.ﬁ.

,,-

   

 

 

“ -5“; sur—n—

 

  

w

47’“

.éfé-n

  

 

 

.7 . 4-44—1Y_.__

EHN)‘ «(I - ‘11”1‘ Y "‘" 3"!“
‘11‘111111111i11‘f‘;1.?!‘311. “‘1 , )1! It'll?!"
,I \.: $‘1I‘I’;""Yf’ﬁ“‘ ‘11,;1
. l 1‘

 
 
    

2;

 

3'53
3:71;:

 
 

2 N

2“. ,ot‘h 1‘17‘1

    
 

cup—p
--

    
   
 

jam
"’5‘

    
   

  

. .1111..,41.11}.1u 1‘ 114111;, 14;: (~1- , 1:.
...€~ :1." t' "J" a 1.:I u ff. . 5.411.212 ' 1.
v. ' v ‘ ‘1“ ’1, :.| .’ a“
I? “ 1" 4 . “1&1! 1::
'2 - D‘

. a "I
. 1-3111.

  
 
   

 
 

9n 4,

 
    

Mr- . .0-
uvtr
-.,._
:1.

r - ‘5”. u4~

        

Y

  
   

n :0” It" :4; ‘
“I amp. A' " i‘; ‘1

\L‘Lv’h“, ‘4‘?
. \H‘ W!“

 
        
    
   
  
 
  

        
  
 
  

         

   
  
 
  

    
 
 
 
   
  

   
 
 
  
  
  
 

       
 

  

   

 

    
  

n"! ‘1 " ‘ ’4‘“
y 1 141mm}. 1,‘ L'l;}‘::;s! :
l»: 4' .11'2'31'7' ‘ l-Ji
'2?" 1 “$5 '
.. 3i {2‘37} . 31.3.; :2": 75:22.55
it: ’ u‘fr'. ' .3” :8? ~,' ‘ 3t {1 3.2.!
i‘ﬁllf' , . '1 'l (_ "1 1 , i "in: it”
. 1&3“ 9-“ L‘ 4‘15 f LQ: x:a_ . ,QE .
4.2.4411 1 1 , .-
.. -4. ‘: :gsic ' :gl‘8 ~ 1‘ I. 7

 
   
   
    
   

3:4

 
 
  
  

W
... 7

9’0" Pd‘“ " {fry 3“}:

1471;! .~.
‘91“, ,1"? ti} 2' 1:3,)?

.4- 2“” . q-J‘l
{.1"i¢,.(i 41,4
_, ,1 4i“ 7.

(35’ £53.11": §;:(

- v
15111111,? 2:332 “"
4.5%?"

Erik-i; viii 3:51.

”.4;
a: #93

  
 
  
 

  

..C-c—vn ‘

.4 a’;

         
 

4

 
 

      

 

 
   
 
 

 
 

           
    
 

 
 
 

   
 
    
  
     

 

‘A
. 1t
‘1 $213;
3% 7i,“
3!“ 1‘?!
£95 9.1 .4 .
$3 I. '
it 1 1 .
_n 7 .;A_
‘33. 7 6»:
i‘_ 7n“: .
.12 21:"
I" ' :z.
'1'! JI‘
1?:

us-

    

 
 
  
  

    
     
 

;- r

1 ' "2

2 ‘ 12.
I i: :4"
. 11 1,

.15 -s

1.

'V

577'-

IIUHHUIHIlllUllWHHWIHIIIHIIHHIIHNIINHIIWI

1293 00759 1658

 

 

Thisistocertifythatthe
dissertation entitled
AN INVESTIGATION OF THE INFLUENCE
OF SELF-OTHER FEEDBACK CONGRUENCE
0N FEEDBACK ACCEPTANCE

presented by

Mary Beth DeGregorio

has been accepted towards fulﬁllment
of the requirements for

Ph-D- degreeinhxnhalngL

 

 

 

MS U is an Afﬁrmative Action/Equal Opportunity Institution 0- 12771

PLACE IN RETURN BOX to remove this checkout from your record.
TO AVOID FINES return on or before due due.

DATE DUE DATE DUE DATE DUE

 

 

 

U3
3’”

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

by
4-

FEE—fl
TLIIJLJ

 

 

 

 

 

 

 

 

 

 

 

 

 

MSU Is An Amman,- ~33quan Opportunity Institution
chna-p.‘

 

 

 

AN INVESTIGATION OF THE INFLUENCE
OF SELF-OTHER FEEDBACK CONGRUENCE
ON FEEDBACK ACCEPTANCE
By

MaryBeth DeGregorio

A DISSERTATION

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

DOCTOR OF PHILOSOPHY

Department of Psychology

1990

I l

A c.
l

boﬁq

ABSTRACT
AN INVESTIGATION OF THE INFLUENCE
OF SELF-OTHER FEEDBACK CONGRUENCE
0N FEEDBACK ACCEPTANCE
By

MaryBeth DeGregorio

Performance feedback acceptance is generally considered an integral
component in the performance feedback process. Little research has
considered the factors that inﬂuence acceptance of feedback from an
external source or the consequences of external feedback acceptance.

The present research used a control systems perspective to investigate the

i inﬂuence of congruence between external feedback and self-assessments

of performance on feedback acceptance. College students were given
false feedback regarding their performance on an in—basket exercise. This
feedback was varied in terms of feedback sign, feedback discrepancy and
source credibility. Results showed that (a) feedback sign was positively
related to feedback acceptance, (b) feedback discrepancy was negatively
related to feedback acceptance, (c) self-conﬁdence interacted with
feedback discrepancy to inﬂuence feedback acceptance, (d) external
feedback interacted with acceptance of external feedback to inﬂuence
consensus judgments made regarding performance and (e) self-assessment

of performance carried more weight than external judgments of

performance in inﬂuencing consensus performance judgments.
Implications of these ﬁndings as well as future research needs and study

limitations are discussed.

Coryright by
MaryBeth DeGregorio

1990

ACKNOWLEDGEMENTS

Arriving at this point could never have been accomplished without
the support and encouragement of many people. I would ﬁrst like to
thank my committee members John Hollenbeck, Dan llgen and Steve
Kozlowski for their thoughtful commentary in reviewing my work. Their
efforts helped to improve the quality of this product. I would especially
like to express my deepest gratitude to my committee chairperson Kevin
Ford. Kevin’s faith in my ability and desire for my success were
‘, motivating forces in the completion of this project. I am additionally
grateful for the time and effort he put into this dissertation project.
(Great turn around time Kevin!)

I would also like to acknowledge the assistance of Sandra Lee,
Sharon Sosnowski and Sandy Wanek in collecting and entering the data.
You made the boring task of data collection tolerable and provided a
sounding board for the complaints of a frustrated doctoral student. I
hope this experience was educational and didn’t discourage you from
pursuing advanced degrees.

My sincere appreciation also goes to my friends in East Lansing,

Chicago and elsewhere. The dissertation process can be a solitary one,

iv

however, I always knew I had friends to turn to when the going got
tough. You saved my sanity with matinee movies, evening coffee and
conversation, jaunts to Chicago and of course constant encouragement.
(Jill, Howard, MJ, John, Mary - Thanks!)

A special thanks goes to Ed Mongillo, the best friend and conﬁdant
anyone could hope to have. He was with me at every obstacle I faced,
both personal and professional, to nudge me over each hurdle. The
interest he expressed in my professional endeavors was both encouraging
and motivating. His tremendous patience, love and support greatly
contributed to the success of this project and to my personal growth and
strength. It was a long time coming, but I ﬁnally made it to Chicago and
1 I couldn’t be happier.

Finally, I would like to thank my family without whom this
accomplishment would never have been achieved. My family provided a
safe haven during the most trying times of my academic career. The
unconditional love they have given so freely has provided me with
comfort and security throughout my life.

My family has taught me that life is a series of choices and that I
can choose to be whatever I would like. I thank them for their support
in any decision I made and any direction I chose to pursue. I am sure

that at times they had more conﬁdence in me than I did in myself. They

have made a signiﬁcant contribution to the achievement of both my

personal and professional goals and to the completion of this project.

TABLE OF CONTENTS

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

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

CHAPTER 1: INTRODUCTION ..............................................................

Problem Overview ..... _ - .....
Performance Feedback

 

 

Performance Feedback Acceptance
Feedback Acceptance Outcomes .....

 

Self-Evaluation Change- - - -- --

 

Behavior Change -- .......

 

Feedback Acceptance Determinants .......... -
Feedback Sign .......................................

Source Credibility .............

“‘ “ 3‘ 0000

 

vv::=:.

Psychological Closeness ................ .............................
Self-Processes .............................................................................

Summary--

 

 

Feedback Source congruence---

Congruence Def CIency - -- - - --

Cause of Congruence Deﬁciency

 

Motivational Theories ...................................................
Cognitive Theories .........................................................
Outcomes of Congruence Deﬁciencies ................................

Control Theory

 

_----- --__-
-vvvv ' -v—vvvv.

Primary Components of Control Theory .......................................

Organizational Applications of Control Theory.

 

Additional Control Theory Components ----- - --
Control Hierarchies - -
Controlled Quantity Status----

CHAPTER 2: MODEL AND HYPOTHESES ........
vii

OOOOOOOOOO

xi

xxi

15
I7
18
18
20
2]
23
26
29
29
32
35
35
38

Goals and Performance

 

Feedback

 

Feedback Discrepancy

 

 

Feedback Acceptance

Source Credibility

 

Self-Certainty

 

Consensus Performance Judgment

 

CHAPTER 3: METHOD

 

Pilot Study 1 - ...................

...........

 

Subiects

va

Tack

 

Procedure--

 

Results

 

Pilot Study 11

 

Subjects

 

Task

 

Procedure

 

Results

 

Primary Study

 

Subjects
Task-
Procedure

 

Variables and Measures

 

Feedback Sign

 

Feedback Discrepancy

 

 

Perceived Feedback Congruence
Source Credibility

 

Perceived Source Credibility

 

Self-Rating

 

 

External Rating
Consensus Performance Judgment

 

External Feedback Acceptance

 

 

Self-Feedback Acceptance
Self-Certainty

 

Self-Conﬁdence

 

viii

46
47
48

49

50

50

52

53

53
53
54
54
55

57
57
57
57
59
59
59
63
63
66
66
66
67
67
68
68
68
69
69
69
70
7O

CHAPTER 4: RESULTS- -- - -- - - 71

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Manipulation Checks - -- ........ 73
Hypotheses - -- -- ----- -- 75
Hypothesis 1 -- - - 75
Hypothesis 2 ------ -- -- - ....................... 76
Hypothesis 3 - - ------ 79
Hypothesis 4 ........................................................................................ 82
Hypothesis 5 ........ - - 85
Hypothesis 6 ........................................................................................ 88
Hypothesis 7 ........................................................................................ 90
Additional Results - - - - -- ............. 93
CHAPTER 5: DISCUSSION ...................................................................... 98
Determinants of Feedback Acceptance ................................................... 101
Inﬂuences on Consensus Performance Judgment .................................. 104
i Research Implications .................................................................................. 110
Limitations ...................................................................................................... 114
APPENDICES
Appendix A - In-Basket Exercise ................................................................ 117
Appendix B - Measures Completed ........................................................... 129
Informed Consent Form ............................................................................ 129
Introduction to Task ................................................................................... 130
Performance Dimensions ........................................................................... 131
In-Basket Self-Rating Form ...................................................................... 132
Self-Feedback Acceptance--- - -- ------ 133
Self-Certainty - - ..... - -- 134
In-Basket Evaluator’s Rating- ..................................... 135
Attitude Survey--- .......................................... 136
Perceived Congruence ................................................................................ 142
Perceived Credibility-------- - - 143
External Feedback Acceptance ................................................................ 144
Self-Rating Time IL..-“ - ..................................................... 145
Debriefing.- - -- ...... . ................................................................... 146

 

Appendix C - Credibility Manipulations

 

Appendix D - Results of factor analysis of self-certainty
and self-feedback acceptance scales

LIST OF REFERENCES

 

147

148

149

LIST OF TABLES

Table 1: Means and Standard Deviations for Manipulated
Variables ..... - -- - ..... 6O

 

Table 2: Descriptive Statistics, Intercorrelations and
Reliabilities of Dependent Measures ............................................................ 61

Table 3: Descriptive Statistics, Intercorrelations and
Reliabilities of Dependent Measures ----- - - 72

 

Table 4: Means and Standard Deviations for Manipulated Variables 74

Table 5: Means, Standard Deviations and Cell Sizes for External
. Feedback Acceptance by Feedback Sign, Feedback Discrepancy and
Source Credibility -- - - -- 77

 

Table 6: Regression of External Feedback Acceptance on Feedback
Sign and Feedback Discrepancy. - - -- - 81

 

Table 7: Regression of External Feedback Acceptance on Feedback
Sign, Source Credibility and Feedback Discrepancy- - - - 84

 

Table 8: Regression of External Feedback Acceptance on Feedback
Sign, Feedback Discrepancy and Self-Conﬁdence ...................................... 87

Table 9: Regression of Consensus Performance Judgment on Self-
Rating and Self-Conﬁdence - - - - ---------- - - 91

 

Table 10: Regression of Consensus Performance Judgment on
External Rating and External Feedback Acceptance ............................... 94

Table 11: Regression of Consensus Performance Judgment on Self-
and External Ratings of Performance ------ - - -------------- - 97

 

LIST OF FIGURES

 

 

 

Figure 1: The TOTE Unit- - ............................................... 30
Figure 2: Control System Model of Motivation ......................................... 33
Figure 3: Hierarchy of Control for Lighting 3 Lamp -- -- 36
Figure 4: Powers’ Hierarchy of Control ...................................................... 37
Figure 5: Self-Other Feedback Congruence Model.....- -- ------ . 44
Figure 6: Experimental Design- - ----- - - - 64

 

Figure 7: Plot of the Relationship Between Feedback Sign and
External Feedback Acceptance...-.-...- - - - 78

 

Figure 8: Plot of the Relationship Between Feedback Discrepancy
and External Feedback Acceptance ............................................................... 80

Figure 9: Plot of the Relationship Between Feedback Sign, Feedback
Discrepancy and External Feedback Acceptance ....................................... 83

Figure 10: Plots of the Relationship Between Feedback Sign, Feedback
Discrepancy, Source Credibility and External Feedback and
External Feedback Acceptance -- - ------------ 86

 

Figure 1]: Plot of the Interaction of Self-Conﬁdence and Feedback
Discrepancy on External Feedback Acceptance 89

 

Figure 12: Plot of the Relationship Between Self-Rating, Self-
Conﬁdence and Consensus Performance Judgment ................................... 92

Figure 13: Plot of the Interaction Between External Feedback and
External Feedback Acceptance on Consensus Performance
Judgment .............................................................................................................. 95

Figure 14: Modiﬁed Model of Self-Other Feedback Congruence .......... 107

xii

CHAPTER 1: INTRODUCTION

Problem Overview

Performance feedback is well established as an important
determinant of effective performance (e.g. see llgen, Fisher and Taylor,
1979 for a review). Investigations of performance feedback have
demonstrated its importance in areas such as job design (Hackman &
Oldham, 1976), organizational development and motivation (Locke, 1968).

llgen et a1. (1979) contend that the inﬂuence of feedback on behavior
is a complex process. They propose several cognitive variables that
intervene between feedback and performance. An important variable
which mediates the feedback-performance relationship is feedback
acceptance. Feedback acceptance refers to the recipients’ evaluation of
the accuracy of the feedback that is provided.

llgen et a1. (1979) point out that feedback acceptance is inﬂuenced
by characteristics of the feedback source, message and recipient. One
issue that has not been addressed by feedback researchers is the inﬂuence
of congruence of the message from different feedback sources. It is likely

that incongruent feedback from different sources inﬂuences feedback

2

acceptance (perceived feedback accuracy).

The majority of organizational research investigating the inﬂuence of
performance feedback has been concerned with the joint inﬂuence of
feedback and goals on performance. Becker (1978) investigated the
inﬂuence of goals and feedback on energy conservation. He found that
participants with a difﬁcult goal who were given information regarding the
number of kilowatt hours they used, reduced their consumption
signiﬁcantly more than participants with difﬁcult goals only. Other studies
of .the joint effect of goals and feedback have reached similar conclusions
(Erez, 1977; Komaki, Barwick & Scott, 1978; Strang, Lawrence & Fowler,
3 1978). Upon reviewing the literature regarding the relationship between
goals and feedback, Locke and his colleagues concluded that "neither
knowledge of results nor goals alone are sufﬁcient to improve
performance" (Locke, Shaw, Saari & Latham, 1981, pg. 135).

Investigations of the inﬂuence of performance feedback from a goal
setting perspective have generally overlooked the inﬂuence of variables
which mediate the relationship between feedback and performance. One
reason this issue has been overlooked is that these studies have primarily
been conducted with tasks for which the criterion is quantiﬁable (llgen &
Moore, 1987). Therefore, the feedback could be easily compared to goals

in determining the adequacy of performance. As such, perceptual

3

processes on the part of the feedback recipient become less relevant.
However, for many jobs, especially those for which few objective
measures of performance exist, interpersonal feedback is relied upon
(Larson, 1984). Much research attests to the fact that performance
feedback from others is an important determinant of job performance and
job-related attitudes (See llgen et al., 1979 for a review). The cognitive
evaluation of the feedback, speciﬁcally feedback acceptance, becomes an
important consideration when interpersonal feedback is considered and
when objective information is not available (Festinger, 1954).
A main assumption of self-management theorists is that individuals

.- evaluate their gv_vn_ performance with respect to standards and provide
self-feedback (Manz, 1986). Investigations have found the self to be the
most available, useful and trustworthy source of feedback (Greller &
Herold, 1975; Hanser & Muchinsky, 1978; Herold, Liden, & Leatherwood,
1987). Self-assessments of performance have also been found to inﬂuence
affective, cognitive and behavioral reactions to feedback from other
sources (Brickman, 1972; Shrauger, 1975; Swann & Read, 1981).

The feedback individuals perceive at any given time will likely be a
composite of information selectively obtained from a variety of sources
and weighted by its perceived utility (Taylor, Fisher & llgen, 1984). The

interaction between feedback provided by oneself and feedback provided

 

4

by other interpersonal sources could have important implications for the
acceptance of performance feedback. A lack of congruence between
these feedback sources could inﬂuence the relationships between the
feedback itself, characteristics of the feedback source, characteristics of
the feedback recipient and feedback acceptance.

The purpose of this study is to investigate the inﬂuence of self-other
feedback congruence on feedback acceptance. The congruence between
self-evaluations of performance and other evaluations of performance is
expected to interact with source credibility and individual self-certainty to
inﬂuence feedback acceptance and the ultimate judgment one makes
regarding his/her performance.

The following section discusses the previous work which has been
conducted on feedback acceptance. Limitations to this line of research
are noted. Next, literature relevant to self-other feedback congruence is
reviewed. Control theory is presented to provide a theoretical perspective
for investigating the inﬂuence of feedback source congruence on feedback
acceptance. A model is outlined which delineates the expected
relationships and the predicted hypotheses are speciﬁed. Finally, the
method, results and conclusions drawn from studying these relationships

are presented.

 

5

Performance Feedback

Performance feedback is information regarding how well goals are
being met (Ashford & Cummings, 1983). While performance feedback is
a relatively simple concept, it can be thought of in different contexts.
Performance feedback can be assessed in terms of its purpose, in terms
of individual cognitive and behavioral reactions to the feedback or in a
number of other ways.

The provision of performance feedback serves two primary functions.
First, it provides information regarding the adequacy of performance.
Informational feedback directs behavior toward a desired state or goal
(Locke, et. al., 1981; Locke, 1968). According to Ashford and Cummings
(1983) the informational value of performance feedback will vary as a
function of the uncertainty an individual experiences about issues
surrounding goal attainment.

The second function of performance feedback is motivational (llgen
et al., 1979). From this viewpoint feedback can act as an incentive,
reward or sanction. From the perspective of the organization the
.u_lt_in_1_a_t£ purpose of providing performance feedback is to enhance

performance.

6

The simple provision of performance feedback may M lead to
desired performance improvements. llgen et a]. (1979) posit feedback
acceptance as an important mediator of the performance feedback-
performance relationship. The degree to which performance feedback
has performance enhancing effects is fundamentally a function of its
acceptance.

The following section discusses feedback acceptance outcomes and
determinants. Although goal acceptance has received considerable
attention as an important moderator of the goal setting-performance
relationship (Barley, 1985; Erez & Zidon, 1984) feedback acceptance has
received relatively little research attention as an important variable in the
performance feedback-performance relationship. The research that has
addressed feedback acceptance is discussed next and areas in need of
more research attention are noted.

Performance Feedback Acceptance

Feedback acceptance refers to the feedback recipient’s belief that the
performance evaluation provided is an accurate portrayal of his/her
accomplishments (llgen et al., 1979). Feedback acceptance is an
important organizational objective in so far as it leads to goal attainment.
Outcomes associated with feedback acceptance are discussed next. Once

it has been established that feedback acceptance leads to important

7

organizational outcomes, determinants of feedback acceptance will be
outlined.
Feedback Acceptance Outcomes

llgen et al. (1979) propose that feedback acceptance plays an integral
role in the relationship between the provision of performance feedback
and performance improvements. They also posit the desire to respond to
the feedback and intention to respond to the feedback as two important
mediators of the feedback acceptance-performance relationship.
Organizational research has not investigated the cognitive outcomes
associated with feedback acceptance (desire or intention to respond to the
E feedback) nor the behavioral outcomes of feedback acceptance
(performance improvement). Stone and Stone (1984) suggested that
greater feedback acceptance should result in more changes in self-
perceptions and behaviors but did not empirically test this proposition.

Self-Evaluation Change. Social psychological research has
demonstrated that evaluative feedback that is perceived as accurate
inﬂuences many dimensions of self-perception (see Shrauger, 1975 for a
review). For example, Harvey and Clapp (1965) asked subjects to
describe themselves, how they expected others to rate them,'and how they
hoped others would rate them. They found that subjects with positive

expectancies raised their self-ratings more following feedback that was

8

more positive than expected. Subjects expecting negative evaluations
lowered their self-ratings more following feedback which was more
negative than expected. Following strongly discrepant feedback, there
were no differential changes in self-assessments. While feedback
acceptance was not assessed in this study, their ﬁndings suggest that
feedback acceptance can lead to changes in self-evaluations.

Self-assessment change may be desirable from an organizational
perspective. If self-views were completely unchanging, people would
never gain new knowledge about themselves. A primary function of
performance feedback is to provide information. When performance
. feedback is accepted and changes in self-assessments occur, it is likely
that the performance information will inﬂuence subordinates’ future
performance.

Behavior Change. Research has also overlooked the relationship
between feedback acceptance and behaviors or behavioral intentions.
llgen et al. (1979) suggest that once feedback is accepted, the intention to
respond based on the feedback should be strong. The inclusion of the
behavioral intention in their model recognizes the strong evidence that
performance improvements are contingent on speciﬁc, difficult goals
(Locke, 1968; Locke et al., 1981). If the feedback which has been

accepted relates to speciﬁc, difﬁcult goals, and the intention to improve

 

9

performance is evident, then barring any situational constraints,
performance improvements should ensue. O’Reilly and Anderson (1980)
reported that employees’ perceptions about the accuracy of their
performance appraisal evaluations directly moderated the relationship
between the feedback received and subsequent performance.

Important organizational outcomes that can result from performance
feedback acceptance have just been delineated. Since feedback
acceptance can lead to cognitive and behavioral reactions by feedback
recipients, the factors which stimulate feedback acceptance must be
considered. There has been very little research attention identifying
performance feedback acceptance determinants.

Feedback Acceptance Determinants

Eeedback Sign. Much research on performance feedback has been
concerned with feedback sign (e.g. DeNisi, Randolf & Blencoe, 1983;
Stone & Stone, 1984). A well accepted ﬁnding is that positive feedback
is more likely to be accepted and more accurately recalled than negative
feedback (llgen et al., 1979).

For example, Stone, Guetal and McIntosh (1984) found sign of the
feedback to be an important inﬂuence on feedback acceptance. They
investigated the sequence of positive and negative feedback as it

inﬂuences feedback acceptance. Their ﬁndings supported the hypothesis

10

that performance feedback conveyed in a positive-negative versus a
negative-positive sequence is more accepted. They suggest that the
defensiveness provoked by the initial provision of negative feedback (Kay,
Meyer and French, 1965) blocks the likelihood of feedback acceptance.

llgen et a]. (1979) suggest that the consistency of the feedback sign
inﬂuences feedback acceptance. Consistency refers to whether the
feedback presented in a performance evaluation session is all positive or
negative (consistent) or a combination of the two (inconsistent). llgen et
al. (1979) contend that when feedback is consistent, internal attributions
are made and individuals are more likely to accept the feedback provided.
A This proposal, however, has not been empirically tested.

Source Credibilig. Ilgen et al. (1979) also identiﬁed source
credibility as an important determinant of feedback acceptance.
Credibility can be broken down into a variety of dimensions but an
important one for feedback acceptance appears to be expertise of the
source (Gifﬁn, 1967). Investigations of rater expertise which are
described below, reveal relatively consistent ﬁndings.

Bannister (1986) carried out a study assessing the inﬂuence of
performance outcome feedback on responses to feedback. One variable
of interest in this study was credibility of the feedback source. Credibility

was manipulated by telling undergraduate subjects that they were either

11

being evaluated by a 43 year old personnel vice president or by a college
sophomore intern. Results indicated a small but signiﬁcant difference
between high and low source credibility on feedback acceptance, such that
high credIbility led to higher feedback acceptance ratings.

Stone et al. (1984) investigated the inﬂuence of feedback sequence
and rater expertise on feedback acceptance. Source expertise was
manipulated by varying the name, title and amount of experience
possessed by the rater. Their ﬁndings revealed that when subjects
believed ratings were provided by a high expertise source, greater
feedback acceptance resulted than when subjects believed their rater had
i a low level of expertise.

Egchologjca] Closeness. Research indicates that not only the
credibility of the source but the psychological closeness of the feedback
source inﬂuences reactions to performance feedback. (Greller, 1980;
Greller & Herold, 1975; Hanser & Muchinsky, 1978; Herold, et al. 1987).
Performance feedback can be provided from a variety of potential sources
ranging from the company, supervisor, co-workers, task and self (Hanser
& Muchinsky, 1978; Herold & Greller, 1977). The formal provision of
supervisory feedback is the most prevalent approach for disseminating
performance related information that is used in organizations (Cedarblom,

1982). This type of feedback is relied upon for many jobs, however other

12

feedback sources may be perceived as useful.

Studies have investigated the frequency or amount, consistency and
perceived usefulness of feedback generated by a given source. Greller
and Herold (1975) found that individuals relied most upon feedback
sources that were psychologically close to them. According to Greller and
Herold (1975), psychological closeness refers to the intrinsic nature of the
feedback. Herold and Greller (1977) suggest that feedback sources can
be arranged hierarchically according to psychological closeness from self,
task, and co-worker to supervisor and organization respectively. Their
ﬁndings suggest that individuals viewed feedback sources which were
psychologically close as providing feedback that was more frequent, more
consistent and more useful than psychologically distant sources of
feedback.

In a related study, Herold et a1. (1987) investigated how feedback
recipients rate feedback sources on frequency or amount, consistency and
perceived usefulness of feedback information. They found, consistent with
previous research, that individuals viewed feedback from themselves (self-
generated feedback) as more frequent, more consistent and more useful
than any other source of feedback. Feedback from the task was next,
followed by feedback from supervisors, co-workers and the organization

respectively. As Herold et. al. (1987) note, little attention is paid to how

13

individuals can be their own sources of feedback because most
organizational feedback systems are based on organizational and
supervisory feedback. In addition, no research has yet looked at the
inﬂuence of source of feedback on feedback acceptance.

Barley (1988) conducted one of the few studies assessing
performance as a function of feedback source. Subjects in his study were
telephone salespeople working for a magazine publisher. In this
investigation he manipulated feedback source (self, supervisor) and
feedback speciﬁcity (told exceeded or underproduced goal level, told exact
number sold). The feedback source which Barley identiﬁes as self-

; generated feedback would be better termed task feedback. It was
computer generated feedback sought by salespeople which reports how
many subscriptions they have sold. Barley (1988) found that feedback
from the computer led to superior performance compared to feedback
provided from a supervisor. This ﬁnding provides preliminary evidence
that feedback from psychologically closer sources may lead to greater
performance improvements than feedback from more distant sources.

The research on determinants of feedback acceptance leads to the
following conclusions. Feedback acceptance is enhanced when the
feedback provided is positive, when the feedback source is credible and

when the source is psychologically close. j These research ﬁndings leave
I

14

several issues unresolved. First, as Taylor, Fisher and llgen (1984) point
out, organizations are most concerned with situations in which employees
are performing below standard. Therefore, they should be most
concerned with acceptance of negative feedback. Organizational research
has not addressed how the acceptance of negative feedback can be
enhanced. In addition, research has not addressed the possibility that
feedback recipients may receive feedback regarding the same behavioral
event from two or more sources. If this occurs, the potential for conﬂict
exists.

The majority of organizational feedback systems are based on
i supervisory feedback, yet individual employees can and do assess their
own performance. They also provide themselves with feedback regarding
their performance which could potentially conﬂict with other feedback
sources. The research by Herold, Greller and their associates indicates
that individuals rely on self-feedback more than supervisory feedback.
These investigations have found the self-feedback to be evaluated more
positively than other sources of feedback (Greller & Herold, 1975; Hanser
& Muchinsky, 1978; Herold, et al., 1987). Finally, the interaction of
source credibility with other variables has received little attention. llgen
et al. (1979) suggest that negative feedback may be more acceptable

when the feedback source is credible but research has yet to be

15

conducted on this issue.

Self-Processes. Research conducted in social psychology may help us
to address issues left unresolved by organizational research. Research in
social psychology suggests that individuals are most willing to accept
feedback which is similar to their own self-view. According to self-
consistency theory (Secord & Backman, 1961) this tendency results from
an individual’s desire to maintain a consistent view of themselves. In fact,
Swann and his colleagues (Swarm & Read, 1981, 1982) have found that
individuals bias their search of self-
relevant information toward the acquisition of conﬁrmatory data. In
' addition, they found that the retention of self-conﬁrmatory information
was longer than for disconﬁrmatory information. Finally, conﬁrmatory
information was viewed as more accurate (i.e. more acceptable) than
disconﬁrmatory information.

Social psychologists have also identiﬁed self-certainty (Jones &
Schneider, 1967; Marek & Mettee, 1972) as a potentially important
determinant of feedback acceptance. Self-certainty refers to how
conﬁdent an individual is in his/her self-assessment of performance. Jones
and Schneider (1967) proposed that uncertainty regarding a particular
ability results from unstructured cognitions about the ability. Therefore,

individuals with low levels of self-certainty will be more receptive to

16

information (feedback) about their own competencies.

Jones and Schneider (1967) induced subjects to adopt low self-
appraisals of performance on a social sensitivity task. Certainty was
manipulated by either providing subjects with detailed negative feedback
(high certainty condition) or limited negative feedback (low certainty
condition). They found that desire to receive self-conﬁrmatory
information was higher for those in the high certainty group than for
those in the low certainty group. Implications from this study are limited
as the manipulation of certainty used in this investigation is confounded
with speciﬁcity of feedback, so it is unclear whether subjects preferred
I. feedback because it was speciﬁc or because they were more self-certain.

Maracek and Mettee (1972) also investigated the inﬂuence of self-
certainty on reactions to feedback. They conducted a study in which
subjects were asked to complete self-esteem questionnaires and to
indicate their level of certainty about their responses. Individuals were
then provided with success feedback. Subjects with low self-esteem who
were certain about their self-assessment rejected or minimized success
feedback and avoided future success while subjects who were uncertain
accepted the success feedback. This ﬁnding suggests that self-certainty
may differentially inﬂuence individuals’ receptivity to feedback which is

incongruent with their self-views.

17
Summary. A consideration of the research ﬁndings in the

organizational sciences and social psychology leads to several conclusions.
First, characteristics of the feedback message, especially the inﬂuence of
feedback sign, are well established as inﬂuencing feedback acceptance, but
the acceptance of negative feedback has received little attention. The
inﬂuence of both feedback recipient and source characteristics on
feedback acceptance has received scant empirical investigation. In
addition, the abundant evidence regarding superior and subordinate
disagreements in performance assessment (See Fisher & Russ, 1987 for a
review) suggests a need for studying the joint inﬂuence of source and
recipient characteristics on feedback acceptance.

Social psychology research suggests that disagreement between self
and other sources may be a function of preference for self-confirmatory
information (Swann, 1982). This preference is inﬂuenced by the certainty
with which an individual holds his or her self-view (Marek & Mettee,
1972). Research has shown that feedback which is discrepant from one’s
own self-view is less likely-to be accepted than feedback which is
congruent with one’s self-view (Crary, 1966; Shrauger, 1975). A lack of
consensus between self and other feedback sources should significantly
inﬂuence the positive relationship traditionally found between feedback

and performance. An understanding of the causes of feedback source

18

incongruence is needed to make predictions regarding the joint inﬂuence
of the feedback recipient’s self-feedback and feedback provided by
another.
Feedback Source Congruence

This section considers the issue of self-other congruence in assessing
and evaluating performance. First, the need for an inquiry into self-other
feedback congruence is described. Second, several theories that may
provide explanations for a lack of congruence are noted. Finally, the
inﬂuence of self-other feedback congruence on outcomes such as feedback

acceptance is discussed.

Congr_uence Deﬁciency

Information regarding the adequacy of one’s own performance can be
acquired in a variety of ways. The most common assumption is that
performance information derives from the reaction of others to one’s own
behavior. A critical component of an individual’s self-deﬁnition is the
response of others to an individual’s actions. This is the premise upon
which most organizational feedback systems are based. This presumption
is also the concept from which symbolic interactionism (Cooley, 1902;
Mead, 1934) derives. Symbolic interactionists contend that individuals’

self-perceptions develop from how others view them (the looking glass

self).

 

19

While self-deﬁnition may derive from the responses of others, this
does not necessarily mean that the self and any other single individual will
agree. In organizational settings, agreement between superior and
subordinate perceptions of subordinate performance is critical (Taylor,
Fisher & llgen, 1984).

Research indicates that these individuals often do not agree.

For example, llgen, Peterson, Martin & Boeschen (1981) assessed the
correspondence between supervisors and subordinates reactions to
performance feedback sessions. They found that supervisor and
subordinate perceptions of these feedback sessions show lit_tl_e agreement.
Supervisors’ evaluations of the session were much more favorable than
those of subordinates. In addition, their results demonstrated that overall
written descriptions completed by superiors and subordinates immediately
after the feedback session of how the appraisal session had progressed
were signiﬁcantly different.

Other evidence of self-other rating discrepancies comes from the self-
appraisal literature. Levine, Flory and Ash (1977) found that self-
appraisals have less variability than supervisor-based assessments.
Numerous studies have also found that self-ratings tend to have higher
mean values (leniency) than other appraisal methods (Prien & Liske,

1962; Thornton, 1968, 1980). In addition, self-assessments appear to

20

exhibit less halo error (deﬁned as the intercorrelation among dimensions)
than superior ratings of performance (Lawler, 1967; Parker, Taylor,
Barrett & Martens, 1959). Harris and Schaubroeck (1988) recently
conducted a meta-analysis of the relationship between self- and superior
ratings of performance. Their results revealed only a moderate (.35)
correlation between self- and superior assessments of performance.

In summary, these studies demonstrate that self- and superior ratings
of performance are only somewhat related (Harris & Schaubroeck, 1988).
If self-assessments are not solely based on the viewpoints of others, it
would be useful to explore other bases for making self-evaluations.
b Causes for Congr_uence Deﬁcieng

Several theories have been proposed as explanations for the low
congruence between self- and other assessments of performance (See
Fisher and Russ, 1986 for a review of superior-subordinate agreement).
The following section brieﬂy describes several theories of self-assessment.
The purpose of this discussion is not to identify which theory provides the
"true" explanation for a lack of congruence between self and other
feedback sources. Instead it lays the groundwork for understanding the
joint inﬂuence of self- and other feedback on outcomes such as feedback

acceptance.

21

The basic premise underlying the theories of self-assessment is that
individuals utilize either self-protective strategies (Strube & Roemmele,
1985) or make self-serving attributions for their performance (Bradley,
1978). The cause of this egocentric bias have been ascribed to either
motivational or cognitive processes. In either case, this bias inﬂuences
the nature of self-evaluations which are made.

Motivational Theories. Selfoconsistency theorists contend that
individuals attempt to maintain a ﬁxed self-view (Korman, 1970).
According to Korman (1970) individuals’ self-assessments of performance
are based on their level of task related selfoesteem. If they have high
task related self-esteem people will be motivated to seek feedback which
indicates success. If individuals have low task related self-esteem they will
be motivated to seek feedback which indicates failure, thereby conﬁrming
their initial self-view.

Dipboye (1977) questions this strong consistency point of view. He
suggests that research purported to support the self-consistency view can
also support a self-enhancement view of work motivation. A crucial test
of self-consistency predictions would involve persons who would seek
consistency even at the cost of failure or other negative consequences.

Some research evidence supports the view that individuals desire

feedback which confirms initial self-expectancies (Shrauger, 1975). Swann

22

and Read (1981) found that people are more likely to attend to, encode
and retrieve information if it is self-verifying. In addition, Swarm and
Read (1981) gave subjects an opportunity to seek feedback from
interaction partners that would either conﬁrm or disconﬁrm their self-
conceptions. Subjects displayed a clear preference for feedback that
would conﬁrm their self-perceived emotionality and assertiveness.
Traditional conceptualizationss of self-enhancement theory (Smith, 1968)
have suggested that people have a basic desire to think of themselves
favorably. Self-consistency theory is a similar theoretical position which
posits that people are motivated to sustain their self-conceptions (Swarm,
' 1982). The difference between these two positions can best be
demonstrated in a negative feedback situation. Consistency theory states
that negative feedback is accepted if an individual has negative
expectancies (conﬁrms expectancy of failure). Self-enhancement theory
contends that negative feedback is not readily accepted because
individuals are motivated to think of themselves favorably.

Research also demonstrates that individuals may avoid evaluative
information about ability if this information is damaging to one’s self-
esteem. Subjects who believed that they were high in task related ability
chose to engage in highly diagnostic tasks, but people who believed that

they were low in ability avoided the opportunity to gather diagnostic

23

information (Meyer & Starke, 1982). In another study, subjects who were
led to doubt their ability to maintain high scores similarly avoided
diagnostic test items (Sachs, 1982).

Cognitive Theog’es. Another theory that may help to explain the
differences between self- and other ratings of performance is attribution
theory. Self-attrrbutional bias is the tendency to attribute positive
outcomes to oneself and negative outcomes externally to aspects of the
situation beyond the individual’s control. Jones and Nisbett (1971)
provide evidence that "actors" and "observers" differ in their perceptions
of the causes of behavior, therefore their evaluations of behavior will
differ. They assert that "there is a pervasive tendency for actors to
attribute their actions to situational requirements, whereas observers tend
to attribute the same actions to personal dispositions" (Jones & Nisbett,
1971 p. 2). Actors are likely to emphasize situational constraints,
conﬂicting pressures and role limitations while observers are more inclined
to personalize the cause for success or failure.

Miller (1976) tested this proposition. Subjects were asked to take a
social perceptiveness test. After testing was completed, subjects were
either told that the test was a valid well established test or that it was a
new, experimental and unvalidated examination. The experimenter then

provided subjects with false feedback indicating that they had either done

24

well or poorly on the test. The results of the study indicated that
successful outcomes were attn'buted more to internal factors and
unsuccessful outcomes more to external factors. In addition, these
tendencies were greater under high than under low ego involvement
conditions. That is, high-involvement failure subjects (when test was
presumably valid) made greater attributions for their performance to luck
and less to ability and effort than did low-involvement failure subjects
(when the test was presumably unvalidated). Similarly, high-involvement
success subjects made greater attributions for their performance to ability
than did low-involvement success subjects.

Finally, investigations of the role of self-schemas in self-assessments
also reveal a rationale for differences found between self-assessment and
external assessments of behavior. Markus (1977) takes a cognitive
perspective postulating that an individual utilizes "self-schemata" (cognitive
generalizations of information about the self) to organize personal data.
These schemata are thought to bias the type of information attended to
and encoded into memory. In her study, some subjects were chosen who
had well-developed schemas of themselves as being dependent or
independent (schematic) and others who had not articulated self-schema
on either of these dimensions. Those possessing schemas for

independence processed schema relevant information (independence

25

related adjectives) more quickly than information that was unrelated to
the schema (dependence related adjectives). Subjects with schemas for
dependence showed exactly the opposite effect. Aschematics did not differ
in their processing of either type of information.

Rogers, Kuiper and Kirker (1977) also investigated the existence of
self-schemas and the capacity of self-schemas to inﬂuence the encoding
process. He asked some subjects to determine whether or not a series of
sentences were self-descriptive and to use that decision as an aid to
memory, while other subjects were not given those instructions. Subjects
with these instructions had greater recollection of the sentences than
those without the instructions. In a related study, Kuiper and Rogers
(1979) investigated the processing of self- and other-referential
information. Subjects in their study were presented with adjectives and
asked to rate whether or not they were self-descriptive or descriptive of
another. Their study demonstrated the self-referential adjectives were
more quickly and easily recalled than other-referential adjectives. They
concluded that self-referential data may be more cognitively available than
other-referential information.

The previous review of these theories suggests that the reason self-
other feedback may differ can be explained in motivational or cognitive

terms. First, individuals may wish to protect or enhance their own self-

26

esteem. Therefore, their self-evaluations would tend to be higher than
those provided by other sources. This conclusion is supported by research
which has found self-assessments to be more lenient that other
assessments. A second explanation for self-other feedback differences
may be due to information processing differences. Self-relevant
information has been found to be easier to encode and recall than other
information. There are clearly many explanations for the lack of
congruence between self- and other assessments. Given this information
we now turn our attention to outcomes associated with the lack of
congruence between self and other feedback sources.
Outcomes of Congruence Deﬁciencies

Research supports the view that people prefer events that are
predictable and consistent with their expectations (Snyder & Swarm,
1978). Of greatest interest here are data suggesting that the preference
for information which conﬁrms individuals’ view of themselves, inﬂuences
reactions to self-relevant feedback from others. Shrauger (1975) reviewed
numerous studies which showed that individuals accept information which
conﬁrms their self-views much more than disconﬁrmatory information.
External evaluations consistent with an individuals’ self-view were also
more accurately retained and given more credence than inconsistent

feedback. It is interesting to note however, that affective reactions to

27

feedback did not necessarily indicate a preference for conﬁrmatory
feedback. Individuals showed higher levels of satisfaction with positive
feedback than negative feedback even if initial self-evaluations were low.

Feedback which conﬁrms individuals’ self-perceptions is not only
viewed as more accurate when received, but is actually sought after.
Swarm and Read (1981) conducted a study in which subjects assessed
their own assertiveness and emotionality. They hypothesized that self-
assertive individuals would solicit assertive feedback and self-unassertives
would solicit unassertive feedback. Similarly, those who viewed
themselves as emotional would seek consistent feedback, while those who
believed they were unemotional would seek self-unemotional feedback.
Consistent with these predictions subjects were more likely to seek social
feedback which conﬁrmed their own view of themselves than information
which disconﬁrmed their view. Follow-up investigations (Swarm & Read,
1981) revealed that individuals were even willing to pay for self-
conﬁrmatory feedback and regarded this type of information as especially
informative.

Taken together, the research which has investigated individuals’
responses to feedback which is incongruent with self-views suggests that
this type of feedback may be rejected. People expect to receive feedback

which conﬁrms their self-expectancies. In fact, individuals seek

28

conﬁrmatory feedback and may go to great lengths, such as paying for
such information, in order to receive it. Swarm and Hill (1982) found
that if feedback conﬁrmed self-conceptions passive acceptance of the
feedback resulted. However, if the feedback disconﬁrmed self-
conceptions, they fervently resisted the feedback by demonstrating they
were not as the feedback made them out to be. The fact that individuals
tend to reject feedback that does not coincide with the feedback they
provide for themselves has important implications for organizational
feedback systems. This implies that subordinates will reject performance
feedback that is inconsistent with how they believe they have performed.
From an organizational perspective it may be desirable for subordinates
to accept and act upon the feedback which is received from superiors.

The next section presents the control theory model of self-regulation.
Control theory is presented to provide an integrative framework for
viewing message congruence among feedback sources. First, the basic
tenets of control theory are presented. Then, organizational applications
of control theory are discussed. Finally, control theory concepts are used
to develop a model of individual reactions to self-other feedback

congruence.

29

Control Theog

Cybernetics (Wiener, 1948) or control theory is a general approach to

the understanding of self-regulating systems. It is based on the premise
that regulation occurs between a standard and external feedback. The
application of control theory to psychological issues has been slow in
coming (Miller, Galanter & Pribaum, 1960; Powers, 1973). Only recently
has control theory been utilized in the organizational literature (Campion
& Lord, 1982; Hollenbeck & Brief, 1987; Klein, 1988; Lord & Hanges,
1987; Taylor, et al., 1984) to provide an explanation for the motivating
inﬂuence of goals and performance feedback.
" Primag Comp_onents of Control fljheog

The negative feedback loop is the fundamental element of control
theory. Miller et al. (1960) present the TOTE unit to describe the
functioning of the negative feedback loop (See Figure 1). TOTE is an
acronym for Test-Operate-Test-Exit. This model suggests that
determining the appropriateness of an existing state is achieved by first
t_esti_ng for this condition. Testing is accomplished by comparing the
current state with a referent state or goal. If a discrepancy exists then
some ogration is required to bring the current state in line with the
referent. Testing is repeated until the test is passed. When the test is

passed, control is no longer needed so control exits from this system.

30

Figure 1. The TOTE Unit. Adopted from Miller, Galanter and Pribaum
(1960).

v- Q»

4 TEST Congruity

 

\t

 

 

 

Incongruity

 

 

 

OPERATE

 

 

 

 

31

The feedback loop described above is said to be a negative feedback
loop because the response to an error is an attempt to reduce the
discrepancy (Powers, 1973). A positive feedback loop is a response to an
error that attempts to increase the perceived discrepancy. Taylor et al.
(1984) suggest that investigation of self-regulating systems in the
workplace focus on discrepancy reducing systems. They point out that
when considering the attainment of a performance goal, being above
standard is desirable, however being below standard is undesirable.

The typical illustration utilized to portray the logic of control theory
uses a thermostat. A thermostat is a device which controls the
._ temperature in a given environment. The thermostat is set by its
operator at a speciﬁc temperature such as 68 degrees. This temperature
serves as the referent. 'Ihe thermostat then senses the environment to
determine if the temperature is consistent with the previously set
standard. If a discrepancy is detected between the referent or standard
and the sensed temperature, the thermostat operates by turning on the
heat

When a discrepancy is eliminated, control of the system is not
relinquished as could be deduced from the examination of a single TOTE
sequence. The environment is continually monitored or "sensed" to

determine if conditions have changed resulting in additional discrepancies.

32

Carver and Scheier (1981) term these changes to the existing environment
"disturbances". For instance in the thermostat example, assume the
thermostat operated and brought the room temperature in line with the
previously set standard of 68degrees. Then someone opens a window
which cools the room off considerably. This disturbance reactivates the
control system to reduce the discrepancy between the sensed environment
and the referent.
Organizational Applications of Control Theog

Organizational research utilizing control theory as a framework has
been primarily concerned with control theory applications to motivation.
Speciﬁcally, control theory has been presented as an explanation for the
motivational properties of the interaction of goals and feedback (Campion
& Lord, 1982; Lord & Hanges, 1987). Campion and Lord (1982) suggest
that people regulate their own behavior by monitoring the environment
(See Figure 2). Individuals compare referents or goals with sensed
environmental (task) feedback. Any discrepancy perceived between the
sensor and referent leads to self-correction.

They identify two potential responses to goal-feedback discrepancies.
A behavioral response would most likely result in increased effort or
potentially a change in the strategy used to attain the goal. A cognitive

response would be lowering the goal. The recognition that discrepancies

33

Figure 2. Control systems model of motivation (adopted from Campion
and Lord, 1982).

Referent
(Goal)

 

 

 

 

 

 

 

 

 

 

 

Referent Cognitive
Signal Change
_ error?
"———"'__{Comparator Decision
1 - Mechanism
Sensor
Signal Behavior
Change
,__J__.

 

 

 

sensor l PERSON geffector

— — —
ENVIRONMENT
Feedback Output
(e.g.
action)

 

 

.L--————_ environment

 

34

can be reduced by altering the goal suggests that goal commitment is a
necessary precursor to the performance enhancing effects of the
interaction of goals and feedback. If goal commitment is present, a
cognitive change or change of goal is unlikely.

Campion and Lord (1982) investigated a control theory model of the
relationship between goals and feedback. In their study they asked
students to set goals for performance on ﬁve tests throughout the course
of a term. Their ﬁndings indicated that the comparator was a key
component in triggering changes in behavior or cognitions. Magnitude
and number of failures were associated with behavioral changes such as
increased effort and number and consistency of failures were associated
with cognitive changes such as goal reduction.

Their investigation incorporated control theory in it’s simplest form.
The primary goal of the study was to test the viability of a negative
feedback loop for explaining the relationship between goals and feedback.
However, some of their ﬁndings cannot be explained by this elementary
perspective. Speciﬁcally, a number of students raised their goals after
initial failure suggesting that lower level goals may fulﬁll strategic
functions in attaining more important higher level goals. This ﬁnding
suggests the existence of a hierarchy of control. The notions of goal

hierarchies, multiple goals and controlled quantity status are crucial to the

35

understanding of control theory. A discussion of these additional
components to control theory is next.
Additional Control Theog Components

Qntrol Hierarchies. Complex control sequences can be described by
hierarchies of feedback loops. Control theory research should attend to
the hierarchical level at which behavior is operating. The combination of
feedback loops can be illustrated by an example which uses the
terminology set forth by Powers (1973) (See Figure 3).

Consider attempting to turn on a lamp which will not light. The
standard, or desired state, is for the light to be on. In determining how
to "operate", to turn the light on, one may go through a series of tests.
The ﬁrst step in the sequence may be to test the light bulb in a different
lamp. If the bulb works, one proceeds to another possibility. The second
step could be to test the electrical outlet to determine if it is working
properly. This could be done by testing a lamp in this outlet that you are
certain works. If the working lamp lights then one moves on to the next
possibility and so forth. It is apparent from this example that simple
decisions can be linked in an lengthy chain of potential operations.

The interconnection of feedback loops also suggest that a single

behavior may aid in achieving a variety of referents. Powers (1973)

36.

Figure 3. Hierarchy of control for lighting a lamp. Adapted from
Powers (1973).

 

[Is lamp on?

no

Check light bulb?

F—fﬁorks in another]
Get new NO . - -
lightbulb

 

 

 

 

 

'5'
B
"U

c

 

 

E

I Check outlet J

 

 

L

Another lamp works
in this outlet?

 

 

 

 

Hi}

 

LGet new lamp

 

 

37

Figure 4. Power’s hierarchy of control. Adapted from Powers (1973).

GOAL

Becoming
Educated

Reading a
Book

Turning the
Light Switch

Hand on
, Switch

Turn of
Switch

Hand Grasps
Switch

Grasping

Muscle
Tensions

 

Systems - Organized entities; models; beings.

Principles Generalizations drawn from many
different examples of lower-order perceptions;
facts, heuristics, laws, beliefs.

Programs - A network of choice-points
characterized by tests at the nodes.

Relationships - A regularity in the simultaneous
space-time behavior of two or more independent
lower-order elements.

Sequence or - A ﬁxed succession of lower-order
Event elements; ordering.

Transition - Time and space changes; partial
derivatives.

Conﬁguration- An object, pattern, arrangement, or
invariant of the present moment.

Sensation - Quality of intensity; vector.

Intensity - Magnitude of stimulation of sensory
receptor; energy ﬂow.

38

hierarchy provides the standard for the next lower level in the hierarchy.

That is, turning on the lamp could be a standard in the middle range
of a hierarchy of standards. A subordinate standard at the lowest level
would be the muscle tension required to turn the switch. At higher
levels, turning the lamp on could be part of a plan of operation to
achieve a superordinate standard such as reading a book. Similarly,
reading a book could be part of an operation designed to achieve an
even higher order goal of becoming educated. At each level in the
hierarchy, the outcomes of behavior are assessed by monitoring input
information and comparing it with the referent which derives from the
‘ level above. At the lowest level the sensors indicate the present level of
muscle tension and the comparator assesses whether the desired muscle
tensions (referent) are being created. Similar processes occur at
successively higher levels.

Controlled Quantity Status. The notion of controlled quantity status
concerns the issue of what discrepancy behavior is attempting to reduce.
That is, attention should be paid to the level in the hierarchy of control
at which behavior is operating. Behavior geared toward lighting a room
will have no inﬂuence on the room temperature. The controlled quantity
status (that which is being controlled) of the actor must be identiﬁed to

explain behavior in control theory terms. Campion and Lord (1982)

39

assumed that the controlled quantity for subjects in their study was
individual test score not overall course grade. This assumption led to the
unexpected ﬁnding that subjects who failed to achieve individual test goals
actually raised their subsequent goals. It may be that the controlled
quantity for these individuals was overall course grade.

While Campion and Lord (1982) were primarily concerned with the
initial application of control theory to organizational issues, Hollenbeck
and Williams (1987) addressed the issue of control quantity status in their
investigation of control theory. They identiﬁed six facets of work which
could serve as controlled quantities for department store salespeople.
These facets were: 1) base pay, 2) job security, 3) the nature of the work
itself, 4) co-worker relations, 5) job performance and 6) supervisory
practices. Their ﬁndings indicate that the inﬂuence of goal level on
performance was signiﬁcantly higher for individuals with high performance
control. That is, when job performance was the controlled quantity, the
relationship between goal and performance was strengthened. These
results support the view that not all perceptions require behavioral
control, therefore it is necessary to consider the performance control
variable which is operating.

This ﬁnding suggests the possibility of expanding the control theory

perspective to organizational issues outside of motivation. The research

40

by Hollenbeck and Williams (1987) suggests that standards other than
performance can be controlled quantities. For example, individuals
receiving performance feedback may wish to control the discrepancy
between self-generated feedback and feedback received from external
sources.

Taylor et al. (1984) discuss the effects of feedback on individual
reactions from a control theory perspective. They consider discrepancies
between personal and organizational goals, the causes for these
differences and potential ways of reducing these differences. They
contend that performance feedback must closely correspond with the goal
for performance feedback to result in desired performance improvements.
In addition, the feedback recipient and feedback provider must agree on
the goal or desired behavior. Taylor et al. (1984) posit that any
incongruity between the goal of the feedback recipient and the goal of
the feedback provider could result in feedback which is perceived by the
recipient as inaccurate. However, an issue left unaddressed is the
existence of self-other feedback congruence and the inﬂuence of
congruence on feedback acceptance issues.

Note that the model and proposed hypotheses presented next, do not
attempt to provide a test of control theory. Instead, control theory is

utilized as a framework for examining the performance feedback process.

41

The fundamental position of control theorists is that individuals regulate
their behavior in relation to standards (Powers, 1973). They determine
their match to standards through external feedback. Discrepancy reducing
mechanisms are posited by control theory to operate between behavioral
standards and external feedback. Individuals may also attempt to reduce
discrepancies which exist between self-assessments of performance and
external assessments of performance. This can be accomplished through
the acceptance or rejection of external feedback.

Lord and Hanges (1987) posit feedback acceptance as a boundary
condition to a control theory model of the relation between goals and
feedback. However, it may be that feedback acceptance is controlled at a
lower level in the hierarchy of the control system operating to reduce the
discrepancy between performance goal and performance feedback control.
Control theory explanations of behavior are still applicable when
considering the discrepancy between self-generated feedback and other
generated feedback as that which behavior attempts to reduce. The
perceived discrepancy between these two feedback sources will inﬂuence
feedback acceptance and potentially cognitive and behavioral reaction to
feedback. In control theory terms, self-feedback is the referent against
which feedback from other sources is compared. Instead of concentrating

on the regulation between a standard and feedback, the controlled

42

quantity in this utilization of control theory is the feedback itself.

The following section describes a model which utilizes control theory
to view the performance feedback process. External feedback is
compared to self-feedback in this control system. The model details
individual responses to incongruence between self and other feedback
sources. Finally, this model will be used as a basis for advancing

hypotheses regarding responses to feedback incongruence.

CHAPTER 2: MODEL AND HYPOTHESBS

Figure 5 represents a conceptual model illustrating the relationships
among variables relevant to self-other feedback congruence. In line with
a control theory perspective this model suggests that individuals regulate
their behavior (box 2) in relation to a goal (box 1). In Campion and
Lord’s control theory formulation, performance judgment or feedback
from an external source (box 33) is perceived at the sensor (box 3b).

A unique contribution of the control system model of self-other
feedback congruence is the explicit addition of a self-generated
performance judgment or self-feedback (box 4). The self-generated
performance judgment can be thought of as a standard or referent in
control theory terms. People strive to maintain a consistent view of
themselves and often bias search for self-relevant information toward self-
conﬁrmation (Swarm & Read, 1981). Individuals often believe that their
self-evaluations are the correct ones, therefore their attention to self-
relevant information is biased.

Feedback provided by an external source is compared with the
referent of self-feedback through the comparator function (box 5). The
feedback is compared in terms of discrepancy and sign of the feedback.
If no discrepancy is detected (box 6) between the self-feedback and

external feedback and external feedback is positive, feedback acceptance

43

Figure 5. Self-other feedback congruence model

 

 

 

 

 

 

 

 

 

 

 

 

 

 

    

 

 

 

 

 

 
   

oweqnu wou>enom
owaocu
~ouoa>ozen .
uoq>ocom u=o«>uum cu :HJJuMII. IlmLtw acu>unon
on xuanvoum
cocaBMOuuom
azmzzomH>ZM vuuouocoo
aaaocuouxm
I I I, I
Asaumsmv q<=oH>HazH ul-
IM~:II oz I! - oz
uoa>ozon owcozu III 0“
cu coaucoucu sucaua«c~uu

       

 

o Abouuov

 

ouuaom
a“ Auouuov mo* >ucoauuonwn

 

 

 

  

 

  
 

 

 

  

 

       

 

   
 

 
  

 

  

 

 

 

 

  

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

no
>ucoaouuuao n monogamou< an
III. . sooavoum uonoom
nuaueoEomvoh vououucoo n
.eoecou\aooov aaaocuouxm - can»:
IIII heuounnaco xumaewom zoeoaouomwvt
aaﬁcacomz III casoumcn Auunuo\CHomv
ccwnauon \ououonuoucn neueueaaou q
anacmzuoz . .III xuoavoum
- neumﬁuon ouceEhouuom .
ucaaowvah IIII vououucuu
GUGIIhOwHUAH mn HHQMI
noncoecoo oueeunouu<
xuoaeoom acuuouuu
wououocuo m
«mango uaom suacauuoo «Hum
o>wuucwou , — IIII
. H anon-
ed uncozu ueouuuum
Heoc Ill

 

 

 

45

will be more likely to result (box 7). If a discrepancy between the
external and self-generated feedback is detected, feedback acceptance is
less likely to occur.

The decision of whether to incorporate external feedback or not (box
8) will be inﬂuenced by self-certainty (box 9) and source credibility (box
10). If self-certainty is high, acceptance of self-generated feedback (box
11) should be high. Similarly, if source credibility is high acceptance of
externally generated feedback (box 7) should be high. The extent to
which self-generated feedback (box 4) and externally generated feedback
(box 33) inﬂuence consensus performance judgment (box 12) is a function
‘ of self-generated and externally generated feedback acceptance.
Consensus performance judgment is a clinical combination of the
information which is received from oneself and an external source of
feedback. It is this consensus performance judgment that is compared to
the goal through a comparator function (box 13) to determine the
adequacy of performance.

It is at this point where the controlled quantity changes. The system
is no longer controlling for a self/other feedback discrepancy, but controls
for a goal/consensus performance judgment feedback discrepancy. This
comparison represents the traditional way control theory has been applied

in organizations. A goal is compared to the feedback (consensus

46

performance judgment) to determine if a discrepancy (box 14) exists. If
no discrepancy exists, previous behavior is continued. If a discrepancy
exists, a decision mechanism (box 15) is engaged and the individual makes
either a cognitive (goal change box 16) or behavioral change (intention to
change behavior box 17).

A complete understanding of the preceding formulation requires a
detailed explanation of the proposed relationships among the variables
which have been presented. The focus of this study is on the control
theory formulation of self-other feedback discrepancy, because traditional
control theory applications of goal/feedback discrepancy have been
previously studied (e.g. See Campion & Lord, 1982; Hollenbeck &
Williams, 1987; Klein, 1987). The following section provides descriptions
of each of the components to be assessed. The unique contribution of
each of the variables is discussed. Hypotheses regarding the proposed
relationships among these variables are then presented.

Goals and Performance

Previous research has clearly established the positive relationship
between goals and performance (Locke, et al., 1981). Goals which are
speciﬁc and difﬁcult direct behavior toward a desired outcome, such as

task performance.

47
Feedback

After a task has been performed, feedback is provided. Feedback is
a message an individual receives from a source, which provides
information regarding the adequacy of task performance (llgen et al.
1979). Feedback can originate from a variety of sources, the task, other
individuals or from the individual him/herself (Greller & Herold, 1975).

Self-generated feedback is a primary component in self-assessment.
Theories of self-regulation (Bandura, 1977) suggest that individuals
observe and evaluate their own behavior. Self-generated feedback is a
personal judgment regarding performance. This self-evaluation of
performance serves as a referent against which all other assessments of
performance are compared.

External feedback is the type of feedback which is generally discussed
in the literature. It is feedback which is not intrinsic to the performer.
External feedback can come from supervisors, co-worker, subordinates or
from the task itself. Research has focused on interpersonal sources of
feedback, speciﬁcally supervisory feedback. While feedback is always
present in some form, all sources of feedback are not perceived equally.
Self-feedback is the standard against which other forms of feedback are
compared. Feedback which is not perceived as accurate will be rejected

(llgen et al., 1979; Taylor et al., 1984).

Feedback Discrepancy

Feedback discrepancy refers to the lack of agreement among
feedback sources. While much research has investigated the inﬂuence of
various sources of feedback (organizational, self, supervisor, co-worker,
subordinate and task), little research has addressed the effect of
disagreement among feedback sources.

Fisher & Russ (1987) discuss several reasons why self-superior
feedback disagreements occur. Stone and Stone (1985) addressed the
issue of whether consistent feedback from multiple sources is more
effective in altering self-perceptions than the provision of consistent
feedback on multiple occasions by a single feedback source. In addition,
they investigated the effects of single versus multiple feedback sources on
self-perceived task competence. However, no research has investigated
the inﬂuence of agreement among the two most common feedback
sources, self- and supervisory feedback.

Agreement can be assessed in absolute terms and in terms of
feedback sign. External feedback can be similar or dissimilar to self-
feedback and can be above (positive sign) or below (negative sign) self-

feedback. Both of these variables will inﬂuence feedback acceptance.

49

Eeedback Acceptance
A variable identiﬁed by llgen et al. (1979) as an important mediator

between objective feedback and performance is feedback acceptance.
Feedback acceptance is deﬁned as the perceived accuracy of the
feedback. Whether or not the feedback is actually an accurate portrayal
of performance is insigniﬁcant to feedback acceptance. Since feedback
acceptance can be linked to key organizational outcomes, its study is
important.

Individuals are often unwilling to accept information about
themselves which is inconsistent with what they know about themselves
' (Shrauger, 1975; Swarm & Read, 1982). Individuals tend to reject
information that is inconsistent with their own self-view.

Another aspect of acceptance which has not been discussed in the
literature is acceptance of or perceived accuracy of self-generated
feedback. Self-generated feedback derives from the performance
judgment that one makes regarding his/her own performance. Therefore,
the following hypotheses regarding factors affecting feedback acceptance
are posited:

H1: Feedback sign will be positively related to acceptance of
externally generated feedback.

H2: Feedback discrepancy will be negatively related to acceptance of
externally generated feedback.

50

H3: There will be an interaction between feedback sign and feedback
discrepancy on acceptance of external feedback, such that the
relationship between feedback sign and feedback acceptance will
be stronger when self-other discrepancy is high.

Source Credibility

A characteristic of the feedback source that can inﬂuence the
relationship between objective feedback and feedback acceptance is the
credrbility of the source. Source credibility is deﬁned as the perceived
expertise of the feedback source. While a direct relationship between
source credibility and feedback acceptance has been established (see e.g.
Halperin, Snyder, Shenkel & Houston, 1976; Stone, et al., 1984), no
research has addressed the interactive effect of source credibility and
feedback discrepancies. In addition, llgen et al. (1979) suggest that
acceptance of external feedback that is negative will be moderated by
source credibility. However, research to date has not investigated this
issue. The present research hypothesizes that:

H4: The relationship between the feedback comparator function
(feedback discrepancy, feedback sign) and acceptance of
externally generated feedback will be moderated by source
credibility. The relationship will be stronger when source
credibility is high.

Seigcminu
Self-certainty refers to the conﬁdence one has in his/her self-view.

When one is certain about possessing an attribute, the perceived accuracy

of self-generated feedback should be high (Swarm & Read, 1981).

51
Markus (1977) compared schematics (those who had well developed self-
views) to aschematics (those who did not have well developed self-views)
on the attribute of assertiveness. She found that schematic individuals
were likely to reject feedback that was inconsistent with their self-schema.
Those who believed they were assertive rejected feedback indicating they
were not assertive and those who believed they were unassertive rejected
feedback indicating they were assertive.

When self-certainty is low individuals accept information about
themselves whether it be discrepant from their own self-view or not. This
suggests that when individuals do not have a well established self-view
they will be less likely to perceive their self-generated feedback as
accurate, therefore acceptance of self-generated feedback should be
diminished. New employees prefer feedback from others as opposed to
self-generated feedback (Hillery & Wexley, 1974). Acceptance of self-
relevant information is also dependent on its information value, i.e.; the
incremental increase in knowledge about the self that the information
provides (llgen et al., 1979). When people are not self-certain, self-
generated feedback may be perceived as less valuable. Therefore, the
following hypothesis is proposed:

H5: The relationship between the comparator (feedback discrepancy,

feedback sign) and acceptance of self-generated feedback will be

moderated by self-certainty. The relationship will be stronger
when self-certainty is high.

52

Congnsus Performance Judgment

Consensus performance judgment refers to a clinical combination of
performance relevant feedback from varying sources. In this study, we
are interested in the combination of feedback from oneself and feedback
from an external source. The importance placed on these two sources of
feedback will be inﬂuenced by the acceptance (perceived accuracy) of
feedback from each source:

H6: The relationship between self-generated feedback and consensus
performance judgment will be moderated by acceptance of self-
generated feedback. The relationship will be stronger when
acceptance of self-generated feedback is high.

H7:The relationship between externally generated feedback and
consensus performance judgment will be moderated by acceptance of
external feedback. This relationship will be stronger when
acceptance of external feedback is high.

No speciﬁc hypotheses are proposed to predict the interaction of self-and

other generated feedback on consensus performance judgment. This

relationship will be investigated in an exploratory fashion. No basis exists

for predicting the weight that either self or other generated feedback will

be given in such an analysis.

CHAPTER 3: METHOD

Prior to conducting the main study of this project it was necessary to
conduct two pilot studies. These pilot studies addressed several key issues.
This study was interested in assessing reactions to feedback discrepant from
one’s own self-view. The provision of highly discrepant feedback requires that
self-ratings are neither too low nor too high. Therefore, it was necessary to
choose a task for the main study that would not result in either a self-
assessment ﬂoor or ceiling effect. The primary purpose of Pilot Study I was to
test two tasks and choose the one resulting in a moderate mean self-assessment
rating. In addition, the strength of the credibility and feedback congruence
manipulations were Checked. Finally, the reliabilities of some of the dependent
measures were assessed in this pilot study.

Due to the results of Pilot Study 1, a second pilot study was conducted.
In this pilot study, the strength of the manipulations were checked and the

reliabilities of the measures assessed.

Pilot Stud

Subjects

Subjects in the ﬁrst pilot study were 33 undergraduate students enrolled
in an introductory psychology course Fall term, 1988. Students received course

credit in return for their participation.

53

54
flask

The ﬁrst pilot study was conducted in a laboratory setting. Two tasks
were developed for this study. The ﬁrst task was a sales task in which subjects
were asked to role play selling a nutritional supplement to a confederate.
Subjects were provided with a detailed description of a ﬁctitious nutritional
supplement. They were also given information regarding the steps to an
effective sale as well as fabricated testimonials from satisﬁed customers. They
were given 15 minutes to familiarize themselves with the material and 15
minutes to sell the product to the confederate.

The second task was an in-basket exercise. The in-basket exercise used
in this study was developed for use in an assessment center for a large
midwestem computer organization. The original in-basket was designed for
salespeople who were allowed 75 minutes to complete it. The in-basket was
shortened for the purposes of this study. Subjects were allowed 20 minutes to
complete the in-basket exercise. This shortened version of the in-basket was
piloted to determine the amount of time needed to complete the exercise.
11%

Pilot study I subjects arrived at the experimental session individually.
Subjects were escorted to the room where the experimental session was to take
place. They were asked to read and sign an informed consent form. The task

in which they were engaging (sales or in-basket) was then explained to them

55

and the rater credibility manipulation was given. This manipulation consisted
of telling subjects they would be evaluated by either an undergraduate student
who was also participating in the study for course credit or by a graduate
student. Subjects were given 20 minutes to complete the task.

After completion of the task the experimenter returned and asked the
subject to ﬁll out a self-assessment form and a self-certainty questionnaire
(These questionnaires are described in greater detail in the section to follow
titled "Variables and Measures"). These items were collected from the subject
and a ﬁller task consisting of attitude and personality questionnaires was given.
In the meantime the rater ostensrhly rated the subjects’ performance. Subjects
were shown the rater’s evaluation and asked to complete a feedback
acceptance questionnaire. Subjects were debriefed and thanked for their
participation.

Banks

The ﬁrst pilot study was conducted before a determination was made to
include both positive and negative feedback in the study. Therefore, the sales
and in-basket tasks were pre-tested to determine which was least likely to result
in a self-rating ﬂpo; effect. A ﬂoor effect would prevent the possibility of
providing feedback which was negatively discrepant from self-assessments. The
sales task resulted in mean self-assessments of 4.3 (on a 10-point scale), while

the average self-rating for in-basket exercise performance was 5.0. In addition,

56

subjects displayed a great deal of resistance to participating in the sales task.
Therefore, the in-basket exercise was chosen as the task to be completed in
this study.

Comments from subjects who completed the in-basket exercise indicated
that the time allowed to complete the exercise was insufﬁcient. Therefore, it
was decided that subjects would be allowed 45 minutes to complete the
exercise rather than 20 minutes.

The results of the rater credibility manipulation indicate that the
manipulation was not strong enough. A chi-square statistics was calculated to
be 10.51 df = 10 and the signiﬁcance level was .40. Due to these results, the
credrhility manipulation was made stronger in pilot study 11. Not only were
subjects informed that the rater was either an undergraduate or
industrial/organizational graduate student, but they were told that the
undergraduate had absolutely no experience in rating performance while the
graduate student had extensive experience in rating in-basket exercise
performance. In addition, the manipulation was presented after completion of
the task to make it more salient and to eliminate any inﬂuence the
manipulation might have had on performance.

The reliabilities of the feedback acceptance and the self-certainty scales
were also assessed. It was determined that the feedback acceptance scale was

reliable (alpha = .83; n=33) but the self-certainty scale needed slight

57

modiﬁcations (alpha = .72; n=33). Also, the variance on self-certainty was

relatively low (Mean = 4.36 SD = .61).

Pilot Studfﬂl

meets

Subjects in the second pilot study were 33 undergraduate students
enrolled in a psychology course winter term, 1988. Students received extra
credit in return for their participation.
Task

The second pilot study was also conducted in a laboratory setting. The
task subjects were asked to complete was an in—basket exercise. Subjects were
allowed 45 minutes to complete the exercise. The in-basket was a modified
version of one used to select salespeople at a large midwestern computer
organization.
Procedure

Pilot study 11 subjects arrived at the experimental sessions individually.
They were escorted to the room where the experimental session was to take
place. Subjects were asked to read and sign an informed consent form. The
in-basket exercise was then explained to them and they were allowed 45

minutes to complete the exercise.

 

58

After 45 minutes the experimenter collected the in-basket materials and
asked the participants to complete a self-evaluation and self-certainty
questionnaire (Questionnaires are described in greater detail in the section to
follow titled "Variables and Measures"). At this time, the subjects were given
the credibility manipulation. They were told that their performance was either
being evaluated by an undergraduate student with no previous experience in
performance assessment or in-basket exercises, 9; that they were being
evaluated by an industrial/organizational psychology student with much
experience in the evaluation of in-basket exercise performance. The
participants were asked to complete an attitude questionnaire while the rater
ostensibly rated their performance.

The experimenter returned 15 minutes later and showed the participant
the evaluation that the rater had provided. The ratings were manipulated to
be 0, 10 or 20 points (based on pilot study I data) below the participants’ self-
rating, depending on the feedback congruence condition to which the subject
had been assigned. The subject was then asked to complete a questionnaire
including the feedback acceptance, perceived credibility and perceived feedback
congruence scales. Finally, they were asked to complete a second self-
evaluation and indicate whether or not they intended to change their behavior
in a second in-basket exercise. Subjects were debriefed, and thanked for their

participation.

59
Basic

The results of the manipulation checks on source credibility and feedback
discrepancy, indicated that there were statistically signiﬁcant differences in
subjects perceptions of l) the credibility of the feedback source F(1,31)=12.39,
p<.001 and 2) feedback congruence F(2,30)=27.48, p<.001. A post-hoe
Scheffe test indicated that the high, moderate and low feedback congruence
conditions were signiﬁcantly different and in the expected direction (See Table
1).

Another purpose of the second pilot study was to test the reliabilities of
the dependent measures. Descriptive statistics, intercorrelations and reliabilities
for these scales are provided in Table 2. It was determined that the
reliabilities of the dependent variable scales were adequate.

rima tud
m

Subjects were non-freshman undergraduate students enrolled in
psychology courses in March and April of 1989. Subjects received course credit
in return for their participation. A power analysis was conducted to determine
the sample size needed with a statistical power of .80. The analysis requiring
the greatest statistical power attempts to explain a variance of .4 and the effect
size associated with the interaction term explains an incremental increase of at

least 2.5% of the variance. The number of subjects needed to detect this

60

Table 1. Means and Standard Deviations for Manipulated Variables.

 

DV Source Credibilitvﬁ
(High Cred) (Low Cred)
Mean SD Mean _S__12

 

 

 

Perceived

Source

Credibility 5.47 1.14 3.90 1.41

DV Feedback Discrepancy

(High Disc) (Low Disc) (No Disc)
Mean SD Mean SD Mean SD

Perceived

Feedback

Congruence 2.63 .97 4.95 1.19 6.30 1.38

 

 

 

'il.‘ “
-..‘7-.

 

61

Table 2. Descriptive Statistics, Intercorrelations and Reliabilities
of Dependent Measures.

 

 

DV Mean SD 1 2 3 4
1) Perceived
Source
Credibility 4.71 .49 (.89)*
2) Perceived
Feedback
Congruence 4.51 .92 .27 (.96)
3) Feedback
Acceptance 4.31 .53 .39 .74 (.79)
4) Self
Certainty 4.61 .10 -.01 -.17 -.14 (.79)

"Values in parentheses indicate alpha reliability coefﬁcients.
Correlations less than -.28 or greater than .28 are signiﬁcant at the
.05 level.

 

62

effect with a desired statistical power of .80 at the .05 level was calculated to
be 194.

Data was collected from 280 subjects who were assigned-to one of three
feedback discrepancy conditions (high, low, no), one of three
feedback sign conditions (positive, neutral, negative), and one of two source

credibility conditions (high, low). It was discovered that one of the research

 

assistants participating in this project was providing subjects with incorrect ’7
instructions. It was therefore decided that the data from the 24 subjects who ..
received the erroneous information would be deleted from the study. The ﬁnal
sample size was reduced to 256.

The design appears to be a 3 x 3 x 2 factorial design but this is not the
case. Feedback discrepancy is deﬁned as feedback which is discrepant from
one’s own self-view and feedback sign is deﬁned as being above (positive),
below (negative) or the same as (neutral) one’s self-rating. Therefore,
whenever non-discrepant feedback was provided, this feedback was also
neutral in sign. Similarly, feedback discrepant from self-ratings could not have
a neutral sign (See Figure 6 for the experimental design).

Due to an error in subject assignments to experimental conditions, data

collection occurred in two stages. The ﬁrst group of subjects were randomly

 

assigned to one of six conditions. These were conditions 1, 3, 5, 7, 9, and 10

as depicted in Figure 6. Subjects in the second group were then randomly

63

assigned to the remaining four conditions. These were conditions 2, 4, 6, and 8

(See Figure 6).

To assess the similarity between subjects in the ﬁrst and second phases of
the study, comparisons were made between these two subject groups on
individual difference variables of self-esteem, need for achievement, need for
autonomy, need for afﬁliation, need for
dominance and locus of control. Subjects in these two groups did not
signiﬁcantly differ on any of these individual difference variables.

Task

The task used in this study was the in-basket exercise described above
(See Appendix A). An in-basket exercise represents a simulation in which
subjects are presented with a variety of in-basket items (memos, reports and
notes) which might be found in a typical manager’s in-basket. It has been
suggested that the skills subjects display in analyzing and evaluating the data is
representative of how they will resolve similar tasks on the job (Bray & Grant,
1966).

Procedure

Upon arrival at the experimental session, subjects were asked to read and

sign an informed consent declaration (See Appendix B for measures used in

this study). Subjects were read a preliminary script that introduced the task.

 

 

Figure 6. Experimental Design.

Feedback
Sign

High Credibility
Low Credibility

Feedback
Sign

Feedback Discrepancy Condition

 

 

 

 

 

 

 

High Low None
1 3
n=26 n=29
5
n=29
2 4
n=22 n=22
6 8
n=23 n=22
10
n=27
7 9
n=28 n=28

 

 

 

 

65

The in-basket evaluation procedure and an explanation of the performance
dimensions to be used for evaluation was then explained to subjects. Subjects
were given 45 minutes to complete the in-basket exercise. When time had
expired, they were asked to assess their own performance and to complete the
self-certainty and acceptance of self-generated feedback measures both of which
are described below. The self-assessment and self-certainty/self-generated
feedback acceptance questionnaire were collected and subjects were given the
credibility manipulation (See Appendix C) and asked to complete a series of ..
personality questionnaires while the rater ostensibly evaluated their
performance.

Subjects were given false feedback regarding their performance according
to the feedback discrepancy and feedback sign conditions to which they were
assigned. Subjects then completed a questionnaire assessing their reactions to
the feedback. This measure included the external feedback acceptance
perceived credibility of the feedback source and perceived feedback congruence
scales. Subjects were then told they would be completing a second in-basket
exercise. They were asked to again assess their own performance after careful

consideration of the previously completed self- and rater’s evaluation of

 

performance. Subjects were debriefed and thanked for their participation.

 

 

Variables and Measures

Provided below is a detailed description of the variables which were
assessed in this study. See Appendix B for the actual measures which were
used.

Feedback Sigp. Feedback sign was operationalized in a way that is
unique to this study. Traditional performance feedback studies have used a
goal as the referent against which the evaluator’s performance feedback is
compared. However in this study, the referent against which the evaluator’s
performance feedback is compared is self-feedback n_o_t a performance goal.
Sign of the feedback is deﬁned as whether the feedback provided by the
evaluator is above, below or equal to the referent of self-feedback. Feedback
sign was manipulated in this study resulting in three levels of feedback sign
(positive, negative, neutral). Positive feedback was above the referent, negative
feedback was below the referent and neutral feedback was exactly equal to the
referent of self-feedback.

Feedback Discrepang. Feedback discrepancy was deﬁned as the degree
of disagreement between one’s self-assessment and an evaluator’s external
assessment of performance. This variable was manipulated resulting in three
levels of feedback discrepancy (high, low, no). In the high feedback
discrepancy condition, the evaluator’s external assessment was 20 points above

or below the subject’s self-assessment. In the low feedback discrepancy

 

67

condition the evaluator-’s external assessment was 10 points above or below the
subject’s self-evaluation.

The point values for the high and low discrepancy conditions were based
on pilot study data. These values represented feedback that was approximately
two standard deviations away from the average self-assessment in the high

feedback discrepancy condition, or approximately one standard deviation away

1 1 £135.- ::0.r.qnu-ou ‘ p

from the average self-assessment in the low feedback discrepancy condition.

In the no feedback discrepancy condition the evaluator’s external I.
assessment exactly matched the subject’s self-assessment. This discrepancy
condition is confounded with the neutral feedback sign condition. When the
exialuator’s feedback matches self-feedback (no discrepancy), feedback sign
MUST be neutral (See Figure 6).

Perceived Feedback Congr_uence. The perceived degree of congruence
between self- and external assessments of performance, was assessed with a
self-report measure developed speciﬁcally for this study. This measure assessed
the effectiveness of the discrepancy manipulation. A sample item from this
scale is, ’The evaluation 1 received was similar to my self-evaluation’. Subjects
were asked to indicate, on a 1 - 7 scale, the extent to which they agreed with
the statements presented.

Source Credibility. Source credibility has been deﬁned as the rater’s

expertise in evaluating performance. This variable was manipulated resulting in

 

68

two levels of source credibility (high, low). In the high credibility condition,
subjects were told they were being evaluated by an industrial/organizational
psychology graduate student with 3 years of previous experience in evaluating
in-basket exercises. In the low credibility condition, subjects were told they
were being evaluated by an undergraduate student also participating in the
study for course credit, with no experience with in-basket exercise evaluation.
See Appendix C for the complete script of the source credibility manipulation.

Perceived Source Credibilig. The perceived source credibility was
assessed with a self-report measure. This measure was used to determine the
effectiveness of the credibility manipulation. A sample item from this scale is,
The rater who evaluated my performance was knowledgeable’. Subjects were
asked to indicate, on a 1 - 7 scale, the extent to which they agreed with the
statements presented.

Self-Rating. Subjects were asked to provide one overall rating of their in-
basket exercise performance. They were asked to recall the performance goal
of 70, prior to making their self-ratings. The form subjects used to evaluate
their performance appears in Appendix B.

External Rating. False feedback from an external rater was provided to
subjects. This feedback provided one overall rating of the subjects’ in-basket
exercise performance. The rating was provided on a form that was essentially

identical to the self-rating form (See Appendix B).

69

Consensus Performance Judgr_n_ent. Subjects were led to believe they
would be participating in second in-basket exercise. Prior to completing this
exercise they were asked to complete a second self-assessment measure. This
measure was identical to the ﬁrst except, subjects were asked to consider both
their self-evaluation of performance as well as the rater’s evaluation of their
performance in providing a second assessment of their previous performance.
The weight subjects assigned to either their self-rating or the external rating
was left to them.

@nemal Feedback Acceptance. External feedback acceptance is the
perceived accuracy of the feedback provided by an external source. This
variable was assessed with a modiﬁed version of the feedback acceptance
measure used by Stone, Guetal and McIntosh (1984). A sample item from this
scale is ’The feedback did not truly depict my performance on the sales task.’
Subjects were asked to indicate, on a 1 - 7 scale, the extent to which they
agreed with the statements presented.

Self-Feedback Acceptance. Acceptance of performance feedback
provided to oneself was assessed in a manner similar to acceptance of
externally generated feedback. The same measure was used but subjects were
asked to consider the feedback they provided for themselves when they rated
the perceived accuracy of the feedback. One item from the external feedback

acceptance scale had to be eliminated from the self-feedback acceptance scale

70

because when the item was converted from acceptance of another’s feedback to
acceptance of self-feedback it no longer made sense. Subjects were asked to
indicate, on a 1 - 7 scale, the extent to which they agreed with the statements
presented.

Self-Certainty. Self-certainty refers to the faith one has in his/her self-
evaluation of performance. This variable was assessed by a scale developed for

the purposes of this study. A sample item from this scale is, ’I am conﬁdent

 

about the self-evaluation I provided.’ Subjects were asked to indicate, on a 1 - I.
7 scale, the extent to which they agreed with the statements presented.
Self-Conﬁdence. The self-certainty and the self-feedback acceptance
séales were so highly correlated that a decision was made to combine the two
scales and call it self-conﬁdence. The basis for this combination is described in

detail in the results section.

 

 

CHAPTER 4: RESULTS

Prior to testing the proposed hypotheses, descriptive statistics were
calculated for the variables of interest in this study. The means, standard
deviations and reliabilities for the variables assessed in this study are provided
in Table 3.

First, these data reveal a mean self-rating of approximately 58 with a
standard deviation of approximately 15. This indicates that there was neither a
ﬂoor nor a ceiling effect for self-ratings of performance. Perceived rater
expertise was strongly correlated with manipulated rater expertise. Similarly,
perceived congruence between self- and external ratings of performance was
strongly negatively correlated with the manipulated discrepancy between these
two ratings.

Note the strong correlation between self-certainty and self-feedback
acceptance. Both self-certainty and self-feedback acceptance were measured
using a single questionnaire. Due to the strong correlation between these
scales, a factor analysis was completed for this questionnaire. The rotated
factor loadings indicated that the two factors extracted, with eigenvalues greater
than one, corresponded to one factor for positively worded items and one for
negatively worded items (See Appendix D). Upon examination of the items
from the two scales, it was determined that they were quite similar. An

examination of the intercorrelations between the items of these scales also

71

'72!

Table 3. Descriptive Statistics, Intercorrelations and Reliabilities of Measures.

!gge leer 50 1 2 3 1 5 6 1 8 9 10 ll 12

11 Feedback
Sign 1.00 0.88

21 leedbecl
lircrepercy 2.

H

6 0.16 -.006

31 Perceived
leedbtcl
Congruence 1.56 1.68 .153 -.185 (.921

11 Source
Credibility 1.50 0.50 .097 -.026 .036

51 Perceived

Source

Credibility 1.16 1.11 .261 -.011 .218 .581 1.8311
61 Sell-Rating 58.36 11.16 .032 -.050 .091 .103 .155

11 Irterrtl
Rating 58.32 20.51 .669 -.016 .168 .133 .211 .133

81 Correrrur irri.
lldxeent 59.81 16.51 .301 -.018 .153 .113 .227 .870 .811

91 External Feedback
Acceptance 1.50 1.38 .383 -.353 .506 .185 .116 .212 .118 .283 1.861

101 Sell-Feedback
lccepterce 1.53 1.21 -.005 -.013 .139 .015 .059 .362 .

V“.
‘8

' .351 .105 1.951
11) Self-Certeitty 1.11 1.15 .011 .000 .111 -.002 .092 .393 .321 .361 .299 .615 (.19;

12} Sell-
Coriidelce 1.50 1.09 .021 -.016 .151 .018 .086 .111 .335 .392 .371 .8’3 .919 1.8:?

 

38!]!98 in parentheses indicate tlplt reliebility COEIIICICDIS. Correlations (renter than .100 are
titlilicert it the p<.05 level (n=256).

73

indicated a signiﬁcant correspondence between these two scales. The scales
were therefore combined resulting in one scale representing self-conﬁdence in
self-ratings. The reliability coefﬁcient for this combined scale was .86 which
was higher than the reliability for either of the scales separately (See Table 3).

Self-ratings were positively correlated with self-certainty, self-feedback
acceptance and with the combined self-conﬁdence scale. This indicates that the
higher subjects self-ratings, the higher their faith or conﬁdence in those ratings.

Alpha reliability coefficients were calculated for scales used in this study
when possrhle. The coefﬁcients for the scales measured were well within
acceptable ranges (.75 - .92).
Manipulation Check_s

Descriptive statistics were analyzed to determine the effectiveness of the
source credibility and feedback discrepancy manipulations. First, the correlation
between source credibility and perceived source credibility was positive and
signiﬁcant (r=.587). This correlation indicates that high credibility sources of
feedback were perceived as signiﬁcantly more expert than low credibility
sources of feedback. Note, however that subjects in the low credibility
condition still perceived their feedback source as somewhat credible. The
mean perceived credibility rating for the low credibility group was 4.08 on a 7-

point scale (See Table 4).

 

 

74

Table 4. Means and standard deviations for manipulated variables.

 

Dependent Variable =

Perceived
Source
Credibility

Source Credibility

(High Cred)
Mean SD
5.44 .94

(Low Cred)
Mean S_Q
4.08 .95

 

Dependent Variable=

Perceived
Feedback
Congruence

Feedback Discrepancy

(High Disc)

Mean

3.22

_S__I2

1.09

(Low Disc) (No Disc)

Mean

4.73

S12

1.18

M

6.74

_S__12

.54

. -. c L...“

 

75

A one-way ANOVA was also computed to determine the effectiveness of
the feedback discrepancy manipulation. Manipulated feedback discrepancy
(high, low, no) served as the independent variable and perceived congruence
between self- and external feedback served as the dependent variable. This
analysis revealed statistically signiﬁcant differences in subject’s perceptions of
congruence between self- and external performance feedback F(2,252)=206.38,
p<.001 dependent upon the discrepancy condition to which they had been
assigned. A post hoc Scheffe test indicated that all three feedback discrepancy

conditions were signiﬁcantly different from each other (See Table 4).

Hypotheses
thgpothesis 1

The ﬁrst hypothesis posited a positive relationship between feedback sign
and external feedback acceptance. A zero-order correlation coefﬁcient was
calculated between feedback sign and external feedback acceptance to test this
hypothesis. In support of the hypothesis, the correlation coefﬁcient between
feedback sign and external feedback acceptance was positive and statistically
signiﬁcant (r = .383, p<.001).

To determine the mean differences which occurred in external feedback
acceptance at positive, negative and neutral sign levels, a post-hoc Scheffe test
was conducted. The result of this analysis indicated, as expected, positive

feedback was more accepted than negative feedback and neutral feedback

76

(feedback identical to self-ratings) was more accepted than negative feedback.
A signiﬁcant difference in external feedback acceptance was not found between
positive and neutral feedback. Descriptive statistics for this analysis appear in
Table 5 and the relationship between feedback sign and external feedback
acceptance is portrayed in Figure 7.
Hypothesis 2

The second hypothesis suggested a negative relationship would exist
between feedback discrepancy and external feedback acceptance. The zero
order correlation coefﬁcient between feedback discrepancy and external
feedback acceptance was negative and signiﬁcant (r = -.353, p<.001). A
partial correlation coefﬁcient was also calculated between feedback discrepancy
and external feedback acceptance, controlling for the effects of feedback sign.
The partial correlation coefﬁcient between feedback discrepancy and external
feedback acceptance partialling out feedback sign was r = -.383 (p<.001).
This partial correlation indicates that feedback discrepancy had a unique
inﬂuence on external feedback acceptance, over and above the inﬂuence of
feedback sign.

A post-hoe Scheffe test was completed to determine mean differences in
external feedback acceptance at the three feedback discrepancy levels (none,
low, high). This test indicated that external feedback acceptance signiﬁcantly

differed across all three feedback discrepancy conditions. That is,

 

77

Table 5. Means, standard deviations and cell sizes for external feedback
acceptance by feedback sign, feedback discrepancy and source credrhility.

 

External Feedback Acceptance

 

Independent

Variable Mean 82 N

External

Feedback

Sign .
Positive 4.87 1.10 100
Neutral 5.32 1.37 56
Negative 3.68 1.22 100

External

Feedback

Discrepancy
High 4.01 1.40 99
Low 4.54 1.14 101
None 5.32 1.37 56

Source

Credibility
High 4.76 1.37 128
Low 4.25 1.35 128

78

Figure 7. Plot of relationship between feedback sign and external
feedback acceptance.

. Q&A.-‘_l“h‘l'g-m
. .
. . A

External
Feedback
Acceptance

 

Negative Neutral Positive

Feedback Sign

 

 

79

non-discrepant external feedback was signiﬁcantly more accepted than feedback
which was slightly discrepant from self-feedback, which was in turn signiﬁcantly

more accepted than highly discrepant feedback (See Table 5). The relationship
between feedback discrepancy and external feedback acceptance is illustrated in
Figure 8.

Eco—“1&3

Hypothesis three stated that feedback discrepancy would moderate the
relationship between feedback sign and external feedback acceptance. To test
this hypothesis, moderated regression analysis was used. The analysis involved
regressing external feedback acceptance on a) feedback sign, b) feedback
discrepancy and c) an interaction of feedback sign x feedback discrepancy.
Feedback sign was entered ﬁrst followed by feedback discrepancy. The cross-
product of these two terms was then entered.

Table 6 shows the standardized regression coefﬁcients and signiﬁcance
levels for each of the independent variables in the equation. The results
revealed an overall R-squared for external feedback acceptance of .277,
F(3,252) = 32.18, p<.001. Hypothesis three stated that feedback discrepancy
would moderate the relationship between feedback sign and external feedback
acceptance. However, as can be seen in Table 6, the regression coefficient
representing the interaction of feedback sign and feedback discrepancy was not

significant. Post-hoc analyses conﬁrmed these results. The plot of the

 

80

Figure 8. Plot of the relationship between feedback discrepancy and
external feedback acceptance.

External
Feedback
Acceptance

 

Negative Neutral Positive

Feedback Sign

81

Table 6. Regression of External Feedback Acceptance on Feedback Sign and
Feedback Discrepancy.

 

 

Independent Beta Multiple R-Square R-Square Signiﬁcance

Variable R Change of Change 1*
Feedback Sign .3829 .3829 .1466 .1466 .000 3.
Feedback Discrepancy .3511 .5195 .2699 .1233 .000 9
Feedback Sign x

Feedback Discrepancy .4513 .5263 .2770 .0071 .117

 

N=256

 

 

82

relationship between feedback sign, feedback discrepancy and external feedback
acceptance appears in Figure 9.
Hypothesis 4

Hypothesis four suggested that the relationship between the comparator
(feedback discrepancy and feedback sign) and external feedback acceptance
would be moderated by source credibility. Moderated regression analysis was
used in which feedback sign, feedback discrepancy and source credibility were
entered into the regression equation. The cross products of feedback sign x
feedback discrepancy, feedback sign x source credibility, feedback discrepancy x
sourceicredibility and feedback sign x feedback discrepancy x source credibility
were then entered respectively.

Results revealed an overall R-squared for external feedback acceptance of
.301, F(7, 248) = 15.29 p< .001. No signiﬁcant interactions were found. Main
effects for the independent variables of feedback sign, feedback discrepancy
and source credibility were found. Table 7 shows the standardized regression
coefﬁcients and signiﬁcance levels of each of the independent variables in the
equaﬁon.

These ﬁndings suggest that in addition to feedback sign and feedback
discrepancy, source credibility made a signiﬁcant independent contribution to
the prediction of external feedback acceptance. The main effect for source

credibility indicates that high credibility sources were associated with higher

 

83

Figure 9. Plot of the relationship between feedback sign, feedback discrepancy
and external feedback acceptance.‘

 

External
Feedback 4
Acceptance

 

Low High

Feedback Discrepancy

"When feedback sign and feedback discrepancy are crossed, subjects in the no
discrepancy/no feedback conditions are eliminated. This is because sign and
discrepancy are completely confounded in these conditions.

 

84

Table 7. Regression of External Feedback Acceptance on Feedback Sign,
Source Credibility and Feedback Discrepancy.

 

 

Independent Beta Multiple R-Square R-Square Signiﬁcance
Variable R Change of Change
Feedback Sign .3829 .3829 .1466 .1466 .000
Feedback Discrepancy .3511 .5195 .2699 .1233 .000
Source Credibility .1401 .5379 .2893 .0194 .009

Sign x Discrepancy .4640 .5448 .2968 .0075 .103

Sign x Credibility .0727 .5452 .2972 .0004 .693
Discrepancy x

Credibility .1083 .5458 .2979 .0007 .630

Sign x Discrepancy x

Credibility 1.0749 .5490 .3014 .0035 .264

 

N=256

1

85

levels of external feedback acceptance than low credibility sources. Plots of the
relationship between feedback sign, feedback discrepancy, source credibility and
external feedback acceptance appear in Figure 10.

Exam;

Hypothesis ﬁve suggested that the relationship between the comparator
and self-feedback acceptance would be moderated by self-certainty. The self-
certainty and self-feedback acceptance scales were combined, therefore this
hypothesis was not tested. However, a test of the moderating effect of self-
conﬁdence on the relationship between the comparator and external feedback
acceptance was conducted. Moderated regression analyses were used in which
feedback sign, feedback discrepancy, self-conﬁdence and their interactions were
regressed on external feedback acceptance.

This regression analysis revealed signiﬁcant main effects for each of the
independent variables of feedback sign, feedback discrepancy and self-
conﬁdence. The main effects for sign and discrepancy have already been
discussed. The main effect for self-conﬁdence suggests a positive relationship
between self-conﬁdence and external feedback acceptance such that when self-
conﬁdence is high, external feedback acceptance is higher than when self-
conﬁdence is low. In partial support for Hypothesis 5, a signiﬁcant feedback
discrepancy x self-conﬁdence interaction on external feedback acceptance was

found. Regression coefﬁcients for this analysis appear in Table 8.

 

86

Figure 10. Plots of the relationship between feedback sign, feedback
discrepancy, source credibility and external feedback acceptance.

Low Credibility

 

 

 

 

 

7
Positive
6
5
External 4
Feedback
Acceptance3
2
Negative
1 Sign
Low High
Feedback Discrepancy
High Credibility
7 Positive
Sign
6
X=5.00 X=5.06
5 (n :29) (n—‘26)
External 4 X=4.33
Feedback (n=22) =3.52
Acceptance3 (n=22)
2 Negative
Sign
1

 

Low High
Feedback Discrepancy

87

Table 8. Regression of External Feedback Acceptance on Feedback Sign,
Feedback Discrepancy and Self-Conﬁdence.

 

 

Independent Beta Multiple R-Square R-Square Signiﬁcance
Variable R Change of Change
Feedback Sign .3829 .3829 .1466 .1466 .000
Feedback Discrepancy .3511 .5195 .2699 .1233 .000
Self-Conﬁdence .3549 .6290 .3957 .1258 .000

Sign x Discrepancy .2857 .6313 .3985 .0028 .279

Sign x Self-

Conﬁdence .3503 .6362 .4047 .0062 .108
Discrepancy x

Self-Conﬁdence .7590 .6522 .4253 .0206 .003

Sign x Discrepancy x

Credibility 1.8094 .6562 .4306 .0053 .129

 

N=256

 

 

88

Post-hoc analyses were completed to determine mean differences which
occurred in external feedback acceptance when feedback discrepancy was
crossed with self-conﬁdence. These analyses revealed mean differences
between high and no, and low and no feedback discrepancy groups when self-
conﬁdence was high. However, no mean differences occurred in external
feedback acceptance when self-conﬁdence was low. Figure 11 portrays the
interaction of self-conﬁdence and feedback discrepancy on external feedback
acceptance.

Hypothesis 6

Hypothesis six stated that an interaction would exist between self-
feedback and self-feedback acceptance on consensus performance judgment.
Since self-feedback acceptance and self-certainty were combined into a self-
conﬁdence scale, self-conﬁdence was used as the second independent variable.
To test the hypothesis that self-feedback and self-conﬁdence would interact to
influence consensus performance judgment, moderated regression analysis was
used. Self-feedback and self-conﬁdence were entered, followed by their cross
product with consensus performance judgment serving as the criterion.

A main effect for self-rating on consensus performance judgment was
found. However, neither a main effect for self-conﬁdence nor an interaction of
self-rating and self-conﬁdence was revealed. The beta weight for self-rating was

positive, indicating a positive relationship between an individual’s self-rating and

89

Figure 11. Plot of the interaction of self-conﬁdence and feedback
discrepancy on external feedback acceptance.

   

 

7
6
5
External
Feedback 4 High Self-
Conﬁdence*
A
3 X=3.25
(n=13)
Low Self-
Confidence

 

N 0 Low High
Discrepancy Discrepancy Discrepancy

Feedback Discrepancy

*Low self-conﬁdence is deﬁned as one standard deviation below the mean self-
conﬁdence rating. High self-conﬁdence is deﬁned as one standard deviation
above the mean self-conﬁdence rating.

90

the consensus performance judgment. That is, when self-rating was high
consensus performance judgment tended to be high. The beta weights for self-
conﬁdence and the cross product term representing the interaction between
self-rating and self-conﬁdence were not signiﬁcant (See Table 9). Figure 12
provides a representation of the relationship between self-rating, self-conﬁdence
and consensus performance judgment.
Hypothesis 7

Hypothesis seven suggested that external feedback and externally
generated feedback acceptance would interact to affect consensus performance
judgment. To test this hypothesis, moderated regression analysis was used.
External feedback and externally generated feedback acceptance were entered
with consensus performance judgment serving as the criterion. In the third step
the cross product of external feedback and externally generated feedback
acceptance was entered.

The data from the present study support the prediction. A
signiﬁcant main effect for external feedback on consensus performance
judgment was revealed. The beta weight for the external rating was positive
indicating a positive relationship between external rating and consensus
performance judgment. That is, when external rating was high, consensus

performance judgment tended to be high.

91

Table 9. Regression of Consensus Performance Judgment on Self-
Rating and Self-Conﬁdence.

 

 

 

Independent Beta Multiple R-Square R-Square Signiﬁcance
Variable R Change of Change
1‘
Self-Rating .8703 .8703 .7575 .7575 .000 .f
Self-Conﬁdence .0385 .8710 .7587 .0012 .258
Self-Conﬁdence X
Self-Rating .0064 .8710 .7587 .0000 .976
N=256

 

 

 

92

Figure 12. Plot of the relationship between self-rating, self-conﬁdence and
consensus performance judgment.

Consensus
Performance
Judgment

   
 
 
 
 
  

100

High"

Self-
90 Conﬁdence
80 Low

Self-

Conﬁdence

70

50

40

30

20

10

Low High

Self-Performance Rating"

" High self-conﬁdence is deﬁned as one standard deviation above the mean
self-conﬁdence rating. Low self-conﬁdence is deﬁned as one standard deviation
below the mean self-conﬁdence rating. "High self-rating is deﬁned as one
standard deviation above the mean self-rating. Low self-rating is deﬁned as
one standard deviation below the mean self-rating.

93

In addition, a signiﬁcant main effect was found for external feedback
acceptance on consensus performance judgment. The beta weight for the
external feedback acceptance was positive indicating a positive relationship
between external feedback acceptance and consensus performance judgment.
That is, when external feedback acceptance was high, consensus performance
judgment tended to be high. Finally, a signiﬁcant interaction was found
between external rating and external feedback acceptance on consensus
performance judgment. Table 10 presents the regression coefﬁcients for this
analysis.

Post-hoe comparisons revealed significant differences occurred in
censensus performance judgment between high and low external ratings, when
external feedback acceptance was high but not when it was low. Figure 13
displays the interaction between external feedback and external feedback
acceptance on consensus performance judgment.

Additional Resultjs

Exploratory analyses were conducted to determine the relative inﬂuence
of self-feedback and external feedback in predicting consensus performance
judgment. A hierarchical regression analysis was completed to determine the
results. Self-feedback and external feedback were allowed to enter the
regression equation based on which variable contributed the most variance to

the prediction of consensus performance judgment. Results revealed that self-

94

Table 10. Regression of Consensus Performance Judgment on External
Rating and External Feedback Acceptance.

 

 

Independent Beta Multiple R-Square R-Square Signiﬁcance
Variable R Change of Change
External Feedback .8413 .8413 .7077 .7077 .000
External Feedback

Acceptance .0827 .8446 .7134 .0056 .027

External Feedback x
External Feedback
Acceptance .6669 .8548 .7307 .0173 .000

 

N=256

 

95

Figure 13. Plot of the interaction between external feedback and external

feedback acceptance on consensus performance judgment.

   
  
  
 
 
  

Consensus
Performance 50
Judgment

30

20

10

Low High
External Performance Rating"

High“
External
Feedback

Acceptance

Low
External
Feedback
Acceptance

"‘ High external feedback acceptance is deﬁned as one standard deviation above
the mean external feedback acceptance rating. Low external feedback
acceptance is deﬁned as one standard deviation below the mean external

feedback acceptance rating.

"High external rating is deﬁned as one standard deviation above the mean
external rating. Low external rating is deﬁned as one standard deviation below

the mean external rating.

96

feedback entered the equation ﬁrst, accounting for more than 75% of the
variance in consensus performance judgment. However, external feedback did
account for a signiﬁcant amount of additional variance in consensus

performance judgment. (See Table 11).

 

 

 

97

Table 11. Regression of Consensus Performance Judgment on Self- and
External Ratings of Performance.

 

 

 

Independent Beta Multiple R-Square R-Square Signiﬁcance
Variable R Change of Change
Self-Rating .8703 .8703 .758 .758 .000
External Feedback .4396 .9203 .847 .090 .000
N=256

 

CHAPTER 5: DISCUSSION

The purpose of this study was to investigate the inﬂuence of discrepancies
between self-assessments of performance and other assessments of performance
on performance feedback acceptance. A model of self-other feedback
congruence was proposed within a control theory framework. Hypotheses
derived from this model were tested and partial support for the model was
found.

The model of self-other feedback congruence suggested that feedback
from an external source is compared to self-feedback on the dimensions of
feedback sign and feedback discrepancy. The results of the present study
revealed that both the direction of the feedback discrepancy (sign) and the
magnitude of the discrepancy inﬂuenced acceptance of feedback provided by an
external source. The ﬁrst two hypotheses that feedback discrepancy would be
negatively related to external feedback acceptance, and feedback sign would be
positively related to external feedback acceptance, were supported. The third
hypothesis that feedback discrepancy and feedback sign would interact to
inﬂuence external feedback acceptance was not supported.

The model proposed that source credibility would play an important role
in the decision to accept or reject feedback provided by an external source.
Further, source credibility was hypothesized to interact with feedback sign and

feedback discrepancy to inﬂuence external feedback acceptance. A direct

98

99

relationship was revealed between perceived credibility of the feedback source
and acceptance of externally generated feedback. However, source credibility
did not moderate either the relationship between feedback sign and external

feedback acceptance or feedback discrepancy and external feedback acceptance.

The model further suggested that self-certainty would inﬂuence the
decision to accept or discount feedback provided to oneself. The hypotheses
regarding the inﬂuence of self-certainty and that of self-generated feedback
acceptance were modiﬁed. This change was necessary due to the strong
relationship between these two variables. In retrospect, the constructs of self-
certainty and self-feedback acceptance both refer to faith or conﬁdence in one’s
own assessment of performance. These variables were therefore combined into
one scale called self-conﬁdence.

The ﬁrst modiﬁed hypothesis proposed that the relationship between the
comparator (feedback sign, feedback discrepancy) and external feedback
acceptance would be moderated by self-conﬁdence. Findings demonstrated that
in partial support of this hypothesis, self-conﬁdence modiﬁed the relationship
between feedback discrepancy and external feedback acceptance. Individuals
high in self-conﬁdence accepted external feedback which was not discrepant
from their self-view signiﬁcantly more than feedback which was highly

discrepant. However, no signiﬁcant mean differences in external feedback

 

 

 

100

acceptance occurred for those who were less conﬁdent in their self-ratings at
any level of discrepancy.

This ﬁnding suggests that when individuals have relatively little conﬁdence
in self-ratings, they may view themselves as low credIhility sources of feedback.
When external sources of feedback agree with individual self-assessments,
external sources of feedback are perceived as less credible. Therefore,
externally-generated feedback is less accepted. In essence, the comparison
between self-feedback and external feedback is less meaningful for individuals
who lack conﬁdence in their self-assessments.

The ﬁnal group of hypotheses derived from the model of self-other
feedback congruence relate to the ﬁnal judgment individuals make regarding
their performance. The model suggested that the relationship between self-
feedback and consensus performance judgment would be moderated by self-
conﬁdence. This hypothesis was not supported. Self-feedback was only found
to have a direct inﬂuence on consensus performance judgment. A parallel
hypothesis indicating that external feedback acceptance would moderate the
relationship between external feedback and consensus performance judgment
was supported.

The model suggested that self- and external feedback would both
contribute to the consensus judgment made regarding performance. However,

no speciﬁc hypotheses regarding the relative weight given to these factors could

 

101

be gleaned from the model. Therefore, exploratory analyses were conducted to
determine the relative inﬂuence self-generated feedback and externally
generated feedback had on consensus performance judgment. Results revealed
that self-feedback contributed to the variance in consensus performance
judgment more than externally generated feedback. However, external
feedback made a signiﬁcant contribution to the variance in consensus
performance judgment above the inﬂuence of self-feedback.

The ﬁndings of this study are discussed in more detail below. First,
results relevant to the ﬁrst part of the model identifying determinants of
feedback acceptance are elaborated. Results related to the second part of the
model identifying determinants of consensus performance judgments are then
discussed. A modiﬁed model of self-other feedback congruence is presented.
Implications of this investigation for application in organizations as well as for
future research are detailed. Finally, limitations to this inquiry are outlined.
Determinants of Feedback Acceptance

llgen et al. (1979) describe the concept of feedback sign as feedback
indicating that performance was either above or below some standard. A
unique contribution of this study to the performance feedback literature is it’s
atypical operationalization of feedback sign. What makes the deﬁnition of
feedback sign distinct is that the standard of comparison in this study was one’s

own self-assessment of performance M an externally set goal. So while the

102

ﬁnding that positive feedback leads to higher levels of external feedback
acceptance is consistent with previous literature (e.g. DeNisi, et al., 1983; Stone
& Stone, 1984; Stone & Stone, 1985), it also provides additional information
that previous studies do not. The present results suggest that no matter if the
standard of comparison is deﬁned as a goal or as one’s self-assessment of
performance, sign is positively related to external feedback acceptance. One
might suggest that feedback above the standard of self-feedback is generally
positive and therefore more accepted than feedback below self-feedback.
However, the average self-rating in this study was 58.36 when the assigned goal
was 70. So, feedback which was slightly discrepant (10 points above self-rating)
and in the positive direction, could still be considered negative in the traditional
sense. This ﬁnding adds to the generalizability of the proposal that feedback
indicating performance is above a standard leads to higher levels of feedback
acceptance than feedback indicating one’s performance is below standard.

The ﬁndings of this study are also consistent with previous work which
found that the credibility of the feedback source directly inﬂuenced reactions to
performance feedback (Bannister, 1986; Halperin et a], 1976; Stone et. a1, 1984,
Podaskoff & Fahr, in press). The results show that when feedback sources
were perceived as more credible, greater acceptance of the feedback provided
by these sources occurred. An interesting yet unexpected ﬁnding was that

source credibility did not interact with feedback sign to inﬂuence acceptance of

103

externally generated feedback. llgen et al. (1979) have suggested that negative
feedback from a credible source would be more accepted than negative
feedback from a less credible source. Our ﬁndings do not support this
proposal however, this could have partly been due to the manipulation of
credibility in this study. While high credibility sources were perceived as
signiﬁcantly more credible than low credibility sources of feedback, both
feedback sources were viewed as credible. That is, the average perceived
credibility rating was above the midpoint of the scale for both the high and low
credibility groups.

In addition, restriction of range occurred on the source credibility scale
which was used. Perceived source credibility was assessed using a 7-point scale.
The standard deviation of perceived source credibility at both high and low
credibility was just less than one. This may have limited the possrhility of
ﬁnding signiﬁcant relationships. These ﬁndings regarding source credibility ar_e
consistent with that of Podaskoff and Fahr (in press) who found no sign x
credibility effect on feedback acceptance but did ﬁnd a sign x credibility effect
on performance.

This study also investigated the inﬂuence of the magnitude of self-other
feedback discrepancies on acceptance of externally generated feedback. The
results clearly demonstrate that the extent to which performance feedback from

an external source is incongruent with self-generated feedback regarding

104

performance inﬂuences acceptance of external feedback. The negative
relationship between feedback discrepancy and external feedback acceptance
supports the prediction that more discrepant external feedback is less likely to
be accepted than feedback similar to self-feedback.

This ﬁnding supports the view that individuals prefer self-relevant
information which is consistent with self-assessments (Korman, 1970; Shrauger,
1975). Even after the inﬂuence of feedback sign was partialled out, the
negative relationship between feedback discrepancy and acceptance of external
feedback remained strong. These results appear to counter the self-
enhancement view of work motivation (Baumeister, 1982) which suggests that
individuals prefer to see themselves in the most favorable light. However,
these ﬁndings may in part be due to the unique operationalization of feedback
sign used in this study. Even when sign was negative, it may not have
threatened self-esteem, therefore the results do not necessarily contradict the
self-enhancement view.

Inﬂuences on Consensus Performance Judgment

The relationship between feedback from an external source and one’s
judgment about performance was moderated by the acceptance of externally
generated feedback. This ﬁnding directly addresses the signiﬁcance of feedback
acceptance in the performance feedback process. In support of propositions by

llgen et a1. (1979), feedback acceptance played an integral role in the

 

105

relationship between externally generated feedback and one’s ﬁnal assessment
of the adequacy of performance. It seems likely that feedback acceptance
would also be important to an individual’s motivation to change his/her
behavior based on the feedback that was provided.

Feedback provided to oneself about the adequacy of performance was
also found to signiﬁcantly inﬂuence one’s overall evaluation of performance.
This ﬁnding is consistent with work by Podaskoff and Fahr (in press) which
suggests that self-processes are important to the performance feedback process.
In addition, this ﬁnding supports previous work by Greller and his associates
(Greller, 1980; Greller & Herold, 1975; Herold et al., 1987) indicating that
feedback provided to oneself is perceived to be the most available and useful
source of feedback. This proposition is further supported by the ﬁnding in the
present research that self-feedback contributed more to one’s overall
assessment of performance than external feedback did. These results are
consistent with recent work by Barley (1988) which revealed that feedback from
psychologically closer sources was more acceptable and was more likely to
inﬂuence later performance than feedback from psychologically distant sources.

In summary, the results of this study provide some support for the model
of self-other feedback congruence. The results suggest that self-assessments are
an important component of the performance feedback process. Self-evaluations

serve as a standard against which feedback from external sources is compared.

106

These comparisons were found to inﬂuence the feedback recipient’s
determination of feedback acceptance.

One unexpected yet interesting ﬁnding was that self-feedback regarding
the adequacy of performance inﬂuenced the overall judgment of performance
made, regardless of self-conﬁdence in those ratings. This ﬁnding supports the
view that self-feedback represents a potent standard that is important to
individuals. On the other hand, the ﬁndings indicate that feedback from an
external source must be accepted before it is incorporated into one’s overall
judgment of performance. Therefore, conﬁdence or faith in the ratings
provided appears to be an important factor when the feedback source is
external but not when it is internal.

Finally, self-generated feedback was shown to inﬂuence consensus
performance judgment more than external feedback. This outcome provided
further support that while external feedback inﬂuenced ﬁnal assessments, self-
evaluation plays an essential role in the performance feedback process.

The model of self-other feedback congruence proposed in this study has
been revised (See Figure 14). The changes that have been made in this model
do not completely reﬂect the results of this study. Methodological difficulties
occurred in the measurement of a few variables, therefore, it was not
considered appropriate to modify the model to reﬂect the results relevant to

these variables. Revisions to the self-other model of feedback congruence are

107

Figure 14. Modified Model of Self-Other Feedback Congruence.

ownenu . owaesu uow>unom
3328

 

 

 

~33 Suﬁsm

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

   

 

 

 

  

 

 

 

 

       

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

scented“
ouoeehouuom
quzzomw>zu voueuoeoc
uoaauauQEIIII- .IIIIIIIIIIIII .IIIIIIII IIILuummmwmwwmlelIlll I
omoenu
ou nowuuuuou -
oz A<DQH>HQZH —
oz
2; 1838.832 _i- 53933.6 H"
. ouuaom
H on» ~hucnnouunwa
xuepvoom
Aneuawvsn Hocuouxm -
.ooeoou\«eouv \udom —1uoncom
nouuuuoaoo xuonvooh -
anaconda
\uuouoauoucu w
EV a «m
awﬁuwﬁ 3888....
sequence: N uoueuuoeou
oouuquun ucoamuan ooueunuuu< xuoaeoou I111 avenues»
oooeauouuom sooaouu unsounan euoeauouuom
noeuoooou voyeuocoo \ounuoououou headcouuoan voucuocuu
anouueuxu enwoeeuux suenvoom Naomi
63368 .— uououooloo noun—owed
r A.
. shuuonouuuwa
_ sueavoom
owcegu no» Hecuuuxm
gauuenou .. nun-nuance \uaom
«new I11
_ oz
2.... - I“-..
aeou I. u cued

 

 

 

 

owoeauaeoo I

108
discussed in detail below.

First, feedback sign and feedback discrepancy should be regarded as two
distinct ways of comparing self- and other-generated feedback. They did not
interact in inﬂuencing external feedback acceptance, but each had a unique
effect on this dependent variable. In control theory terms, each can be thought
of as a comparator, or a dimension of comparison between self- and external
feedback. Feedback discrepancy and feedback sign independently inﬂuenced
the decision to accept or discount feedback from an external source. As such,
each uniquely inﬂuenced external feedback acceptance. These results are most
likely due to the unique conceptualization of feedback sign. Since sign did not
necessarily reﬂect an evaluation of being above or below standard, it did not
interact with feedback discrepancy as predicted. Future research should
investigate the inﬂuence of traditional feedback sign and feedback discrepancy
on feedback acceptance.

Second, neither self-conﬁdence nor source credibility interacted with 9913
feedback sign and feedback discrepancy to inﬂuence external feedback
acceptance. Source credibility did not interacted with either of these variables
and self-conﬁdence only interacted with feedback discrepancy. Restriction of
range occurred for both of these variables and the self-conﬁdence variable was
developed post-hoc. Therefore, the model has not been modiﬁed to directly

reﬂect these results.

109

A change has been made to indicate the direct inﬂuence of self-
conﬁdence and source credibility on external feedback acceptance. The
moderating effects of self-conﬁdence and source credrhility on the relationship
between feedback discrepancy and external feedback acceptance and feedback
sign and external feedback acceptance remain in the model. The problems
which occurred in this study, in measuring source credibility and self-conﬁdence
may have signiﬁcantly inﬂuenced the results. Changes made in the model
related to these variables should therefore be considered tentative, and in need
of replication.

Note that the variable of self-conﬁdence replaced self-certainty in the
revised model. In retrospect, self-conﬁdence appears to reﬂect the faith one
has in self-ratings and in one’s own ability to rate performance. Self-conﬁdence
can be thought of as a variable that parallels source credibility. Source
credibility indicates the feedback recipient’s faith in and perceived ability of, the
external source of feedback. Therefore, self-conﬁdence should moderate the
relationship between self-feedback and feedback acceptance just as source
credibility should moderate the relationship between external feedback and
feedback acceptance.

In an attempt to better understand the inﬂuence of self— conﬁdence and
source credibility in the model, post-hoe hypotheses were proposed regarding

these variables. First, the possibility that self-conﬁdence moderated the

110

relationship between self-feedback and consensus performance judgment was
tested. This hypothesis suggested that self-feedback would inﬂuence the ﬁnal
judgment made regarding performance more for those high in self-conﬁdence,
than for those low in self-conﬁdence. Similarly, the possibility that source
credibility moderated the relationship between external feedback and consensus
performance judgment was tested. This hypothesis suggested that the external
feedback would inﬂuence the ﬁnal judgment made regarding performance more
when source credibility was high, than when source credibility was low. No
support was found for either of these suppositions. Self-feedback was found to
directly inﬂuence consensus performance judgment and as hypothesized,
external feedback was moderated by external feedback acceptance in
inﬂuencing consensus performance judgment.

The ﬁnal change made in the model was the elimination of the variable
of self-feedback acceptance. If one has faith in the performance ratings he/she
has provided, as reﬂected by self-conﬁdence, self feedback acceptance becomes
a moot issue. Therefore, this variable does not appear in the revised model of
self-other feedback congruence.

Research Implications

Results of the present study demonstrate the importance of feedback

discrepancy for acceptance of external feedback. Note however that feedback

acceptance is not an end in and of itself. According to llgen et al. (1979)

111

feedback acceptance acts as a mediator in the relationship between the
provision of feedback and the intention to act based on the feedback. The
present research can be extended by investigating the inﬂuence of feedback
discrepancy on the motivation to act based on the feedback.

In addition, the relationship between feedback discrepancy and
performance deserves further attention. Barley (1988) has found that feedback
acceptance is related to performance improvements and has demonstrated the
importance of self-generated feedback. No research has yet looked at the
effects of feedback discrepancy on performance. Due to the inﬂuence of
feedback discrepancy on feedback acceptance and the importance placed on
self-feedback, feedback discrepancy should be negatively related to motivation
to improve and to actual performance improvements.

Feedback discrepancy was manipulated in the present research. Future
research should be conducted in real work settings to determine the extent to
which discrepancies actually occur between self-feedback and feedback
generated by an external source. A review by Fisher and Rusk (1987),
proposed that self-superior disagreement occur relatively frequently. They also
reviewed literature which provide explanations for these disagreements. Future
research should utilize these theories in an attempt to identify methods of
reducing discrepancies between self- and superior assessments of performance.

Such research should avoid the assumption that feedback provided by a

112

superior is "correct". Both self- and superior assessments should be compared
to more objective sources of feedback (e.g. task feedback) to determine their
respective validities.

Previous research related to self-assessments in organizational settings has
focused on outcome issues such as the validity or accuracy of self-assessments.
Some researchers report that individuals are not skilled at provided self-
assessments which are accurate (e.g. Levine, Flory & Ash, 1977) while other
researchers suggest individuals have limited capabilities for accurate self-
assessments (Mabe & West, 1982; Shrauger & Osberg, 1981). Results of the
present research indicate that self-assessments are key components to overall
evaluations people make regarding their performance. Results also reveal that
differences between self- and other assessments inﬂuence acceptance of
external feedback. Therefore, future research is needed which investigates
methods of making self-assessments more accurate.

One way of enhancing the accuracy of self-assessments is by ensuring a
similar frame of reference with regard to the standards of performance. Taylor
et a1. (1984) suggest that superiors and subordinates should have a clear
understanding of the criteria for effective performance. They should also agree
that these criteria are valid. This can be accomplished through encouraging
participative communication between the two parties (O’Reilly & Anderson,

1980). In addition, feedback should be provided by a credible source. Taylor

113

et al. (1984) further recommend frequent feedback sessions which would
increase the perceived accuracy of performance feedback or feedback
acceptance.

This study supports the view that individuals are more willing to accept
feedback which is consistent with their self-view than feedback which is
inconsistent with self-perceptions (Shrauger, 1975; Swarm & Read, 1981, Swarm
& Hill, 1982). From an organizational perspective, these ﬁndings imply that
any incongruity between the feedback provider’s and feedback recipient’s
perception of the adequacy of performance would be detrimental to acceptance
of external feedback and should be reduced.

Finally, the results of this study have important implications for the
performance feedback process which takes place in organizations. First, it
suggests that self-processes, speciﬁcally self-feedback, inﬂuence individual
reactions to performance feedback provided by others. This ﬁnding is
consistent with models of self-regulation.

The integration of the control theory and self-management perspectives
suggests that individuals attempt to regulate behavior in relation to a standard
based on feedback from multiple sources (e.g. self, other). The control
theoretic perspective suggests that discrepancies between perceived
environmental feedback and standards (in this case self-feedback) are regulated

(Powers, 1973). Previous studies utilizing control theory as a framework have

114
not addressed the possibility that individuals provide self-feedback, however
control theory can be applied in these situations.

Self-management theorists recognize the fact that individuals observe and
evaluate their own behavior (Bandura, 1977, 1978; Kanfer & Karoly, 1982).
They further believe that individual attempt to regulate their behavior in
relation to a standard. Our ﬁndings indicate that selﬂevaluations of
performance strongly inﬂuence judgments made regarding the adequacy of
performance. These results suggest that self-feedback can serve as a standard
against which feedback from other sources in compared.

Limitations

‘ The limitations of this study deserve mention. First, due to an error in
assignment of participants to experimental conditions, the design was not
completely randomized. Data was collected in six out of ten conditions, after
which, data was collected in the remaining four conditions. An analysis of
individual differences between the two groups was completed and revealed no
significant differences between them. Any differences which may have occurred
in experimenter behavior from early to late in the study were indeterminable.
Participants who were assigned to the ﬁnal four experimental conditions

received instructions from experimenters who over time became more

experienced in the experimental procedure.

115

An issue which is often advanced in discussing experiments conducted in
laboratory settings in the generalizability of ﬁndings. Clearly this study lacks
what has been termed "mundane realism" which is the degree to which the
experimental setting matches things which occur in organizations (Berkowitz &
Donnerstein, 1982). No study is completely generalizable (Fromkin &
Struefert, 1976) but the strength of laboratory investigation lies in it’s ability to
control irrelevant variables and test causal hypotheses.

llgen (1986) points out that high ﬁdelity (physical similarity between
laboratory and ﬁeld settings) can increase the generalizability of laboratory
ﬁndings. However, he further suggests that if laboratory subjects attribute the
same meaning to the variables of interest as subjects in organizations would,
high "psychological" ﬁdelity is achieved. The purpose of this study was to
determine how individuals react to performance feedback. The process
individuals undertake in determining the acceptability of feedback was of
primary interest. Therefore, high psychological ﬁdelity was considered more
important than high physical ﬁdelity.

One issue related to the generalizability of the ﬁndings is that the
feedback recipient typically has an ongoing relationship with the feedback
provider in real work settings. Since there was no one actually providing
feedback in this study, the influence of this type of relationship was not

assessed. Research conducted in an organizational setting could ascertain the

116

inﬂuence of the quality and length of the feedback recipient/feedback provider
relationship on reactions to performance feedback.

The lack of experience subjects had with the task they were asked to
complete also limits the generalizability of this study. The task was chosen to
maximize variance on self-certainty. Future research should assess self-
assessments of performance on familiar tasks to determine if they inﬂuence

consensus performance judgments more than that found in the present study.

Appendices

117
Appendix A
IN-BASKET EXERCISE

Your name is 1&9 Your work for the Bennett Corporation. You will be
transferred in one week to replace John Quitt who left suddenly. Your
employer, the Bennett Corporation, has decided to send you to the Midwest
ofﬁce to go through John Quitt’s (your predecessor’s) in-basket. It is Sunday
evening April 13th, and no one else is in the ofﬁce. In 45 minutes you must
catch a plane to a training center where you will be for one week and cannot
be contacted.

Read through the information, memos and reports. Respond to each
item in writing. Responses to these items may include letter writing, memos to
others or to your self and scheduling meetings. You may write your response
on the same memo you received or no the memo pad provided writing paper is
also provided for letter writing. §e_s_urp to attach any letters or memos to the
appropriate item.

An organizational chart is provided for your reference. You have 20
minutes to go through your in-basket.

IT IS ADVISABLE TO READ THROUGH THE ENTIRE IN-BASKET

BEFORE TAKING ANY ACTION.

1 18
BENNETT CORPORATION

Organization Chart for Midwest Branch

Fred Waters
Branch Manager

Betsy Ferris
Secretarial

Tom Smith
Zone Sales Manager

Carla Dwyer
Manager,Computer
Install/Repair

Lee Stanley Dave Thompson

Sales Repair

Tina Merloff Cathy French

Sales Repair

Susan Krane Vince Turen

Sales Repair

Dan Larson

Sales

Roy Turner

Sales

Mary Shop

Sales

Phyllis Whipple
Secretarial

119

Approximate Time Required for Various Types of Sales Calls

Initial Sales Call ime
Regular calculators 2 hrs.
Programmable calculators 2 hrs.
Computer system 4 hrs.
Demonstration

Regular calculators 2 hrs.
Programmable calculators 3 hrs.
Computer systems B10—B20 4 hrs.

B40-B55 6 hrs.

B70-B90 8 hrs.

Regular Calculators

Deluxe Calculator

Regular Programmable Calculator
Deluxe Programmable Calculator
Programs - pre-packaged
Programs - tailor made

Computers

Mini Computers

Medium Size Computers

Large Computers

120

Brim

Model No.
B2

B3
X-R.5
X-R.7

Mn

B10
B12
B15
B19

B40
B44
B48
B50
B55

B70
B74
B75
B77
B80
B84
B88
B90

.Cﬁt.
200.00
290.00

1400.00

1850.00
125.00
200.00

Basic Cost

26,000
28,000
36,000
41,000

75,000
80,000
97,000
102,000
110,000

375,000
390,000
450,000
470,000
725,000
810,000
915,000
1,325,000

121

3......“ Corporation 3
a1; ' V7 '
14¢» a: ,4/7g Jami.
(067:4 we...” my .1 ﬂux...-
«A; M. 'Jm maﬁa“; ./»C..
[/47 «A W 7»wa
J8 M J M A eye” {A
W7 71% %/ M3 era-£6.
ﬂy...) .1 My A»: ‘65qu
136%)”: _ '
ﬂy

Eon... T. SHITH

122

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 
 

 

 

 

 

5.6 582533 .wreomcu 92¢ 25555». 20 520
LIPIDQZUO «I.»
wdaouowuocn when. 3.3.25 203. 9330 Yet itcrropwoé we: 3.... downs. wczuiJEmaH
:66 516C $1.3 . 43323.9 rezoned 2,5 885.95.26 2 8a szo 288: Saw
.E IE... r92: ‘15: .5333: .qaoyEot or... 0:8: 232.2282
3:. $5.5: .5 79...? 35576.0 «2. 825:3 cash £32 «3313 ore v. 35;. $93 1:37
7 In: .3: 71C .0; $0579 02 Us: 3;: 32.18 #6650 {2:
c {at 20 >36: 99% Isa: .8125 E. 28:: E E: 2:8 0:5 no 58 So 3... adage 32:33
watt. . 23333.3 w _. 553 33433 .C «3.3 9:983:13 31¢ I: .707 35.3. v83
.7300 3,3ro nECSDUuto
4.520%. 57.34: use 3.20 303: #5239: So .63: 526 or... aoeutau 4.855
33. ulna $272123: 2... 936595 35.8 £25619 02 r. 2.6 one 283.52.
ID“- Q t LEHSQMW. 0383 he“... batﬁbl- .
”guitaciwountnx “SEQ MCI IS 23...}: 3.:
- s .-
mcwuaﬂbowfﬂwn do”? J..C.S.d 33¢ foe... Cuban
I .239 501965 who... .133 .t. .392 . a 2.2
.Q 33 5.13 Democracy mu 731655239 5.53 .30 >59 bro 2.... 1525.. 98. .56 E
.35: we.“ 524:3 63:33:62 out accelcexw .5360 ﬂ 2300 55¢ a. at 22:. 29:5:
.076. c «05303. Eu 0 H
>3 9:039? .ﬂpucwasn C .6306 use 2.5.; we 9033.5... 9 :89. or... 1.29% . E 39..
out“ an .43an 45:33. you? a mH zeiaou .wt Satousvo was. on :32: >53. 95323:
.ﬂwcameu 20 SEEMn .336 .33 . «no-» .3
Um... 330m .wuzaoiou 8 «atomic USDEEO
r55 1. on 01 €th 621.595 323 u:
u.» 6.6:. "Zen {:1 .266 EECV 356.. 0N9 5:; 9.3.. 95553332
~30 . 1.5.9532” yearn 250 {2,513 533 22.3533 032.. .9 n 3.... Ed 3332:
.ZchcuuoEvto .mcDrCIueioU
33:3 .8 I388 hereunto cur.) 33cm «H 3233 «SE d .230 2.: 05125 Cauck (50
w..41:3.3e. .25.E~u2¢sco 4:233:25 >50 EEDJSFEEV .mp6 9. 2.5.... 4.50 I
CU? p3 (ecu 77C 8 craves: L02 $33.53: or: 8.384 Jon—L233...
re: :6 «32.555. E3523 5530.5 .353 3.52 951539: ST: _ :92... yeah 3::
.Joﬁiruu 1.552 4 .9020 .56.. bwzauﬁowwc
.2 «2H: EH3 .303 (Zoe an: 20 53:38.03 5622306 55¢ . or... 2 2F E:
3 3:5 .tyzﬁuo 3 tree .3: . Binnie .55 «one: 3218 .39: n :23: «£3 22 NWEPrSJ
. 23 2.6a. am. .
76510 Eur. 3b .3 “5).—I .5
AL. U E Iver: #03556 C

123

Defunct Corpornt 10!: Q

‘. we.

Waéa’mwmﬁ
. agar/W M ,4...
.IWMWK'IO- m

 

 

 

124

”til .11 .12 .73 .w 44"

 

.100

3.0513.“- Tucson ummrsnm' Tawnsmv rmon'
M

 

 

30:00

 

IlcOU

 

 

“00

 

1100

 

3:00 ,

 

0:00

 

SIOO

 

5100

 

 

 

 

 

 

125

Klingon lndustries
lobk Spock Street
Iesteen. Kl

Harch 21

lennett Corporation
1555 15th Street
lesteen, Hl

hr. Quitt:

Regarding our conversation of Harch 20. l have been
considering your offer to demonstrate the 390 and
would like to set up an appointment. 1‘9111 be at
a conference for the next tun weeks, but as soon as
1 return 1 would like to see the demonstration. l

have, therefore. set aside the norning of April 21
for the denonstration.

I am looking forward to seeing you at that ticee

Yours truly. .
Kathy Cornan
Vice President

 

 

126

 

 

 

‘AV.
ne‘eéa/m c. any 0’2" m (to, *ldé
7’“ #011 tau/J 5 It! A”: ‘4‘“) M

JWJKLJMM

 

F T? -

a»... Q seeeo-o-terree 3 Infamy-H

 

 

127

um mmnos certs-onto: coeecsenuoeuc

 

 

 

 

 

oooaea ”.0 . ]eotauo- osev.
. . Don
t CPL! April 9
. .O’I. O b.‘.'0..
ns Horrison, Plant 89erintendent Stratton Assembly Plant
.81. C.C-
I80 Printer haliuoctions A ' . °

berial embers ”-5753 through‘ 114755

Us have received reports of proble-s vith the line printer (or the I00.
a checking a realised one o! our our technicians has been enabling
the printer in an incorrect manner. which results in continuous oaliunc-

tioning. The technician was responsible for asso-bling the printers with
the above Dentioned serial embers.

II you have any reports of problem with these printers. please replace
than ad return the eali’u-etionieg printer to es.

128

_ bennett Corporation
1535 13th St

Pesteen. on d 75.0,

uster— .bate zym /§

‘57.," J137-FI‘M J”“r
Address ﬁtting 1”! I

”r" M—
‘ Signature

951 M Description Unit Price 1”“
’ 3'75 Cmr’"" ' (370/12 (3%”

Sub Total {.220 ill)
Sales Tax lZﬂD

Total I Y0? [2W

:Au‘érised Signature; I

_ Renal/UT

(Title)

129

Appendix B
MEASURES COMPLETED
INFORMED CONSENT DECLARATION

The purpose of this project is to examine the determinants of sales
performance. In order to comply with professional standards as well as those
established by Michigan State University, it is essential that you be made aware
of the nature of this research and the rights and responsibilities incurred by
both you and the researcher.

The nature of this projects involves the investigation of what qualities a
good salesperson has and what behaviors lead to effective selling. You will be
asked to complete an in-basket exercise and will be provided with an evaluation
of your performance. You will also be asked to evaluate your own
performance. Finally, you will be asked to react to the evaluation you have
given yourself as well as the evaluation which will be provided.

The study will take approximately 1.5 hours of time for which you will
receive course credit. Please note that the information you supply will be kept
completely conﬁdential. The principal investigator, MaryBeth DeGregorio will
assume complete responsxbility for maintaining the anonymity of all you
responses.

. It should be stressed that you can decline to participate or terminate your
participation in this project at any time without recrimination. In return for
your participation, you will be given course credit as outlined by the Psychology
Department Human Subjects Committee and your instructor. If you wish to
receive this credit, but do not wish to participate in this study, or decide to
terminate your participation prior to the study’s completion, an alternative
activity requiring similar commitments of time and effort is available. At the
end of the study, a more complete written description of this investigation and
its ﬁndings will be available from the researcher. Along with the rights listed
above, you also incur certain responsibilities. Primarily among these
responsibilities is that you must provide information that is, as far as possible,
accurate and complete. We also ask that you DO NOT disclose any
information related to this project to any other persons until this project is
completed (e.g., until the end of Spring Quarter, 1989).

I certify that I have read and understand the rights and responsibilities
incurred by both me and the researchers in this project as outlined above.

 

(Signature)

 

(Print Name)

 

(Date)

130

Appendix B (cont)
Introduction

The purpose of this study is to identify what qualities a good salesperson
has and what behaviors lead to effective selling. In this study you will be asked
to complete an in-basket exercise. An in-basket exercise is a frequently used
technique for evaluating sales ability. The in-basket itself, is a collection or
group of materials which might be found in the incoming mail of a supervisor
or manager who has just taken over the position. A secretary is not available,
phones are disconnected and ﬁles are locked. Your task is to determine what
to do with the various situations presented in the in-basket materials.

Please keep in mind that it is not important whether or not you have
previous experience in sales. We are interested in the raw talent people
possess to be salespersons. It is therefore very important that you try to do
your very best at the in-basket exercise.

Attached you will ﬁnd a description of skills that the in-basket exercise
measures. Preliminary data indicates that the average in-basket exercise score
received by undergraduate students working on an in-basket for the ﬁrst time is
a 50 on a 100 point scale. We would like you to attempt to receive a score of
70. This is a difﬁcult but attainable goal for students with no in-basket exercise
experience.

131

Appendix B (cont)
IN-BASKET PERFORMANCE

Organization and Planning - The participant should develop a plan of attack.
Priorities should be set. Time must be used efﬁciently. Delegation should be
used. Conﬂicts should be recognized and dealt with.

Decision Making - Decisions should be made quickly. Procrastination should
not take place. The decision should be of high quality given the available
information.

Problem Analysis - Interrelationships between items should be noted and
handled. Additional information should be sought. Customer signiﬁcance
should be noted. Missing information must be recognized and sought out.

social Skills - Memos, letters, etc. should ﬁt the situation. Consideration
should be demonstrated by the use of thank-you’s and polite language where
appropriate. Firmness should be conveyed in writing where appropriate.

Unsatis- Average Outstanding
factory
O 10 20 30 40 50 60 70 80 90 100

132

Appendix B (cont.)

In-Basket Performance Evaluation Form
(Self Rating)

Please provide an OVERALL rating of performance. Your goal for
performance was a score of 70. You should CONSIDER the performance
dimensions listed below in making ONE overall rating. Please respond by
placing any number from 0 to 100, which best describes your performance, in
the space provided. Use the 83$ below for assistance in making your rating.

1. Dimension 1 - Organization and Planning

A plan of attack was deve10ped. Conﬂicts were recognized and
dealt with.

2. Dimension 2 - Decision Making

Decisions were made quickly. Decisions were of high quality given
the available information.

3. Dimension 3 - Problem Analysis

Interrelationships between items were noted and handled. Missing
information was recognized and sought out.

4. Dimension 4 - Social Skills

Memos, letters ﬁt the situation. Consideration was demonstrated.
Firmness and tact was used when appropriate.

Unsatis- Average Outstanding

factory
0 10 20 30 4O 50 60 70 8O 90 100

Overall Rating

133

Appendix B (cont.)

Self Feedback Acceptance

Please consider the ratings that you provided for yourself. Answer the
following questions using the scale provided below. Write your responses in
the space provided at the left.

Strongly Neither Agree Strongly
Disagree Nor Disagree Agree
1 2 3 4 5 6 7
1) My self feedback did not truly depict my performance
on the sales task.
2) My self evaluation reﬂected my true performance.
3) My self feedback was an accurate evaluation of my

performance on the task.

134

Appendix B (cont.)

Self Certainty

The following questions address how conﬁdent you are in the ratings you have
just provided. Please use the following scale to answer the questions below.
Fill in your response in the space provided at the left.

Strongly Neither Agree Strongly
Disagree Nor Disagree Agree
1 2 3 4 5 6 7

1) I am conﬁdent about the self evaluation I provided.
2) I am not sure how I did on the sales task.

3) I am certain that my self-ratings reﬂect my true
performance.

4) I have faith in my self-ratings of performance.

5) I doubt the self-ratings I gave.

135

Appendix B (cont.)
In-Basket Performance Evaluation Form

(Evaluator Rating)
Please provide an OVERALL rating of performance. You should CONSIDER
the performance dimensions listed below in making ONE overall rating. Please
respond by placing any number from O to 100, which best describes the
student’s performance, in the space provided. Use the scale below for
assistance in making your rating.

1. Dimension 1 - Organization and Planning

A plan of attack was developed. Conﬂicts were recognized and
dealt with.

2. Dimension 2 - Decision Making

Decisions were made quickly. Decisions were of high quality given
the available information.

3. Dimension 3 - Problem Analysis

Interrelationships between items were noted and handled. Missing
information was recognized and sought out.

4. Dimension 4 - Social Skills

Memos, letters ﬁt the situation. Consideration was demonstrated.
Firmness and tact was used when appropriate.

Unsatis- Average Outstanding

factory .
0 10 20 30 4O 50 60 7O 8O 90 100

Overall Ratin

136

Appendix B (cont.)

Shown below are a number of statements concerning your attitudes regarding
various topics. Carefully consider each of the following statements with respect
to your personal opinions. Then, decide the extent to which each statement
describes you. Use the response options shown below. Write the number of
your response to the left of each statement.

 

1 2 3 4 5 6 7
Always Almost Usually Sometimes Seldom Almost Never

Always Never

1. I do my best work when my job assignments are fairly
difficult.

2. When I have a choice, I try to work in a group instead of
by myself.

3. In my work assignments, I try to be my own boss.

4. I seek an active role in the leadership of a group.

5. I try very hard to improve on my past performance.

6. I pay a good deal of attention to the feelings of others I
work with.

7. I go my own way when I work, regardless of the opinions of
others.

8. I avoid trying to inﬂuence those around me to see things
my way.

9. I take moderate risks and stick my neck out to get ahead in
my work.

10. I prefer to do my own work and let others do theirs.

11.

12.

13.

14.

15.

16.

17.

18.

19.

20.

137

I disregard rules and regulations that hamper my personal
freedom.

I ﬁnd myself organizing and directing the activities of
others.

I try to avoid any added responsibilities when I work.

I express my disagreements with others openly.

I consider myself a "team player" when I work.

I strive to gain more control over the events around me.
I try to perform better than my fellow students.

I ﬁnd myself talking to those around me about non-school
matters.

I try my best to work alone on a job.

I strive to be "in command" when I am working in a group.

 

138

Shown below are several more statements concerning your attitudes.
Carefully consider each of the following statements and decide the degree to
which you agree or disagree with each statement. Use the response options
shown below. Write the number of your response to the left of each
statement.

1 2 3 4 5 6 7
Strongly Moderately Slightly Neither Slightly Moderately Strongly
Disagree Disagree Disagree Disagree Agree Agree Agree

 

Nor

Agree
_1. I feel that I’m a person of worth, at least on an equal

basis with others.

__2. I feel that I have a number of good qualities.
_3. All in all, I am inclined to feel that I am a failure.
__4. I am able to do things as well as most other people.
_5. I feel I do not have much to be proud of.
__6. I take a positive attitude toward myself.
__7. On the whole, I am satisﬁed with myself.
__8. I wish I could have more respect for myself.
_9. I certainly feel useless at times.

10. At times I think I am no good at all.

139

Consider each of the following statements. Choose the statement in each
pair of statements that you feel is closest to your opinion. Write the letter
indicating your response to the left of each pair of statements. THEN, check
the box which best describes how much closer the statement you choose is to
your true opinion than the statement you did not choose.

 

Statement
closer to
my opinion
(a or b?)

 

a)Children get into trouble because their parents punish
them too much.

b)The trouble with most children nowadays is their parents
are too easy.

Slightly Closer Much Closer

a)In the long run people get the respect they deserve in
the world.

b)Unfortunately, an individual’s worth often passes
unrecognized no matter how hard he/she tries.

Slightly Closer Much Closer

a)The idea that teachers are unfair to students is
nonsense.

b)Most students don’t realize the extent to which their
grades are inﬂuenced by accidental happenings.

Slightly Closer Much Closer

a)Becoming a success is a matter of hard work, luck has
little to do with it.

b)Getting a good job depends on being in the right place at
the right time.

Slightly Closer Much Closer

 

140

a)The average citizen can have an inﬂuence in government
decisions.

b)This world is run by the few people in power, and there
is not much the little guy can do about it.

Slightly Closer Much Closer
a)In my case getting what I want has little or nothing to
do with luck.

b)Many times we might just as well decide what to do by
ﬂipping a coin.

Slightly Closer Much Closer

a)Who gets to be the boss often depends on who was lucky
enough to be in the right place ﬁrst.

b)Getting people to do the right thing depends upon
ability, luck has little to do with it.

Slightly Closer Much Closer

a)Most people don’t realize the extent to which their lives
are controlled by accidental happenings.

b)There really is no such thing as "luck".

Slightly Closer Much Closer
a)In the long run the bad things that happen to us are
balanced by the good ones.

b)Most misfortunes are the result of lack of ability,
ignorance, laziness or all of these.

Slightly Closer Much Closer

10.

11.

141

a)Many times I feel that I have little inﬂuence over the
things that happen to me.

b)It is impossible for me to believe that chance or luck
plays an important role in my life.

Slightly Closer Much Closer
a)What happens to me is my own doing.

b)Sometimes I feel that I don’t have enough control over
the direction my life is taking.

Slightly Closer Much Closer

 

142

Appendix B (cont.)

Perceived Congruence

Please consider the ratings that your evaluator gave you. Answer the following
questions using the scale provided below. Write your responses in the space
provided at the left.

Strongly Neither Agree Strongly
Disagree Nor Disagree Agree
1 2 3 4 5 6 7

The evaluation I received was similar to my self
evaluation.

The rater’s evaluation was inconsistent with my self
rating.

My self ratings agreed with the ratings I received from
the evaluator.

My self assessment and the evaluation I received from the
rater matched.

There was a close correspondence between the rating I
received from the evaluator and my own self rating.

143

Appendix B (cont.)

Perceived Credibility Items

Please consider the ratings that your evaluator gave you. Answer the following
questions using the scale provided below. Write your responses in the space
provided at the left.

Strongly Neither Agree Strongly
Disagree Nor Disagree Agree
1 2 3 4 5 6 7

The rater who evaluated my performance was knowledgeable.
The rater was incompetent.
The rater could be considered an expert evaluator.

The rater was skilled in the evaluation of in-basket
exercise performance.

The rater was inexperienced in performance assessment.

144

Appendix B (cont.)

External Feedback Acceptance Items

Please consider the ratings that your evaluator gave you. Answer the following
questions using the scale provided below. Write your responses in the space
provided at the left.

Strongly Neither Agree Strongly
Disagree Nor Disagree Agree
1 ” 2 3 4 5 6 7

The feedback did not truly depict my performance on the
sales task.

The rater’s evaluation reﬂected my true performance.

The feedback the rater gave me was an accurate evaluation
of my performance on the task.

I do not feel the feedback reﬂected my true performance.

145

Appendix B (cont.)
Self Rating 11

You will soon be asked to complete a second in-basket exercise. Prior to
doing so we ask that you again think about your performance on the previous
in-basket exercise. We would like you to CAREFULLY consider your self
evaluation of performance as well as the rater’s evaluation of your
performance.

Please provide an OVERALL rating of performance. Your goal for
performance was a score of 70. You should CONSIDER the performance
dimensions listed below in making ONE overall rating. Please respond by
placing any number from 0 to 100, which best describes your performance, in
the space provided. Use the scale below for assistance in making your rating.

1. Dimension 1 - Organization and Planning

" A plan of attack was developed. Conﬂicts were recognized and
dealt with.

2. Dimension 2 - Decision Making

Decisions were made quickly. Decisions were of high quality given
the available information.

3. Dimension 3 - Problem Analysis

Interrelationships between items were noted and handled. Missing
information was recognized and sought out.

4. Dimension 4 - Social Skills

Memos, letters ﬁt the situation. Consideration was demonstrated.
Firmness and tact was used when appropriate.

Unsatis- Average Outstanding

factory
0 10 20 3O 40 50 6O 7O 80 90 100

Overall Rating

146

Appendix B (cont.)

DeBrieﬁng

This study investigated reactions to performance feedback. You received
feedback which was either very similar to your own self-evaluation or dissimilar
from your self-evaluation. Half of those who received feedback which was
dissimilar from their own self-evaluation received feedback indicating that their
performance was poor, the other half received feedback indicating that their
performance was good. In addition, you were led to believe that the individual
who provided the feedback had either a great deal of expertise
(industrial/organizational graduate student) or no expertise in evaluating sales
performance (undergraduate student). In fact, the feedback you received was
predetermined. We expect that both similarity of other assessments to self
assessments and the expertise of the rater will inﬂuence feedback acceptance.

It should be stressed that the performance feedback you received IN NO
WAY reﬂects your true performance. Your performance was in NO WAY
being evaluated. You were randomly assigned to either a group who received
feedback similar to your self-view or feedback which was dissimilar from your
self-view.

Thank you for your time.

147

Appendix C
LOW CREDIBILITY (Experimenter Script)

Now that you have completed your self-evaluation I am going to take your
in-basket to be evaluated by another person. Your performance will be
evaluated by a Michigan State University Freshman psychology student who is
also participating in this study for course credit. The student has NO previous
experience in evaluating performance. The student has no familiarity with in-
basket exercises either. In fact, no exposure to in-basket exercises will be
provided to this evaluator until the student is asked to assess your performance.
The freshman evaluator will be assessing your performance based on the same
dimensions you used for self-evaluation. While your performance is being
evaluated, please ﬁll out this attitude questionnaire.

HIGH CREDIBILITY (Experimenter Script)

. Now that you have completed your self-evaluation I am going to take your
in-basket to be evaluated by another person. Your performance will be
evaluated by the Industrial/Organizational psychology graduate student who is
conducting this study to complete requirements for the degree of PhD. This
graduate student has worked as a manager for ﬁve years prior to entering
graduate school, therefore has a great deal of experience in evaluating
performance. The graduate student has also received extensive training in the
measurement of in-basket exercise performance. In-basket exercises were
frequently used at the organization where the student worked. This evaluator
now conducts seminars on the use and evaluation of in-basket exercises. The
graduate, student evaluator will be assessing your performance based on the
same dimensions you used for self-evaluation. While your performance is
being evaluated, please ﬁll out this attitude questionnaire.

148
Appendix D

Results of factor analysis of self certainty and self feedback
acceptance scales.

 

Item
Factor 1: Positively Eigenvalue= 4.11
Worded Items
Factor Loadings
l .2.
My self feedback did not truly depict my
performance on the sales task. .655 .167
I am conﬁdent about the self evaluation I
provided. .649 .357
My self evaluation reﬂected my true
performance. .809 .091
I am certain that my self-ratings reﬂect my
true performance. .839 .181
My self feedback was an accurate evaluation
of my performance on the task. .817 .163
I have faith in my self-ratings of performance. .618 .569
Factor 2: Negatively Eigenvalue= 1.08
Worded Items
I am not sure how I did on the sales task. -.025 .849

I doubt the self-ratings I gave. .395 .717

LIST OF REFERENCES

149
LIST OF REFERENCES

Ashford, SJ. & Cummings,LL.(1983).Feedback as an individual
resource: Personal strategies of creating information. Organizational Behavior
prld Human Performancje, 3;, 370-398.

Bandura, A. (1978). The self system in reciprocal determinism.
ﬁneﬁcan Psychologist, 33, 344-357.

Bandura, A. (1977). Self-efﬁcacy: Toward a unifying theory of
behavioral change. Psychological Review. 84, 191-215.

Bannister, B.D. (1987). Performance outcome feedback and attn’butional
feedback: Interactive effects on recipient responses. Journal 91' Applied
Psychology, 11, 203-210.

Baumeister, RF. (1982). A self-presentational view of social phenomena.
Psycholgical Bulletin 21, 3-26.

 

Becker, LJ. (1978). Joint effect of feedback and goal-setting on
performance: A ﬁeld study of residential energy conservation.
Journal o_f Applied Psychology, Q, 428-433.

Berkowitz, L. & Doonnerstein, E. (1982). External validity is more
than skin deep: Some answers to criticisms of laboratory experiments.
American Psychologist, 12, 245-237.-

Bradley, G.W. (1978). Self-serving biases in the attn’bution process. A
reexamination of the fact or ﬁction question. Journal o_f Personality m
Social Psycholcgy, 36, 56-71.

 

Bray, D.W. & Grant, D. (1966). The assessment center in the
measurement of potential for business management. Psychological
Monoggaphs, 66, 80, (17, Whole No. 625).

Brickman, P. (1972). Rational and nonrational elements in reactions to
disconﬁrrnation of performance expectancies. Journal pf E_xpgrimental

ﬁgia] Bsycholog, 8, 112-123.

Campion, M.A. & Lord, R.G. (1982). A control systems
conceptualization of the goal-setting and changing process. Organizational

Behavior am Human Eeﬁonpapce, 10, 265-288.

150

Carver, C.S. & Scheier, M.F. (1981). Attention apt! Self-regplation: A
Control flfheog Approach 39 Human Behayior. New York: Springer-Verlag.

Cedarblom, D. (1982). The performance appraisal interview: A review,
implications and suggestions. Academy 91' Management Review 1, 219-227.

 

Cohen, J. (1988). Statistical power analysis M _t_h_e behavioral
sciences. Hillsdale, NJ: Lawrence Erlbaum Associates.

Cooley, CH. (1902). Human Nature £138 13 Social Order. New York:
Scribner.

 

 

Crary, W.G. (1966). Reactions to incongruent self-experiences.
Journal pf Consulting chhology, ﬂ, 246-252.

DeNisi, A.S., Randolph, W.A. & Blencoe, AG. (1983). Potential
problems with peer ratings. Academy o_f Management Journal, _2_§, 457-464.

Dipboye, R.C. (1977). A critical review of Korman’s self-consistency
theory of work motivation and occupational Choice. Organizational

Behavior m Human Bgrfpmancp, 18, 108-126.

Barley, P.C. (1988). Computer-generated performance feedback in the
magazine-subscription industry. Organizational Behavior app Human
Decision Processes, 4_1, 50-64.

Earley, P.C. (1985). Inﬂuence of information, choice, and task
complexity on goal acceptance, performance, and personal goals.

Journal pf Applied Psychology, _7_0_, 481-491.

 

Erez, M. & Zidon, I. (1984). Effect of goal acceptance on the relation
of goal difﬁculty to performance. Journal pf Applied Psychology, 82, 69-78.

 

Erez, M. (1977). Feedback: A necessary condition for the goal-setting
performance relationship. Journal pf Applied Psychology, 82, 624-627.

 

Festinger, L. A theory of social comparison processes. uman
Relations, 1, 117-140.

Fisher, C.D. & Russ, G. (1987). Self and superior assessment.
Technical Report for the Ofﬁce of Naval Research: N00014-85-k-0289:
TR-ONR-4. College Station, TX, Department of Management.

151

Gifﬁn, K. (1967). The contribution of studies of source credibility
to a theory of interpersonal trust in the communication process.

[gchological Bulletin Q, 104-120.

Greller, M.M. & Herold, D.M. (1975). Sources of feedback: A
preliminary investigation. Organizational Behavior m Human
Performance 18, 244-256.

 

 

Greller, M.M. (1980). Evaluation of feedback sources as a function
of role and organizational level. Journal 8f Applied P cholo , Q, 16-23.

Hackman, J.R. & Oldham, GR. (1976). Motivation through the design of
work. Organizational Behavior and Human Performance 18, 250-279.

 

Halperin, K., Snyder, C.R., Shenkel, R.J., & Houston, B.I(. (1976).
Effectiof source status and message favorability on acceptance of
personality feedback. Journal 8f Applied chhology, 81, 85-88.

Hanser, L.M. & Muchinsky, RM. (1978). Work as an information
environment. Organizational Behavior apg Human Performance, 81, 47-60.

Harris, M.M. & Schaubroeck, J. (1988). A meta-analysis of self-
supervisor, self-peer, and peer supervisor ratings. Egrsonnel Psychology, ﬂ,
43-62.

Harvey, OJ. & Clapp, W.F. (1965). Hope, expectancy, and reactions to
the unexpected. Journal 81’ Eersonality g8 Social Psychology, 2, 45-52.

Herold, D.M. & Greller, M.M. (1977). Feedback: The deﬁnition of a
construct. Academy 91‘ Management Journal, 28, 142-147.

 

Herold, D.M., Liden, R.C. & Leatherwood, MJ... (1987). Using multiple
attributes to assess sources of performance feedback. Academy pf
Management Journal 89, 826-835.

 

Hillery, J.M. & Wexley, K.N. (1974). Participation effects in
appraisal interviews conducted in a training situation. Journal

91' Applied Psychology, 89, 168-171.

Hollenbeck, J.R. & Brief, AP. (1988). Self-regulation in the
workplace: Towards a uniﬁed approach to understanding worker
attitudes and behaviors. In R. Schuler (Ed.) Readings i_n
Personnel 898 Human Resources Management. St. Paul, MN.: West.

 

152

Hollenbeck, J. R. & Williams, CR. (1987). Goal importance, self focus
and the goal setting process. Journal pf Applied Psychology, 22, 204-211.

llgen, DR. (1986). Laboratory research: A question of when, not if.
In E.A. Locke (Ed). Generalizing 80m laboratory 19 8818 settings.

llgen, D.R., Fisher, C.D. & Taylor, MS. (1979). Consequences of
individual feedback on behavior in organizations. Journal _o_f

Applied Psychology, ﬂ, 349-371.

 

llgen, D.R. & Moore, CF. (1987). Types and choices of performance
feedback. Journal 81' Applied Psychology, 1;, 401-406.

llgen, D.R., Peterson, R.B., Martin, B.A. & Boeschen, DA. (1981).
Supervisor and subordinate reactions to performance appraisal sessions.
Organizational Behavior _a_n_g Human Performance, _2_1, 437-450.

 

Jones, E.E. & Nisbett, RE. (1971). DLe actor 8n_d 818 observer:

General Learning Press.

 

Jones, S.C. & Schneider, DJ. (1968). Certainty of self-appraisals and
reactions to evaluations from others. Sociometg, ,3_1, 395-403.

Kanfer, F.H. & Karoly, P. (1982). The psychology of self-management:
Abiding issues and tentative directions. In Paul Karoly & Fredrick H. Kanfer
(Eds) 8ng Management m Behavior Qhange Em Theogy _tp Practice. NY:
Pergamon Press.

 

Kay, E., Meyer, H.H. & French, J.R.P., Jr. (1965). Effects of threat in a
performance appraisal interview. Journal o_f Applied Psychology, Q, 311-317.

Klein, HJ. (1987). Reactions to goal setting and feedback: A test of a
control theory model of work motivation. Unpublished doctoral dissertation.
Michigan State University.

Komaki, J., Barwick, K.D. & Scott, I... R. (1978). A behavioral safety
approach to occupational safety: Pinpointing and reinforcing safe
performance in a food manufacturing plant. Journal 8f Applied
Psychology, 83_, 434-445.

153

Korman, A. (1970). Toward a hypothesis of work behavior. Journal 8f
Applied Bgchology, 53, 31-41.

Kuiper, N.A. & Rogers, T.B. (1979). Encoding of personal information:

Self-other differences. Joumal 8f Bersonalig 888 Social Psychology, ﬂ,
499-514.

Larson, J.R., Jr. (1984). The performance feedback process: A
preliminary model. Organizational Behavior m Human Performance, 88,
42-76.

Lawler, BE. (1967). the multitrait-multirater approach to measuring
managerial job performance. Journal 8f Applied Esychology, 81, 369-381.

Levine, E.L., Flory, A. & Ash, RA (1977). Self-assessment in personnel
selection. Journal 81‘ Applied Esychology, 82, 428-435.

Locke, EA. (1968). Toward a theory of task motivation and incentives.
_Cﬂganizational Behavior 888 Human Eerformance, 8, 157-189.

 

Locke, E.A., Shaw, K.N., Saari, L.M., & Latham, GP. (1981). Goal
setting and task performance: 1969-1980. Egchologjcal Bulletin, 98, 125-152.

Lord, R.G. & Hanges, PJ. (1987). A control system model of
organizational motivation: Theoretical development and applied

implications. Behavioral ﬁjence, 82, 161-178.

Manz, CC. (1986). Self-leadership: Toward an expanded theory of
elf-inﬂuence processes in organizations. Academy o_f Management
Review, 11, 585-600.

Marek, J. & Mettee, DR. (1972). Avoidance of continued success as a
function of self-esteem, level of esteem certainty and responsibility for
success. Journal pf Personalig 8118 Social Psychology, Q, 98-107.

 

Markus, H. (1977). Self schemata and processing information about the
self. Journal o_f Eersonalig gig Social Psychology, 3_5_, 63-78.

Mead, OH. (1934). Mind, §£Lf 3.119. Society. Chicago: University of
Chicago Press.

154

Meyer, W.U. & Starke, E. (1982). Own ability in relation to self-concept
of ability: A ﬁeld study of information seeking. Bersonalig 88d M81

mm ﬁ. 501-507-

Miller, D.T. (1976). Ego involvement and attributions for success and
failures. Journal pf mm £119 m1 Eachology, 85, 901-906.

Miller, G.A., Galanter, E., & Pribaum, K.H. (1960). Blans _a_ng 189
structure pf behavior. New York: Holt, Reinhart, and Winston.

O’Reilly, C.A. & Anderson, J. (1980). Trust and the communication of
performance appraisal information: The effects of feedback on performance
and job satisfaction. Human Communication Research 8, 290-298.

 

Parker, J.W., Taylor, E.K., Barrett, R.S. & Martens, L. (1959). Rating
scale content: 111. Relationship between supervisory- and self-ratings.

Personnel Egchology, 18, 49-63.

Podaskoff, P.M. & Farh, J. (In Press). Effects of feedback sign and
credibility on goal setting and task performance. Organizational Behavior
m Human Performance 43.

 

Powers, W.T. (1973). Behavior: 1h; Control 8f Perception. Chicago:
Aldine Publishing Company.

Prien, E.P. & Liske, RE. (1962). Assessments of higher level
personnel: III. A comparative analysis of supervisor ratings and
incumbent self-ratings of job performance. Bersonnel chhology, 18, 187-194.

Rogers, T.B., Kuiper, N.A., & Kirker, W.S. (1977). Self reference and
the encoding of personal information. Journal o_f Personalig 888 Social
Psychology, 88, 677-688.

 

Sachs, P.R. (1982). Avoidance of diagnostic information in self-
evaluation of ability. Personality 13ch Social Psychology Bulletin, 8, 242-246.

 

 

Secord, P.F. & Backman, CW. (1961). Personality theory and the
problems of stability and change in individual behavior. An interpersonal

approach. Psychological Review, 88, 21-32.

Shrauger, 1.5. (1975). Responses to evaluation as a function of initial
self-perceptions. Esychological Bulletin, 88, 581-196.

155

Shrauger, J.S. & Rosenberg, SE. (1970). Self-esteem and effects of
success and failure feedback on performance. Journal pf Eggsonality, 88,
404-417.

Shrauger, J.S. & Schoeneman, J. (1979). Symbolic interactionist view of

self-concept: Through the looking glass darkly. Rsygllglpglgal Bulletip, 88,
549-5 73.

Smith, MB. (1968). The self and cognitive consistency. In R.B.
Abelson et al. (Eds). Theories 8f cogpitive consistepgy: A sourcebook.
Chicago: Rand McNally.

 

Snyder, M. & Swarm, W.B., Jr. (1978). Behavioral conﬁrmation in social
interaction: From social perception to social reality. Journal 91 Emrimental

Social Egghglggy, 11, 148-162.

Stone, D.L., Guetal, H.G. & McIntosh, B. (1984). The effects of
feedback sequence and expertise of the rater on perceived feedback

accuracy. Bersonnel chhology, L7. 487-506.

Stone, D.L. & Stone, ER (1985). The effects of feedback consistency
and feedback favorability on self-perceived task competence and perceived

feedback accuracy. W Behavip; an_d Hm mcision Brocesses,
E, 167-185.

Stone, E.F. & Stone, DJ... (1984). The effects of multiple sources of
performance feedback favorability on self-perceived task competence and
perceived feedback accuracy. Journal 91' Managemeng, 11), 371-378.

 

Strang, H.R., Lawrence, E.C., & Fowler, P.C. (1978). Effects of assigned
goal level and knowledge of results on arithmetic computation: A laboratory
study. Journal _o_f Applied Psychology, _68, 446-450.

Strube, M.J. & Roemmele, LA. (1985). Self-enhancement, self-
assessment and self-evaluative task choice. Journal _o_f Personalig a_ng Social
Psychology, 89, 981-993.

Swann, W.B. (1982). Self-Veriﬁcation: Bringing social reality into
harmony with the self. In J. Suls & A.G. Greenwald (Eds) Psychological
Perspectives pp 188 ,S_el_f, 8, 33-66. Hillsdale, NJ: Lawrence Erlbaum
Associates.

156

Swann, W.B. & Hill, CA. (1982). When our identities are mistaken:
Reafﬁrming self-conceptions through social interaction. Journal _o_f

Eersonalig g8 Social chhology, 48, 59-66.

Swann, W.B. & Read, SJ. (1982). Self-veriﬁcation processes: How we

sustain our self-conceptions. Journal pf ﬂpgrimental Social chhology, 11,
351-372.

Swarm, W.B. & Read, SJ. (1981). Acquiring self-knowledge: The search

for feedback that ﬁts. Journal 91' Personalig an_d _S8c1_'8l chhology, 4_1,
1119-1128.

Taylor, S.M., Fisher, C.D. & llgen, DR. (1984). Individuals’
reactions to performance feedback in organizations: A control theory
perspective. In K. M. Rowland & G. R. Ferris (Eds), Research in Bersonnel
a_n_d_ Human Resources Management, 8, 81-124. Greenwich, Conn: JAI Press.

Thornton, GO (1968). The relationship between supervisory and
self-appraisals of executive performance. Bersonnel chhology, a. 441-455.

Thornton, G.C. III. (1980). Psychometric properties of self-appraisals
of job performance. Bersonnel Esychology, 88, 263-272.

 

animal and 81; machine. Cambridge, MA: MIT Press.

l'. L) “la :02 ‘1

-'-ﬂm

 

111041an STATE UNIV. LIBRnRIEs
mW"Wll‘l‘lilllHllllllllllUlllWWIIIWI'HI
31293007591658