. V . v . . 4. . .. o. .. . . ... ,..... o _ . . . . . . .. . . .. o. . . ....T . ... ‘....o. . ..¢... 0 .. ¢ undo ....Jo Q...
. . . . ., ... . . . . .. . .....I . ._ . .. ...... . ,. .... ... .. ... . .. _ . ...... ...Jo. TM.”
_ A ... ., . ..H . .. . . .. ... . .... . . .‘ . . . . . ._ . . . ... .. ._ ... . ..«4‘ . . . \ o_._._¢......... ‘..........A;..»
. . V . . JR. ..,..... o . . ... _ . .. ..f u. . _. ,_ ._ . _ _ .‘ . . , ...... . ... ..... ..'....c..1.7..oobp.:.__...“..‘”
...... . .. _. . . .. . . . .. \ . ‘ ... . . . . . . . ._ . .. .. _. .4 ... r...«.‘. ..p.....4...o.t.. I
. .... . . ... o . . . .. . . . ‘ . ... . .. _ . . . . .. ,.. ¢ . . .... ..v..o..oa.o../J. ...tr.’
. . . . .. . V .. . . . . .. ... ..... .... ...
w y. . .... . .. .. .. .. . _. .. .v . . . u. ... . o. .. .....v.,......o.f...a
“ ... . . . . . . .., _ . _ _ . . . . V. . _ . . . _ . c ..u... 331.33.)...“
“ .... . .. ‘ . .. .. . . . . .. .. . . , . . .. .. _ _ . . u .. no. ;. , .JH. .. . .... . whiff...
« . . . ......o . .,. . ~ .‘o ... . . . . . . .. . H. ‘.. . 0.. I .... ”.../OJ “.... v.04; .00
_< . _. . . n. . _ .. .. . . . . . . .. n _ 0..WJ4.. 0...: ......V. .J
m . . ... \ . ... . .. . ... __ . . . _., .. . . . a... . 35...... 31"..Wu3u
. . . . . . . . . . . v . . . . ... .t . ool.’...)..c.. ., .1...—...
a . . .. .. . . . . _. ... .. _ ... ye. ........r ...; . a‘ fro“.-.
w . . . .. . . _ . V.” . .. ,_ .. . .....p......”..u_u..,....fl,.fi._.
._ . . . ....... v- ‘ .4.,
. _ . : . ...,7 _HU./..,H.~...u.
W . . . . . . . _ . ..hsz. ......m,». f,
. . . . . .0 ... ... .
w . . . m ...o .
C , . . . . . . . .
‘ . . . . .
~ . . . _
m On 3 . . .
u . U o . .
m m A. . A ‘
. ,
u ..
w M m m H. W ,
. II. M I...
u . A, . R U .
m D .Mw W ..L
H _., . N
m. E N 0 .e l U .... . _
a D A I: e. N H _ M
w ......u .. mlu my ...U C ,. .. a" L
A . r.. _ . . ...; .97... .d
, . . i .
. m S In .s 0 . . ...-rs ...... w... . ...H
o .I i t ‘ J _ ,.. ....I . .3... ._
. 0 B _ r. N H. . . ...“
PIII. F . E ._ u ‘ ( .n.»u
. . N N W a C, .J. _ .. h.
_ mm m R m M _ .
.u T c A . . . .. . .. a... ..
D N A. . . ..v .
. M E i ..
. E 0.. . .
4 W F . 1
... E3 . .. .
o . . u
H_ . ..
m
. H
u
7”
—.
_
_,
..
v
F... . _ _ . ... . . . . . 1 . ... . . o. ... f o J . . K ’..‘.’:m50., v 91"...31“..- o
.' M ..o.. v _. . ... . .... .. ..iJ: on... .... 2. .. . ... ...... .... l . _ o. .. a. . .. . ... .. . .. . . ... ,n ..,-u ._ .4. ..- a I .. ).1..4F!..‘ A::olb.~."
T r 11': f-’ I III! -Illl. I,

 

        

 

  

 

 

 

   
 
 

  
    
   

3g LIBRARY
“Michigan Scare
4-“ University

 

!

ABSTRACT

THE RELATIONSHIP OF ATTITUDE, AS MEASURED BY SELF-PREDICTION
OF BEHAVIOR, AND ACTUAL BEHAVIOR IN A
REWARD DISTRIBUTION SITUATION

by Joseph John Schuller

This study examines the relationship between attitudes, as
measured by self-prediction of behavior, and actual behavior in a
reward distribution situation. Previous research in the area of
attitude-behavior relationships indicates that there is little
relationship between the two.‘ A number of authors feel that the
situation determines what behavior will be emitted. Attitude may be
related to behavior in some instances, but the situation is often a
more important determinant. At any rate, it is felt that the rela-
tionship between attitude and behavior must be examined in each
particular situation. One cannot generalize from one situation to
another unless the situations are virtually identical.

The present research evaluates this theoretical position with
respect to reward distribution behavior. In this study, subjects
were divided into three groups: 36 choosers, 38 receivers, and 129
hypotheticals. Choosers and receivers worked for one hour on a
questionnaire about attitudes toward industry, while hypotheticals
were told to imagine that they had. Choosers made choices on a

reward distribution booklet which determined the amount of money they

Joseph John Schuller

and one of the receivers would be paid. Hypotheticals were told to
imagine that their choices would actually determine the wage for
themselves and another person. Receivers also made choices, but they
were told they would have no effect.

Three types of data were collected and analyzed. (a) A measure
of self-interest (SI) which indicated a subject was maximizing his
gain. (b) The amount of money, of a total $4.00, the subject would
give to himself. (c) The number of errors on infbrmation questions
which preceded the forced-alternative questions measuring amount of
SI. The scores of hypotheticals for (a) and (b) were taken as a
measure of attitude, whereas the scores of choosers for (a) and (b)
were taken as a measure of behavior. Like those for hypotheticals,
scores of receivers were also taken to reflect attitudes, but in this
condition these attitudes were measured in a setting that should have
maximized their salience.

It was predicted that choosers would be higher on SI and the
proportion of the $4.00 kept than receivers, and likewise receivers
would be higher than hypotheticals. However, the results indicated
that there were no differences among the three groups for these two
measures, but there was a significant difference for number of
errors, indicating that the receivers were less careful than the other
two groups in filling out the reward distribution booklets.

These results were taken as evidence that in this particular
reward distribution situation, attitudes are related to behavior, due
to the fact that the concern for equity is within the realm of normal,

everyday consideration for people.

THE RELATIONSHIP OF ATTITUDE, AS MEASURED BY SELF-PREDICTION
OF BEHAVIOR, AND ACTUAL BEHAVIOR IN A
REWARD DISTRIBUTION SITUATION

By

Joseph John Schuller

A THESIS

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

MASTER OF ARTS

Department of Psychology

1971

TABLE OF CONTENTS

CHAPTER

I. INTRODUCTION .
Empirical studies .
Theoretical articles . .
Purpose and hypotheses .

I I O METHO D O O 0 0 O O O O O O O O O 0
Subjects . . .
Instruments
Procedure

III. RESULTS

IV. DISCUSSION .

BIBLIOGRAPHY.

APPENDICES

PAGE

16
16
16
17
22

26

35

38

CHAPTER I

INTRODUCTION

The purpose of this study is to examine the relationship between
attitude, as measured by self-prediction of behavior, and the actual
behavior that is displayed. In this study, the situation involved a
distribution of reward as chosen by one member of a pair of subjects

who had been involved in a real or hypothetical work situation.

Empirical studies

 

Previous empirical studies concerning the relationship between
attitude and behavior have mainly indicated little or no relationship,
and various theoretical papers have attempted to account for this
seeming‘discrepancy.

The first study done in this area was by LaPiere (1934). In a
two year time span, LaPiere and a young Chinese couple traveled
across the United States and back, and up and down the Pacific coast.
During these travels they stayed at 66 motels or hotels and ate at
184 restaurants and cafes. Only once were they refused service, this
being at a motel. However, a questionnaire sent out six months later
to the same establishments obtained vastly different results. Of 47
hotels and 81 restaurants replying to the questionnaire, over 90%

said they would refuse service to a Chinese couple, with the rest

being undecided or indicating it would depend upon the circumstances.
Only one motel said they would definitely accept a Chinese couple,
this acceptance being due to the favorable impression LaPiere's
Chinese couple had made six months earlier. As a validity check on
the attitude questionnaire, LaPiere sent out questionnaires to 32
hotels and 96 restaurants in the same general regions as the ones
they had visited, and the results were virtually identical, indicating
that this was a widespread attitude, and not due to having been
exposed to LaPiere's Chinese couple. One problem with LaPiere's
methodology was that he, a Caucasian, was physically present during
nearly all the confrontations with managers, clerks, waitresses, and
hostesses, but in the questionnaire that was sent out, the question
asked if the establishment would serve a Chinese couple, and said
nothing about Caucasians being in the group. LaPiere attempted to
control for this somewhat in about 20 percent of the confrontations
by letting the Chinese couple deal with the establishments themselves,
and not appearing himself for several minutes. Results of these
situations were essentially the same as for those when he was present.
(The Chinese couple was not aware that LaPiere was doing a study.
This was in order not to influence the behavior of the Chinese couple
and also because the primary purpose of the trips was for pleasure).

LaPiere writes:

Only a verbal reaction to an entirely symbolic
situation can be secured by the questionnaire. It

may indicate what the reSponder would actually do
when confronted with the situation symbolized in

the question, but there is no assurance that it

will. . . If social attitudes are to be concep-

tualized as partially integrated habit sets which

will become Operative under specific circumstances

and lead to a particular pattern of adjustment they

must, in the main, be derived from a study of humans

behaving in actual social situations. They must

not be imputed on the basis of questionnaire data.

(pp. 236-237).

A study by Kutner, Wilkens, and Yarrow (1952) showed essentially
the same results with Negroes in a restaurant setting. A group of
two white and one Negro woman was served normally in 11 restaurants
in a Northeastern suburb. However, in response to a letter sent two
weeks later asking for reservations fer a small social affair at
which several of the participants would be Negro, no replies were
received in 17 days. Phone calls were then placed to the 11
restaurants asking for reservations, with the result that 8 of the 11
restaurants denied receiving the letters, and only 5 gave some type of
agreement, three of these requesting someone to come in person to
make the reservations. Calls were made one day later requesting
reservations, with no mention made of race, and all were accepted.

The problem with this study, however, was that in the behavioral
measure, the black woman did not enter the restaurants with the white
woman, but came in a few minutes later, so that the behavioral measure
and attitudinal measure may not have been comparable.

A study by Linn (1964) showed the same lack of relationship.

Subjects in this experiment were 34 college females who were admin-

istered attitude questionnaires, among which were items asking in

what situations they would consent to be photographed with a Negro
male, the situations being couched in terms of what kind and how
extensive an audience would see the photographs. Four weeks after
the attitude questionnaire administration, the subjects were asked
to help the Psychological Testing Company to devise a TAT-like test.
For those who volunteered, the procedure involved an interview, at
which they were confronted by a white and a black experimenter, and
told that they were being asked to volunteer to pose with a Negro
male and then given release forms to sign indicating under what
circumstances, in terms of the type and number of the audience to see
the pictures, they would be willing to be photographed. The seven
release forms were identical to the seven statements on the attitude
questionnaire. The volunteers were given an appointment time, in
order to see how many would actually keep the appointment, and at
this time they were told it was an experiment. Results indicated
there was no statistical relationship between attitude and behavior,
18 of the 34 subjects showed a discrepancy, 16 of these being less
willing behaviorally than attitudinally. The explanation given is
that the verbal attitude is a cultural phenomenon of the university
setting, but that few of the subjects have actually had face-to-face
confrontation with a Negro, and when doing so they evaluate their
behavior in terms of the general population norms rather than the

university norms.

Bray's study (1950) concerned the attitude toward Negroes and
Jews and a behavioral measure of the amount of agreement with a
stooge, designated as Gentile, Jewish, or Negro, on an autokinetic
apparatus. There was no significant correlation between attitude and
behavior. However, Bray notes that by first dividing subjects on the
basis of attitude and then by looking at personality factors, signif-
icant relationships were found between scores on a personality test
and behavior. Bray believes that situations are complex and that
attitudes alone do not determine behavior.

Berg (1966) conducted an experiment similar to Bray's in
which 60 subjects were given an attitude scale to determine prejudice
toward Negroes, and then were asked to make judgments of movement on
the autokinetic effect in the company of a white and a Negro
confederate. There was feund to be no relationship between the
attitude scale and the amount of agreement with the Negro confederate,
though Jewish subjects scored significantly lower on the attitude
scale, but agreed less with the Negro confederate on the autokinetic
judgments. Berg also believes that attitudes don't completely
determine action, but that action is also a function of the particular
situation or social context.

McGrew (1967) had Negro and white couples visit 10 apartments
with vacancies and attempt to rent an apartment, and a few weeks

later called the 10 landlords and asked them if they would rent to a

Negro couple. The results were that four landlords refused to show
the vacant apartments to the Negro couple, and six landlords refused
to rent the apartments to Negroes. McGrew's report is a bit unclear,
but he does note that verbal attitudes were not consistent with
behavior. Fishman (1961) feund that whites moving away from a newly
integrating neighborhood did not necessarily depend on attitude,

but rather on the occupational status of the people involved, those in
white collar occupations moved away even when they possessed positive
racial attitudes, while blue collar workers stayed, even with negative
attitudes. Fishman's explanation is that white collar workers

could financially afford to move, while blue collar workers could
not; he suggests that status needs, fears, and pressures determined
the actions rather than the racial prejudices. Brookover and Holland
(1952), in a field study done in a rural, midwestern county, found
that there were differences between attitude and behavior in regards
to minority group members (Negroes). They explain these differences
in terms of the specific situation in which the behavior occurs and
also in terms of the roles and expectancies connected with roles.

The example is given of a Negro family moving into a white neigh-
borhood, where attitudes were unfavorable to Negroes, but nothing

was done behaviorally, because this family had all the proper rural,
middle-class behaviors and attitudes, and so they fit in except fer
skin color. Saenger and Gilbert (1950) interviewed customers of a
department store employing Negro clerks and found that the amount of

prejudice was the same fer customers of white clerks and Negro

clerks, and the same as a random sample of people drawn from the street.
They also noted that several customers said that they would never
buy from a Negro clerk, when they had been observed doing so several
minutes previously.

Freeman and Ata6v (1960) conducted a study on cheating, where
they fbund there was no relationship between an attitude measure of
cheating plus a question asking if they ever had cheated and actual
observed cheating on a test. Zunich (1962) found that there was
little relationship between attitudes of mothers towards their
children and actual behavior toward the children.

While the previously cited studies indicate little or no
congruence between attitude and behavior, several studies do indicate
at least a low level relationship. Fendrich (1967a, 1967b) did two
similar studies in which he measured attitudes toward Negroes and
then obtained a behavioral measure of willingness to participate in
interracial group discussions. In the first study (1967a), he
measured reference group support by asking the degree to which five
reference groups: 1) same sex friends, 2) opposite sex friends,

(3) parents, 4) roommates or spouses, 5) older persons, would partic-
ipate in interracial activities. He found that over two—thirds of
the subjects with favorable racial attitudes and reference group
support were unwilling to commit themselves to interracial discussion
groups. He found that perceived reference group support predicted

behavior better than attitudes. In the second study (1967b),

Fendrich led half of subjects to believe they would be expected to
participate in interracial group discussions, while the other half
were not told this. He fbund that when subjects were led to believe
they would participate in interracial discussions, attitudes were
predictive of behavior, while this was not true for the other half

of the subjects. Fendrich concludes that since behavior is a function
of personality and environment, the situation plays a large part in
mediating the relation between attitudes and behavior. In particular,
if the attitude measurement process is seen as real and having possible
behavior consequences, attitude can be predictive. Also in this
study, the attitude measure involved a scale of commitment as to what
circumstances they would interact with Negroes; it was feund that

this scale of commitment was a better predictor of behavior than the
normal attitude measure.

A study by DeFleur and Westie (1958) was similar to Linn's
(1965) study. However, they took their subjects from the upper and
lower quartiles on an attitude scale toward Negroes and using this
procedure, they found a significant relationship between racial
attitude and willingness to be photographed with a Negro of the
opposite sex. Also, this study included males and females, whereas
Linn's only used females. Also it was noted that one-third of the
subjects' behavior were in opposition to attitudes. Subjects gave
as a reason for their behavior the consideration of peer-group

pressure and beliefs, and the authors conclude as did Fendrich,

that peer groups significant to the subject must be considered in making
predictions of behavior from attitudes.

Finally, Mann (1959) used a sociometric questionnaire as an
affective indicator of prejudice, a part of the California Public
Opinion Survey as a cognitive indicator, and a ranking by group
members on the amount of prejudice of other group members as a
behavioral indicator of prejudice. He found a relationship between
these three measures of racial prejudice, but when broken down by
race, whites in the groups had a negative relationship between
affective and behavioral measures, while Negroes had a strong
relationship between all three.

Overall then, there is little empirical evidence for a direct
relationship between attitude and behavior. While attitude may help
to determine behavior, other mechanisms seem more important in being
able to predict behavior. Presented below are a number of theoretical
articles which attempt to account for this discrepancy between

attitude and behavior.

Theoretical articles

 

Rose (1956) argues that racial attitudes and intergroup
relations are fairly unrelated, though they do influence each other
somewhat. He feels that behavior between races is to be explained
in terms of legal, economic, political, and social forces. Even
though Rose sees attitudes and behavior as separate entities, he

still believes that attitudes are worth studying.

10

Deutscher (1966) reviewed the area of attitudes and actions
and concluded that the preponderance of studies, many cited,
indicates there is no relations between the two. He feels that the
social situations and pressures inherent in them often determine
behavior. He also notes that there is no reason to believe that
changing attitudes will change behavior nor vice versa. Deutscher
also takes social scientists to task for acting as if attitudes do
determine behavior even though nearly every study since LaPiere shows
they do not.

Tittle and Hill (1967) in another review article feel that no
conclusion can be made about the relation between attitudes and
behavior. They do feel that the degree of correspondence between
the two is partly a function of (a) the measurement techniques
involved, Cb) the degree to which the criterion behavior is part of
the individual's common range of experience, and (c) the degree to
which the criterion behavior represents a repetitive behavioral
configuration. They then did a study validating 5 measurement
scales: a Thurstone scale, a semantic differential procedure, a
Likert technique, a Guttman scale, and a self-rating scale against
five criteria of behavior. The attitude and behavior scales were
about personal participation in student political activity. They
feund only a moderate amount of agreement between attitude and
behavior, but that this was also a function of the attitude scale
used; the Likert scale showed the greatest relationship to behavior,
and in descending order were the Guttman Scale, self-report, semantic

differential, and the Thurstone scale.

ll

DeFleur and Westie (1963) note the inconsistency between
attitude and behavior, and agree that, "There are clearly situational
factors such as group norms, roles, definitions of situations, and

other social constraints which materially 'mediate' responses in

 

situations involving either verbal behavior or overt action. Such
'mediating' social constraints appear to explain lack of corres-
pondence better than such a conception as 'true attitudes' conceived
of as latent variables." This statement refers to evidence the
authors cite in support of the position that attitude is a probability
conception, rather than a latent process which somehow "mediates"
observable behavior. Also DeFleur and Westie believe that contributing
to the discrepancy between attitude and behavior is the fact that the
attitude measurements are not entirely accurate because of the
inability of subjects to self-analyze themselves and because

subjects are often incapable of making discriminations between self-
concept and items on an attitude questionnaire.

Cook and Selltiz (1964) believe that attitudes are underlying
di5positions which partly determine behavior, but that various
internal characteristics, such as motivational state or expressive
style, and situational characteristics, such as expectations of
others or possible consequences of behavior, also determine behavior.
They also mention that social desirability and response set lead to
inaccurate questionnaire responses, which in turn lead to attitude-

behavior discrepancies.

12

Ballard (1948) states that the validity of opinions in predicting
behavior must be established in each new situation. He also V
lists seven conditions which affect the validity of opinion-action
situations. He postulates that prediction should be better fer people
who have greater verbal ability, people who normally act only after
careful thinking, people who have experienced the behavioral
situation befere, and people in situations in which the behavior is
made explicit. On the other hand, prediction should be worse when
the unconscious is involved in determining behavior, when extraneous
variables such as fear intrude, and when variables intervene between
the attitude measure and the opportunity for action.

Festinger (1964) suggests that attitudes do determine behavior.
He cites the DeFleur and Westie study as proof, but then reviews
three studies which indicate that changing opinion does not change
behavior. He explains these results by hypothesizing that the
environment was not changed, and so behavioral change could not come
about. So even though Festinger believes that attitude does
determine behavior, he also notes the importance of environment or
situation.

Fishbein (1966) hypothesizes that attitude determine action,
but that the response to an attitude must be learned in addition to
learning the attitude itself. Also, he feels that beliefs and
behavioral intentions account for the discrepancy between attitudes

and behavior.

13

Campbell (1963) does not believe that inconsistency has been
shown between attitudes and behavior, but rather that these are
different situational thresholds. For instance, in discussing
LaPiere's study, he notes that it is easier to discriminate by mail
rather than in a face-to-face situation. Inconsistency would have
involved those who refused face-to-face accepting by questionnaires,
or if those who accepted by questionnaire refused face-to-face.
Campbell's position therefore also advocates the importance of the
situation in determining behavior.

Two additional theoretical articles (Kelman, 1967, and Freedman,
1969) discuss the use of role-playing in psychology. They are
relevant to the focus of the present research because the attitude
measure in this experiment and in other experiments has involved
the use of a role-playing instrument. Kelman has taken the position
that deception has been used too much in research and that role-
playing can often be an acceptable alternative. He admits there are
problems with role-playing, but feels many of them can be worked out.
Freedman, on the other hand, feels that people are poor at predicting
behavior in ordinary everyday situations, and that they are almost
totally incapable of predicting behavior in situations which are
out of the ordinary. He does not feel that role—playing has any
place as a research technique except to generate hypotheses which
will be tested empirically, in situations where it is impossible to
induce the actual experimental manipulations, or when long experi-
mentation has shown that role-playing produces the same results as

experimentation.

14

To summarize, then, nearly all the empirical and theoretical
articles have indicated that there is no necessary connection between
attitude and behavior, at least not from the standpoint of being
able to predict behavior from attitude. A number of the authors
have noted that the situation plays a large part in determining
behavior. In particular, a number of the studies involved behavior
which was not in the everyday realm of the subjects involved, and
hence they could not readily predict what their behavior would be.

As Dollard indicates, this then involves testing every new situation
to see if the attitudes and behavior do correspond.

Finally, Lane and Messé (1969) found that there was a discrep-
ancy between choosers and receivers in a wage distribution situation.
The receivers in this situation were seen as giving attitudinal
data, since they knew that their choices of a distribution of rewards
had no effect on the actual distribution of wages, whereas the choices
of the choosers were an actual behavioral measure, since their choices
determined the distribution of wages. However, the responses of the
receivers of the wage distribution may have been more of an
expectancy response rather than a measure of attitude, since their
responses may have reflected the way they expected the choosers to
divide the money, rather than their own attitude as to how they
would have divided the money if given the chance. Lane and Messé
in two separate studies, feund that in certain situations, choosers

had higher self-interest scores in terms of their responses on the

15

reward distribution booklet, indicating they wished to keep more of
the wage fer themselves. In one study, a number of variables were
measured, and this discrepancy occurred when choices of the

choosers of the reward distribution were kept in private rather than
being made public. In the other study, work inputs on the task
were varied and this discrepancy occurred only when choosers and
receivers had high work inputs and the alternatives of the receiver
getting more money or the money being divided equally, occurred

first.

Purpose and hypotheses

 

The purpose of the present study was to examine a measure of
self-interest for three groups, a chooser group and a receiver group
similar to the ones in Lane and Messé (1969), and also a third group
of subjects who were placed in a hypothetical situation where they
did not work, but were told to imagine that they did and that their
choices would actually count in distributing the wage. Also
examined was the amount of money kept by each group on a question
asking how much of a total $4.00 would they keep for themselves.

By means of these comparisons, evidence was gathered relevant to
whether or not a paper-and-pencil test was a valid indicator of the
actual behavior in this situation.

The main hypothesis in this study is that choosers will show
more self-interest scores and will keep a larger percentage of the
$4.00 than the receivers. Likewise, receivers will have more self—
interest scores and keep a larger percentage of the $4.00 than the

subjects in the hypothetical situation.

CHAPTER 11

METHOD

Subjects

Subjects in this experiment were recruited through advertise-
ments in the Michigan State University student newspaper. This
advertisement asked for male students who would be paid for partici-
pating in a motivational research program. Altogether, in this
study, there were 203 male students which were randomly assigned to
three groups, 129 in the hypothetical group, 38 in the receiver
group, and 36 in the chooser group. The reason for the unequal
numbers in each condition was that this data were collected in the
course of a larger series of studies by Lane and Messé, and the
comparison among these three groups was not planned beforehand, but

was an after-the—fact comparison.

Instruments

 

Two instruments were used in this study. One was a Reward
Distribution booklet, (see appendix A) presented in Messé (1968)
and shortened by Lane and Messé (1969). The second instrument,
which was only given to the 38 receivers and 36 choosers, was an
Industrial Opinion questionnaire (see appendix B) developed by
Lane (1970).

16

17

Procedure

The procedure of the experiment involved the 38 receivers and
the 36 choosers working for one hour on the Industrial Opinion
questionnaire, which asked them to write eight short essays on
various topics relating to industry. The responses to the Industrial
Opinion questionnaire were not important in this study in terms of
analysis of the data, but served only as a work task on which both
receivers and choosers worked fer an equal amount of time.

The receivers and choosers were then given the Reward Distri-
bution booklet, but prior to filling it out the 36 choosers were
told that their responses would actually determine the wage received
by themselves and one anonymous other person randomly chosen from
the receivers group. The 38 receivers were told that their wage
would be determined by one of the choosers, again unspecified and
randomly chosen, but that they were to fill out the Reward Distri-
bution booklet merely to see how they would have divided the money,
even though their choices had no effect on the actual distribution
of money. Choosers and receivers were paid in sealed envelopes
according to the choices made by the choosers on the Reward
Distribution booklet. (The purpose of the sealed envelopes was to
keep the choices private). This was done by randomly choosing one
of the first six pages of the booklet, which had the forced-

alternative questions and paying the pairs of choosers and receivers

18

according to the randomly picked choice of each of the choosers in
each of the pairs.1

The 129 subjects in the hypothetical group were not given the
Industrial Opinion questionnaire or any additional task, but they
were paid $1.00 for completing the Reward Distribution booklet.
They were told to imagine that they and another person had each
worked for one hour filling out a questionnaire about problems in
industry, and that they had been picked as one of the choosers.
They were asked to fill out the Reward Distribution booklet as if
their choices would actually determine the wage they and the other
person would receive.

The Reward Distribution booklet consisted of seven pages.
The first six pages of this booklet consisted of three pairs of
pages (X1, Y1, X2, Y2, X3, Y3). which were randomized in presentation
to the subjects to counteract any order effects. On each page, the
subject had a choice of the distribution of reward that he would
prefer to see used. The A alternative on each page always involved
a total of $4.00 which.was divided either with the subject receiving
$3.00 and the other person in the experiment $1.00 (X pages) or
the reverse, the subject receiving $1.00 and the other person $3.00
(Y pages). The B alternative always involved an equal split of the

total amount of money, which was $2.50 each for one pair of pages

 

l . . .
Two choosers actually made ch01ces for two receivers, Since
there were two more receivers than there were choosers.

19

(X1 and Y1 pages), $2.00 each for the second pair (X2 and Y2
pages), and $1.50 each for the third pair (X3 and Y3 pages). For
example, on the pair of pages labelled X1 and Y1, the subject had
the choice on page X1 of his receiving $3.00 and the other person
$1.00 or his receiving $2.50 and the other person receiving $2.50.
On page Yl, the subject had the choice of his receiving $1.00 and
the other person $3.00 or his receiving $2.50 and the other person
receiving $2.50. Pages X2 and Y2 followed this same paradigm with
the B alternative being $2.00 for each person, and likewise pages
X3 and Y3 with the B alternative there being $1.50 for each person.

Before subjects made the six reward distribution choices they
first filled out three infermation questions on each page. These
questions were designed to examine if the subject understood the
alternatives that were presented. The questions asked which alter-
native gave the subject more money, which alternative gave the
subject and the other person the greater total amount of money, and
which alternative gave equal amounts to both.

Finally, the last page of the booklet asked how the subject
would divide $4.00 between himself and the other person. This was
an open-ended question, which the subject could answer in any way
he wished.

Each subject then had three scores obtained from the Reward
Distribution booklet. One score was the amount of money he would

keep for himself out of a total of $4.00 available to himself and

20

another person, this score coming from the last page of the test
booklet. A second score was the number of mistakes each person

made on the information questions preceding the forced-alternative
choices. The third score was the number of self-interest (SI) scores
the subject received. This third score was obtained by looking at
the pairs of pages on the Reward Distribution booklet, where the
subject could obtain one of four classifications of responses. If

a subject chose an A response on page X1 and a B response on page Yl,
he was acting in a self-interested manner (SI), by maximizing his
gain. In other words he chose the alternative on each page that
gave himself the greater amount of money. If a subject chose the

B response on both pages of a pair of pages, he was acting a role-
symmetric manner (RS), meaning he divided the money equally between
himself and the other person. If a subject chose the B alternative
on the first page of a pair and the A response on the second page of
a pair, he was acting in an altruistic manner (ALT), by maximizing
the gain of the other person. In other words he chose the alter-
natives that gave the other person the greater amount of money.
Finally, if a subject chose the A response on each page of a pair,
he was acting in a manner whereby he maximized the total reward

from the experiment, regardless of which person received the greater
amount. However, the last two types of responses were so infrequent
that we are interested only in the role-symmetric (RS) and self—

interest (SI) scores, and this accounts for the fact that the third

21

score represents the number of SI responses that a subject received,
since in analyzing the SI scores all subjects giving ALT and
maximizing of total reward scores were discarded, therefore meaning
that if a subject had two SI scores, he also had one RS score. In
effect this third score has been reduced to a dichotomous response
and we are looking at the number of SI scores, which could range

from zero to three, there being three pairs of pages.

CHAPTER III

RESULTS

To test the hypothesis that choosers would show more self-
interest (SI) scores than receivers, and receivers more than the
hypothetical group, a 1X3 AOV with unequal sample sizes was performed
on the data (Winer, 1962, pp. 96-104). The mean SI score for the 36
choosers was 1.28, for the 38 receivers it was 1.21, and for the 125
hypothetical group (4 of the 129 in this group did not make choices
on the booklet) it was 1.34. The results in the three conditions were
not significantly different (E_= .17; d£_= 2,196), and therefore the
hypothesis was not supported.

To test the hypothesis that the choosers would keep a larger
percentage of $4.00 than receivers, and that receivers would keep a
larger percentage than the hypothetical group, a second 1X3 AOV with
unequal sample sizes was performed. The mean amount of the $4.00
kept by choosers was $2.28, by receivers was $2.26, and by the
hypothetical group was $2.33. Again, the results were not significant
(§_= .15; d£_= 2,200). Thus, the second hypothesis was not supported,

again indicating no difference between the 3 groups.

After examination of the data, several other analyses were also

performed. In the first of these analyses the number of errors on the

22

23

information questions preceding the forced alternative questions was
computed. The mean number of mistakes for the 3 groups was .33 for
the 129 subjects in the hypothetical group, .97 for the 38 receivers,
and .28 for the 36 choosers. A 1X3 AOV with unequal sample sizes
indicated there was a significant difference (E_= 8.24; df_= 2,200;
p.< .01). Using the Newman-Keuls method of comparison of individual
means, the mean of the 38 receivers was found to be significantly
different from the other two means (p_< .05), which did not signifi-
cantly differ from each other.

It was also decided to look at the number of self-interest (SI)
scores for each of the three conditions as a function of the choices
the subjects had available to them. In other words, did the number
of SI scores increase as the amount of money in the B response, the
equity response, decreased. In this case, all subjects who gave
altruistic or social welfare responses were thrown out, reducing the
data to dichotomous data, since each subject either made a SI
response or an R5 response. For the 28 receivers this became 11 SI
scores for the X1, Y1 pair ($4 vs $5), 17 SI scores for the X2, Y2
pair ($4 vs $4), and 17 SI scores for the X3, Y3 pair ($4 vs $3).
For the 29 choosers this became 8, 15, and 20 respectively and for
the 117 subjects in the hypothetical group this became 42, 50, and
70 respectively. Using a 1X3 AOV for dichotomous data with
repeated measures (Winer, 1962, pp. 138-139), significant results

were found in all three groups, indicating that in all three cases,

 

24

the amount of self-interest increased as the amount of money in the
equity response decreased. For the three groups the results were
receivers (E_= 13.00; df_= 2,54; p_< .01, choosers (§_= 5.86;

d£_= 2.56; p_< .01), subjects in the hypothetical condition (E_= 22.25;

df_ 2,232; p_< .01). Using the Newman-Keuls test, it was found that
for the receivers, pairs 2 and 3 were significantly different from
pair 1 (p_< .05), but not from each other. For the choosers it
was again found that pairs 2 and 3 were significantly different from
pair 1 (p_< .01), but not from each other. However, for the subjects
in the hypothetical group, pairs 1 and 2 were significantly different
from pair 3 (p_< .01), but not from each other. This tends to
disconfirm further the original hypothesis that amount of SI in all
three groups would not be equal, since in all three groups the amount
of SI increased significantly, whereas if the groups had differed
on amount of SI scores, it might have been expected that SI scores
would increase in the choosers or receivers group as amount of money
in the equity response decreases.

The last analysis involved a 2-factor, unweighted-means, 3X3
AOV with repeated measures and unequal sample sizes (Winter, 1962,
pp. 374-378). This involved combining the data from the above three
1x3 AOV's with repeated measures and dichotomous data into one large
AOV table. This analysis indicated that the factor of amount of
money in the equity alternative was significantly different (f_=

16.95; df_= 2,342; p'< .01) across all three conditions and that the

25

factor of group, receivers, choosers, or hypothetical was not signifi-
cant in terms of 81 scores, nor was the interaction effect significant.
This further disconfirms the hypothesis that SI scores differ for the
three groups, choosers, receivers, and hypothetical. Using the

Newman-Keuls test, it was found that all three pairs of money alter-

natives differed significantly from each other CB < .01).

 

CHAPTER IV
DISCUSSION

The results of this study do not support the hypothesis that
choosers would show more self-interest (SI) and keep a larger
percentage of $4.00 than receivers. Nor is there support for the
hypothesis that receivers would show more self-interest (SI) and
keep a larger percentage of $4.00 than subjects in the hypothetical
group. Rather, the results indicated that in this particular situation,
there was a fairly close relationship between attitude--as measured
by a hypothetical situation where subjects are asked to imagine they
are in a situation and then tell what they would do-- and behavior--
where the actual behavior of subjects is measured in a real situation
that is identical to the situation subjects were to imagine in the
hypothetical situation. The subjects in the hypothetical group may
be considered to have given an attitudinal measure, since they were
merely asked to imagine what they would do, and their responses had
no behavioral consequences.

The findings of this study are contrary to the findings of
nearly all the other studies done in this area. However, as Dollard
(1948) noted, the relationship between attitude and behavior must

be determined in each particular situation, and since the only study

26

27

that was similar to the present situation was that of Lane and Messé
(1969)--at least in terms of its being a study involving the distri-
bution of reward--it is not particularly surprising that the results
were different. Likewise, we would be justified in thinking these
results might be similar only to those generated in situations which
are very similar to the one presented here. 4
The problem then is to determine what the difference is between
this particular situation and the situations in the studies that
were cited that will account for the differences in the relationship
of attitudes and behavior. As an example, Linn's study (1965) was
similar to this study in that he used one measuring instrument as
both a measure of attitude and as a measure of behavior. This
measuring instrument was embedded in an attitude test and was then
given later in a behavioral situation. Linn found no relationship
between the two measures, whereas the present study did. Dollard's
seven conditions affecting the validity of opinion-action situations
offers a possible reason why attitudes were identical to behavior in

this situation, but not in Linn's. One of his conditions is that

 

people who have experienced the behavioral situation will better be
able to predict their behavior. Tittle and Hill (1967) and also

Linn himself, note in their articles that if the behavior is one that
is experienced as being normal and part of everyday experience,

there is a much greater chance of congruence between attitude and

behavior. It would seem that the present situation meets these

 

28

requirements, as we are always involved in work situations, and if
not actively involved ourselves in dividing the money being paid, we
are conscious of the amount being paid and how it is being distributed.
In other words, we are always evaluating ourselves in relation to
others in the amount of work we do, and the wage we are paid for
working. However, in Linn‘s study, this was a situation involving
face-to-face confrontation with Negroes, with whom most university
students have very little experience, particularly in this case,

where white females were interacting with Negro males. In addition,
the fact that the behavioral situation of allowing one's photograph

to be seen nationally is, by itself, not a normal, everyday situation,
and when put in conjunction with being photographed with a person of
opposite sex and different race, this situation then becomes very

out of the ordinary. This same type of argument can also be made for
a number of the other studies, since they used the study of racial
attitudes and racial behavior, and many people have infrequent

contact with members of other races, making the behavior not a part of
normal, everyday life.

Rose (1956) noted the lack of correspondence between racial
attitudes and behavior and accounted for this in terms of legal,
economic, political, and social forces. If this position is true,
it would help to explain the congruence between attitude and behavior
in this situation, since we are measuring a combination of an economic

and a social attitude with the attitude scale.

29

DeFleur and Westie (1963) cite the inconsistency between
attitude and behavior, and explain the inconsistency in terms of
situational factors such as group norms, roles, and definitions of
situations which 'mediate' behavior. This explanation also seems to
fit the present study because of the frequent exposure that people
have to reward distribution situations which should have enabled
them to formulate roles for themselves, that in most cases, fit
some type of group norm. Since the group norms and roles are so
established because of frequent exposure and use of norms, it is not
surprising that there is attitude-behavior congruence.

Cook and Selltiz (1964) state that situational characteristics
such as expectations of others or possible consequences of behavior
can determine behavior. This would help explain the present situation
since, again, through frequent exposure to work and wage situations,
expectations and consequences are well known and therefbre they are
well-internalized into a person's attitude structure besides showing
in his behavior.

Fishbein (1966) felt that not only must an attitude be learned,
but the response to the attitude had to be learned. Again, through
frequent exposure to wage and work situations, most people have
learned the re5ponses to the attitudes they hold.

There is another of Dollard's seven conditions that accounts
for the results of this experiment, this being the one which postulates

that prediction will be better in a situation where the behavior is

30

made explicit. This condition seems to be met by this experiment,
since explanations were very clear as to the amount of work
done, and wage alternatives on the reward distribution booklet were
very explicit. It is possible that this condition of Dollard's
accounts for the discrepancies in the studies by LaPiere (1934);
Kutner, Wilkens, and Yarrow (1952); Linn (1965); and Brookover and
Holland (1952). These four studies had to do with racial attitudes
and behavior, where subjects had stereotyped attitudes toward other
races, but when they actually came in contact with these members of
other races in a behavioral situation, the attitude descriptions had
not been explicit enough to prepare them fer the actual people they
met. For instance, in the case of LaPiere’s study, the subjects were
asked whether they would serve a Chinese couple, but subjects' stereo-
types of Chinese people are no doubt greatly different from the actual
behavior and appearance of LaPiere's Chinese couple. If LaPiere's
attitude questionnaire had described the Chinese couple, it is
possible there would have been greater congruence, though probably
by no means perfect, since many subjects may have rejected or
distorted the descriptions. I believe the same argument can be made
for the other three studies cited above, all of which had to do with
attitudes and behaviors towards Negroes.

As mentioned earlier, the only studies done that were similar
to the present one were those by Lane and Messé (1969). In these
studies they did find a few significant differences between receivers

and choosers on self-interest scores, suggesting a difference between

31

attitude and behavior. (As noted previously, receivers' responses
may not actually be an attitude measure, but rather a measure of
expectancy). However, these differences were not consistent, and
only occurred in specific situations. In one of the studies that
they report, this discrepancy occurred when the choices were kept in
private rather than being made public, and in the other study
reported by them, the discrepancy occurred when choosers and
receivers had high inputs and Series Y responses were presented first.
Both Lane and Messé believe that their findings provide some positive
support for a difference between hypothetical and real choices.
However, as already noted, these findings did occur in more specific
situations than in the present study and this is a possible explana-
tion for the discrepancy between studies. What this does suggest is
that there is a need for more research that investigates other
variables, to see if the congruence between attitude and behavior is
a general one for reward distribution situations, or whether the
introduction of other variables, as in Lane's and Messé's studies
will have an effect on this congruence. At any rate, this uncertainty
lends support to Freedman's position, that for role-playing to be
accepted as a substitute for actual experimentation, as Kelman
suggests, a relationship would have to be shown for a variety of
similar reward distribution studies before we could have confidence
that role-playing gives the same results as actual behavior.

The last two analyses that were performed on the data, which

looked at the results as a function of the total amount of money

32

involved, indicate that this variable influences a person's decision
to be equitable or self—interested, and this decision is not dependent
upon which group (chooser, receiver, or hypothetical) one is in.

This further provides support for the position advanced by both

Weick (1966, Weick and Nesset, 1968) and Lane and Messé (1971) that
there is a minimum wage people think they are worth, and while many
people will be equitable when this minimum wage is met, they will
move to a position of self-interest once their amount of reward falls
below this. Lane and Messé (1971) suggested that for college
students performing in psychological research this minimum wage was
in the area of $2.00 - $2.50. It should be noticed that on the
question of how much money a person would give to himself out of a
total of $4.00, the mean amounts for the three groups were $2.26,
$2.28, $2.33.

One problem to be noted in this study is that the responses to
the final question on the reward distribution booklet, which asked
how each subject would distribute $4.00, may not be able to differ-
entiate among the three groups. The problem is that the choosers
knew that one of their choices on the first six pages of the reward
distribution booklet would actually determine the amount of money they
and one of the receivers would be paid. However, the question about
the $4.00 did not in any way affect the reward distribution outcome,
so therefbre, it is doubtful to know whether it actually could have
differentiated between groups, since it had no behavioral conse-

quences in any of the groups.

33

This study also gives evidence that receivers are giving the
same responses as subjects in the hypothetical group, indicating that
receivers‘responses actually are attitudinal responses rather than
expectancy responses. However, as noted, the results did show that
the receivers had significantly more errors than the other two
groups. This may be explained by the fact that receivers knew their
choices would have no effect on the amount of money they received,
and that they were told to role-play and imagine that their choices
would count. Even though they had been told to imagine that their
responses would count as the subjects in the hypothetical group had,
they were in the actual wage situation and had contradictory evidence
to indicate that their responses would not count, therefore negating
any instructions to role—play.

There is another way of explaining the differences in the
number of errors among the three groups. The choosers knew that
their choices would count, so they were fairly careful in filling
out the reward distribution booklets. The subjects in the hypo-
thetical group only worked a few minutes in filling out the reward
distribution booklet, fer which they were paid $1.00, thus creating a
situation where they also felt constrained to fill out the booklet
carefully in order to feel that they really deserved the wage.
However, the receivers, as already noted, knew that their responses
would have no effect on the distribution of the reward, so they did
not feel constrained to be as careful on the test booklet as the

other two groups.

34

As many past studies have noted, factors inherent in the
particular situation can play a large part in determining behavior,
rather than the attitude itself. In the present research it is possible
that there may have been a number of situational variables at work.
However, since the results of all three groups were congruent with
each other, except for the number of errors on the information questions,
it seems doubtful that situational variables actually did play a
part in determining behavior, but that rather the situation was such
as to allow the attitude to determine behavior. However, it is
also possible that since equity is a fairly salient behavior, as can
be noted by the number of RS responses in the choosers' group, it
could not be expected that attitude would be that different from
behavior. The only way attitude could significantly deviate would be
if there were less RS responses on the attitude measure than on the
behavioral measure, an unlikely occurrence, since this would mean that
people would express attitudes of self-interest, but in behavioral
situations would act in an equitable manner. This might be possible,
since there might be situational variables which would compel a
person to be equitable when his attitudes were of self-interest, but
this seems unlikely or at least more unlikely than having attitudes
indicate equity more than behaviors do.

This study then indicates that in this particular reward
distribution there is congruence between attitude and behavior,
perhaps because of the behavioral salience of equity. However, this
is only one specific situation, and much further research needs to

be done before we can accept this as a general statement of fact.

BIBLIOGRAPHY

Berg, K. Ethnic attitudes and agreement with a Negro person. Journal
of Personality and Social Psychology, 1966, 4, 215-220.

Bray, D. The prediction of behavior from two attitude scales.
Journal of Abnormal and Social Psychology, 1950, 45. 64-84.

Brookover, W. B., and Holland, J. B. An inquiry into the meaning of
minority group attitude expressions. American Sociological
Review, 1952, 17, 196-202.

Campbell, D. T. Social attitudes and other acquired behavioral
dispositions. In S. Koch, (Ed.), Psychology: A Study of a
Science (Vol. 6), New York, McGraw-Hill, 1964.

Cook, S. W., and Selltiz, C. A multiple indicator approach to
attitude measurement. Psychological Bulletin, 1964, 62, 36-55.

 

DeFleur, M. L., and Westie, F. R. Verbal attitudes and overt acts.
American Sociological Review, 1958, 23, 667-673.

 

. Attitudes as a scientific concept. Social Forces, 1963,
42 (1), 17-31.

 

Deutscher, I. Words and deeds: Social science and social policy.
Social Problems, 1966, 13 (3), 235-254.

 

Dollard, J. Under what conditions do opinions predict behavior?
Public Opinion Quarterly, 1948, 12, 623-632.

Fendrich, J. M. Perceived reference group support: Racial attitudes
and overt behavior. American Sociological Review, 1967a, 32
(6), 960-970.

. A study of the association among verbal attitudes,
commitment, and overt behavior in different experimental
situations. Social Forces, 1967b, 45, 347-355.

 

Festinger, L. Behavioral support for opinion change. Public Opinion

Quarterly, 1964, 28 (3), 404-417.

Fishbein, M. The relationship between beliefs, attitudes, and
behavior. In B. Feldman (Ed.), Cogpitive Consistengy, New
York, Academic Press, 1966.

35

36

Fishman, J. Some social and psychological determinants of inter-
group relations in changing neighborhoods: An introduction to
the Bridgeview study. Social Forces, 1961, 40, 42-51.

 

Freedman, J. L. Roleplaying: Psychology by consensus. Journal of
Personality and Social Psychology, 1969, 13 (2), 107-114.

Freeman, L. C., and Atabv, T. Invalidity of indirect and direct
measures toward cheating. Journal of Personality, 1960, 28,
443-447.

Kelman, H. C. Human use of human subjects: The problem of deception
in social psychological experiments. Psychological Bulletin,
1967, 67 (l), l-ll.

Kutner, B., Wilkens, C., and Yarrow, P. R. Verbal attitudes and overt
behavior involving racial prejudice. Journal of Abnormal and
Social Psychology, 1952, 47, 649-652.

 

Lane, I. M. Behavioral strategies in wage distribution. Unpublished
doctoral dissertation. Michigan State University, 1970.

Lane, I. M. and Messé, L. A. Equity and the Distribution of Rewards.
(Mimeo) Cooperation/Conflict Research Group, Michigan State
University, 1969.

The distribution of insufficient, sufficient, oversuffi-
c1ent rewards: A clarification of equity theory. Journal of
Personality and Social Psycholggy, 1971, in press.

 

LaPiere, R. T. Attitudes vs. actions. Social Forces, 1934, 13,
230-237.

 

Linn, L. S. Verbal attitudes and overt behavior: a study of racial
discrimination. Social Forces, 1964, 43, 353-364.

 

Mann, J. H. The relationship between cognitive, affective, and
behavioral aspects of racial prejudice. Journal of Social
Psychology, 1959, 49, 223-228.

McGrew, J. M. How "open" are multiple-dwelling units. Journal of
Social Psychology, 1967, 72, 223-226.

 

Messé, L. A., Parameters of fair play: an investigation of differ-
ences in the just distribution of rewards. unpublished
doctoral dissertation. University of Chicago, 1968.

Rose, A. Intergroup relations vs. prejudice: Pertinent theory for
the study of social change. Social Problems, 1956, 4, 173-176.

 

37

Saenger, G., and Gilbert, E. Customer reactions to the integration of
Negro sales personnel, ‘International Journal of Opinion
Attitude ReSearch, 1950, 4, 57-76.'

 

Tittle, C. R., and Hill, R. Attitude measurement and prediction of
behavior: an evaluation of conditions and measurement tech-
niques. Sociometry, 1967, 30 (2), 199-213.

 

Weick, K. E. The concept of equity in the perception of pay.
Administrative Science Quarterly, 1966, 11, 414-439.

Weick, K. E., and Nesset, B. Preferences among forms of equity,
Organizational Behavior and Human Performange, 1968, 3, 400-416.

Winer, B. J. Statistical Principles in Experimental Design, New York,
McGraw-Hill, 1962.

Zunich, M. Relationship between maternal behavior and attitudes
toward children. Journal of Genetic Psychology, 1962, 100,
155-165.

APPENDICES

APPENDIX A

Alternative A Alternative B

You get $3.00 You get $2.50
and and
He gets $1.00 He gets $2.50

1. Which alternative gives you
more? A or B? (Check one)

 

 

2. Which alternative, if either,
gives the greater total to the
two of you, together? Check
A or B. If the two give the
same total, check "Equal."

 

Equal
3. Which alternative gives the
two of you more nearly the
same? A or B? (Check one)

 

 

SKIP THE QUESTION BELOW: GO ON TO THE NEXT PAGE

 

 

CHOICE X1:

Which alternative set of payments
would you rather see used? A or
B? (Check one)

 

 

Remember: If this choice is the one drawn, you and he will be paid
just the amounts you now choose.

(X1)

38

44
If you were given $4.00 to distribute to yourself and one person in the

other group, how would you distribute this money?

I would give myself

I would give the other individual

 

Remember these two values must add up to exactly $4.00.

APPENDIX B

INDUSTRIAL OPINION QUESTIONNAIRE

45

46

Name

 

Instructions: Please spend about 8 minutes on each question. Try not
to spend too much time on any particular question. For your conveni-
ence, the time remaining will be written periodically on the board in
the front of the room. The questionnaire will be collected in 69_
minutes. Please try to answer all _8_ questions.

 

47

a. If you were asked to take a psychological test for a job you
wanted how likely is it that you would fake your answers to try
to make a very favorable impression?

b. What personality traits do you feel are generally desired in
prospective employees?

c. Why do you feel these qualities are desired in prospective
employees?

d. To what extent do you feel you possess these qualities?

48

While working on your job you are reprimanded by your supervisor
for making an error in your work. The error was really made by a
co-worker of yours.

a. What would you do in such a situation?

b. Why would you do this?

c. What factors would influence your decision?

d. How do you feel the average workingman would react to such a
situation?

49

While working for a large company you make a suggestion to your
supervisior which he reports to a vice-president as his suggestion.
The suggestion proves to be a valuable one. It saves the company
considerable money.

a. What would you do in such a situation?

b. Why would you do this?

c. What factors would influence your decision?

d. How do you feel the average workingman would react to such a
situation?

50

4. a. What factors do you feel are the most important ones for
employees being satisfied with their jobs?

b. Which factor of the ones mentioned in part a. do you feel is
the most important one?

c. Why did you select this one factor? Explain fully.

51

Your company introduces a new plan. Whenever any suggestion by a

worker saves the company money all the employees share in the
benefits.

a. Do you like this plan better than one in which only the
person who thought of the suggestion gets his bonus?

b. Explain the reasons for your answer in part a.

c. Which plan do you feel would be more successful?

52

6. a. In deciding which of two companies to work for what three
factors would be the most important ones influencing your decision?

b. Why did you select these three? Explain fully.

53

Some companies train their promising young executives by having
them work two years on a job and then transferring them to another
job. These job changes often involve moving to another part of
the country.

a. Would you accept such a job?
b. Why?

c. What were the most important factors influencing your decision?

54

It is widely believed that most employees now place little pride
in their work.

a. Do you feel this is true?

b. If you feel this is true, what do you feel is responsible for
this occurring? If you feel this is not true, why do you believe
that so many people feel that most employees place little pride

in their work?

WIWMHLIJMIIMIIIEHEISII'IWITITS