nor conformity pressure. Subjects Due to significant differences in conformity behavior of males and females (Beloff, 1958; Reitan and Shaw, 1964), only females were used as £s The £s were obtained on a volunteer basis from introductory psychology, sociology, and speech courses at the University of Florida. Those in introductory psychology courses received experimental credit for their participation. Ss were requested to volunteer for experimental sessions in groups of four, three or two. Sampling bias was minimized by randomizing the sequence of conditions and by assigning S^s to conditions in the order in which they volunteered for participation in the experiment.

PAGE 25

A total of 251 Ss participated in the experiment. Data from 32 S^s were discarded due to prior knowledge of the experimental treatment and an additional 19 Ss were randomly eliminated from the sample to achieve equal n's. Apparatus The apparatus was similar to that employed by Crutchfield (1955). Five adjacent cubicles were located at one end of a 12' x 20' room. The experimenter's cubicle, which was equipped with a Beseler opaque projector and a master panel of lights and switches, was in the middle position. £s occupied the four other cubicles, each of which contained a panel of twelve lights arranged in four rows of three lights (one row of lights for each of four Ss and one light for each of three possible responses). Mercury switches corresponding to the three columns of lights were mounted on each panel to permit Ss to report their judgments. Each switch activated one light in the fourth row of lights on a given S_'s panel, and a corresponding light on E_'s master response panel. Although each S_ was led to believe that the first three rows of lights reflected the judgments of the other Ss, the lights were actually controlled by E by means of his master switch panel, which was wired to permit lights in identical positions to be illuminated in all cubicles simultaneously. Each S was led to believe that she was Subject 4 and the light responses she saw on her panel for Subjects 1, 2, and 3 represented the judgments of the other Ss

PAGE 26

18 Stimulus Materials Two sets of stimulus cards were used with each card depicting three geometric figures (squares, rectangles, triangles, parallelograms) of different sizes. The first set of ten cards was used for "practice" trials and the second set of twenty cards was used under conformity pressure conditions. The two sets were matched with respect to areas represented but differed with regard to the shape of the figures, i.e., the figures on cards 1, 11, and 21 were identical in area, as were the figures on cards 2, 12, and 22, etc. The cards were projected on a screen at the opposite end of the room and Ss were instructed, in all cases, to select the figure representing the largest area. Table 12 in Appendix A provides the area sizes, order of presentation, and majority selections. A sample stimulus card is shown in Figure 12, Appendix A. Procedure Upon reporting to the laboratory to participate in a "perception" experiment, Ss were requested to select a cubicle position and be seated. The following instructions were read to the Subject: I request that there be no discussion or communication among you while you are seated in the cubicles. If you have any questions during the experiment I will discuss them with you individually. A number of cards will be projected on the screen you see in front of you. On each card will be three figures of differing areas. In each case you are requested to determine which of the three figures represents the largest area. Each card will be shown on the screen for a period of three seconds. Following removal of the card you are requested to make your judgment. (Response sheets were distributed at this point.)

PAGE 27

19 Please note the subject number on your response sheet. This is your subject number and it also represents the cubicle you have selected. Please indicate your judgments as to which figure represents the largest area by checking the appropriate column on your response sheet. Let's try a sample card. In this case, if you determine that figure "C" represents the largest area, you would check "C" on your response sheet for "sample" card. Indicate your judgments in the same manner for cards 1-10. Please be as accurate as possible. Are there any questions? The response sheets were collected by E following the presentation of the tenth stimulus card and E purportedly scored them. The instructions for the second phase of the experiment were then read with appropriate feedback information for each experimental condition: We have now completed the first phase of the experiment. I will tell you how you have performed so far; however, in order that you remain anonymous, I will give you results by subject number, the number which you noted on your response sheet. It should be noted that there is no logical order of subject numbers according to cubicle location. (Competency feedback was provided at this point.) During the second phase of the experiment you will indicate your judgments relative to the figure representing the largest area by means of the switches in each of your cubicles. You will notice that your panel has four rows of lights numbered 1, 2, 3, and k. These numbers represent subjects — each of you — 1, 2, 3, and 4. The columns of lights are lettered A, B, and C. Corresponding to the columns of lights are three switches, A, B, and C. If, for example, you are Subject 2 and you decide that figure "C" represents the largest area, you would turn "on" the "C" switch and the "2C" light would be illuminated on your panel, on the panel(s) of the other subject(s) and on my pane 1 as we 1 1 There are 20 cards in this sequence and they will be presented in the same manner as were the first 10 cards. Do not make your responses until I request them by calling out your subject number. Each card will be presented for three seconds; I will remove the card and will then ask for your judgments by subject number. Please leave your lights on until I have had the opportunity to

PAGE 28

20 record your judgments. After I have done so, I will request you to turn your lights off. Again, please be as accurate as possible. Are there any questions? The two levels of perceived self -competency were manipulated by stating that Subject ^ was correct on either 3 or 7 of the initial 10 trials. Perceived majority-competency was manipulated by stating that Subject 1 (majority of one), Subject 1 and 2 (majority of two), or Subjects 1, 2, and 3 (majority of three) were each correct on 3 or 7 of the initial 10 trials. In all cases Subject 4 was the last to respond. On 15 of the 20 trials in the second phase of the experiment, all experimental Ss, and the control Ss exposed to conformity pressure, were required to respond following the judgments of the unanimously wrong majority. On these trials the majority "selected" the smallest of the three figures in each case. On trials 11, 12, 18, 21, and 28, the majority unanimously "selected" the largest figure. Agreement with the unanimously wrong majority constituted conformity behavior. Following presentation of the stimuli, Ss were requested to complete a questionnaire designed to obtain additional data concerning the effects of conformity behavior. Ss were asked to indicate on a seven-point scale how difficult the task was, how confident they were in the correctness of their judgments, and how confident they were in the correctness of the judgments made by the others. Additionally, they indicated whether they were concerned about the disagreement; were doubtful of their accuracy; were tempted to answer as the others did; and did answer as the others did against their own first choice.

PAGE 29

21 The questionnaires were also used to identify those Ss who had previous information about the experimental condition. The postexperimental questionnaire is reproduced in Table 13, Appendix A. The instructions read to the control Ss, who were not exposed to conformity pressure, were modified to omit any reference to the judgments of Subjects 1, 2, and 3. Naturally, switches representing the judgments of Subjects 1, 2, and 3 were not used while these control Ss made their judgments. All Ss were debriefed and requested not to divulge any information about the experiment to potential participants. Results Scoring Raw conformity scores were the number of times S agreed with the unanimously wrong majority. Raw conformity scores, therefore, had a possible range of 0 to 15. During the initial ten trials the Ss who were subsequently exposed to conformity pressure (all experimental Ss and the Pressure Control Ss) selected the smallest of the three figures a total of ^9 times. Although there was no systematic relationship between these errors and experimental conditions, the conformity scores of the 42 £s who were responsible for these judgments were proportionally reduced to correct for errors of this nature. In a few cases negative conformity scores resulted, so a value of "3" was added to all conformity scores. The scores of the No Pressure

PAGE 30

22 Control Ss were corrected in the same manner. The corrected scores were used for statistical analysis. The postexperimental questionnaire responses were scored as follows: Question 1. The open-ended question provided only qualitative data. Responses indicating the S was aware of the purpose of the experiment served as one basis for the elimination of S from the experiment. The responses to this question by all retained Ss were examined by three judges to determine degrees of suspicion as to the nature of the experiment. Ten Ss were identified by one or more of the judges as expressing some degree of suspicion, including two Ss who were identified by all three judges as indicating a high level of suspicion. No experimental or control condition included more than two of the ten "suspicious" Ss Questions 2, 3, and k The scoring of perceived difficulty, confidence in the judgments of others, and confidence in own judgments was identical. Ss responded to these questions on a sevenpoint scale with the highest score indicating greater difficulty and greater confidence. Control Ss who were not exposed to conformity pressure (No Pressure Control) responded only to Question 2 (level of difficulty) and Question 4 (confidence in own judgments). Questions 5, 6, 7, and 8 Responses to these questions were "Yes" or "No" which were coded "1" and "0," respectively. Question 9 Affirmative responses to this question, about prior knowledge of the experiment, with an appropriate explanation, served to eliminate 28 £s from the experiment.

PAGE 31

23 All conformity scores and questionnaire responses are given in Appendix B, Table Ik. Analysis of Conformity Scores The uncorrected conformity responses showed that the experimental £s and the Pressure Control (PC) Ss agreed with the unanimously wrong majority judgments on 566 occasions, or 20.96 percent of the time. These conforming responses were made by 130 of the 180 Ss (72.22 percent). On the initial ten trials, under no conformity pressure, these £s selected the smallest figure on 49 occasions, or on 2.72 percent of the opportunities. The errors committed under no pressure conditions were random across conditions and were not related to level of conformity behavior. The difference in errors under conformity pressure and under no pressure to conform is significant (p<.001). A comparison of the experimental and PC Ss with the 20 Ss who responded to all 30 stimulus cards under no conformity pressure provides similar results. Four of the No Pressure Control (NPC) Ss made one or more of the erroneous majority selections on a total of six occasions (2 percent). These results are not significantly different from those of the experimental and PC Ss under no pressure conditions, but are significant (p<.001) when compared with the responses of the experimental and PC £s under conformity pressure conditions. A 2 X 3 X 2 factorial design with a single control group (PC) was employed to examine the effects on conformity behavior of perceived majority-competency (high vs low), majority size (3-member vs 2-member vs 1-member), and perceived se If -competency (high vs low).

PAGE 32

24 Due to the relatively simple nature of the task, it was anticipated that there would be no systematic relationship between real competency and conformity behavior. Therefore, real competency was not included as an independent variable. An examination of the data revealed no discernible relationship between real competency and conformity behavior. Table 3 summarizes the analysis of variance results and indicates the percentage of the variance accounted for by each main effect. Means and standard deviations of the conformity scores by experimental conditions are provided in Table 15, Appendix B. The PC group did not differ from the experimental groups; however, a comparison between the PC and NPC groups yielded a t of 21.64 which is significant (p<.001) and indicates the general effectiveness of the conformity pressure. All three independent variables significantly influenced conformity behavior, as hypothesized. Of the three, group size, which accounted for 13.04 percent of the variance, proved to be the the strongest. A comparison of means indicated that the 1-member majorities elicited significantly less conformity behavior than did the 2member majorities (p<.001), but the difference between the 2and 3member majorities was not significant, although it was in the 1_, 2 The control group was not included in the computation of ou since the primary focus of interest was upon the relative contribution of the three experimental variables occurring in combination. This holds true for the computation of (^2 throughout the remainder of Experiment 1 and for Experiment 2.

PAGE 33

25 anticipated direction. Hypothesis 1, therefore, gained partial support. Figure 1 shows the differential effects of the three levels of majority size. Also included are the mean conformity scores for the PC and NPG groups. TABLE 3 SUMMARY OF THE ANALYSIS OF VARIANCE OF CONFORMITY SCORES, EXPERIMENT 1 Source Sums of Squares df Mean Square F P 2* U) (%) Control vs all others 3.82 1 3.82 Majority-competency (M) 170.08 1 170.08 15.62 .001 7.01 Majority size (G) 318.14 2 159.07 14.61 .001 13.04 Self-competency (S) 102.52 1 102.52 9.41 .01 4.02 M X G 52.23 2 26.12 2.40 M X S 2.92 1 2.92 G X S 30.00 2 15.00 1.38 M X G X S 21.07 2 10.53 Error 1819.37 167 10.89 The relevant values for the computation Squares, 2248.11; MS Error, 11.75. 2 of IX are: Total Sums of Hypotheses 2 and 3 were also supported Majorities depicted as possessing high competency elicited significantly more conformity behavior than did majorities having low competency (p<.001), and Ss

PAGE 34

26 1-member 2-member 3-member Control Majority Size ^''^PS Figure 1. — Mean conformity scores for the three levels of majority size, and for the Pressure Control (PC) and No Pressure Control (NPC) groups, Experiment 1. oL Self Majority High Low Perceived Competency Figure 2. --Mean conformity scores for high and low levels of perceived se If -competency and perceived majority-competency, Experiment 1.

PAGE 35

27 who perceived themselves as having low competency conformed less than did Ss who perceived that they had performed at a high level (p<.01). Figure 2 indicates the mean conformity scores at the two levels of these variables. It was predicted that the three independent variables would have equivalent (Hypothesis 4a) and additive (Hypothesis 4b) effects on conformity behavior. To test this two-part hypothesis a comparison was made between the obtained experimental group means and the predicted relationships of these groups (see Table 2). A Spearman rank order correlation of .83 was obtained (p<.01). This would indicate some support for the hypothesis. Figure 3 depicts the predicted and obtained ordering of means by experimental condition. The differential strength of the majority size and perceived selfcompetency variables, the relatively high level of conformity behavior that occurred in the 3-member homogeneous groups, and the consistently low level of conformity in the 2-member groups accounted for most of the discrepancies between the predicted and obtained results Analysis of Questionnaire Responses Responses to Questions 2, 3, and 4 yielded information relative to how difficult S_ perceived the task, how confident S was in the judgments of the others, and how confident S was in her own judgments. These data were examined by analysis of variance.

PAGE 36

Obtained Conformity Predicted Conformity Vmember 2-member 3"member hmember 2-member 3-member Majority Size Figure 3. — Obtained mean conformity scores by experimental conditions compared with the predicted relationships of the experimental groups. Experiment 1.

PAGE 37

29 Question 2. Please circle on the. line below, the number which represents how easy or difficult you feel it is to select the correct choice for this perceptual task. A summary of the 2x3x2 analysis of variance carried out on the perceived difficulty ratings is shown in Table 4. The only variable that achieved significance was perceived self -competency (p<.001), and there were no significant interactions. Figure 4 indicates the mean levels of difficulty ratings for the two levels of perceived selfcompetency as well as the mean ratings of difficulty for the PC and NPC groups. A comparison between the two control groups means yielded a _t of 8.86, which is significant beyond the .001 level of confidence. Question 3. How confident were you in the judgments of the others? The analysis of variance summary table for the ratings of confidence in the judgments of others (Table 5) reflects that such confidence was a function of perceived majority-competency (p<.001), perceived self -competency (p<.001), and an interaction effect between the size of the majority and perceived self -competency (p<.01). A plot of mean confidence in the judgments of others for high and low levels of perceived self -competency and perceived majority-competency (Figure 5) indicates that confidence in the judgments of others varied directly with perceived majority-competency and inversely with perceived self -competency. Figure 5 also indicates the PC group mean rating of confidence in the judgments of others. Figure 6 indicates the interaction between majority size and perceived self -competency. Tests for simple main effects indicated

PAGE 38

30 TABLE k SUMMARY OF THE ANALYSIS OF VARIANCE OF DIFFICULTY RATINGS, EXPERIMENT 1 Source Sums of Squares df Mean Square F p 2* (£ (%) Control vs all others .52 1 .52 Majority-competency (M) .18 1 .18 Majority size (G) 2.73 2 i .JO Se If -competency (S) 47.84 1 47.84 27.81 .001 15.12 M X G 4.04 2 2.02 1.17 M X S 1.17 1 1.17 G X S 9.55 2 4.78 2.78 M X G X S 4.07 2 2.04 1.19 Error 286.89 167 1.72 The relevant values for the computation Squares, 302.83; MS Error, 1.77. 2 or (JO are : Total Sums of

PAGE 39

31 6 Figure 4. — Mean difficulty ratings for high and low levels of perceived se If -competency, and for Pressure Control (PC) and No Pressure Control (NPC) groups. Experiment 1.

PAGE 40

32 TABLE 5 SUMMARY OF THE ANALYSIS OF VARIANCE OF RATINGS OF CONFIDENCE IN JUDGMENTS OF OTHERS, EXPERIMENT 1 Source Sums of Squares df Mean Square F P 2^ UU (%) Control vs all others .19 1 .19 Majority-competency (M) 19.51 1 19.51 14.14 .001 7.85 Majority size (G) 3.60 2 1.80 1.30 Self -competency (S) 21.01 1 21.01 15.22 .001 8.50 M X G .68 2 .34 M X S 2.00 1 2.00 1.45 G X S 13.51 2 6.75 4.89 .01 4.71 M X G X S .60 2 .30 Error 230.92 167 1.38 The relevant values for the computation Squares, 230.84; MS Error, 1.29. 2 of uj are : Total Sums of

PAGE 41

33 o o r 3 o ''^ CD Qj •E E CO cQj.iz i Self Majority PC Higli Low Perceived Competency V/Control Group Figure 5. — Mean ratings of confidence in the judgments of others for high and low levels of perceived self -competency and perceived majority-competency, and for the Pressure Control (PC) group, Experiment 1. cu o -a 03 o O K/i .E| CD cn Qj c: High Self-Competency 4Low Self -Competency ~l hmember y 2-member Majority Size 1 — 3-member Figure 6. — Mean ratings of confidence in the judgments of others by the three levels of majority size for high and low levels of perceived se If -competency. Experiment 1.

PAGE 42

34 that when perceived self -competency was high the confidence in judgments of others was uniformly low regardless of majority size, but when perceived self -competency was low, the confidence in the judgments of others increased as the size of the majority increased (p<.01). The interaction also resulted in a significant difference (p<.001) between the confidence in judgments of others by high-and low-competent Ss opposed by 3-member majorities. Question 4. How confident were you of the correctness of the judgments you made? A 2 X 3 X 2 analysis of variance carried out on the ratings of confidence in own judgments indicated that all three main variables influenced the confidence S had in her own judgments. The strongest variable, in this case, was perceived self -competency which accounted for 13.86 percent of the variance. Table 6 shows the summary of the analysis of variance carried out on confidence in own judgments. Figure 7 indicates graphically the mean ratings of confidence in own judgments for the levels of perceived selfand majority-competency. Also included are the mean ratings of confidence in own judgments for the PC and NPC Ss A t_ test comparing the mean ratings of the control groups indicated that they were not significantly different. Figure 8 portrays the mean ratings of confidence in own judgments for the three levels of majority size. A further examination of the responses to Questions 3 and 4 was made in terms of the frequency with which Ss rated confidence in the judgments of others higher than confidence in own judgments.

PAGE 43

35 TABLE 6 SUMMARY OF THE ANALYSIS OF VARIANCE OF RATINGS OF CONFIDENCE IN OWN JUDGMENTS, EXPERIMENT 1 Source Sums of Squares or Mean Square P (%) Control vs all others .73 1 .73 Majority-competency (M) 7.56 1 7.56 4.47 .05 1.97 Majority size (G) 11.93 2 5.96 3.53 .05 2.88 Self -competency (S) 43.34 1 43.34 25.64 .001 13.86 M X G 4.55 2 2.27 1.34 M X S 2.51 1 2.51 1.49 G X S 5.68 2 2.84 1.68 M X G X S 6.67 2 3.34 1.98 Error 282.81 167 1.69 The relevant values for the computation Squares, 299.16; MS Error, 1.64. 2 of iju are : Total Sums of

PAGE 44

36 Perceived Competency Figure 7. — Mean ratings of confidence in own judgments for high and low levels of perceived self -competency and perceived majoritycompetency, and for the Pressure Control (PC) and No Pressure Control (NPC) groups, Experiment 1. 5 03 1-member 2-member 3-member Majority Size Figure 8. — Mean ratings of confidence in own judgments for the three levels of majority size, Experiment 1.

PAGE 45

37 Chi-square tests revealed that more Ss opposed by high-competency majorities reported a higher level of confidence in majority judgments 2 than in their own (23 vs 9; X. = ^.87; p<.05). Similarly, more Ss who perceived themselves as demonstrating lowversus high-competency reported that they had greater confidence in the judgments of others 2 than in their own judgments (27 vs 5; X = 19.44; p<.001). To determine the relationship between conformity behavior and the tendency to assign greater confidence to the judgments of others, the difference between confidence in own judgments and judgments of others was determined for each of the experimental Ss Three categories were used to classify the relationships of responses: confidence in self greater than confidence in others, equal ratings of confidence in self and others, and confidence in self less than confidence in others. Figure 9 portrays these relationships by five levels of conformity behavior. A chi-square test on these data, with the three cells of greatest conformity pooled to permit sufficiently large expected cell frequencies, indicated that as conformity behavior increased, proportionally fewer Ss reported greater confidence in own judgments and proportionally more Ss reported equal levels of confi2 dence and greater confidence in the judgments of others (x =16.68; p<.005). Table 16 in Appendix B contains the frequency data which provided these results. Chi-square tests were also carried out on the frequencies of "Yes" and "No" responses to Questions 5 through 8. Although more Ss who perceived themselves as having low-rather than high-competency

PAGE 46

38 Conformity Scores Figure 9. — The relationship between ratings of confidence in own judgments and confidence in the judgments of others by levels of conformity. Experiment 1.

PAGE 47

39 indicated that they were concerned about disagreements (28 vs 21), were doubtful of their accuracy (49 vs 42), were tempted to answer as the others did (46 vs 35), and answered as the others did against their own first choice (20 vs 14), none of these differences achieved an acceptable level of significance. The perceived majority-competency variables provided significant results with regard to all four questions. More Ss who perceived themselves opposed by high rather than low majority-competency reported that they were concerned about disagreements (32 vs 17; 2 2 X = 6.80; p<.01), were doubtful of their accuracy (52 vs 39; x =5.04 2 p<.05), were tempted to answer as the others did (48 vs 33; x = 6.35; p<.02), and answered as the others did against their own first choice (22 vs 12; = 3.85; p<.05). The majority size variable also influenced responses to Questions 5, 7, and 8. As majority size increased, more Ss stated that they were tempted to answer as the others did (14 vs 33 vs 34; 2 X =21.52; p<.001), and more reported that they answered as the 2 others did against their own first choice (3 vs 11 vs 20; ^ = 16.69; p<.001). With regard to concern over disagreements, Ss opposed by l-member majorities made the fewest number of affirmative responses, and £s opposed by 2-member majorities made the most (7 vs 24 vs 18; X^ = 13.80; p < .01).

PAGE 48

kO Summary of Results All three independent variables (majority size, perceived majority-competency, and perceived self-competency) were found to significantly influence conformity behavior with majority size being the strongest of the three variables by accounting for 13.04 percent of the variance. Hypothesis 1 received partial support in that l-member majorities evoked significantly less conformity behavior than did 2-member majorities, but 3-member majorities did not elicit significantly more conformity behavior than did the 2-member majorities. Hypotheses 2 and 3 received strong support; majorities perceived as having high competency elicited significantly more conformity than did majorities perceived as possessing low-competency, and Ss who perceived themselves as having low-competency conformed more than did Ss who perceived themselves as having high-competency. Hypothesis 4, which predicted that the three variables were additive and equivalent, received some support by virtue of the fact that the predicted relationships among experimental groups correlated favorably with the obtained experimental group means (£ = .83; p<.01). The differences between the predicted relationships and the obtained experimental group means were primarily due to (1) the relatively high level of conformity elicited by 2-member majorities when perceived majority-competency and perceived se If -competency were equivalent, and (2) the uniformly low level of conformity behavior in response to l-member majority judgments. The questionnaire results were consistent with the more objectively measured conformity behavior. Perceived self -competency was the

PAGE 49

41 only variable that significantly influenced ratings of difficulty; however, all three variables significantly influenced the ratings of confidence in own judgments. Confidence in the judgments of others was primarily a function of perceived self -competency and perceived competency of the majority; the confidence in the judgments of others varied directly with perceived majority-competency and inversely with perceived self -competency. It was also found that a significant inter action occurred between majority size and perceived self -competency; if perceived self -competency was low, confidence in the judgments of others varied directly with the size of the majority. This relationship resulted in a significant difference between perceived high and low se If -competency Ss exposed to 3-member majorities, but not at the other two levels of majority size. It was also found that perceived majority-competency and perceived se If -competency significantly influenced the relationship of confidence in the judgments of others and confidence in own judgments. More Ss opposed by high-competent majorities and more Ss who perceived themselves as having low-competency reported less confidence in their own judgments than they did in the judgments of others. Furthermore, the difference between confidence in own judgments and confidence in the judgments of others was negatively correlated with conformity behavior. The responses to the questionnaire items dealing with concern over disagreements, doubtfulness of accuracy, temptation to answer as the others did, and whether they answered as the others did against

PAGE 50

42 their own first choice, supported the indication of the relative strength of the variables. The levels of perceived self -competency failed to yield any significant differences in responses to these questions; whereas perceived majority-competency reflected significant differences for all four questions. Majority size provided significantly differing frequencies with regard to concern over disagreement and the temptation to respond as the others did. Discuss ion Conformity Behavior Perceived se If -competency, perceived majority-competency, and majority size all significantly influence the level of conformity behavior. As was hypothesized, perceived high se If -competency resulted in less conformity behavior than did perceived low self -competency; perceived high majority-competency elicited more conforming responses than did perceived low majority-competency; and, conformity behavior varied directly with majority size. These results serve to extend and clarify results from other studies dealing with these variables in singular and paired relationships. The prediction that the three variables are equivalent and additive was basically supported; however, relationships among these variables were not consistent. Regardless of majority size, virtually no conformity behavior was demonstrated when S perceived herself as possessing greater competency than the other member (s) of the group.

PAGE 51

43 It is suggested that under such conditions the combined effect of the variables was below threshold. When all group members are perceived as demonstrating the same level of competency, the 1-member majority again failed to elicit conformity behavior. Under these conditions, 2and 3-member majorities elicited significant levels of conformity; however, there was no significant difference in conformity elicited by these larger majorities. Only under conditions when S perceived herself as less competent than the other member (s) of the group did the three variables approach additivity and equivalence of effect. These results, and those mentioned above, indicate that the majority size effect on conformity behavior is, to a great extent, a function of how competent the other members of the group are perceived by S. Quite logically, if S_ has reason to believe that she is more competent than all the other members of the group, regardless of majority size, she is not inclined to rely on the majority judgments under informational influence conditions. If, however, the majority members are each perceived as having levels of competency equal to or greater than S_ the size of the majority becomes salient. The subject apparently perceives that there is greater probability that 2and 3-member majorities would be correct in unanimous judgments than would be a single member of a group. When the majority members are portrayed as more competent than the naive subject, the majority effect is highly differentiated as a function of majority size.

PAGE 52

44 Although the correlation between the obtained experimental group means and the predicted relationships among them was significant (p<.01), the variables did not account for equivalent proportions of the variance. This was in part due to the minimal conformity in response to the 1-member majority. The limited effect that a majority of one has on conformity behavior is well documented (Asch, 1956; Luchins and Luchins, 1955a; Nichols, 1964; Rosenberg, 1963); however, it was anticipated that a high competent majority of one would elicit more conformity behavior than it did. More ambiguous stimulus materials would possibly have increased the conformity influence of the l-member majority without appreciably increasing the influence of the 3-member majority (Bass, 1961). In any event, the relative contribution any relevant situational variable has on conformity behavior depends on the levels employed and the context in which they are tested. MacBride and Tuddenham (1965) concluded that a demonstration of ability to perform a task can prepare an individual better to withstand group pressure and a demonstration of inaccuracy can weaken selfconfidence and reduce ability to withstand pressure. A similar conclusion could be made from the results of the present study with regard to the influence of the perceived se If -competency variable on conform• ity behavior. When the effects of the other variables are considered, however, the conclusion must be qualified. At the levels tested, the absolute level of competency was largely irrelevant. If one is unsuccessful at a task, and everyone else in the group is also unsuccessful, conformity behavior is little different from that in a situation in

PAGE 53

45 which all group members perceive themselves as having high levels of competency. Inaccuracy per se does not reduce one's ability to withstand pressure, nor does a demonstration of ability necessarily dispose one to independent behavior. However, if one is aware of one's own level of competency and the competency of the other members of the group, conformity behavior becomes a function of the individual's perceived relative competency and the size of the group. Questionnaire Respo nses Responses to the questionnaire items were consistent with the behavior exhibited under conformity pressure. This is to be expected since the conformity behavior and the questionnaire responses were not independent responses. It is reasonable to expect that conformity behavior varies directly with perceived difficulty and confidence in the judgments of others, and inversely with confidence in one's own judgments. Such results were obtained in the present study and provide support for similar results reported by Rosenberg (1963). Perceived Difficulty Ratings of the task difficulty were found to be a function of perceived self -competency but were not related to the other independent variables. The significant difference between the difficulty ratings of the PC and NPC Ss would indicate that ratings of task difficulty were also influenced by the experimental treatment. For the experimental £s, the effect of the disagreeing majority was modified by the individual's perceived self -competency. All of this points to

PAGE 54

46 the consideration that perceived difficulty ratings are highly susceptible to influence. Confidence in the Judgments of Others Confidence in the judgments of others was found to be a function of perceived self -competency, perceived majority-competency and an interaction between majority size and perceived self -competency. The findings relative to the relationship of the two competency variables indicate that the Ss were attendant to the reported results on the initial ten "practice" trials. The interaction effect is interesting in that it may reflect the perceived credibility of the majority judgments when the perceived self -competency is low. Under conditions of perceived high se If -competency, majority size had a very slight negative relationship with reported confidence in the judgments of others, i.e., as majority size increased, confidence in the majority judgments decreased. On the other hand, when Ss perceived themselves as having low-competency,, confidence in the judgments of the majority increased as majority size increased. The difference between high and low perceived se If -competency when Ss were opposed by 3 -member majorities, is significant at beyond the .001 level of confidence. Unfortunately, it is not possible to determine to what degree this interaction effect is attributable to the perceived levels of relative competency, the experimental treatment and/or the conformity behavior itself. It can be concluded, however, that reported confidence in the judgments of others is modified to some degree by one's own behavior.

PAGE 55

H7 Confidence in Own Judgments Confidence in S*s judgments varied directly with perceived se If -competency and inversely with perceived majority-competency which would be expected. However, the finding that confidence in own judgments increased as majority size increased was not expected and this finding is incongruent with all other findings in the study. It is suggested that this is a spurious finding and should be disregarded. This is supported by the fact that the mean ratings of confidence in own judgments by Ss opposed by 3-member majorities were not significently different from those made by Ss opposed by 1-member majorities. The only significant difference existed between the ratings made by Ss opposed by 2and 3-menber majorities (p<.05). Conformity Behavior as a Rational Process The examination of the relationship of the difference between the two confidence ratings and conformity behavior provides strong support for Rosenberg's (1963) results in this regard. At the lower levels of conformity behavior, proportionally more Ss indicated greater confidence in their own judgments than in the judgments of others. At the higher levels of conformity, however, proportionally more Ss reported equal levels of confidence in own and other judgments and more confidence in the judgments of others than in their own. These results and the relationship of conformity behavior to perceived self -competency, perceived majority-competency, and majority size, support Samelson's (1957) supposition concerning the cognitive process associated with conformity behavior. In the experimental

PAGE 56

48 session, the "conformist," the "independent," and the "anticonf ormist" all may be considered as demonstrating rational and logical behavior. Each attends to the cues she had determined to be relevant in that particular situation and, depending on the direction and perceived strength of these selected cues, demonstrates conforming or independent behavior. The present study found perceived relative competency and majority size to represent salient cues. The naive subject demonstrated conformity behavior when she perceived herself to be less competent than the other member (s) of the group and when she was outnumbered by a unanimous majority. Independent behavior occurred, regardless of majority size, when the majority members were perceived as less competent than the naive subject perceived herself to be. Furthermore, the individual who perceived herself to have low-competency perceived the task to be relatively difficult, had relatively little confidence in the judgments she made, and relatively high confidence in the judgments made by others. Conversely, the individual who perceived herself as possessing high-competency perceived the task to be relatively simple, was relatively confident in the judgments she made and had relatively little confidence in the judgments made by the majority.

PAGE 57

V CHAPTER III EXPERIMENT 2 Experiment 1 was designed to investigate the effects on conformity of perceived self -competency, perceived majority-competency, and majority size. In all cases the majority members were portrayed as having competency levels equivalent to one another. The purpose of this experiment is to investigate the nature of conformity behavior when majority members are portrayed as possessing differing levels of competency. If majority members are perceived as having homogeneous competency levels, the order in which the majority members report their judgments is immaterial. However, if the majority members are perceived as possessing differing levels of competency, the degree to which the naive S is influenced by the majority may depend to some extent on the order in which the majority members report their judgments and the degree to which the naive S perceives the judgment of one majority member influencing the other (s). As previously mentioned, the studies which have dealt with the composition of the group as an independent variable have provided no conclusive evidence concerning the effects of homogeneous and heterogeneous groups. Gerard (1953) examined the effects on opinion change of small groups (8 14 members) perceived as having homogeneous 49

PAGE 58

50 and heterogeneous levels of competency. He found limited evidence to indicate that groups having homogeneous levels of competency exerted greater pressure toward uniformity and that change toward uniformity occurred only in a homogeneous high-pressure condition. The studies dealing with the same-sex versus mixed-sex variable (Luchins and Luchins 1955b; Tuddenham etal 1958; Reitan and Shaw, 1964) may be interpreted in terms of perceived competency, but the conflicting results indicate that factors in addition to perceived competency are influencing conformity behavior when the sex composition of the group is varied. It is suspected that the heterogeneity of majority-competency, in and of itself, is not a critical factor but that the order of majority member responses is probably a significant variable. Thibaut and Riecken (1965) have demonstrated that the degree to which a person perceives himself as influencing others depends to a significant degree on the perceived power relationships. Ss perceived the causal locus for compliance as "internal" to the high-power stimulus person and "external" to the low-power stimulus person. Although Thibaut and Riecken were concerned with power relationships and the effects of perceived conformity of majority members, their results are, in a sense, applicable to the proposed investigation of order effects of heterogeneous majorities. If, in a 2-member majority, the high-competent majority member is the first to indicate her judgment, the low-competent majority member may be seen as merely

PAGE 59

51 conforming to the judgment of the high-competent member when she agrees with the initial judgment. If S perceives such to be the case, the majority-size effect may be reduced, since would possibly be influenced only by the high-competent member's response. On the other hand, when the low-competent majority member is the first to respond, the high-competent majority member would not likely be seen as conforming to the initial judgment. Instead, she may be viewed as confirming it. The judgments of the majority members, in this case, would quite likely evoke more conformity behavior than the reverse order of majority member responses. Hypotheses The hypotheses offered in Experiment 1 with respect to perceived self -competency and perceived majority-competency were not expected to be altered by the inclusion of the homogeneity of majority competency variable. Thus, again, it was expected that: 1. Ss who perceive themselves as possessing low -competency conform morethan Ss who perceive themselves as having high -competency. 2. Majorities perceived as having high -competency evoke more conformity than majorities perceived as having low -competency. No directional hypothesis is offered with regard to homogeneous majorities versus heterogeneous majorities. Although the bulk of evidence suggests that more conformity occurs in homogeneous groups this evidence is relatively weak. It is expected, however, that conformity under the two conditions may differ and the hypothesis becomes

PAGE 60

52 3. Majorities perceived as having homogeneous and heterogeneous competencies differ in their elicitation of conformity. With regard to Hypothesis 3 an assumption must be made as to what represents equivalent levels of homogeneous and heterogeneous competencies. For purposes of this experiment, it was assumed that if the sums of majority member competency levels are equal, regardless of the individual values, the majorities would be equally competent. It is recognized that this is a rather gross assumption but is considered appropriate at this point. As has been suggested, order effects associated with heterogeneous majorities may influence conformity as a function of perceived locus of causality. The hypothesis is: 4. More conformity is elicited when the low -competent member of a 2-member majority is the first to report her judgment than when the high-competent member reports her judgment initially. Method Experimental Design A 2 X 3 X 2 factorial design was used to test perceived majority-competency (high vs low), homogeneity of perceived majoritycompetency (homogeneous vs heterogeneous vs heterogeneous^) and perceived se If -competency (high vs low). The two levels of heterogeneity refer to the two possible orders in which 2-member majorities can report their judgments. Three-member groups were used and, as in Experiment 1, 12 Ss were tested in each condition.

PAGE 61

53 The data representing the judgments of homogeneous majorities were those which were obtained for 3-member groups in Experiment 1. The 3-member control Ss from Experiment 1 were also used for Experiment 2. Data for the two experiments were collected simultaneously and Ss were randomly assigned to the experimental conditions. Subjects Subjects for this experiment came from the same subject pool as did those who served in Experiment 1. In addition to the 3-member groups and the NPC subjects from Experiment 1, 114 more female undergraduates served in the present experiment. Data from 18 Ss were discarded as follows: 6 Ss were familiar with the experimental treatment, experimenter error invalidated data from three Ss, and 9 Ss were randomly discarded from within conditions to achieve equal n's. Procedure The same apparatus and stimulus materials that were used in Experiment 1 were used in this experiment and all procedures in Experiment 1 pertaining to instructions, presentation of stimuli, manner of making responses, administration of questionnaire, and debriefing, also apply to Experiment 2. Table 7 indicates the levels of homogeneous and heterogeneous majority-competency. The two levels of perceived se If -competency were again manipulated by stating that Subject 4 was correct on either 3 or 7 of the 10 initial trials. The highand low-homogeneous majority-competency

PAGE 62

54 data were the 3-member group data from Experiment 1: high-homogeneous competency was indicated by reporting that Subjects 1 and 2 had both been correct on 7 of the 10 initial trials; low -homogeneous majoritycompetency was indicated by stating that Subjects 1 and 2 were each correct on 3 of the 10 initial trials. The high-competent heterogeneous majority was perceived by S as having competency levels of 4 and 10. Low-competent heterogeneous majorities were attributed with competency levels of 0 and 6. TABLE 7 REPORTED LEVELS OF HOMOGENEITY OF MAJORITY -COMPETENCY Homogeneous Heterogeneous ^ Hetero^ feneous^ High 7, 7 10, 4 ^, 10 Low 3, 3 6, 0 0, 6 Results Conformity Behavior Scoring of conformity and questionnaire responses for Experiment 2 were identical to those used for Experiment 1. The experimental and PC £s conformed 53 6 times out of 2,340 opportunities to do so, or 21.52 percent of the time. Compared with the NPC group, this represented a significant level of conformity behavior (p<.001). Table 8 summarizes the analysis of variance that

PAGE 63

55 was carried out on the corrected conformity scores of the experimental and PC Ss. The table also provides the percentage of variance (uj ) accounted for by the significant main effects and the one significant interaction. Hypotheses 1 and 2 predicted that, as in Experiment 1, the two levels of perceived majority-competency and perceived selfcompetency would again have significant and opposite effects on conformity behavior. These hypotheses were supported (p<.001). The means for the levels of these variables and for the two control groups are shown in Figure 10. Hypothesis 3 stated that majorities perceived as having homogeneous and heterogeneous levels of competency would differ in their elicitation of conformity. The main effect was not considered significant (p<.10); however, a significant triple interaction occurred (p<.05) which indicated that when both se If -competency and majoritycompetency were perceived to be high, the homogeneous majority elicited significantly more conformity than did the heterogeneous majorities (p< .01). Based on an extension of Thibaut and Riecken's (1965) results. Hypothesis 4 predicted that when the low-competent member of a 2-member heterogeneous majority was the first to report her judgment, conformity behavior on the part of a naive subject would be greater than when the high-competent member was the first to respond. This hypothesis was not supported. In fact, when perceived se If -competency and perceived majority-competency were both high, the opposite results were obtained (p<.05).

PAGE 64

56 TABLE 8 SUMMARY OF THE ANALYSIS OF VARIANCE OF CONFORMITY SCORES EXPERIMENT 2 Source Sums of Squares df Mean Square F P 2^ (JU (%) Control vs all others .70 1 .70 Majority-competency (M) 178.78 1 178.89 19.51 .001 9.43 Majority-homogeneity (H) k9.l5 2 24.58 2.68 .10 1.72 Self-competency (S) 276.39 1 276.39 30.14 .001 14.86 M X H 2.70 2 1.35 M X S 5.64 1 5.64 H X S 26.33 2 13.17 1.14 M X H X S 60.82 2 30.41 3.32 .05 2.38 Error 1311.73 143 9.17 The relevant values for the computation Squares, 1790.23; MS Error, 9.02. 2 of u) are : Total Sums of

PAGE 65

57 Self High Low Control Groups Perceived Competency Figure 10 Mean conformity scores for high and low levels of perceived se If -competency and perceived majority-competency, and for the Pressure Control (PC) and No Pressure Control (NFC) groups, Experiment 2.

PAGE 66

58 A graphic representation of the triple interaction effects is provided in Figure 11. Results of the triple interaction, in addition to those mentioned above, are as follows: 1. A significant two-way interaction between perceived selfcompetency and perceived majority-competency occurred when the low-competent member of the heterogeneous majority responded first (p<.05). 2. A significant two-way interaction between homogeneity of majority-competency and perceived self-competency occurred when perceived ma jority -competency was high (p<.05). 3. When the low-competent member of a heterogeneous majority was indicated as the first to respond, significantly more conformity occurred when perceived majority-competency was high and perceived self -competency was low, than under any other conditions at this level of ma jorityhomogeneity (p< .001). 4. When the high-competent member of a high-competent heterogeneous majority was portrayed as the first to report her judgment, Ss who perceived themselves as having lowcompetency conformed significantly more than did Ss who perceived themselves as possessing high -competency^ (p<.05). Questionnaire Responses The responses to the questionnaire items, as in Experiment 1, provided strong support for the demonstrated conformity behavior. Analysis of variance summary tables for ratings of difficulty, confidence in own judgments, and confidence in the judgments of others are provided in Tables 9, 10, and 11, respectively. The significant findings from these analyses are: 1. Ss who perceived themselves as having low -competency rated the task as more difficult than did Ss who perceived themselves as possessing high-competency (p< .001).

PAGE 67

59 10 9 8 "e 3 5 4 o C_3 3 2 High Majority-Competency and Low SelfCompetency High Majority-Competency and High Self-Competency Low Majority-Competency and Low Self-Competency Low Majority-Competency and High SelfCompetency I \ Homogeneous Heterogeneous^ Heterogeneous^ (High, Low) (Low, High) ^ Homogeneity of Perceived Major ity Competency Figure 11. — Mean conformity scores for the triple interaction effects. Experiment 2.

PAGE 68

60 TABLE 9 SUMMARY OF THE ANALYSIS OF VARIANCE OF DIFFICULTY RATINGS, EXPERIMENT 2 Source Sums of Squares df Mean ^ Square 2 p ^* Control vs all others .36 1 .36 Majority-competency (M) .70 1 .70 Ma jorityhomogeneity (H) 1.27 2 .63 Self -competency (S) 34.03 1 34.03 20.50 .001 12.12 M X H 2.09 2 1.05 M X S 6.25 1 6.25 3.76 .10 H X S .18 2 .09 M X H X S 1.29 2 .65 Error 237.75 143 1.66 2 The relevant values for the computation of uj are: Total Sums of Squares, 265.31; MS Error, 1.66.

PAGE 69

TABLE 10 SUMMARY OF THE ANALYSIS OF VARIANCE OF RATINGS OF CONFIDENCE IN 0V7N JUDGMENTS, EXPERIMENT 2 Source Sums of Squares Mean Square E cu P ^ (%) Control vs all others .21 1 .21 Majority-competency (M) 3.36 1 3.36 2.35 Majority-homogeneity (H) 9.39 2 4.70 3.29 .05 2.66 Self -competency (S) 34.02 1 34.02 23.84 .001 13.22 M X H 3.72 2 1.86 1.30 M X S 1.01 1 1.01 H X S 5.06 2 2.53 1.77 M X H X S .66 2 .33 Error 204.00 143 1.43 The relevant values for the computation of uj are: Total Sums of Squares, 245.22; MS Error 1.42.

PAGE 70

62 TABLE 11 SUMMARY OF THE ANALYSIS OF VARIANCE OF RATINGS OF CONFIDENCE IN THE JUDGMENTS OF OTHERS, EXPERIMENT 2 Source Sums of Squares df Mean Square F 2* p ^ (%) Control vs all others .09 1 .09 Majority-competency (M) 15.34 1 15.34 11.12 .01 5.77 Majority-homogeneity (H) .52 2 .26 Self -competency (S) 22.57 1 22.57 16.36 .001 8.77 M X H 3.43 2 1.72 1.25 M X S .84 1 .84 H X S 7.03 2 3.52 2.55 .10 M X H X S 4.18 2 2.09 1.51 Error 197.92 143 1.38 2 The relevant values for the computation of cu are: Total Sums of Squares, 239.83; MS Error 1.41. ~

PAGE 71

63 2. Confidence in own judgments varied directly with perceived self -competency (p<.001) and Ss opposed by majorities having heterogeneous levels of competency were more confident in their judgments than were Ss opposed by majorities having homogeneous competency (p< .05). 3. Confidence in the judgments of others varied directly with perceived majority-competency (p<.01) and inversely with perceived self-competency (p<.001). The only variable that significantly influenced responses to the questionnaire items requiring "Yes" or "No" responses was perceived self -competency. More Ss who perceived themselves as having low-compe2 tency were doubtful of their accuracy (55 vs 44; x = 3.91; p<.05)j 2 were tempted to answer as the others did (57 vs 36; x = 13.39; p<.001), and answered as the others did against their own first choice (32 vs 12; = 13.09; p < .001). Summary of Results Perceived majority-competency had a direct relationship and perceived se If -competency had an inverse relationship with conformity behaviorThese results provided strong support for Hypotheses 1 and 2. Hypothesis 3, which predicted that majorities perceived as having homogeneous levels of competency would differ in their effect on conformity behavior from majorities perceived as having heterogeneous levels of competency, gained partial support as a result of a triple interaction effect. At high levels of perceived self -competency and perceived majority-competency, majorities portrayed as having homogeneous levels of competency evoked significantly more conformity behavior than did majorities having heterogeneous competency levels.

PAGE 72

64 Hypothesis k suggested that the influence of 2-inember majorities having heterogeneous competency levels would be greater when the low-competent member was the first to respond. This hypothesis not only was not supported but the opposite occurred when perceived selfand majority-competency were both high. This was also a function of the triple interaction. The questionnaire results were in accord with the conformity results. Perceived difficulty was inversely related to perceived self -competency. Confidence in the judgments of others was directly related to perceived majority-competency and inversely related to perceived self-competency. Confidence in own judgments was directly related to perceived self -competency and was lower in groups having homogeneous versus heterogeneous levels of majority-competency. More Ss with perceived low versus high se If -competency reported that they were doubtful of their accuracy, were tempted to answer as the others did, and answered as the others did against their own first choice. Discuss ion The findings concerning the effects of perceived self -competency and perceived majority-competency on conformity behavior are similar to the results obtained in Experiment 1 with regard to these variables. These results further indicate the strength of these variables and the relevance they have to conformity behavior. Under conditions of disagreement, individuals must determine to what degree they will rely on their own judgments and to what extent they will depend on the judgments

PAGE 73

65 of others. If information is available to establish relative levels of competency the reliance upon the judgments of the majority tends to be modified accordingly. In the present experiment, when the naive subject had evidence to indicate that she was better able than the majority members to make the type of judgments called for, she demonstrated virtually no tendency to rely on the majority judgment. However, when she perceived herself to have the same level of competency as each of the other members of the group she was inclined to rely on the majority judgment. Such reliance became more pronounced when she viewed herself as being less competent than the majority members. The results found with regard to the homogeneity of majoritycompetency are of primary interest as the main focus of the experiment was upon this variable and its interaction with perceived selfand majority-competency. The main effect of homogeneity of majoritycompetency was not statistically significant (p<.10); however, the trend was in the same direction as that reported by Gerard (1953) in another context. In both cases majorities perceived as having homogeneous-competency levels tended to elicit somewhat more conformity behavior than did heterogeneous majorities. Gerard speculated that when others in the group are perceived as having equal ability, the subjects depend upon all equally and, therefore, upon the group as a whole. In the heterogeneous condition, dependency is not upon the group as a whole but upon the expert within the majority. This may represent a partial explanation of what occurs but implicitly it states that a 2-member heterogeneous majority would,

PAGE 74

66 in effect, be reduced to a majority of one. The results of the present experiment bear this out to a degree. However, the order effects which were associated with heterogeneous majority responses cannot be fully accounted for by suggesting that the dependency is upon the expert within the majority and not upon the majority as a whole. The results of this study indicate that the tendency to perceive a heterogeneous majority as a monolithic whole depends in part upon the naive subject's perceived level of relative competency. It was predicted that when the low-competent member of a heterogeneous majority was perceived to be the first to respond, and the most competent majority member's response reflected agreement, the confirmation would serve to provide more informational influence than when the response order was reversed. This was not the case, particularly when the perceived selfand majority-competencies were both at high levels. Under these conditions, significant results were obtained which were opposite from what was predicted. Similar, but nonsignificant, results occurred when perceived selfand majoritycompetencies were both at low levels. In these conditions the naive subject was, in effect, led to believe that her level of competency was intermediate to the competency levels of the two majority members. When the most competent majority member was seen as responding first and the least competent member followed, the agreement of the majority members was in accord with what would possibly be expected by the naive subject. Gerard and Greenbaum (1962) report that interviewed naive subjects indicated that they tended to perceive Subject "1" as the "leader" and Subject "2" as

PAGE 75

67 a "spineless sheep." Such an observation would be congruent with the reported levels of competency attributed to Subjects "1" and "2." When the least competent member of the majority is perceived to be the first to respond, the situation is considerably different. In this case, the "leader" is perceived as relatively incompetent and the "spineless sheep" is seen as the most competent member of the group. This incongruence apparently reduces the credibility of the majority judgment and encourages the naive subject to demonstrate independent behavior. When the majority members are perceived as possessing levels of competency equal to that of the naive subject, the credibility of the majority judgment does not suffer and the influence of the majority is maintained. The order effects of heterogeneous majority responses did not differentially influence conformity behavior when both members of the majority were perceived as having higher levels of competency than the naive subject. That is, when Subjects "1," "2," and "4" were perceived as having competency levels of 4, 10, and 3, respectively, conformity behavior was not different from when the respective levels of perceived competency were 10, k, and 3. In this case the naive subject possibly sees the majority as a whole and is apparently not inclined to view the differing levels of majority member competencies as incongruent with their agreement when the low-competent member is perceived as the first to respond. The extension of Thibaut and Riecken's (1965) results to third person behavior did not hold up. Apparently when a perceived low-

PAGE 76

68 competent third person is viewed as "influencing" a high-competent individual, the perception with regard to locus of causality is considerably different from that which occurs when the subject perceives herself as the influencing agent. The overriding factor seems to be the perceived credibility of the situation which is influenced by tte subject's perceived relative competency and the response order of the majority members.

PAGE 77

CHAPTER IV SUMMARY The purpose of the research reported here was to investigate the relationships to conformity behavior of majority size, perceived self -competency, perceived majority-competency, and homogeneity of majority -competency. Two experiments were performed to evaluate these situational variables. Experiment 1 was designed to investigate the effects of majority size, perceived self-competency and perceived majoritycompetency. Previous research has indicated that conformity varies directly with perceived majority-competency and majority size (to a limit of three members), and inversely with perceived self -competency. Similar results were predicted in the present research. Additionally, it was suggested that the three variables are additive in their effects on conformity and, as employed, approximately equivalent, y A 2 X 3 X 2 factorial design with a single control group was employed to test the relationships of the variables. Two levels (high vs low) of the competency variables were used and three levels (3-members vs 2-members vs 1-member) of majority size were employed. Levels of selfand majority-competency were conveyed to the subjects by false feedback indicating performance on a judgmental task concerning relative area represented by geometric figures. A Crutchf ield(1955) 69

PAGE 78

70 type apparatus was used and 12 Ss served in each condition. Following the experimental session Ss completed a questionnaire indicating their ratings of task difficulty, levels of confidence in judgments, and the like. It was found that all three variables significantly influenced conformity behavior in the predicted directions. The variables were found to be additive in their effects; however, they could not be considered equivalent due primarily to the disproportionate conformity that occurred in response to 2-member majority judgments and the uniformly low level of conformity elicited by majorities of one. The questionnaire responses indicate the following: 1. Perceived difficulty was inversely related to perceived self -competency. 2. Confidence in own judgments varied directly with perceived self -competency and inversely with perceived majoritycompetency. 3. Confidence in the judgments of others varied directly with perceived majority-competency and inversely with perceived self-corape tency. 4. Confidence in the judgments of others was also found to be a function of an interaction between majority size and perceived self-competency. When perceived se If -competency was high, majority size had no influence on confidence in the judgments of others; however, when perceived selfcompe tency was low, confidence in the judgments of others increased in proportion to the size of the majority. Experiment 2 investigated the interrelated influence of perceived majority-competency (high vs low), homogeneity of majority-competency (homogeneous vs heterogeneous^ vs heterogeneous^) and perceived selfcompetency (high vs low). The two heterogeneous levels refer to the two

PAGE 79

71 possible orders in which 2-member majorities can report their judgments. The data for the two experiments were collected simultaneously and the data for the 3 -member groups having homogeneous majorities were used in both analyses. It was hypothesized that perceived majority-competency and perceived self-competency would have the same relationship to conformity behavior in the context of Experiment 2 as these variables had in the first experiment. Results supporting these hypotheses were obtained. It was also hypothesized that the level of conformity in groups having majorities with homogeneous levels of competency would differ from that elicited by heterogeneous majorities. Results indicated that only when both perceived selfand majority-competency were high was there a difference. In that case, homogeneous majorities elicited more conformity. Based on an extension of Thibaut and Riecken's (1965) results concerning perceived locus of causality, it was predicted that when the least competent member of a heterogeneous majority was perceived as the first to report her judgment, conformity behavior would be greater than when the response order was reversed. Opposite results were obtained when perceived selfand majority-competency were at high levels. Similar findings occurred when both perceived selfand majority-competency were at low levels; however, these results were not statistically significant.

PAGE 80

72 It was concluded that perceived relative competency does indeed influence conformity behavior and serves to modify the influence of majority size. The order effects associated with responses of heterogeneous majorities were interpreted as being a function of the perceived credibility of the situation.

PAGE 81

APPENDICES

PAGE 82

APPENDIX A EXPERIMENTAL MATERIALS

PAGE 83

TABLE 12 DESCRIPTION OF STIMULI 75 Stimulus Number Area Size 1 R 8 T 10 T 12.50 2 T 6 T 6.75 S 9 3 T 9 R 6 T 7 R 12 T 10 T 13.50 5 P 12 R 10 T 15.75 6 T 9 R 10 T 12 7 T 13.75 T 11 R 10.50 8 T 16 T 20 P 18 9 T 6 R 5 T 7.50 10 T 10 R 17.50 S 12.25 11 T 10 R 12.50 T 8 12 S 9 T 6.75 T 6 13* T 7 R 6 T 9 Ik* T 13 .50 R 10 T X 1 7 15* T 15.75 T 12 R 10 16* P 12 T 9 R 10 17* R 10.5 T 13.75 T 11 18 S 16 R 18 R 20 19* T 5 T 7.50 R 6 20* T 10 S 12.25 R 17.50 21 T 12.50 T 8 R 10 22* T 6.75 S 9 T 6 23* T 9 T 7 R 6 2k* P 10 T 12 T 13.50 25* R 10 T 15.75 P 12 26* R 10 P 12 T 9 27* T 11 P 10.50 T 13.75 28 P 20 R 18 P 16 29* T 5 R 6 T 7.50 30* R 17.5 T 10 P 12.25 Critical stimuli. Key : The shape of geometric figures is indicated by P (parallelogram), R (rectangle), S (square), or T (triangle). Underlined values represent the majority response. The values given are area sizes in square units.

PAGE 84

76 Figure 12. --A sample stimulus card.

PAGE 85

77 TABLE 13 THE POSTEXPERIMENTAL QUESTIONNAIRE Please describe in your own words your experience during this experiment. 2. Please circle on the line below, the number which represents how easy or difficult you feel it is to select the correct choice for this perceptual task EASY 1 2 3 4 5 6 7 DIFFICULT 3. How confident were you others? of the correctness of the judgments of the NOT AT ALL CONFIDENT 3 1+ 5 6 7 VERY How confident were you you made? of the correctness of CONFIDEOT the judgments NOT AT ALL CONFIDENT 3 4 5 6 7 CONFIDENT 5. Would you say that you were quite concerned about those times that you disagreed with the others? YES NO 6. Did the others make you doubtful of your accuracy? YES NO 7. Would you say that you were tempted to answer as the others did on some trials? YES NO 8. Did you ever answer as the others did against your own first choice? YES NO 9. The results of this study depend to a large extent upon identifying those individuals who have prior knowledge about this particular experiment or other similar experiments. Please indicate below whether you had prior information of any type. You will, of course, receive experimental credit in either case. YES, I did have prior knowledge of this type of experiment. NO, I did not have prior knowledge of this type of experiment. If you answered yes, please indicate below what you knew or had heard, and where or how you received the information, e.g., from previously serving in a similar experiment, from class discussion, a textbook, another student, etc.

PAGE 86

APPENDIX B RAW DATA AND SUPPLEMENTARY ANALYSIS, EXPERIMENT 1

PAGE 87

TABLE 14 79 INDIVIDUAL CONFORMITY SCORES AND QUESTIONNAIRE RESPONSES BY EXPERIMENTAL CONDITIONS, EXPERIMENT 1 HIGH SELF-COMPETENCY, LOW MAJORITY-COMPETENCY Corrected Postexperimental Questionnaire Items Conformity Conformity 2 3 4 5 6 7 1 -Member Majority U 3 4 4 5 0 0 0 0 0 3 5 4 3 0 0 0 0 0 3 5 1 2 0 0 0 0 0 3 6 2 5 0 1 0 0 0 3 2 2 6 0 0 0 0 0 3 5 2 4 0 0 0 0 0 3 6 2 3 0 0 0 0 1 4 2 2 6 0 1 1 0 1 2.5 7 3 2 0 1 0 0 3 6 5 2 4 0 1 0 0 3 4.5 3 4 5 0 0 0 0 3 4.5 6 2 2 0 1 0 0 2 -Member Majority 0 3 4 2 3 0 1 1 0 0 3 3 3 5 0 0 0 0 0 3 3 2 5 0 0 0 0 1 2.5 5 3 3 0 I 1 1 4 2 5 0 1 1 0 1 fa 1 4 4 2 4 0 0 1 0 1 2.5 5 2 2 1 I 1 0 1 4 3 1 4 0 0 0 0 1 4 6 2 2 0 1 0 0 3 6 3 3 5 1 1 1 0 3 6 4 3 3 0 1 0 0 7 10 4 3 5 0 0 0 0 i -Member Majority 0 3 6 3 4 0 0 1 0 0 3 2 3 5 0 0 0 0 0 3 2 1 7 0 0 0 0 0 3 1 1 7 0 0 0 0 1 4 4 3 3 1 1 1 0 1 1 6 2 6 0 0 1 0 1 4 1 2 6 0 1 0 0 1 4 1 1 7 0 0 0 0 1 4 3 2 5 1 1 1 1 1 4 2 2 6 0 0 0 0 1 4 3 2 5 0 0 0 0 2 5 5 4 6 1 1 1 1

PAGE 88

80 TABLE 14 (Continued) HIGH SELF-COMPETENCY, HIGH MAJOR TT Y-COMPETENCY Raw Corrected Conformity Conformity Postexperimental Questionnaire Items 2 3 4 5 J f. u 7 Q O 4 2 5 0 1 0 0 k 2 5 1 1 X X A w 3 4 6 0 \ 0 0 4 2 5 0 0 n V 0 4 4 4 0 0 n 0 5 3 3 0 1 X 0 0 H 3 3 1 0 1 X X 5 4 6 0 1 X 1 X X 3 1 6 \ 0 0 0 5 4 3 0 1 0 n 5 3 4 0 1 X 0 4 4 4 0 0 1 X n u 5 2 4 I 1 1 0 3 2 5 0 1 X n u 6 2 2 1 1 0 0 5 1 5 1 0 n u A U 1 2 3 1 l L 1 1 1 I 4 4 4 1 X 1 X n u 4 4 4 1 X i 1 4 4 4 1 1 X 1 X U 6 2 2 0 1 X X X 3 4 4 1 n n U 5 3 3 n w X X I 5 5 K 1 u i 0 6 3 4 1 1 X 1 X 1 2 1 6 0 0 0 0 4 1 6 0 1 0 0 6 5 4 0 1 1 1 4 3 4 0 1 1 0 3 3 6 0 1 1 0 3 1 5 0 0 1 0 5 1 5 0 1 1 1 3 2 3 1 1 1 0 6 3 3 0 1 1 0 3 4 4 1 1 1 1 3 4 3 1 1 1 I 1 -Member Majority 0 0 0 0 0 0 1 1 1 I 2 9 3 3 3 3 3 0 1 4 4 4 2 10.5 2-Member Majority 0 3 1 4 1 4 1 4 1 4 3 6 4 5.5 5 6.5 7 10 9 12 11 14 15 18 3-Member Majority 0 0 2 3 3 4 4 6 6 8 11 15 3 0 5 6 6 7 7 9 9 9.5 14 18

PAGE 89

81 TABLE 14 (Continued) LOW SELF-COMPETENCY, LOW MAJORITY-COMPETENCY Raw Corrected Pos texperimental Questionnaire [tems Conformity Conformity 2 u c 0 7 / 8 1-Member Majority 0 3 6 1 5 0 0 0 0 0 3 5 2 5 0 0 0 0 0 3 1 X X A U A u A u 0 0 3 4 1 e n u 1 A u 0 0 1.5 3 X X U 0 0 1.5 h 1 X c J n u A U A U 0 1 2.5 7 1 X X n u A U A U 0 1 1 w •J c J u A U A U 0 1 4 /I •1 X Jl •t A U A 0 0 3 4.5 #1 *T #1 H A A u A u 0 7 o c J A u 1 1 0 5 6.5 /I ll A u 1 0 2-Member Majority 0 3 3 2 4 1 1 1 0 1 4 5 3 4 1 1 1 0 3 6 5 Zi u t A 1/ i 0 3 6 6 4 5 0 1 0 3 6 6 2 c n u 1 0 3 4.5 7 2 n A u 0 7 3 3 •J 1 X X 1 4 7 7 2 2 A U 1 i 0 5 6.5 6 2 3 1 X 1 X 0 7 8.5 U o z 1 X 1 0 8 9.5 c X 1 1 11 14 C 1 X t 1 0 2-Member Majority 0 3 i. A \J i 0 0 3 6 4 3 1 0 0 0 3 6 1 5 0 0 0 0 3 1 4 7 0 0 0 1 4 6 4 3 1 4 7 6 5 3 1 4 7 5 4 4 0 4 5.5 6 3 3 0 6 9 6 3 3 1 9 12 4 3 3 0 10 11.5 6 3 3 0 14 17 6 4 4 0

PAGE 90

82 TABLE m (Continued) LOW SELF-COMPETENCY, HIGH MAJORITY-COMPETENCY Raw Corrected Postexper imental Questionnaire Items onf ormity Conformity 2 3 4 5 6 7 8 1 -Member Majority 0 3 6 2 2 0 0 1 0 0 3 6 5 3 0 1 0 0 0 1.5 5 5 2 0 1 0 0 1 4 6 5 2 1 1 0 0 2 5 5 4 2 0 0 0 0 2 5 7 1 1 0 1 1 0 2 5 3 4 4 1 0 1 0 3 4.5 5 2 2 0 1 0 0 4 7 4 4 4 0 1 1 0 H 7 5 3 5 0 1 0 0 5 6.5 3 4 3 0 0 1 0 6 6 7 2 2 1 1 1 1 2 -Member Majority 0 3 3 6 1 1 1 1 0 0 3 5 2 5 0 0 0 0 1 4 3 2 2 0 0 0 0 3 6 5 3 4 0 0 1 0 3 6 5 5 4 0 0 1 1 4 7 6 4 2 1 1 0 6 7.5 3 4 4 1 J 1 1 9 12 4 4 3 0 I 1 1 9 12 7 2 1 1 1 0 10 13 5 5 4 1 0 0 10 11.5 5 4 2 1 1 0 11 14 6 5 4 0 0 1 3 -Member Majority 0 3 5 4 2 0 1 0 2 5 6 5 3 0 1 0 3 4.5 7 4 1 1 1 0 3 6 5 3 5 0 0 0 4 7 5 7 1 1 0 4 5.5 5 4 2 1 1 6 9 5 4 5 1 1 10 13 6 3 4 1 1 12 15 5 3 3 1 1 12 15 6 5 4 0 1 0 15 18 4 4 4 1 1 1 15 18 6 5 2 0 0 1 1

PAGE 91

83 TABLE 14 (Continued) PRESSURE CONTROL Raw Corrected Postexperimental Questionnaire Items Conformity Conformity 2 3 4 5 6 7 8 1 Member Majority 0 3 5 0 3 5 0 3 4 0 3 4 1 4 3 1 4 6 1 2.5 6 1 4 5 1 4 6 1 2.5 4 2 5 4 3 6 6 2 Member Majority 0 3 6 0 3 4 0 3 5 1 4 3 1 4 6 2 5 6 2 3.5 4 4 7 5 5 8 2 5 6.5 5 8 115 10 13 6 3Member Majority 0 3 2 0 3 7 0-3 6 1 4 5 2 3.5 5 4 5.5 2 6 9 5 6 9 5 6 7.5 6 6 9 5 6 9 5 8 9.5 5 5 3 1 1 0 0 5 3 0 1 0 0 4 4 1 0 0 0 3 4 0 0 0 0 2 6 0 1 1 0 2 5 1 0 1 0 1 1 1 1 1 1 4 6 0 0 0 0 1 1 0 0 1 0 1 5 0 0 0 0 4 4 0 0 0 0 2 4 1 1 I 1 3 5 1 1 1 0 2 4 1 1 1 0 1 4 1 1 0 0 I 6 0 1 0 2 2 0 1 1 0 4 2 0 1 1 0 3 5 0 1 0 0 4 4 0 1 1 0 3 4 1 I 0 4 4 0 1 I 0 6 3 1 1 1 1 3 5 1 1 1 1 6 0 0 1 0 2 6 0 0 0 0 4 4 1 1 0 2 4 1 0 0 3 5 0 1 0 4 5 1 0 0 4 4 0 1 3 5 0 1 2 4 1 1 1 2 0 1 3 5 0 1 3 2 1 0 0

PAGE 92

84 TABLE 14 (Continued) NO PRESSURE CONTROL Raw Corrected Postexperimental Questionnaire Items Conformity Conformity 2 4 u 9 3 c c D U 1 c 0 o U 1 c 2 c n U c i ii t It U <* J e D 2 3 5 5 0 3 6 1 1 4 2 5 0 3 2 5 0 3 2 6 0 3 1 5 0 3 3 5 0 3 4 5 0 3 2 6 0 3 3 4 2 3.5 4 3 0 3 2 5 0 1.5 6 2 0 1.5 5 2

PAGE 93

85 TABLE 15 MEANS AND STANDARD DEVIATIONS OF CORRECTED CONFORMITY SCORES BY EXPERIMENTAL CONDITIONS, EXPERIMENT 1 (N = 12 IN EACH CONDITION EXCEPT IN THE NO PRESSURE CONTROL GROUP) Majority Size (1-Member) Self -Competency HIGH LOW Majority-Competency High Low High Low X 3.33 3.54 4.79 3.33 S.D. 2.50 .97 1.65 1.87 Majority Size (2-Members) Self -Competency HIGH LOW Majority-Competency High Low High Low X 7.58 4.33 8.25 6.83 S.D. 4.59 2.05 3.87 2.77 Majority Size (3 -Members) Self -Competency HIGH LOW Majority-Competency High Low High Low X 7.79 3.50 9.91 7.08 S.D. 4.55 .96 5.31 4.33 Control Groups No Pressure Pressure (n = 20) 2 -Member 3-tfember 4-Member X 2.92 3.67 5.92 6.25 S.D. .04 .99 3.17 2.70

PAGE 94

TABLE 16 FREQUENCY ANALYSIS OF CONFIDENCE D VALUES BY LEVELS OF CONFORMITY Confidence d Value 0 Conformity Level Total 0 3 31 10 7 48 4-7 31 22 14 67 8 + 5 _13 _U_ _29 Total 67 45 32 144 X'^ = 16.68, p< .005 Rating of confidence in own judgments minus rating of confidence in the judgments of others.

PAGE 95

APPENDIX C RAW DATA AND SUPPLEMENTARY ANALYSIS, EXPERIMENT 2

PAGE 96

88 TABLE 17 INDIVIDUAL CONFORMITY SCORES AND QUESTIONNAIRE RESPONSES BY EXPERIMENTAL CONDITIONS, EXPERIMENT 2 HIGH SELF-COMPETENCY, LOW MAJORITY-COMPETENCY Raw Corrected Postexperiraenta 1 Questionnaire Items Conformity Conformity 2 3 4 5 6 7 8 Heterogeneous ^ 1 0 3 3 1 2 0 0 0 0 0 3 k 1 5 1 0 1 0 0 3 3 1 5 0 0 0 0 0 3 1 0 0 0 0 0 3 6 2 5 1 0 0 0 0 3 2 4 4 0 0 0 0 0 1,5 3 3 4 0 0 0 0 0 3 6 3 5 0 1 0 0 1 k 3 1 4 0 1 0 0 1 4 3 5 0 0 0 0 1* 5.5 3 3 4 0 1 1 1 5 8 3 3 4 1 1 1 1 He terogeneous ^ 0 3 3 3 6 0 0 0 0 0 3 6 2 6 0 0 0 0 0 3 5 1 3 0 0 0 0 3 3 3 6 1 0 0 0 3 5 1 6 0 0 0 0 1.5 5 3 5 0 1 0 0 1.5 3 2 6 1 1 0 1 k 2 2 6 0 1 0 1 k 4 5 5 0 1 0 1 2.5 2 5 5 0 0 0 4 7 3 3 4 0 0 0 5 8 5 3 4 1 1 0

PAGE 97

TABLE 17 (Continued) HIGH SELF-COMPETENCY, HIGH MAJORITY-COMPETENCY Raw Corrected Postexperimental Questionnaire Items Conformity Conformity 2 3 4 5 6 7 8 Heterogeneous 0 3 2 1 6 1 0 0 0 0 3 4 2 6 1 0 0 0 0 3 4 3 6 0 0 1 0 1 3 5 4 0 \ I 0 1 5 3 4 1 I I 0 1 2.5 5 2 5 1 1 1 1 3 6 5 2 4 0 1 1 0 3 6 6 3 4 1 0 4 7 4 3 4 1 0 5 6.5 4 2 2 1 0 6 9 6 4 6 1 0 10 13 4 3 4 1 1 Heterogeneous^ 0 3 2 3 5 1 1 1 0 0 3 5 3 5 0 1 0 0 0 1.5 3 2 5 0 0 0 0 0 3 5 4 4 0 0 0 0 1 4 5 2 4 0 1 1 0 1 7 1 3 0 0 0 0 1 k 5 5 4 0 1 0 0 1 2.5 2 1 5 0 0 0 0 2 5 5 5 6 0 1 1 1 2 5 4 3 5 0 0 0 0 3 4.5 6 6 2 1 1 1 1 5 8 5 3 5 0 1 1 1

PAGE 98

90 TABLE 17 (Continued) LOW SELF -COMPETENCY, LOW MAJOR IT Y-CCMPETENCY Raw Corrected Postexperimental Questionnaire Items Conformity Conformity 2 3 4 5 6 7 8 Heterogeneous U 3 7 1 4 0 0 0 0 2 5 6 3 3 0 1 1 1 2 3.5 6 4 5 1 1 1 1 3 6 6 5 4 1 1 1 1 3 6 3 3 5 0 1 0 5 8 4 5 4 1 1 1 1 5 8 5 4 4 0 1 1 0 ^ Q J O 6 4 5 0 0 6 9 6 6 2 1 1 6 9 5 4 4 0 1 7 10 5 3 2 1 1 9 12 4 4 4 0 1 Heterogeneous^ 0 3 5 3 5 0 0 1 0 0 3 2 1 5 0 0 0 1 1 4 6 2 3 0 0 0 0 1 4 4 1 3 1 0 0 0 2 5 5 4 2 0 1 1 2 5 6 2 2 1 1 0 2 3.5 7 2 4 1 1 1 2 3.5 4 4 3 1 0 0 3 6 6 2 3 0 1 0 3 6 4 2 6 0 1 1 3 6 6 4 3 1 1 0 5 6.5 6 2 5 0 1 0 0

PAGE 99

91 TABLE 17 (Continued) LOW SELF-COMPETENCY, HIGH MAJORITY-COMPETENCY Raw Corrected Postexperimental Questionnaire Items Confomity Conformity 2 3 4 5 6 7 8 Heterogeneous 1 k 4 2 4 0 0 0 0 2 5 1 2 7 0 0 1 0 2 5 5 2 2 0 1 1 1 H 7 6 4 2 0 0 1 0 7 6 5 2 1 1 1 0 7 6 4 3 0 1 1 0 5 8 5 5 2 1 1 1 1 6 9 4 5 5 1 1 8 11 5 4 3 1 0 11 m 6 4 3 1 1 11 14 5 4 4 0 0 0 15 18 3 5 5 0 1 1 Heterogeneous 2 2 5 5 3 5 0 0 0 2 3.5 6 4 3 0 1 4 7 6 4 2 1 0 5 6.5 4 6 2 1 6 9 5 2 2 1 7 10 7 4 2 1 8 11 7 4 1 0 8 11 4 4 3 0 9 12 4 3 5 0 9 12 6 5 3 1 10 13 5 4 3 0 10 11.5 5 5 4 0

PAGE 100

92 TABLE 18 MEANS AND S.TANDARD DEVIATIONS OF CORRECTED CONFORMITY SCORES BY EXPERIMENTAL CONDITIONS, EXPERIMENT 2 (N = 12 IN EACH CONDITION EXCEPT IN THE NO PRESSURE CONTROL GROUP) Majority -Homogeneity (Homogeneous ) Self-Competency HIGH LOW Majority-Competency High Low High Low X 7.58 4.33 8.25 6.83 S.D. 4.59 2.05 3.87 2.77 Majority -Homogeneity (Heterogeneous Self -Competency HIGH LOW Majority-Competency High Low High Low X. 5.58 3.67 9.08 7.29 S.D. 2.96 1.58 4.14 2.56 Ma j or ity -Homogeneity (Heterogeneous ^) Self -Competency HIGH LOW Majority-Competency High Low High Low X 3.96 3.62 9.29 4.62 S.D. 1.57 1.91 3.05 1.25 Control Groups No Pressure Pressure (n = 20) X 2.92 5.92 S.D. .04 3.17 The homogeneous majority data and control group data are common to Experiment 1 and Experiment 2.

PAGE 101

BIBLIOGRAPHY Allen, V. L. Situational factors in conformity. In L. Berkowitz (Ed.), Advances in experimental social psychology, Volume 2. New York: Academic Press, 1965, pp. 133-175. Asch, S. E. Effects of group pressure upon the modification and distortion of judgments. In H. Guetzkow (Ed.), Groups, leader ship and men Pittsburgh, Penn.: Carnegie Press, 1951, pp. 177190. Asch, S. E. Studies of independence and conformity: I. A minority of one against a unanimous majority. Psychol.. Monogr. 1956, 70, No. 9 (Whole No. ^16} Bass, B. M. Conformity, deviation and a general theory of interpersonal behavior. In I. A. Berg and B. M. Bass (Eds.), Conformity and Deviation New York: Harper and Brothers, 1961, pp. 38-100. Beloff, H. Two forms of social conformity; acquiescence and conventionality. J. abnorm. soc. Psychol ., 1958, 56, 99-104. Blake, R. R. and Mouton, Jane S. Conformity, resistance, and conver-/ sion. In I. A. Berg and B. M. Gass (Eds.), Conformity and Deviation New York: Harper and Brothers, 1961, pp. 1-37. Crutchfield, R. S.' Conformity and character. Amer. Psychologist, 1955, ]J0, 191-198. Deutsch, M. and Gerard, H. B. A study of normative and informational social influences upon individual judgment. J. abnorm. soc Psychol 1955, 5_1, 629-636. Di Vesta, F. J. Effects of confidence and motivation on susceptibility to informational social influence. J. abnorm. soc Psychol., 1959, 59.> 204-209. Fagen, S. A. Conformity and the relations between others* competence and own competence. (Doctoral dissertation, University of Pennsylvania), Ann Arbor, Mich.: University Microfilms, 1963, No. 63-7037. 93

PAGE 102

94 Festinger, L. Schachter, S., and Back, K. Social pressure in informal groups New York: Harper and Brothers, 1950. Festinger, L., and Thibaut, J. Interpersonal conmunications in small groups. J. abnorm. soc. Psychol ., 1951, 46, 92-100. Fisher, S., Williams, H. L., and Lubin, A. Personal predictors of susceptibility to social influence. Amer. Psychologist 1957, U, 360 (Abst.). Gerard, H. B. The effects of different dimensions of disagreement on the communication process in small groups. Hum. Rel at., 1953, 249-272. Gerard, H. B. The anchorage of opinions in face-to-face groups. Hum. Relat ., 1954, 7_, 313-326. Gerard, H. B. Disagreement with others, their credibility and experienced stress. J. abnorm. soc. Psychol ., 1961, 62, 559-564. Gerard, H. B., and Greenbaum, C. W. Attitudes toward an agent of uncertainty reduction. J. Pers ., 1962, 30, 485-495. Goldberg, S. C. Three situational determinants of conformity to social norms. J. abnorm. soc. Psychol ., 1954, 49, 325-329. Harvey, 0. J., and Consalvi, C. Status and conformity to pressures in informal groups. J. abnorm. soc. Psychol ., 1960, 60, 182-187. Hays, W. L. Statistics for psychologists New York: Holt, Rinehart and Winston, 1963. Hochbaum, G. M. The relation between group members' self-confidence and their reactions to group pressure to conformity. Amer. sociol. Rev ., 1954, _19^, 678-687. Hollander, E. P. Conformity, status, and idiosyncrasy credit. Psychol. Rev ., 1958, 65^, 117-127. Homans, G. C. The human group New York: Harcourt, Brace, 1950. Kelman, H. C. Effects of success and failure on "suggestibility" in the autokinetic situation. J. abnorm. soc. Psychol., 1950, 45, 267-285. Kidd, J. S. Social influence phenomena in a task-oriented group situation. J. abnorm. soc. Psychol ., 1958, 56, 13-17.

PAGE 103

95 Linde, T. F., and Patterson, CH. Influence of orthopedic disability on conforming behavior. J. abnorm. soc. Psychol ., 1964, 68, 115-118. Lett, A. J., and Lott, B. E. Group cohesiveness communication level, and conformity. J. abnorm. soc. Psychol ., 1961, 62_, 408-^4-12. Luchins A. S., and Luchins, E. H. Previous experience with ambiguous and non-ambiguous perceptual stimuli under various social influences. J. soc. Psychol ., 1955a, k2_, 249-270. / Luchins, A. S., and Luchins, E. H. On conformity with true and false / communications. J. soc. Psychol ., 1955b, 42_, 283-303. Mausner, B. The effect of prior reinforcement on the interaction of observer pairs. J. abnorm. soc. Psychol ., 1954a, 49, 65-68. Mausner, B. The effect of one partner's success in a relevant task on the interaction of observer pairs. J. abnorm. soc. Psychol., 1954b, 49, 557-560. Mausner, B., and Bloch, Barbara L. A study of the additivity of variables affecting social Psychol., 1957, 54, 250-256 variables affecting social interaction. J. abnorm. soc. MacBride, P. D., and Tuddenham, R. D. The influence of self-confidence upon resistance of perceptual judgments to group pressure. J. of Psychol ., 1965, 60, 9-23. Newcomb, T. M. The acquaintance process New York: Holt, 1961. Nichols, Shirley A. A study of the additivity of variables influencing conformity. Doctoral dissertation. University of Florida, 1964. ^Reitan, H. T. and Shaw, M. E. Group membership, sex-composition of the group, and conformity behavior. J. soc. Psychol ., 1964, 64, 45-51. Rosenberg, L. A. Conformity as a function of confidence in self and ^ confidence in partner. Hum. Relat ., 1963, 16, 131-139. Samelson, F. Conforming behavior under two conditions of conflict in the cognitive field. J. abnorm. soc. Psychol ., 1957, 55, 181-187. Schachter, S. Deviation, rejection, and communication. J. abnorm soc. Psychol., 1951, 46, 190-207.

PAGE 104

96 Shaw, M. E., Axelberg, F., Costanzo, P., and Reitan, H. T. Conformity and perceived relative competency. Unpublished study, University of Florida, 1966. Shaw, M. E., Rothschild, G. H., and Strickland, J. C. Decision processes in communication nets. J. abnorm. soc. Psychol ., 1957, 5ij-, 323-330. Sherif, M. A study of some social factors in perception. Arch. Psychol ., 1935, 2_7, No. 187. Sherif, M. and Sherif, Carolyn W. An outline of social psychology New York: Harper and Brothers, 1956. Thibaut, J. W., and Riecken, H. W. Some determinants and consequences of the perception of social causality. In H. Proshansky and B. Seidenberg (Eds.), Basic studies in social psychology. New York: Holt, Rinehart, and Winston, 1965, pp. 81-94. Thibaut, J. W. and Strickland, L. H. Psychological set and social ^ conformity. J. Pers ., 1956, £5, 115-129. Tuddenham, R. D., MacBride, P., and Zahn, V. The influence of sex composition of the group upon yielding to a distorted norm. J. Psychol ., 1958, 46_, 243-251. Walker, E. L. and Heyns, R. W. An anatomy for conformity Englewood Cliffs, N. J.: Prentice -Ha 11, 1962. Winer, B. J. Statistical principles in experimental design New York: McGraw-Hill, 1962.

PAGE 105

BIOGRAPHICAL SKETCH Harold T. Reitan was born November 3, 1928, at Max, North Dakota. In June, 1950, he received the Bachelor of Arts degree from St. Olaf College, Northfield, Minnesota. He was commissioned in the United States Air Force in September, 1951, and was rated as a navigator-bombadier the following year. Prior to enrolling in the Graduate School of the University of Florida in June, 1961, under sponsorship of the Air Force Institute of Technology, he served in Korea, in the Strategic Air Command, and at the United States Air Force Academy. He received the degree of Master of Arts in June, 1962 Following a three-year assignment in Germany, he returned to the University of Florida in June, 1965, under Air Force sponsorship to work toward the degree of Doctor of Philosophy. Major Reitan is married to the former Margaret Lucille Bonsac and is the father of four children, Eric, Karen, Chris, and Jon. He is a member of Phi Kappa Phi. 97

PAGE 106

This dissertation was prepared under the direction of the chairman of the candidate's supervisory committee and has been approved by all members of that committee. It was submitted to the Dean of the College of Arts and Sciences and to the Graduate Council, and was approved as partial fulfillment of the requirements for the degree of Doctor of Philosophy. December, 1967 Dean, Graduate School Supervisory Committee:


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID E4KUCX87B_WVU4FQ INGEST_TIME 2017-07-11T21:13:31Z PACKAGE AA00031419_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES