2

.. . _ .fié; q. 1n.

5.... a». 1??
M11311. 1..
.. M a

r?

1? 1.... 11W

 

1.
.1 .
1 14.9%....
. .3 .

rm
5.3

i... _.

I, .—
. 1 ..
_.fu.,m«m~w....fia
.113 4. :
1:»

.33!

 

a?

14“..

.5"
1.1

3511.. .

1.3.x:
1 I
23:11...
.5110: ,

. 1. 1 1
1.151.113: v.51.“

. , . 1.1.11.1...01111. .1
33,1. 1.. 9.11.11??? gafififi 1 11.1.1.1. .11.- «1.1.11.1...»

.1. 2.... $2.25.? .. .... . . I.

 

meats

 

IUNIHIIWIll)NIIIIHIIHIHIIHHIUMllllHHllHllHll

31293 01801 7388

LIBRARY
Michigan State
University

 

 

 

 

This is to certify that the

thesis entitled

The Effects of Coal Orientation on Performance

Evaluation Across Occupations

presented by

Kerry Allison Delbridge

has been accepted towards fulfillment
of the requirements for

Masters degree in Psychology

 

 

fl/QL

Major professor

Date December 17, 1998

 

0.7639 MS U i: an Affirmative Action/Equal Opportunity Institution

 

 

 

PLACE IN RETURN Box to remove this checkout from your record.
To AVOID FINES return on or before date due.
MAY BE RECALLED with earlier due date if requested.

 

DATE DUE

DATE DUE

DATE DUE

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

use W14

THE EFFECTS OF GOAL ORIENTATION ON PERFORMANCE EVALUATION
ACROSS OCCUPATIONS

By

Kerry Allison Delbridge

A Thesis

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

MASTER OF ARTS
Department of Psychology

1998

one

com
pfl'll
or jo
lilin‘.
with

undeI

ABSTRACT

THE EFFECTS OF GOAL ORIENTATION ON PERFORMANCE EVALUATION
ACROSS OCCUPATIONS

By

Kerry Allison Delbridge

This study examines the effect of similarity between rater and ratee on goal
orientation on performance appraisal and the additional roles of self-esteem and job
context. Rater attributions were examined as mediating processes between similarity and
performance appraisal ratings. Results indicated no significant effect of actual similarity
or job context on performance ratings. A model incorporating perceived similarity and
liking as mediating variables, though, was found to be significant. Thus actual similarity
with respect to goal orientation led to increased perceived similarity, rater attributions of

understanding and reinforcement, liking, and in turn, performance ratings.

TON

To Mike and Marisa, who know me better than anyone, often better than I know myself,

and whose love and support give me the courage to be me.

iii

ACKNOWLEDGEMENTS

I would like to first thank Dr. Richard P. DeShon for his tireless effort and many
hours spent evaluating advising and debating the finer points of my thesis. I especially
appreciate first his willingness to accept my ideas, however farfetched, as credible,
allowing me to fully explore and explain them, and second his commitment to demanding
honestly critical thinking of me. Without his efforts, this work would not be possible. I
would also like to thank my committee members, Dr. Daniel Ilgen and Dr. Neal Schmitt
for their individual contributions. Their advice and input were both insightful and
valuable.

I owe a deep debt to many who were both colleagues and friends. They provided
me with the kind of emotional support, understanding and camaraderie that was and is
invaluable. My cohort members, Dee Wasson, Lori Sheppard, Becca Toney and Danielle
Jennings, were extremely supportive and without them I might have packed up and left
town.

I would also like to thank Morrie Mullins and Cathy Clause who let me know that
stress and worry were normal and that everything really would work out all right.
Morrie’s constant emotional support and advice were more important to me than he could
possibly know. Cathy provided me with wisdom both in terms of excellent advice and in
knowing when to drop everything and have real fun.

Finally, I would like to thank my family. My parents, Tom and Jan Delbridge,
who instilled in me a great desire to learn and the confidence to strive for my real dreams,

those that make me happy and fulfilled. Nicky, whether she knows it or not, has always

iv

 

been

peep

been an inspiration to me, and continues to be a reminder that there are truly wonderful

people in the world.

 

 

List c
List 0
Introc
Accur
Sumn‘
Simild
Goal (
Ann'br
COntex
Self-E.
Model .

Method

TABLE OF CONTENTS

List of Tables
List of Figures
Introduction
Accuracy
Summary
Similarity and Attraction
Goal Orientation
Attributions
Context Effects in Performance Appraisal
Self—Esteem
Model and Hypotheses
Method
Subjects
Materials
Measures
Procedure
Results
Supplementary Analyses
Discussion

Appendices

vi

viii

18

22

27

3O

35

35

36

37

39

43

60

84

92

 

is:

 

Referenc.

Appendix A - Job Performance Descriptions

Appendix B — Job Descriptions

Appendix C - Goal Orientation Measures

Appendix D - Self-Esteem Measure

Appendix E - Performance Measure for Brake Manufacturing Job
Appendix F - Performance Measure for Research Engineering Job
Appendix G - General Performance Items

Appendix H - Perceived Similarity and Attraction Measures

References

vii

93

97

98

99

100

109

122

123

124

LIST OF TABLES

Table l - Pilot Study Results

Table 2 - Descriptive Statistics for Measured Variables Across Ratees

Table 3 - Scale Reliabilities

Table 4 - Correlations of Hypothesis-Relevant Variables Across Ratee

Table 5 - Performance Correlations Across Ratee

Table 6 - Repeated Measures AN OVA on Performance Rating

Table 7 - Repeated Measures AN OVA on Understanding

Table 8 - Repeated Measures AN OVA on Reinforcement

Table 9 -Repeated Measures AN OVA - Understanding on Performance Rating

Table 10 - Repeated Measures ANOVA - Reinforcement on Performance
Rating

Table 11 - Supplementary Correlations

Table 12 - Supplementary Repeated Measures AN OVA - Performance Rating
Table 13 - Similarity Correlations

Table 14 - Repeated Measures AN OVA - Perceived Similarity

Table 15 - Correlations - Perceived Similarity, Liking, and Performance Ratings

Table 16 - Repeated Measures AN OVA Between Perceived Similarity and
Attributions

Table 17 - Repeated Measures ANOVA - Attributions on Liking
Table 18 - Repeated Measures ANOVA - Liking on Performance Rating

Table 19 - Fit Indices for Supplementary Model

viii

41

45

47

49

51

53

55

57

59

62

64

68

69

72

74

76

78

82

 

 

 

Figure 1'
Hgml
fiflm3

fiwm4t
PedOnna

fing-
Pedonna

Hgmfi-

Hflm7-
SWhm}

Hgms-
Smmum

d

FlSure9.

LIST OF FIGURES

Figure 1 - Payoff in Prisoner’s Dilemma
Figure 2 - Hypothesized Model
Figure 3 - Ratee by Rater Mastery Orientation Interaction on Understanding

Figure 4 - Perceived Similarity by Self-Esteem Interaction on Global
Performance

Figure 5 - Perceived Similarity by Self-Esteem Interaction on General
Performance

Figure 6 - Perceived Similarity by Self-Esteem Interaction on BES Rating

Figure 7 - Ratee by Rater Mastery Orientation Interaction on Perceived
Similarity

Figure 8 - Ratee by Rater performance Orientation Interaction on Perceived
Similiarity

Figure 9 - Structural Equation Model

2O

31

65

65

66

7O

70

8O

INTRODUCTION

The use of performance appraisal ratings is based on the premise that people can
accurately determine how well others are performing their jobs. The need for such
determinations is reflected in their wide application. Performance appraisals are used as
the basis of promotions and dismissals. They are used to give feedback to employees and
employees use this feedback to make changes in the way they approach their jobs. In
addition to these direct applications, performance appraisals are used as a dependent
variable in all kinds of research. Various factors are examined to determine what effects
they will have on performance, and often this performance is measured by some sort of
rating. Because of this, the quality of performance appraisals are of great concern to both
researchers and practitioners alike.

Factors that affect the performance appraisal process have been widely examined,
and not surprisingly, the focus has been on rating errors. Many “categories” of factors
can be examined, such as characteristics belonging to the rater, characteristics of the
ratee, and the particular rating process used. Rater characteristics such as demographic
variables, psychological characteristics, and job-related variables such as the rater’s job
experience or performance level have been the focus of much performance literature
(Landy & Farr, 1980). Demographic characteristics have been the most commonly
studied rater characteristics (Landy & Farr, 1980). A typical example of this is a study
done by London and Poplowski (1976) who found that female raters gave higher

performance ratings than males.

Many researchers have shown that the personalities of the raters affect
performance appraisals (e.g. Buffardi & Gibson, 1980). Schneier (1977) investigated the
impact that the cognitive complexity of raters has on the accuracy of performance ratings,
and Drory and Ben-Porat (1980) investigated the effects of the leadership style of the
rater. Wexley and Youtz (1985) investigated an entire battery of personality variables,
including self-esteem and adaptability. Based on this research, it is clear that individuals
systematically differ in the way they rate others.

Ratee characteristics that have been examined include demographic variables
(again the most common), personality variables and job-related variables. (Landy & Farr,
1980). Given the historical imbalance between men and women in the workplace, gender
is one of the most obvious choices for study, and has been examined repeatedly. (e. g.
Maurer & Taylor, 1994; Sackett, DuBois, & Noe, 1991; Schmitt & Hill, 1977) Similarly,
researchers have looked to ratee race for effects on performance appraisal (e. g. Sackett,
DuBois, & Noe, 1991; Schmitt & Hill, 1977).

Performance appraisal research has also examined interactions between rater and
ratee characteristics. Most of this research has been limited to matching the sex and/or
race of the raters and ratees (Landy & Farr, 1980). For instance, Pulakos, White, Oppler
and Borman (1989) ran a study in which soldiers in the army were used to examine the
effects of both rater and ratee sex and race on performance ratings.

There are, however, other possibilities for interactions. Instead of concentrating
on one particular variable, Pulakos and Wexley (1983) examined the relationship between

performance appraisal and the general similarity between particular supervisors and

subordinates. The potential also exists for other variables to be studied, especially those
more directly related to work behaviors. Landy and Farr suggested that performance
appraisal research has often neglected the cognitive processes of the rater. They further
indicated a need to determine what mental processes occur as the rater forms his
judgments. Ilgen, Bames-Farrell and McKellin (1993) have more recently stated that
attempts need to be made at combining knowledge from the different domains. For
example, knowledge about raters, ratees, and setting should be combined. Thus, the
recommendations of these researchers indicate that interactions involving thought
processes, motivations and values are needed.

The research has also tended to ignore the content of a specific situation (1] gen et
al., 1993) In other words, researchers have tended to assume that the rating process is the
same in every situation rather than dependent upon any sort of context. In contrast,
Cleveland, Murphy, and Williams (1989) stress the importance that context plays in the
performance appraisal process. For example, the purpose for which the performance
appraisal is conducted can have great impact on the ways in which the raters gather and
use information (Williams, DeNisi, Blencoe, & Cafferty, 1985)

Thus, although the performance appraisal literature has explored a wide variety of
variables, there is room for expansion, especially by combining the effects of several
different types of variables.

Accuracy
Ilgen, Barnes-Farrell, and McKellin (1993) conducted a review of the

performance appraisal literature to determine what changes, if any, had occurred since

Landy and Farr’s (1980) call for more cognitive research. They stated that Landy and
Farr’s article caused a shift from an emphasis on scales and rater training to attempts to
understand that the rater is a decision-maker who processes social cues.

The performance appraisal process is subject to biases such as halo, and leniency
(Bemardin, 1978), and these influences have been the focus of much research (e.g.
Borman, 1975; Feldman, 1986; Havenstein, 1992; Murphy & Balzar, 1986; Murphy &
Reynolds, 1988). Leniency represents a general tendency to give favorable ratings
(Bemardin, 1978). Halo generally refers to “the inability to discriminate among
conceptually distinct and potentially independent aspects of ratee performance” (Murphy
& Reynolds, 1988 pp 235). The latter is seen as failing to distinguish between
dimensions of performance (Feldman 1986).

Correcting a problem with performance appraisal requires a deep understanding of
the underlying influences and processes. While it is clear that accurate performance
appraisals are very important, this focus may limit the types of questions that get asked.
If researchers are only interested in accuracy, this may keep them occupied with the what
of performance appraisal rather than the how. In other words, this may put the focus on
the results rather than the process. A complete understanding may thus be more difficult
to achieve.

Conversely, a researcher who ignores the focus on accuracy is more free to
examine how ratings occur without feeling that each study must directly point out which
raters are more accurate. He can examine the effects that variable X has on how the rater

processes information and can thus gain a greater understanding of how the rater is

thinking and behaving without feeling tied to the idea that this must somehow directly
explain a portion of the accuracy gained or lost.

1] gen et a1. (1993) stated that research on cognitive processes has the potential to
make two contributions: 1) understanding basic cognitive processes, 2) showing how to
improve rating processes. They also emphasize that there are three steps in the rating
process: 1) acquisition of information, 2) organization and storage of information, and 3)
retrieval and integration of information that leads to an evaluation. They stated that it is
time for the performance appraisal literature to go in new directions including
investigating what kind of thought processes affect the evaluation.

It is extremely important to understand the mental activities that underlie the
appraisal process and the cognitions of raters. Once there is understanding, one is in a
better position to evaluate the effects of these processes and cognition in order to
determine the extent to which they are fair, useful or accurate. The knowledge gained
about the deeper processes will allow increased success in improving performance
appraisal.

Further, this understanding will allow researchers to make distinctions in
performance appraisal that are more complex than just between right and wrong or good
and bad. For example, one can address the issue of whether a particular characteristic of
the performance appraisal process makes it more accurate in some situations and worse in

others.

Summary

In summary, the performance appraisal literature has a need for research of a
different type than typically seen in the past. There is a need for research that
incorporates new variables, that focuses on processes, that takes context into account, that
makes use of basic research, and that integrates research from different domains. Also,
while most research involving the effects of rater characteristics on performance
appraisals have centered around the issue of the susceptibility to particular biases, there is
a need for research that explores what makes supervisors differentially interpret and value
particular behaviors in their subordinates.

The purpose of this study is to incorporate some variables and processes that have
not yet been applied to the performance appraisal literature. The performance appraisal
process is influenced by many different factors including the way in which the rater and
ratee define “good” performance. A relatively less explored personality construct, goal
orientation, has direct implications both for the way in which employees perform their
jobs and the way in which supervisors will evaluate these behaviors. The more the
employee and supervisor see eye to eye on how the job should be performed, the better
the performance rating for the employee should be. These perceptions may be
additionally influenced by the supervisor’s feelings about herself. The lower the
supervisor’s self-esteem, the more likely she may be to try to enhance that self-esteem by
rating employees who are similar to herself highly. Finally, the context in which the
performance takes place should affect supervisor perceptions of the appropriateness of

employee behaviors. What would constitute “good” performance in one context might

not in another. Thus, some jobs will be more appropriate for particular goal orientation

profiles than others. The following sections provide justification for this perspective.

Similarity and Attraction

Extensive research has documented that similarity of attitudes causes attraction
and liking (e.g. Byme, 1971; Byme & Griffitt, 1969; Byme & Nelson, 1965; Golightly,
Huffman, & Byme, 1972; Hamilton, & Fallot 1974; Lydon, J amieson, & Zanna 1988;
Tan, & Singh 1995). In other words, the more similar two people are, the more they will
tend to like each other. For the purposes of this discussion, unless otherwise specified,
similarity refers to actual congruence between people as measured or manipulated in
some manner rather than to the perceptions people may have about their similarity to
someone else.

There have been many practical applications of this effect. van Lange, Liebrand,
& Kuhlman (1990) examined the effects of subjects’ strategies on the prisoner’s dilemma
scenario. They found that people presented with the dilemma made positive attributions
of people with similar strategies and negative attributions of people with dissimilar
strategies. Golightly, Huffman and Byme (1972) showed that similarity of attitudes
affected attraction to a person and further, that it affected decisions about this person,
namely how large of a loan business students were willing to approve for that person.
Good and Good (1972) demonstrated that potential counselors were rated more positively
when their attitudes were similar to the raters. Thus, the similarity-attraction effect has

been shown to be robust and generalizable to many different situations. In most of these

studies, a hypothetical other’s similarity to the subject is manipulated and then the subject
is asked to indicate how much they like this person.

There has also been research that applies this phenomenon to different areas in the
work place. Griffitt and Jackson (1970) demonstrated that similarity of attitudes
increased the probability that subjects would hire a job applicant. Baskett (1973) found
that similarity significantly affected the amount subjects would recommend paying a
prospective employee. Good and Good (1974) showed that supervisors were more
positively rated on several work-related dimensions when their attitudes were similar to
the raters.

Finally, the similarity—attraction paradigm has been applied to performance
appraisal. Smith, Meadow, and Sisk (1970) showed that attitude similarity increased the
favorability of performance evaluations when subjects were asked to rate a previously
unknown person and were led to believe that this person was either similar or dissimilar
to themselves. Senger ( 1971) ran a study with actual managers and their subordinates.
Actual measures of value—similarity were taken, and the experimenters found that more
similar subordinates were rated higher. Pulakos and Wexley (1983) used actual manager-
subordinate pairs and a real performance evaluation. Subjects were selected such that
they represented diverse occupations and were asked to rate subordinates on five
dimensions: performance, conformance, dependability, personal adjustment, and general
satisfactoriness. Further, they were asked to rate the degree to which the subordinate was
similar to themselves in general. The researchers found that ratings were related to how

similar to themselves the mangers perceived the subordinates to be.

This finding brings up an important idea: that it might be the perceived similarity
rather than the actual similarity that affects ratings. The problem with the Pulakos and
Wexley study is that because actual manager-subordinate pairs were used, and no
measures of actual similarity were taken, there is no way to determine causality. It is
equally plausible that managers were motivated to believe that they were similar to the
successful subordinates. In fact, there is literature that supports the idea that people will
be more likely to perceive themselves as similar to a person if that person is successful
than if that person is unsuccessful (Krahe, 1983).

Turban and Jones (1988) did a similar study and measured both perceived
similarity and actual similarity. Both types of similarity were found to be related to
ratings, but perceived similarity was more highly related to ratings than was actual
similarity. A problem with this study is that perceived similarity was measured in terms
of outlook, perspective, values and work habits while actual similarity was measured in
terms of demographic variables such as race, educational level, department tenure, and
age. The conclusion can not be drawn, then, that perceived similarity is more important
than actual similarity.

Wetzel and Insko (1982) found that people are attracted to those who are like their
ideal selves. In other words, we like people who are similar to who we want to be rather
than who we are. This does represent a departure form the traditional
similarity/attraction paradigm. It implies that this literature should be concentrating on
people’s values rather than on people’s actual traits. According to this research, managers

should like employees who behave the way that the manager wants to behave rather than

the way that the manager actually behaves. For example, suppose that manager A really
places high value on being focused. He does not believe that workers should converse
about topics that are not related to work. Despite this, he finds himself slipping every
once in a while, and talking about his children. This particular manager should, in
general, like a subordinate that remains focused on work better than one who converses
about his family. This line of thinking has direct implications for the current research,
and will be discussed further.

Although the similarity attraction effect is strong and its extension to performance
appraisal seems promising, the exact underlying process remains cloudy. Byme and
colleagues’ theories of similarity and attraction phenomenon stem from reinforcement
(Byme & Clore, 1967; Byme & Griffitt, 1969; Byme & Nelson, 1965). According to this
logic, a person who is similar to us reinforces our belief that we are correct or
appropriate. Researchers have, for the most part, either ignored this explanation or taken
it for granted, probably due to the fact that testing this as an underlying process would be
difficult or due to a greater interest in the practical implications of the phenomenon itself
rather than the processes. Byme and Griffitt (1969), though, found that this relationship
holds even when the effects of perceived similarity are partialed out. Thus, awareness of
similarity is not essential to increased attraction. Logically, it does not seem likely that a
person who does not see another as similar is being affirrned by that similarity, suggesting
that other processes may be at work either instead of or in addition to the reinforcement
proposed by Byme and his colleagues. Wetzel and Insko’s (1982) findings that people

like those who are similar to their ideals also suggests other mechanisms at work. An

10

F12.- _ -. ..

 

 

ideal cannot serve as reinforcement about one’s own correctness because it implies that
the idea] has not been reached. In fact it might serve to reinforce the discrepancy between
that ideal and oneself.

There remains, then, questions as to the underlying processes that cause this
phenomenon. There is also a need to explore the effects of similarity with respect to
particular attitudes or personality constructs instead of just as a general concept.

Goal Orientation

While it is a start to say that actual similarity affects attraction and performance
appraisal ratings, this statement is meaningless unless one denotes a frame of reference
(Medin, Goldstone, & Gentner, 1993). One person cannot be generically similar to
another, they must be similar with respect to specific characteristics. One individual
difference that is especially applicable to the workplace is goal orientation. Goal
orientation encompasses performance orientation and mastery orientation.

Dweck (1986) defined a learning (mastery) orientation as those where
“individuals seek to increase their competence, to understand or master something new,”
and a performance orientation as those where “individuals seek to gain favorable
judgments of their competence or avoid negative judgments of their competence.” More
recently, Dweck (1992) conceptualized learning (mastery) as “seeking to improve one’s
competence,” and performance as “seeking to prove one’s competence.” Thus, each
orientation represents the degree to which a person values a particular goal and is driven

to accomplish it, specifically the goals of learning or achieving.

ll

Performance orientation is based on great concern over ability level, and
avoidance of negative judgments of competence (Button, Mathieu, & Zajac, 1996;
Dweck, 1986; Elliot & Dweck, 1988). Performance oriented people derive satisfaction
from displaying their ability, and have a preference for tasks that maximize competence
judgments (Dweck, 1986; Dweck & Elliot, 1983). When they choose a task, they will
tend to avoid challenge, and if possible, will choose a task that appears most difficult to
outsiders and is easiest for them (Button, Mathieu, & Zajac, 1996; Dweck, 1986; Dweck
& Leggett, 1988; Dweck & Elliot, 1983). Thus the reasons they approach tasks and the
motivation behind the way they perform them are different from non-performance
oriented people.

These characteristics should affect work habits as they have implications for the
types of tasks that are enjoyed and the ways in which mastery oriented people go about
performing them. Performance oriented people should be concentrating on how they can
display their ability to others. If they choose assignments that fit their particular strengths
and avoid their weaknesses, they can perform really well and impress their coworkers and
supervisors. If they are confident that they can perform well, they should pick high-
profile tasks, and they should work very hard at perfecting the tasks they do work on.

When faced with obstacles or difficulty during a task, high performance people
exhibit low persistence, negative affect and self-cognition, withdrawal of effort, and
deterioration of performance (Button, Mathieu, & Zajac, 1996; Dweck, 1986; Dweck &
Leggett, 1988; Elliot & Dweck, 1988). They have a tendency to evaluate negative

outcomes in terms of ability (Button, Mathieu, & Zajac, 1996; Dweck, 1986). They

12

evaluate their competence by using normative standards, do not believe that success is
repeatable, tend to over estimate their failures, and have difficulty remembering their
successes (Diener & Dweck, 1980; Dweck & Elliot, 1983).

These negative reactions should affect future workplace actions. Highly
performance oriented people should try to discontinue or avoid tasks that are difficult for
them. If they are in a position of power, they would probably delegate these tasks to
others while choosing to confine themselves to the tasks at which they excel. They
should also spend time comparing their performance to that of others doing the same job.

Due to the fact that performance oriented people place such a high value on
performing well, their tendency to succeed and/or fail should have broader consequences
on their feelings of self-worth, or self-esteem. Difficulty and failure are common in life
for everyone, regardless of their orientation. Nobody does everything perfectly on the first
try. Very few children ride their bike perfectly down the street the first time they get on,
and very few adults perform as well on the first day of their first job as well as other
employees performing that job. For performance oriented people, these failures should
take on greater importance, and since failures and difficulties occur in all aspects of life,
they will experience lower satisfaction with themselves and lowered worth.

Mastery orientation is based on a desire to learn and master. People with high
mastery orientation enjoy exerting effort in the process of mastering a task, increasing
ability level over time, striving to understand something new, attempting to acquiring
knowledge or skills, and seeking of tasks that maximize learning (Button, Mathieu, &

Zajac, 1996; Dweck, 1986; Dweck & Elliot, 1983; Elliot & Dweck, 1988). Mastery

l3

oriented people also tend to seek challenge, and derive satisfaction from exerting effort
(Button, Mathieu, & Zajac, 1996; Dweck, 1986; Dweck & Leggett, 1988).

Again, there are implications for the workplace. People high on mastery
orientation should try to improve their skills, understand the tasks they are working on,
and learn about their jobs, whether about the specific tasks themselves or about the
relationships between those tasks and the jobs of other employees. They should also
show a tendency towards picking tasks that will allow them to learn the most, which
would usually be the more challenging, or diverse assignments.

When faced with obstacles or difficulty during a task, mastery oriented peOple
tend to exhibit persistence, continued positive affect, optimism about future outcomes,
and maintenance of effective efforts (Button, Mathieu, & Zajac, 1996; Diener & Dweck,
1980; Dweck, 1986; Dweck & Leggett, 1988; Elliot & Dweck, 1988). They treat failure
as useful feedback, and have a tendency to use difficulty as a sign that strategies should
be changed or effort should be increased (Button, Mathieu, & Zajac, 1996; Dweck, 1986;
Elliot & Dweck, 1988). They will also place some value on partial attainment of a goal
(Dweck & Elliot, 1983). Since almost everyone experiences some sort of failure or
difficulty on the job, these reactions should occur with some frequency and should have
effects on future tasks. People high on mastery orientation should be more willing to
stick with an endeavor with which they are having difficulty, and to change the way they
are approaching a particular assignment on which they are working if it is not going well.

Mastery oriented people base judgments of their competence on their own

performance, attribute success to effort and are optimistic about future outcomes (Diener

14

& Dweck, 1980; Dweck & Elliot, 1983). These attitudes have less direct implications for
workplace behavior, but affect attitudes and should affect work behavior through these
attitudes.

Since mastery oriented people place such a high value on learning, their feelings
of self-worth, or self-esteem should be heavily based on the learning they experience.
Leaming, unlike high performance, occurs in almost all situations. “Success” and
“failure” do not determine whether or not a person learns because they will learn in both
situations if they desire to do so. In fact, they probably learn more when they fail than
they do when they succeed. Because their learning is what they measure themselves by,
they should have high self-esteem.

Originally, goal orientation was conceptualized as one variable. Mastery-oriented
people were on one end of the continuum, and performance-oriented people were on the
other (Dweck, 1986). More recently, researchers have come to the understanding that
mastery and performance values are not mutually exclusive, and that they should be
measured separately (Bouffard, Boisvert, Vezeau, & Larouche, 1995; Button, Mathieu &
Zajac, 1996). While some hypothesized characteristics of mastery and performance
orientation would appear to be mutually exclusive, this may be an artifact of the fact that
in most of the literature, people were classified as one or the other. For example,
traditionally, mastery oriented people were theorized to choose difficult tasks while
performance oriented people were theorized to choose easy ones. When examining this
from the continuum perspective, it makes perfect sense: Mastery oriented people pick the

difficult tasks because they are looking for opportunities to learn, while performance

15

oriented people pick the easy tasks because they want to perform well. Applying this to
the two-variable perspective appears difficult at first because a person who is high on
both mastery and performance should want to pick tasks that are both easy and difficult.
The problem is easily reconciled, though, when one considers the true motivations. This
person would like a task that offers a chance to learn a great deal, but is also easy.
Obviously, finding these types of tasks might be difficult, and this person may have to
compromise by choosing a medium-difficulty task. This problem would thus not be
detected by the continuum perspective, because the high performance high mastery
person would score in the middle of the continuum and would therefore be hypothesized
to choose a medium-difficulty task anyway. Because the theory that goal orientation is
really two variables is new, there may be issues to be worked out, but they should present
little difficulty in the way of matching raters and ratees.

The research which applies the similarity-attraction paradigm to performance
appraisal can be applied to mastery and performance orientations. As discussed, many
researchers have shown that personality affects performance appraisals, but generally this
research either deals with the personality of the rater, or the personality of the ratee, but
not both (e. g. Buffardi & Gibson, 1980; Drory & Ben-Porat, 1980; Harte], 1993;
Schneier, 1977; Wexley & Youtz, 1985). If the personality characteristics of both the
rater and the ratee are taken into account, more complex processes can be explored.
Because of goal orientation’s implications for the workplace, it is a natural choice for
matching raters and ratees in the performance appraisal setting. According to the above

descriptions, mastery orientation and performance orientation have strong implications

l6

for how people approach tasks. Since people must perform many tasks on most jobs, this
means that the way people approach their jobs is also affected. Performance oriented
people attempt to turn out a perfect product and mastery oriented people attempt to learn
while they work and thereby improve their ability to do future work. People pursuing
mastery goals should want to spend time gathering information that is less directly
relevant to the task at hand because of their desire to learn, and people pursuing
performance goals should want to spend energy on the specific task on which they are
working because it is more directly relevant to the performance that will be evaluated.

A supervisor should be able to witness first hand the kinds of actions caused by
goal orientation because goal orientation directly affects how people attack their work.
These actions are either similar or dissimilar to the ways in which the supervisor would
perform those tasks. Thus, the supervisor should have lots of information readily
available to him about how similar this subordinate is to himself on this particular
personality trait. The behaviors that are affected by goal orientation are not only
applicable to both the rater and the ratee, but are exhibited in the performance context and
may be felt by most raters to be completely reasonable indications of performance. They
should thus have a potentially great impact on performance ratings.

Wetzel and Insko’s (1982) earlier cited hypothesis that we like our ideals also has
applications here. While a distinction between liking people who are similar to us and
liking people who are similar to who we want to be is meaningful and should affect most
research that deals with similarity, goal orientation is unique. Goal orientation focuses on

the values of the person in question. A mastery oriented person values learning while a

17

performance oriented person values performing well. The degree to which a supervisor is
successful at learning or performing is irrelevant to the measurement of his goal
orientation. Thus, supervisors with a mastery orientation should like subordinates who
are exhibiting mastery-oriented actions based on the fact that the subordinate is being
what the supervisor would like to be, regardless of way the supervisor behaves. In this
case, the similar subordinate is one who is successfully exhibiting the characteristics that
the supervisor values. Regardless of how successful the supervisors are at achieving their
mastery or performance goals, they should be attracted to people who are successful in
this sense.

Thus, while raters and ratees can be similar or dissimilar according to many
variables, goal orientation is particularly relevant because it is a value held by the rater
and ratee that also affects behavior in the workplace. It should therefore be both more
visible and more important in the workplace.

Attributions

Supervisors should make different attributions about the way in which their
subordinates go about performing their jobs. Any particular behavior has many possible
interpretations that can be paired with it (Feldman, 1981). Perceptions and judgments of
people reflect two things: the actual stimuli that are present and the perceiver’s inferences
and expectations (Fiedler, 1993). For instance, it has been shown that people’s
personalities can affect the ways they interpret other people’s motives and actions
(Dweck 1995). Thus, person A and person B can both witness person C exhibiting a

behavior, but A may interpret the behavior as “thinking” while B interprets it as “staring

18

off into space and not getting anything done.” If A and B are asked to evaluated how
productive C is being, A is likely to evaluate C more highly than B does. These
attributions have direct implications for performance appraisal.

Research has also shown that the match between personalities can color
perceptions and interpretations. van Lange, Liebrand, and Kuhlman (1990) examined the
causal attributions of people faced with prisoner’s dilemmas. In a prisoner’s dilemma,
two people must each choose between cooperating or defecting. The “game” is set up so
that each person’s payoff is determined by the combination of their own choice and the
choice of the other person. Regardless of whether the other person chooses to cooperate
or defect, a person’s payoff is larger if he/she chooses to defect. The catch is that if both
“players” defect, they both get much lower payoffs than if they both cooperated. (See
Figure 1)

Based on the choice made by a player, he or she can thus be categorized as a
“cooperator” or a “defector.” van Lange et al. (1990) found that cooperators tended to
attribute defection to ignorance while defectors tended to attribute cooperation to
ignorance. This illustrates a discrepancy in understanding and thus interpretation.
Individuals made different attributions depending on whether the target approached the
task in the same way the evaluator did. If the target handled the situation the way that the
evaluator did, the evaluator understood the action and thus evaluated the decision and the
individual positively. Conversely, if the target handled the situation in a different way
than did the evaluator, the evaluator could not understand the action and thus attributed

this choice to ignorance, an obviously lower evaluation.

19

A’s Choice

 

 

 

 

 

 

Cooperate Defect
Cooperate A: 7 B: 7 A: 10 B: l
B’s Choice
Defect A: l B: 10 A: 3 B: 3
Figure l

Payoff in Prisoner’s Dilemma

20

This process can be applied to the workplace. A supervisor who witnesses his
subordinate performing her job in the same way the supervisor would (or does) should
understand better what the subordinate is doing. The supervisor himself has chosen (or
would choose) that same strategy, and thus has already decided that it is the “best” way.
Conversely, a subordinate performing the job in a different way is more likely to be seen
as “ignorant” or less competent because she has chosen another option. Thus, the
subordinate who performs the job in line with the way the supervisor believes it should be
done will tend to be evaluated more positively while the subordinate who tackles the job
in a different way will tend to be evaluated less positively.

Mount and Thompson’s (1987) subjects rated the degree to which their
supervisors behaviors were congruent with the subject’s expectations. In other words,
they rated the degree to which the manager performed responsibilities “in the way the
subordinate believed they should be performed.” (pp 241) They found that ratings of
those supervisors who were behaving as expected contained more leniency and halo than
the ratings of those supervisors who were not behaving as expected. This can be seen as
the same process. Congruent supervisor’s actions are understood and evaluated
positively.

Goal orientation affects work strategies, and the performance appraisal process
requires supervisors to make judgments about subordinates who are either similar or
dissimilar to themselves on these strategies. In accordance with the above research,

subordinates who have different goal orientations use different strategies and are likely to

21

be seen as making less than optimal choices. Thus they should receive lower
performance evaluations than subordinates who are similar to their supervisors.
Combining these literatures produces the result that goal orientation should have
an effect on performance appraisal through two routes. First, it provides a personality
construct on which supervisors and subordinates can be similar or dissimilar, and the
similarity-attraction paradigm would indicate that this should sway performance
evaluations based on reinforcement. Second, supervisors who differ on goal orientation
will make different attributions about the same actions performed by a subordinate, and
these attributions will directly affect the performance appraisal process. It is completely
reasonable, though, that both may be operating to produce higher ratings for subordinates

who are similar to their supervisors on goal orientation.

Context Effects in Performance Appraisal

Another key concept in performance evaluation is the context in which it is made
(Cleveland, Murphy, & Williams, 1989). Performance evaluations are not made in a
vacuum. They are made about people in a particular setting. Cleveland et al. pointed out
that performance appraisals can be done for different purposes such as making promotion
decisions, making salary decisions, assessing training needs and guiding employee
development, and that there are logical reasons why the rating processes that occur for
each of these is different. The purpose for which observation is done can have an effect
on the information gathering process and also on the ratings that are obtained (Murphy,

Philbin, & Adams, 1989). Raters are also likely to use the information they do gather in a

22

different way (Williams, DeNisi, Blencoe, & Cafferty, 1985). For example, if the
purpose is to distinguish between people, (as in the promotion situation), the raters are
likely to search for comparative information so that they can distinguish between pe0ple.
On the other hand, if the performance appraisals are being conducted as part of a training
needs assessment, the raters will make within-individual comparisons (Cleveland et al.,
1989).

If there are differences in the cognitive activities of raters that are dependent upon
purpose within the same setting (i.e. same job), then it makes sense that varying the
setting should also have effects. In other words, the specific job that is the context for the
performance appraisal affects the rating process. If a rater is evaluating a person’s
performance as a cartoonist, the rater certainly uses different types of information that if
she is evaluating a computer programmer. It is important, then, to investigate the effects
that different contexts have on the hypothesized rating processes.

In order for the type of job to make a difference in performance appraisals, raters
must have ideas about what different jobs entail and what kinds of people should be in
them. Research has shown that people have occupational stereotypes. In other words,
they believe that people in different jobs have different personality traits (Hakel,
Hollmann & Dunnette, 1970; Jackson, Peakcock & Smith, 1980; Levy, Kaler & Schall,
1988; Paunonen & Jackson, 1987). Further, applicants who most closely resemble the
stereotype of the job are viewed more favorably and are thus most likely to be hired

(Jackson, Peakcock & Smith, 1980; Paunonen & Jackson 1987).

23

For example, Paunonen, Jackson, and Oberman ( 1987) reviewed the literature and
found that accountants are more likely to be high on order and cognitive structure and low
on autonomy, change and impulsivity. Similarly, engineers are more likely to be high on
endurance, achievement, and understanding and low on social recognition and
succorance. The researchers then manipulated hypothetical interviewees on these traits
and found that subjects rate interviewees more highly when their personalities match the
jobs than when they are inconsistent with the job. This is person-job fit, not from the
typical perspective of the incumbent or in an objective sense, but from the perspective of
the rater.

Because of the previously mentioned effects that goal orientation has on work
strategies and the fact that different jobs require different strategies, it is important to
examine the effects of goal orientation in more than one setting. Jobs vary in the degree
to which they emphasize learning and performance. Therefore jobs should be
differentially suited for mastery and performance approaches. Raters should rate highly
mastery oriented ratees favorably when in the context of a job that is suited for a mastery
approach, and should rate highly performance oriented ratees favorably when in the
context of a job that is suited for a performance approach. This represents not only a
tendency on the part of the rater to desire the person to match the job, but also the fact
that this type of matching probably does, in fact, increase performance.

Because goal orientation has only recently been applied to the work setting, no
research directly tests which jobs are more suitable for performance oriented people, and

which are more suitable for mastery oriented people. The literature does, however,

24

provide us with information about the types of behaviors that will be exhibited, and we
can match these with jobs that benefit from those behaviors.

Jobs that require that a person do the same thing over and over do not require the
constant learning that more changing and diverse jobs do. Instead, they require that an
incumbent learn how to do the job well initially and then use the skills he or she has to
consistently perform the job well. They may require a strict attention to detail and
perfection, but “perfection” is static in this case. For example, a worker on an auto
assembly line that is in charge of making sure that the brakes are assembled correctly
does not need to constantly change and improve. Ideally then, he should be performance
oriented. It is much more important that this worker come as close as possible to
perfection than it is that he learn more about the brakes, how they work, why they are
designed that way, what other systems they are related to, etc. That is not to say that
learning is not important to his job, but that it is less important than focusing on making
sure that a car doesn’t leave the factory with faulty brakes.

On the other hand, some jobs have tasks that are constantly changing and
therefore require that the worker change their strategies and methods. For example, a
research engineer must do a variety of ever-changing, but related tasks. Thus, it is
important that she learn as much as possible about her tasks, her job, and how these relate
to other ideas, other peoples’ jobs, etc. Concentrating on performing computations better
than the other people around her is not as important as being able to come up with new

ideas and being able to put ideas together creatively. Perfect performance is not the

25

primary goal in this job. Instead, the emphasis should be on constantly learning and being
creative, so that the incumbent can keep up with the changing requirements of the job.

Because “good performance” in these two jobs requires completely different
strategies, the incumbents in these two positions are likely to have different personality
traits, and raters are likely to perceive that these differences are advantageous (Jackson,
Peakcock & Smith, 1980; Paunonen & Jackson, 1987; Paunonen, Jackson, & Oberman,
1987). If a high performance, low mastery oriented person were in the research
engineering position, he or she would probably not be able to adapt very well as the job
evolved and fluctuated whereas a high mastery, low performance oriented person would
have less trouble. On the other hand, if a high mastery oriented, low performance
oriented person were in the assembly-line job, he or she would probably be bored after an
initial learning period and would not concentrate enough on the perfection required of
such a position.

Adding this information to the previous sections produces a theory that the highest
performance ratings should be made when the goal orientations of the rater, and the ratee
match and when the job context fits that orientation. Thus, if the ratee is high
performance and low mastery, his ratings will tend to benefit when he’s being evaluated
for a job that rewards performance behaviors such as an assembly line job. Conversely, if
the ratee is low performance and high mastery, he will tend to receive the highest ratings
when being evaluated by a low performance, high mastery rater and his ratings will tend
to benefit when he’s being evaluated for a job that rewards mastery behaviors such as a

research and development job.

26

Self-Esteem

Self-esteem is a variable that has been widely studied in psychology (Rosenberg,
Schooler, Schoenback, & Rosenberg, 1995). Simply put, global self-esteem is “the
individual’s positive or negative attitude toward the self as a totality.” (Rosenberg et. al,
1995, pp 141) It is thus a mixture of self-acceptance and self-respect, a general feeling of
self—worth. This particular variable is important to this study for several reasons.

First, as discussed previously, self-esteem may be related to goal orientation.
Dweck and Elliot (1983) describe mastery oriented people as asking, “How can I do it?”
while they describe performance oriented people as asking, “Can I do it?” Performance
oriented people suffer when they experience failure, and they are not buffered from this
failure by previous success (Diener & Dweck, 1980). When they experience success,
they do not expect it to continue, and when they experience failure, they think that it
indicates that they have low ability. As Elliot and Dweck (1988) put it, performance
oriented people are “vulnerable to the helpless response in the face of failure, setting up
low ability attributions, negative affect, and impaired performance” (p 5). Mastery
orientation, on the other hand, does not have this pattern associated with it (Diener &
Dweck, 1980). All people experience failures during their lives. Since good performance
is highly valued by performance oriented subjects, these feelings of failure and negative
affect should influence their general feelings of self worth and thus lower their self-
esteem. Similarly, since mastery oriented people value learning, which should be

obtainable in most situations, mastery oriented people should have high self-esteem.

27

In this context, self-esteem needs to be distinguished from self-efficacy.
According to Bandura (1986), self-efficacy refers to people’s judgments about their
capabilities. Generally, this concept has been used in a very specific sense. In other
words, people have a self-efficacy with respect to a specific task. Self-efficacy has also
been applied as a general concept (Bandura, 1986). In other words, people can believe
that they’re capable of performing well on tasks in general, rather than limiting it to a
specific task. Generalized self-efficacy, then appears to be very much like self-esteem in
that they both apply to more than one situation and describe the feelings that a person has
about himself. The critical difference between generalized self-efficacy and self-esteem
is that self-efficacy still refers to capabilities while self-esteem refers to self-worth. The
arguments made above are based on people’s perceptions of their own worth, and not
their perceptions of capabilities, in either a general or specific sense.

Research has supported the hypotheses linking self-esteem to mastery and
performance orientation. Positive correlations between self-esteem and mastery
orientation and negative correlations between self-esteem and performance orientation
have both been found (Button, Mathieu & Zajac, 1996; Fisher, Delbridge & DeShon,
1997).

The similarity—attraction research would also seem to naturally be influenced by
how well people like themselves. Subjects suffering from low self-esteem may be
motivated to enhance their feelings about themselves. They might thus be more inclined
to give similar people high ratings in order to boost their feelings about themselves. In

other words, a person who is low on self-esteem would be more motivated to exhibit the

28

similarity/attraction phenomenon simply because they can feel better about themselves if
similar others are rated highly. In agreement with this idea, Weiss (1978) found that the
relationship between value similarity and perceived success and perceived competence of
a supervisor was moderated by self-esteem. Specifically, for the low self-esteem subjects,
there was a positive correlation between similarity of values and both perceived success
and perceived competence. For the high self-esteem subjects, this correlation was
actually negative. At first glance, the negative correlation for the high self-esteem
subjects may be puzzling, but Weiss broke this down by value category and the negative
correlations here seem to be mostly caused by a specific value category, people
orientation, which deals with a value placed on interactions with others.

It is possible that self—esteem might be directly related to how similar ratees
perceive raters to be. In other words, self-esteem might enable, or disable a person to see
that another person is similar. There does not appear to be a large logical connection
between the two, and Weiss (1978) found no correlation between self-esteem and
perceived similarity of values.

Thus, if low self-esteem increases the similarity attraction phenomenon, low self-
esteem raters will show a stronger preference for ratees of the same goal orientation. If
performance oriented people have lower self-esteem, performance oriented raters should
display a stronger difference in their ratings of performance and mastery oriented ratees

than would mastery oriented raters.

29

Because of the effects that performance orientation may have on self-esteem and
the effects that self-esteem may have on the similarity-attraction effect, self-esteem will

also be measured to determine whether or not it is affecting the hypothesized results.

Model and Hypotheses

The model representing the conceptual relationships between the goal orientations
of the rater and ratee, the job context, attributions, and the rater’s self-esteem are
presented in Figure 2. In this figure, the interaction of the rater’s performance orientation
and mastery orientation with that of the ratee, represent actual similarity. Likewise, the
interaction of the ratee’s mastery and performance orientations with job represent person-
job fit. Person—job fit has a direct effect on the performance ratings given. Actual
similarity has an effect on performance rating through both types of attributions and also
interacts with the rater’s self-esteem to affect the performance rating. These relationships

are detailed by the following hypotheses.

30

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I r
Rater Ratee Ratee Rater
Performance Performance Mastery Mastery
Orientation Orientation Orientation Orientation
V L
Self-esteem
Understanding Reinforcement
Attribution Attribution
l v
Performance
Rating
Figure 2
Hypothesized Model

31

 

 

J ob
Context

 

 

 

 

The similarity-attraction literature indicates that the similarity between rater and
ratee should cause an increase in performance ratings. Two respects for similarity,
mastery orientation and performance orientation have been chosen due to their direct
implications for work place behavior. Similarity is thus defined as the degree to which
the rater and ratee match on mastery and performance orientations.

Hypothesis 1a: A subordinate exhibiting performance oriented behaviors will be
more highly evaluated by a high performance rater than by a low performance
rater.

Hypothesis 1b: A subordinate exhibiting mastery oriented behaviors will be more
highly evaluated by a high mastery rater than by a low mastery rater.

People have stereotypes about what the incumbents in certain positions are
generally like. These occupational stereotypes affect how well suited raters judge people
to be for particular positions such that a person who seems most similar to the prototype
will be judged more favorably (Paunonen, Jackson, and Oberman, 1987). Because goal
orientation has direct implications for work-place behavior, particular jobs should be
more suited to mastery or performance oriented incumbents. Person-job fit, then, is
defined as the degree to which the ratee’s goal orientation profile matches that of the job.
Hypothesis 2a: A ratee in a performance-focused job will be evaluated more
positively if his actions reflect a performance orientation.

Hypothesis 2b: A ratee in a mastery-focused job will be evaluated more positively if

his actions reflect a mastery orientation.

32

People experience failures and difficulty as part of their everyday lives. High
performance oriented people place more value on performance, and thus more importance
on a lack of performance than do low performance oriented people. Because of this value
placement, failures are more important to their feelings of self-worth than they are to
people who place less importance on performance. Thus, their self-esteem will be lower,
on average, than low performance oriented people.

Hypothesis 3: Performance orientation will be negatively correlated with self-
esteem. Thus, strongly performance oriented subjects will have lower self-esteem
than non-performance oriented subjects.

People who try to learn from situations should be successful in that endeavor,
regardless of their objective performance. Whether the person fails or succeeds at tasks,
information is gathered, and thus success is attained. High mastery oriented people place
a high value on learning and doing so should increase their feelings of self-worth.
Because of this, high mastery oriented people should on average have higher self-esteem
than low mastery oriented people.

Hypothesis 4: Mastery orientation will be positively correlated with self-esteem.
Thus, strongly mastery oriented subjects will have higher self-esteem than non-
mastery oriented subjects.

People with low self-esteem are likely to try to boost their self-image through
positive comparisons. Subjects suffering from low self-esteem may be motivated to
enhance their feelings about themselves, and one way in which they can accomplish this

is by believing that people who are similar to themselves are good performers. Thus, a

33

person who is low on self-esteem may exaggerate the similarity/attraction literature
simply because they can feel better about themselves if similar others are rated highly.
Hypothesis 5: Self-esteem will moderate the relationship between actual similarity
and ratings. Raters with low self-esteem will be more affected by the similarity-
attraction phenomenon than will high self-esteem raters.

Actual similarity, for the purposes of this study, revolves around mastery and
performance orientation. These constructs are especially relevant for work behaviors. As
such, the similarity between rater and ratee on these constructs should be reflected in a
correspondence in the way they approach their job. Based on van Lange et a1 (1990),
people who behave similarly understand the logic behind each other’s actions. They
better understand a similar person’s behavior and choices because those are the behaviors
they themselves would engage in and the choices they would make.

Hypothesis 6: When evaluating ratees who are similar to themselves, raters will
make attributions that indicate an understanding of the ratees’ behavior.

Byme and colleagues’ similarity-attraction theory is based on reinforcement
(Byme & Clore, 1967; Byme & Griffitt, 1969; Byme & Nelson, 1965). A ratee who is
behaving similarly to a rater thus reinforces the raters’ belief that he is correct or
appropriate.

Hypothesis 7: Raters will report that rating similar ratees highly will cause them to
experience positive affect.

These attributions, that the rater understands the ratees’ actions, and that rating

that ratee highly would make them feel good should in turn be related to the performance

34

ratings given. Mount and Thompson’s (1987) study indicates that when ratees were
consistent with the expectations of the rater and were consistent with the way the rater
believed they should be performed, performance ratings were higher. The ratees actions
were understood and therefore evaluated more positively. By the same token, a rater who
has chosen or would choose the same work strategies has decided that those strategies are
best. Thus, when a ratee exhibits these strategies, they should be evaluated more
positively.

Hypothesis 8: The attributions indicating understanding of the ratee will be
positively correlated with performance ratings.

Reinforcement theory indicates that we engage in behaviors that are pleasing. Thus,
raters’ agreement that rating a particular ratee highly would be reinforcing should lead to
a corresponding actual increase in the performance rating.

Hypothesis 9: Feeling good about rating a ratee highly will be positively correlated

with performance ratings.

Method

Subjects

Subjects were 201 psychology students at a large mid-Westem university. They
received partial course credit for their participation in the study. One subject’s data was

incomplete, and was thus dropped from any affected analyses.

35

Materials

Descriptions of fictional subordinates’ job performance were written to reflect
three combinations of mastery and performance orientations: high-mastery/high-
performance; high-mastery/low-performance; and low-mastery/high-performance. Each
of these was based on the kinds of choices and behaviors that are exhibited by people
with the corresponding motivations. Because a low-mastery/low-perforrnance
combination would not present any information as to what motivated the person or how
that person behaved, this particular combination was not evaluated. Care was taken to
ensure that each of the three descriptions covered similar aspects of the job and that they
contained both positive and negative statements. The high-mastery/high-performance
description was created by combining the characteristics of both types of personalities in
the literature, and reflects statements included in both the low-mastery/high-performance
and high-mastery/low—performance descriptions. The descriptions were written as general
statements that would apply to most jobs. In other words, they describe general
approaches and strategies to work, and not the specific job behaviors. Also, two filler
performance descriptions were written to be presented between the other three. These
reflect the same general content areas as the focal descriptions in order to make them
consistent. For example, the statement “He is pleased when he performs well.” in the
high performance / low mastery description and its counterpart, “He is pleased when he
works hard and learns a lot.” in the high mastery / low performance description, are

replaced by “He is pleased when he finishes quickly.” and “He is pleased when he

36

completes a task without any help.” in the two distractor descriptions. These descriptions
are found in Appendix A.

Descriptions of the two particular jobs studied include the duties associated with
that job and some information about the relevant organization. Half of the subjects
received information that stated that the employees they were rating work in a factory,
working on truck brakes. The remaining subjects were told that the employees were
research engineers for a chemical company. The factory job was chosen because it
seemed most typical of a performance orientation, and the research job was chosen for its
similar affiliation with the mastery orientation. These descriptions are found in Appendix
B.

Measures

Goal orientation was measured using Button et al.’s (1996) 8-question mastery
scale and 8-question performance scale. (See Appendix C). Self-esteem was measured
using Rosenberg et al.’s (1989) lO-item scale. (See Appendix D)

Three performance appraisal measures were used. The first was one of two
behavioral expectation scales (BES) which were created by slightly adapting existing
performance measures so that they were appropriate for the jobs being studied. The one
used to measure the research engineers’ performance was originally developed to measure
engineers’ performance and has been used on research and development engineers. The
measure used for the truck brake assembly job was originally created for a similar
position, a truck mechanic and slight modifications were made to ensure applicability.

These measures can be found in Appendices E and F.

37

The second performance rating consisted of five general performance statements
that are identical for both jobs so that comparisons can be made across jobs. Subjects
were asked to indicate the degree to which they agreed with the statements on a five point
Likert scale. These items can be found in Appendix G.

The third performance appraisal measure was a global measure consisting of a
single question asking how well the employee is performing his/her job. (“How well is
this person performing his/her job?”) The subjects were asked to give the employee a
rating between 1 and 5.

In order to assess the attributions made by the rater, two types of attributions were
measured. First, subjects were asked to indicate how much they believed that rating this
person highly would make them feel good. This measure was aimed at addressing Byme
and colleagues’ (Byme & Clore, 1967; Byme & Griffitt, 1969; Byme & Nelson, 1965)
contention that the similarity-attraction effect is caused by the reinforcement model.
Second, subjects were asked to indicate the degree to which they understood the
employee, thought that the employee’s choices made sense to them, and would approach
the job in the same way. These questions were aimed at determining the remaining
attributional processes that mediate the relationship between similarity and liking.

Manipulation checks were included to determine the extent to which the subjects
perceived the employees as similar to themselves, and how much they liked the
employees. Subjects were asked to rate how similar each employee was to themselves
and were asked to respond to a modified portion of the Interpersonal Judgment Scale

(US) which has been used to assess interpersonal attraction. The two US items ask how

38

likely it is that the subject would like this person and how much they would want to work
with them. Two other perceived similarity items and one more attraction item were asked
of the subjects. All of these were assessed with a 5 point Likert scale from Strongly
disagree to Strongly agree. These statements can be found in Appendix H.

Procedure

Pilot sessions were conducted in which raters were trained on mastery and
performance orientation and were then asked to give each performance description a
rating on both the mastery and performance orientation on 8 point scales. Pilot subjects
were also asked to indicate how appropriate mastery and performance oriented
approaches would be for each job on a 5 point scale.

The results of the pilot sessions are presented in Table 1. The results indicated
that the performance descriptions adequately reflected the differences in performance and
mastery orientation. The low mastery-high performance ratee received an average of 1.10
for mastery orientation and 7.90 for performance orientation on the eight-point scales.
Conversely, the high mastery-low performance ratee received an average of 8.00 for
mastery orientation and 1.10 for performance orientation. The high mastery-high
performance ratee received high ratings for both mastery orientation and performance
orientation, 6.50 and 6.30 respectively. The distractor ratees were given middle-range
scores on both mastery orientation and performance orientation. The first received 3.10
on mastery orientation and 4.20 on performance orientation while the second received
scores of 4.50 and 3.30 respectively. The results also indicated that pilot subjects

perceived the engineering job as primarily mastery oriented, giving it a mastery

39

orientation score of 3.5 and a performance orientation score of 1.90 on the five-point
scales, and that they perceived the truck break assembly job as primarily performance
oriented, giving it a mastery orientation score of 1.40 and a performance orientation score
of 3.00. Thus, the pilot subjects indicated that the mastery and performance orientation

manipulations were appropriately made.

40

Table 1

Pilot Study Results

 

 

 

 

 

 

 

 

 

Stimulus Material Mean MO Rating SD Mean PO Rating SD
Low MO / High PO Ratee 1.10 .32 7.90 .32
High MO / Low PO Ratee 8.00 .00 1.10 .32
High MO / High PO Ratee 6.50 .53 6.30 .82
Distractor Ratee l 3.10 1.60 4.20 1.87
Distractor Ratee 2 4.50 1.43 3.30 1.06
Research Job 3.50 1.27 1.90 1.20
Assembly Eb 1.40 .52 3.00 .943

 

 

 

 

 

Note: M0 = Mastery Orientation, P0 = Performance Orientation

4l

 

For the main study, subjects were asked to assume that they were a supervisor
evaluating five employees’ performance in a single job. Subjects were informed that they
may not have the exact information presented in the performance evaluation form, but
that they should use their best judgments based on the information they were given. First,
the subjects were given one of the descriptions of the job, and were asked to read it
carefully. Next, the subjects were given one of the descriptions of job performance and
were instructed to also read it carefully. They were then asked to rate this employee
according to how well he or she was performing the job. They were asked to complete the
global performance measure first, then the general performance appraisal measure and
finally the BBS measure. The three focal performance descriptions were separated by the
two filler descriptions. The order in which the vignettes were presented to the subjects
was varied to ensure that order effects were controlled. To this end, the order of the three
focus vignettes was reversed for half of the subjects. Next, the subjects were told that
the researcher was interested in the other types of impressions that the subject had formed
about the employee, and the subjects were asked to answer the similarity, liking and
attribution questions. This procedure was repeated for all five performance descriptions,
with the three experimental descriptions being separated by the two distractor
descriptions. The subjects were then asked to answer the goal orientation and self-esteem
questionnaires. Finally, the subjects were debriefed and thanked for their participation.

This design thus contained four between subjects variables: rater mastery

orientation, rater performance orientation, job context, rater self-esteem. All of these

42

were continuous variables except for job context, which had two levels. The design also
contained three within subjects variables, ratee profile, understanding attribution, and
reinforcement attribution. The attributions were continuous variables, and the ratee

profile had three levels.

Results

Table 2 presents the means and standard deviations of each of the measured
variables. Since perceived similarity, liking and the performance ratings were assessed
for each ratee, each is treated separately. Also, the BES scores in this table are separated
by job but represent the raw scores. In all remaining analyses, the BES scale scores were
standardized so that they could be combined. None of the analyses included the two
distractor ratees. Each of the scale scores was created by computing a mean of the scale
items. Thus each could range between 1.00 and 5.00. The means and standard deviations

do not suggest that range restriction was a serious concern for any of the variables.

43

Table 2

Descriptive Statistics for Measured Variables Across Ratees

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variable HmLp SD Hme SD L,,.Hp SD Mean SD
Mean Mean Mean
MO 2.03 .48
PO 1.94 .48
Self-Esteem 2.03 .64
Psim 3.09 .89 2.83 .90 3.41 .86
Liking 2.80 .77 2.61 .79 3.24 .76
Global 2.40 .91 2.02 .87 2.55 .84
BES — Research 2.17 .62 2.10 .71 3.00 .74
BES - Assembly 2.20 .66 1.76 .50 2.33 .65
General 2.13 .71 1.96 .69 2.34 .62
Understanding 2.82 .80 2.51 .85 3.09 .73
Reinforcement 3.05 .98 2.76 .96 3.34 90

 

Note: HmLp = High mastery/low performance ratee, Hme = High mastery/high
performance ratee, L,,,Hp = Low mastery/high performance ratee, M0 = Mastery
Orientation, P0 = Performance Orientation, Psim = Perceived similarity, Global = Global
Performance Measure, BES = Behavioral Expectation Scale, General = General

Performance Measure

 

The internal consistency reliabilities of each of the scales are listed in Table 3.
Three different estimations for reliability were available for perceived similarity, liking,
general performance and BES due to the fact that subjects rated three different ratees.
Reliabilities for the scales were calculated for each ratee and then averaged together. The
reliabilities of these scales are all considerably high, especially when one takes the small

number of items into account.

Table 3

§_c_2_rle Reliabilities

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Scale # of L.,,Hp Hme HmLp Average
items Reliability Reliability Reliability Reliability
Mastery Orientation 8 .82
Performance Orientation 8 .76
Self-esteem 10 .88
Perceived similarity 3 .87 .91 .90 .89
Liking 3 .88 .86 .84 .86
Generamerformance 5 .79 .87 .85 .84
BES - Research 9 .85 .80 .86 .83
BES - Assembly 13 .89 .90 .86 .88
UnderstandingAttributions 3 .86 .79 .73 .79

 

Note: BES = Behavioral Expectation Scale

45

Table 4 presents the correlations between the variables in the model. Most of the
significant correlations are between variables pertaining to the same ratee. Correlations
between variables measured with respect to different ratees do not correlate as frequently

or as highly.

46

Table 4

Correlations of Hypothesis-Relevant Variables Across Ratee

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Variable 1 2 3 4 5 6 7 8 9
lMO 1.00
2P0 .20** 1.00
3Self-Esteem .38** .02 1.00
4 Lme Global .04 .03 .15* 1.00
SLEHDBES .11 .10 .07 .53** 1.00
6L,,,Hp General .03 .08 .04 .61** .59** 1.00
7Lmlfi, Understand -.10 .12 -.09 .26** .48** .41** 1.00
8LE,Hp Reinforce —.01 .15* -.04 .24** .42** .34** .61** 1.00
9HmeGlobal .15* .04 .18* .21** .17* .22** .10 -.01 1.00
lOHmlfiBES .15* .03 .14 .11 .20** .15* .07 .00 .70**
11H,_,1Hp General .14* .08 .19** .19** .13 .25** .03 .01 .70**
12 HmHnUnderstand .25** .10 .06 .03 .13 .16* .24** .08 .62**
13 Hme Reinforce .14 .10 .06 .09 .08 .15* .10 .16* .53**
14HmliGlobal .06 .02 .03 -.04 -.01 .04 .07 .07 .19**
15 HmLp BES .08 .07 .06 -.05 .01 -.03 .02 .07 .20**
16 HELD General .12 .15* .13 -.03 .04 .06 .07 .105 .21**
17 HmLp Understand .12 .08 .03 -.ll .09 .02 .19** .19* .15*
18 HmLp Reinforce .05 .09 -.01 .04 .17* .07 .15* .24** .08

 

 

 

 

 

 

 

 

 

 

 

 

Note: M0 = Mastery Orientation, P0 = Performance Orientation, HmLp = High
mastery/low performance ratee, Hme = High mastery/high performance ratee, 1.,an =
Low mastery/high performance ratee, Global = Global performance measure, BES =
Behavioral Expectation Scale, General = General performance measure, Understand =
Understanding attribution, Reinforce = Reinforcement attribution

47

Table 4 (cont’d).

 

Variable 10 ll 12 13 14 15 16 17 18

 

lMO

 

2PO

 

3 Self-Esteem

 

4 Lme Global

 

5 Lmfip BES

 

6 L,,,Hp General

 

7 L,,,Hp Understand

 

8 Lme Reinforce

 

9 mm, Global

 

10 HEHp BES 1.00

 

11 Hme General .63** 1.00

 

12 Hme Understand .55** .55** 1.00

 

13 Hme Reinforce .50** .50** .60** 1.00

 

14HflLpGlobal .16* .24** .19** .24** 1.00

 

15 HELpBES .27** .22** .17* .26** .62** 1.00

 

16HmImGeneral .19** .35** .25** .24** .71** .73** 1.00

 

17HmLpUnderstand .09 .17* .24** .13 .64** .67** .66** 1.00

 

 

 

 

 

 

 

 

 

 

 

18 HmLp Reinforce .13 .15* .17* .27** .52** .58** .55** .61** 1.00
Note: M0 = Mastery Orientation, P0 = Performance Orientation, HmLp = High
mastery/low performance ratee, Hme = High mastery/high performance ratee, Lme =
Low mastery/high performance ratee, Global = Global performance measure, BES =
Behavioral Expectation Scale, General = General performance measure, Understand =
Understanding attribution, Reinforce = Reinforcement attribution

 

48

 

Repeated measures analysis of variance was used to analyze hypotheses 1, 2, and
5 with the ratee profiles used as a within subjects variable and rater mastery orientation,
rater performance orientation, rater self-esteem and job used as between subjects factors.
All of the analyzed variables were continuous with the exception of ratee profile and job.
Because three performance measures were included in this study, three different
ANOVA’s were run. The correlations between the performance measures are presented
in Table 5. The correlations for each ratee are presented in the first three columns and the
last column represents the average of these correlations. All of the correlations were
significant (p<.01), consistent with the idea that each represents the raters’

conceptualization of performance.

Table 5

Performance Correlations Across Ratees

 

 

 

 

 

14me flme HELD Average
1' lobal & general -61** -70** .71** .67
1' job,“ & BES .53** .20** .62** .45
[general & BES .59** .63** .73** .65

 

 

 

 

 

Note: HmLp = High mastery/low performance ratee, Hme = High mastery/high

performance ratee, LmHp = Low mastery/high performance ratee, ** = p<.01, BES =

Behavioral Expectation Scale

49

 

Hypotheses 1a and 1b predicted interactions between rater performance
orientation and ratee, and between rater mastery orientation and ratee. Specifically,
hypothesis 1a predicted that subordinates exhibiting performance behaviors would be
more highly evaluated by a highly performance oriented rater than by a rater low on
performance orientation. Similarly, hypothesis lb predicted that subordinates exhibiting
mastery behaviors would be more highly evaluated by a highly mastery oriented rater than
by a rater low on mastery orientation. The results of these analyses are presented in Table
6. None of the ANOVA’s indicated significant interactions between the mastery or
performance orientation of the rater and the orientation of the stimulus ratee. Thus,

hypothesis 1 was not supported.

50

Table 6

Repeated Measures AN OVA on Performance Rating

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source Sum of Squares df Mean Square F
DV = Global
Ratee 29.563 2 14.782 21 .678**
Ratee x MO .944 2 .472 .692
Ratee x PO .005 2 .002 .003
Ratee x SE 2.010 2 1.005 1.474
Ratee x Job 1.337 2 .669 .981
Ratee x MO x SE 2.447 2 1.223 1.794
Ratee x PO x SE .335 2 .167 .246
Ratee x Job x SE .149 2 .075 .109
Ratee x PO x MO .070 2 .035 .051
Error 260.474 382 .682
DV = General
Ratee 14.581 2 7.291 20.589**
Ratee x MO .589 2 .295 .832
Ratee x PO .302 2 .151 .427
Ratee x SE .630 2 .315 .890
Ratee x Job .489 2 .245 .691
Ratee x MO x SE 1.572 2 .786 2.220
Ratee x PO x SE .825 2 .413 1.166
Ratee x Job x SE .283 2 .142 .400
Ratee x PO x MO .299 2 .150 .423
Error 135.208 382 .354
DV = BES
Ratee .001 2 .001 .001
Ratee x MO .540 2 .270 .321
Ratee x PO .687 2 .343 .408
Ratee x SE .426 2 .213 .253
Ratee x Job .006 2 .003 .003
Ratee x MO x SE 3.488 2 1.744 2.073
Ratee x PO x SE 2.445 2 1.223 1.453
Ratee x Job x SE 3.772 2 1.886 2.242
Ratee x PO x MO .690 2 .345 .410
Error 321.355 382 .841

 

Note: M0 = Mastery Orientation, P0 = Performance Orientation, SE = Self-esteem, BES
= Behavior Expectation Scale, ** = p<.01

51

 

Hypothesis 2 predicted an interaction between job context and ratee such that the
degree of performance and mastery orientation of the ratee would interact with the job
context in which that ratee was being rated. Because the BES performance measure
differed between jobs, this measure was not used as a dependent measure in this analysis.

Neither of the two remaining ANOVA’s indicated significant interactions between job
context and stimulus ratee (see Table 6). Thus, hypothesis 2 was not supported.

Correlational analyses were run to assess the relationship of rater mastery and
performance orientation with self-esteem. Hypothesis 3 predicted that performance
orientation would be negatively correlated with self-esteem. This hypothesis was not
supported by the data (1:018, ns). Conversely, hypothesis 4 predicted that mastery
orientation would be positively correlated with self-esteem. This hypothesis was
supported ( p.38, p>.01).

Hypothesis 5 predicted a three way interaction between rater mastery orientation,
stimulus ratee, and rater self-esteem, and a second interaction between rater performance
orientation, stimulus ratee, and rater self-esteem. These analyses are also presented in
Table 6. The later interaction was clearly not significant in any of the three ANOVA’s
and thus this hypothesis was not supported. The former interaction was also not
significant, but all three ANOVA’s showed that this interaction was close to significant.
The probabilities for this interaction using the global, BES, and general measures were p
= .157, p = .107, and p = .055 respectively, suggesting the possibility of a weak effect.

Thus, hypothesis 5 remains unsupported, but closer inspection is warranted.

52

Hypotheses 6 and 7 predicted a relationship between the rater-ratee similarity and
the attributions made by the rater. Both of these hypotheses were examined using
repeated measures AN OVA’s to examine the hypothesized effects of the similarity
interactions (mastery orientation by ratee and performance orientation by ratee) on each
attribution.

Hypothesis 6 predicted that the understanding attributions would be more likely
when the ratee was similar to the rater. The ANOVA was used to examine the mastery
orientation by ratee and performance orientation by ratee interactions on this attribution.
The results of this analysis are presented in Table 7. The AN OVA supported the mastery
by ratee interaction but not the performance by ratee interaction. This interaction is

illustrated in Figure 3. The effect size for this interaction was 112 = .04.

Table 7

Repearted Measures ANOVA on Understanding

 

 

 

 

 

 

 

 

 

 

Source Sum of df Mean F
Squares Square
Ratee 32.19 2 16.09 34.05**
Ratee x MO 7.44 2 3.72 7.87**
Ratee x P0 .51 2 .25 .54
Error 186.22 394 .47

 

Note: M0 = Mastery Orientation, P0 = Performance Orientation, ** = p<.01

53

 

.93

 

 

 

5 _
.5 3 ~ 5 r‘
g //
'r: 2 5 ” / +High Mastery/ Low
2’ 2 Performance
3: * + High Mastery/ High
:5 1 5 A Performance
g ' + Low Mastery/ High
a 1 Performance
.. 1
i l
a 0.5

0 T .

Low High
Mastery Orientation

Figure 3
Ratee by Rater Mastery Orientation Interaction on Understanding

54

Hypothesis 7 predicted Byme’s explanation that rating someone similar highly
makes the rater feel good. Thus, the same interactions were hypothesized: a mastery
orientation by ratee interaction and a performance orientation by ratee interaction on this
attribution. This analysis is summarized in Table 8. The ANOVA did not support either

the mastery by ratee interaction or the performance by ratee interaction.

Table 8

Repeated Measures ANOVA on Reinforcement

 

 

 

 

 

 

Source Sum of df Mean F
Squares Square
Ratee 178.95 2 89.46 1 16.39**
Ratee x MO .35 2 .18 .23
Ratee x P0 .76 2 .38 .50
Error 301.30 392 .77

 

 

 

 

 

 

Note: M0 = Mastery Orientation, P0 = Performance Orientation, ** = p<.01

55

Hypotheses 8 and 9 predicted relationships between the attributions and the
performance ratings given by the raters. Repeated measures ANOVA’s were run in order
to examine these effects. Because there were three types of performance measures, these
analyses were run three times, using the global measure, the general performance measure
and the BES measures as the dependant variable.

Hypothesis 8 predicted that raters who report understanding a ratee would give
that ratee a higher performance rating. Thus this attribution should be positively related
to the performance rating. This ANOVA is summarized in Table 9. The hypothesis was
supported using all three performance ratings. For the global performance rating, the

effect size was 112 = .30. When raters indicated low understanding of the ratees’ actions,

the mean global rating was 1.93. When raters indicated high understanding of the ratees’
actions, the mean global rating was 2.76. For the general performance measure, the effect
size was also .30. When raters indicated low understanding of the ratees’ actions, the
mean general rating was 1.83. When raters indicated high understanding of the ratees’
actions, the mean global rating was 2.49. Similarity, for the BES measure, the effect size
was .33. When raters indicated low understanding of the ratees’ actions, the mean

(standardized) rating was -.50. When raters indicated high understanding of the ratees’

actions, the mean rating was .55.

56

Table 9

Repeated Measures ANOVA - Understanding on Performance Ratiflr

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source Sum of df Mean F
Squares Square

DV = Global
Subject 189.84 199 .95 2.00
Ratee 29.49 2 14.75 30.91**
Understanding 77.77 1 77.77 163.04**
Ratee x Understanding 1.36 2 .68 1.43
Error (Subject x Ratee) 189.38 397 .48
DV = General
Subject 131 .27 199 .66 2.66
Ratee 14.50 2 7.25 29.23**
Understanding 41.70 1 41.70 168.15**
Ratee x Understanding .27 2 .14 .54
Error (Subject x Ratee) 98.50 397 .25
DV = BES
Subject 207.88 199 1.05 2.13
Ratee 101.81 2 50.90 103.67**
Understandirm 94.33 1 94.33 192.12**
Ratee x Understanding 1.08 2 .54 1.10
Error (Suly'ect x Ratee) 194.90 397 .49

 

 

 

 

 

Note: BES = Behavior Expectation Scale, ** = p<.01

57

 

Hypothesis 9 predicted that those who report that rating the ratee highly would
make them feel good would, in fact, give that ratee a higher rating. This analysis is
summarized in Table 10. This hypothesis was also supported using all three performance
ratings.

For the global performance rating, the effect size was 112 = .19. When raters
indicated low reinforcement, the mean global rating was 2.10. When raters indicated high
understanding of the ratees’ actions, the mean global rating was 2.81. For the general
performance measure, the effect size was 112 = .21. When raters indicated low
reinforcement, the mean general rating was 1.96. When raters indicated high
understanding of the ratees’ actions, the mean general rating was 2.54. Finally, for the
BES, the effect size was 112 = .23. When raters indicated low reinforcement, the mean

(standardized) rating was -.30. When raters indicated high understanding of the ratees’

actions, the mean rating was .67.

58

Table 10

Repeated Measures ANOVA - Reinforcement on Performance Rating

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source Sum of df Mean F
Squares Square

DV = Global
Subject 189.90 199 .95 1.73
Ratee 29.32 2 14.66 26.51 **
Reinforcement 49.04 1 49.04 88.68**
Ratee x Reinforcement .67 2 .33 .60
Error (Subject x Ratee) 218.80 397 .55
DV = General
Subject 131.23 199 .660 2.35
Ratee 14.73 2 7.37 26.22**
Reinforcement 28.75 1 28.75 102.30**
Ratee x Reinforcement .12 2 .06 .21
Error (Subject x Ratee) l 1 1.20 397 .28
DV = BES
Subject 207.98 199 1.05 1.86
Ratee 101.68 2 50.84 90.63** -
Reinforcement 66.40 1 66.40 1 18.35**
Ratee x Reinforcement 1.60 2 .80 1.43
Error (Subject x Ratee) 222.28 397 .56

 

 

 

 

 

Note: BES = Behavior Expectation Scale, ** = p<.01

59

 

In sum, then, hypotheses la and b, predicting a relationship between actual
similarity and performance rating were not supported. Similarly, hypothesis 2, predicting
a relationship between ratee and job was not supported by the data. Performance
orientation was not found to be correlated with self-esteem, failing to support hypothesis
3; however, mastery orientation was positively correlated with self-esteem, supporting
hypothesis 4. Hypothesis 5 predicted an interaction between actual similarity and self-
esteem on performance rating. This hypothesis was not supported by the data. Partial
support was found for hypothesis 6, indicating that actual similarity was related to the
raters’ indication that he understood the ratee. Hypothesis 7, predicting a relationship
between actual similarity and the reinforcement attribution, was not supported. Both
attributions were significantly related to the performance ratings, supporting hypotheses 8

and 9.

Supplementary Analyses

Two variables not directly involved in the hypotheses were also measured: how
well the rater liked each ratee and how similar they perceived themselves to be to each
ratee. These variables may prove useful in understanding the failure of some of the
hypotheses by allowing us to examine the underlying processes. Actual similarity should
be related to perceived similarity. If they are not related, it would indicate that subjects
may not have been sensitive to the similarity manipulation. It is possible then, to replace

actual similarity with perceived similarity in the hypothesized model. Also, including

60

liking in the model will allow further examination of the processes underlying the
connection between similarity and ratings. A correlation table summarizing the

relationship of these additional variables with the other variables under study is presented

in Table l 1.

61

Table 1 1

Supplementary Correlations

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

LmHp L,.,Hp Hme Hme HmLp HmLp
Psim Liking Psim Liking Psim Liking
th Psim 1.000
EH, Liking .589** 1.000
Mp Psim .187** .075 1.000
__1-1_,,1Hp Liking .289** .209** .699** 1.000
fig Psim .l97** .143* .154* .206** 1.000
£11, Liking .167* .160* .138 .288** .673** 1.000
MO -.O85 .001 .282** . 150* .208** .097
PO .217** .103 .128 .102 .002 .088
SE -.222** -.084 .121 .065 .063 -.030
L",Hp Global .289** .244** -.024 .072 -.057 -.033
_I.,,,_Hp BES .575** .392** .105 .185** .116 .154*
Lme General .440** .371 ** .088 .235** .002 .129
LmHLUnderstand .763** .608** .134 .221 ** . 153* .189**
Lme Reinforce .605** .563** .094 .152* .137 .199**
Hme Global .111 .039 .542** .583** .121 .157*
Hme BES .090 .050 .503** .525** .102 .102
Hme General .039 —.014 .534** .561 ** .116 .162*
HmHgUnderstand .244** .121 .800** .716** .222** .237* *
Hme Reinforce .199** .099 .605** .618** .138 .123
Hmlfl, Global .055 .111 .140* .208** .561** .573**
HmLp BES .039 .028 .162* .206** .550** .558**
HmLp General .079 .077 .207** .243** .535** .614**
HmLp Understand .190** .121 .150* .169* .776** .746**
HmLp Reinforce .210** .109 .139* .120 .547** .580**

 

Note: Psim = Perceived similarity, HmLp = High mastery/low performance ratee, Hme =
High mastery/high performance ratee, Lme = Low mastery/high performance ratee, M0

= Mastery Orientation, P0 = Performance Orientation, SE = Self-esteem, BES =

Behavior Expectation Scale, * = p<.05, **=p<.01, Global = Global performance measure,

BES = Behavioral Expectation Scale, General = General performance measure,
Understand = Understanding attribution, Reinforce = Reinforcement attribution

62

 

To examine the failure of hypothesis 5, which predicted an interaction between
actual similarity (in terms of mastery and performance orientation) and self-esteem on
performance ratings, a similar analysis was conducted using perceived similarity instead
of actual similarity. This ANOVA is summarized in Table 12. In the analyses using the
global performance measure and the general performance measure, the interaction did not
quite reach significance (p=.06). For the BES scales, though, the interaction was

significant. The effect size, however, was small (1]2 = .01). These interactions are

illustrated in Figures 4, 5 and 6, respectively. Thus, substituting perceived similarity for
actual similarity in this case was partially successful. It is also interesting to note that

perceived similarity was significantly related to the performance rating given.

63

Table 12

SupplementagLRepeated Measures ANOVA - Performance Rating

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source Sum of df Mean Square F
Squares
DV = Global
Subject 189.836 199 .954 1.897
Ratee 29.490 2 14.745 29.314**
Psim 62.010 1 62.010 123.280**
SE .667 l .667 1.326
Psim x SE 1.857 1 1.857 3.692
Ratee x Psim 5.363 2 2.681 5.330
Error (Subject x Ratee) 198.614 395 .503
DV = General
Subject 131.272 199 .660 2.578
Ratee 14.496 2 7.248 28.312**
Psim 37.661 1 37.661 147.113**
SE .107 1 .107 0.418
Psim x SE 1.036 1 1.036 4.047
Ratee x Psim .593 2 .297 1.160
Error iSubject x Ratee) 101.074 395 .256
DV = BES
Subject 207.884 199 1.045 .377
Ratee 101.805 2 50.902 18.356**
Psim 81.873 1 81.873 29.525**
SE .550 1 .550 0.198
Psim x SE 2.773 1 2.773 5428*
Ratee x Psim 3.309 2 1.654 3.238*
flor (Subject x Ratee) 201.805 395 .511

 

Note: Psim = Perceived Similarity, SE = Self-esteem, BES = Behavior Expectation Scale

64

 

2.5 i

+ Low Self-esteem
+ High Self-esteem

Global Performance
a or

.0
or

 

 

l

Low High
Perceived Similarity

Figure 4
Perceived Similarity by Self-Esteem Interaction on Global Performance

+ Low Self-esteem
+ High Self-esteem

General Performance
01

.9
m _L

 

0

Low High
Perceived Similarity

Figure 5

Perceived Similarity by Self-Esteem Interaction on General Performance

65

 

P
m
44

0.6 4

P
.h
l_

+ Low Self-esteem
-- High Self-esteem

BES Rating
.<'> .<'> .6
# N O N
1 l 1 1

 

 

9595
com

Low High
Perceived Similarity

Figure 6
Perceived Similarity by Self-Esteem Interaction on BES Rating

66

In order to assess why actual similarity failed to lead to increased performance
ratings, the underlying processes were examined. Several analyses were conducted to
assess the similarity manipulation. First, the correlations between the perceived similarity

of each subject and the mastery and performance orientations of the rater were examined.

If subjects were sensitive to the similarity manipulation, the correlations between their
perceived similarity ratings and the rater’s orientation should follow the pattern indicated
in the top half of Table 13. When the target ratee is highly mastery oriented, the mastery
orientation of the rater should correlate with perceived similarity. Similarly, when the
target ratee is highly performance oriented, the performance orientation of the rater
should correlate with perceived similarity. Conversely, when the target is low on mastery
orientation, the mastery orientation of the rater should lead to a negative correlation with
perceived similarity, and when the target is low on performance orientation, the
performance orientation of the ratee should be negatively correlated with perceived
similarity.

An examination of the actual correlations found between these variables is

summarized in the lower half of Table 13. These correlations indicate that in four out of

the six cases, actual similarity led to perceived similarity. In the two remaining cases, a

negative relationship would be expected, but no relationship was found.

67

Table 13

Similarity Correlations
H
Ex r +
- +

Ex
.00

Actual 1‘ ,28** .21**
.13* .00 .22**

Actual r
Note: Psim = Perceived similarity, HmLp = High mastery/low performance ratee, Hme =
High mastery/high performance ratee, L,,,Hp = Low mastery/high performance ratee, * =

p<.O ,** = p <.01

 

68

Second, a repeated measures ANOVA was run to further examine the link
between actual and perceived similarity. . This ANOVA, summarized in Table 14,
examined the expected mastery orientation by ratee and performance orientation by ratee
interactions on perceived similarity. If subjects were sensitive to the similarity
manipulation, these interactions should be significant. The analysis showed both the
mastery orientation by ratee interaction and performance orientation by ratee interaction

to be significant. These interactions are illustrated in Figures 7 and 8 respectively. The

effect size for the mastery interaction was n2 = .05, and the effect size for the

performance interaction was 112 = .02.

 

 

 

 

 

 

 

 

 

 

Table 14
Remated Measures ANOVA - Perceived Similarity
Source Sum of df Mean Square F
Squares

Ratee 33.250 2 16.625 27.659**
Ratee x MO 11.412 2 5.705 9.493**
Ratee x PO 5.830 2 2.915 4.850**
Error 236.823 394 .601

 

 

 

 

Note: M0 = Mastery Orientation, P0 = Performance Orientation, ** = p <01

69

 

/ + High Mastery/ Low

Performance
+ High Mastery/ High
Performance

+ Low Mastery/ High
Performance

Perceived Similarity
.0 .-* 1° 9’
0 01 d 01 N 01 O) U! 45
1 J 1 l 1 1 l l

 

 

Low High
Mastery Orientation
Figure 7
Ratee by Rater Mastery Orientation Interaction on Perceived Similarity

 

4 l
3.5 l //
3 .1 ‘Vi #4
g | /4 + High Mastery/ Low
"5" 2.5 ‘1 Performance
CD 2 _j +High Mastery/ High
'8 1 Performance
3 1.5 l +Low Mastery/ High
§ Performance
0 1 l
9- l
0.5 i
o l — - i e e 4
Low High
Performance Orientation
Figure 8

Ratee by Rater Performance Orientation Interaction on Perceived Similarity

70

Further analyses were conducted to determine whether perceived similarity was
related to liking and performance ratings. If the failure of hypothesis 1 can be explained
by perceived similarity, then perceived similarity should be positively correlated with
liking and performance ratings. Table 15 illustrates the correlations between perceived

similarity, liking and the three performance measures. The first three columns indicate
the correlation for each ratee, and the final column illustrates the average correlation. The
table shows that the perceived similarity of a ratee was consistently correlated with how
much the subject liked the ratee. Accordingly, the perceived similarity of the ratee was
consistently correlated with the performance score given to that ratee. Finally, the degree
to which the subject liked the ratee was correlated with the performance score given to
that ratee. A note should be added here that the correlations in the low mastery/high
performance column appear to be lower than those for the remaining 2 ratees. This
mirrors the weaker effects found throughout this study with reference to performance
orientation as compared to the effects produced by mastery orientation (See tables 7 and
14)- This suggests that the performance orientation manipulation may have been weaker

than the mastery orientation manipulation.

71

 

Table 15

Correlations - Perceived Similarity. Liking, and Performance Ratings

 

 

 

 

 

 

 

 

 

F Hme Hmlfi, L.,,Hp Average
rPSim & Liking 599’” .673** .589** .654
Esau & Global .542** .561 ** .289** .464
rpsm & 353 .503** .552** .522** .526
rpsim & General .534** .535** .440** .503
rLiking& Global .583** .573** .244** .467
fungal eras .520“ .562** .370** .484
rukinga General .561 ** .614** .371** .515

 

 

 

 

 

 

 

Note: PSim = Perceived similarity, BES = Behavior Expectation Scale, HmLp = High
mastery/low performance ratee, Hme = High mastery/high performance ratee, LmHp =
Low mastery/high performance ratee, ** = p <.01

72

If hypotheses 6 and 7 can be explained in terms of perceived similarity, then
perceived similarity should be related to both attributions. The links between perceived
similarity and the attributions were examined through repeated measures ANOVA’s,
summarized in Table 16. Both of the attributions showed a significant relationship with

perceived similarity (p< .01). The effect size for understanding was n2 = .56, and the

effect size for reinforcement was 112 = .29.

When raters indicated low perceived similarity, the mean understanding rating
was 2.31 and the mean reinforcement rating was 2.59. When raters indicated high
perceived. similarity, the mean understanding rating was 3.39 and the mean reinforcement
rating was 3.58. Thus, the perceived similarity of the ratees to the rater was associated
with the degree to which the rater indicated an understanding of the ratee’s actions and a

perception that rating that ratee would make them feel good.

73

Table 16

Repeated Measures AN OVA Between Perceived SimiariLv and Attributions

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source Sum of df Mean F
Squares Square

Understanding
Subject 183.409 199 .922 4.288
Ratee 32.244 2 16.122 74.986**
Psim 108.599 1 108.599 505.111**
Ratee x Psim .298 2 .149 .693
Error (Subject x Ratee) 85.340 395 .215
Reinforcement:
Subject 230.485 199 1.158 2.838
Ratee 14.901 2 7.450 18.260**
Psim 75.255 1 75.255 184.449**
Ratee x Psim .107 2 .053 .130
Error (Subject x Ratee) 161.237 395 .408

 

Note: Psim = Perceived Similarity, ** = p <01

74

 

If liking mediates the relationship between the attributions made by the rater and
the performance rating given, then these attributions should be significantly related to
liking. The relationship between the attributions and the degree to which the raters
indicated they liked the ratees was similarly examined through a repeated measures
ANOVA. These analyses are summarized in Table 17. Both attributions were
significantly related to liking (p<.01). The effect size for understanding was 112 = .29, and

the effect size for reinforcement was 112 = .34. When raters indicated low understanding,

the mean liking rating was 2.44. When raters indicated high understanding, the mean
liking rating was 3.38. Similarly, when raters indicated low reinforcement, the mean

liking rating was 2.62, and when raters indicated high reinforcement, the mean liking

rating was 3.46.

75

Table 17

Repeated Measures AN OVA - Attributions on Liking

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source Sum of df Mean F
Squares Square

Subject 171.935 199 .864 3.323
Ratee 41.508 2 20.754 79.823**
Understanding 81.733 1 81.733 314.358**
Ratee x Understanding .190 2 .095 .365
Error (Subject x Ratee) 103.385 397 .260
Subject 171.927 199 .864 2.805
Ratee 41.258 2 20.629 66.977* *
Reinforcement 62.662 1 62.662 203.448**
Ratee x Reinforcement .546 2 .273 .886
Error (Subject x Ratee) 122.052 397 .308

 

 

 

Note: ** = p <01

76

 

Further, liking should be related to the performance measure given. Three
analyses were run to examine the relationship between liking and performance,
summarized in Table 18. Each of the performance measures, the global measure, the
general measure and the BES measure were significantly related to liking. The effect size

for the global performance rating was 112 = .21. When raters indicated low liking, the
mean global rating was 1.87. When raters indicated high liking, the mean global rating
was 2.65. For the general performance measure, the effect size was 112 = .24. Raters
indicating low liking gave an average general performance rating of 1.79. Raters
indicating high liking gave an average general performance rating of 2.40. Finally, the
effect size for the BES measure was .22. When raters indicated low liking, the mean

rating was -.53. When raters indicated high liking, the mean rating was .37.

77

Table 18

Repeated Measures ANOVA - Liking on Performance Rating

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Source Sum of df Mean F
Squares Square

DV = Global

Subject 189.836 199 .954 1.790
Ratee 29.490 2 14.745 27.664“
Liking 55.210 1 55.210 103.584**
Ratee x Liking 1.889 2 .944 1.771
Error (Subject x Ratee) 21 1.412 397 .533

DV = General

Subject 131.272 199 .660 2.463

Ratee 14.496 2 7.248 27.045**
Liking 33.716 1 33.716 125.806**
Ratee x Liking .430 2 .215 .802

Error (Subject x Ratee) 106.325 397 .268

DV = BES

Subject 207.884 199 1.045 1.830
Ratee 101.805 2 50.902 89.145**
Liking 62.218 1 62.218 108.963**
Ratee x Liking 1.525 2 .762 1.335
@or (Subject x Ratee) 226.567 397 .571

 

Note: BES = Behavior Expectation Scale, ** = p <01

78

 

Finally, a structural equation model was used to examine the relationships
between these underlying processes. The model encompasses three separate but
structurally identical models, one for each ratee. Figure 9 shows the structure of each
model. In the model, mastery and performance orientation (actual similarity) have a
direct effect on perceived similarity. The direction of these effects should differ by ratee.
For example, mastery orientation should be positively related to perceived similarity
when the target ratee is high mastery and negatively related to perceived similarity when
the target ratee is low mastery. Correspondingly, performance orientation should be
positively related to perceived similarity when the target ratee is high performance
oriented and negatively related to perceived similarity when the target ratee is low
performance oriented. This represents the expected relationship between actual and
perceived similarity. Perceived similarity, in turn, has a direct effect on both of the
attributions made by the rater, which have direct effects on liking. Liking then leads to

the performance rating.

79

Figure 9

 

 

matey—
oogéotom

 

 

 

 

 

 

 

 

waned
cop—«Eaten

 

 

 

 

 

 

 

 

macaw
ooze—Eaton

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

max: 953 was:

5:552 5:353. 5:353. 5:353. 8:352 8:352
EoEoocoEmem $656.52: EoEoocomEom 93.86895 Eon—08853— memeceumcoeeb
eeozém 55:5 eEEa
Guinean 3308?— wozoocom
- + + + + -
cons—Bro eozfleoto 5:955 .8385th 5:358 5:858

eunuch—atom beau—2 ooemncotom .0832 eager—oran— 5832
codex 83M 53m 33m 33m 58%
eager—atom 33 vegetated nmE oofigotom awn.—
Eeaz new E282 :me 3232 33

80

The overall model was originally constrained such that each path between the
variables was equal across all three ratees with the exception of the two paths from
mastery orientation and performance orientation to perceived similarity. These paths
were allowed to vary across models due to the fact that the expected direction of these
effects was different for each model. The overall model, then, tested the extent to which
the data fit the theoretical model across all three ratees.

The links between the hypothesized variables were all significant with the
exception of the link from performance orientation to perceived similarity for the high
mastery-low performance ratee. In that case, the expected negative relationship was not
found. Thus, the relationship between actual and perceived similarity held in five out of
the six possible cases.

The overall model was tested for each of the different performance measures. In
each case, the fit indices indicate that that the fit was reasonable. Table 19 illustrates the
Various fit indices examined for each overall model, the root mean square error of
approximation (RMSEA), the goodness of fit index, (GFI), and the comparative fit index
(CFI). The fit indices suggest that while the model is not perfect, it does a reasonable job

0f describing the data. The RMSEA, GFI, and CFI values are all within reasonable range.

The chi-square values are significant in each case; however, this may be due, in part, to

the large sample size used.

81

 

Table 19

Fit Indices for Supplementary Model

 

 

 

 

Model RMSEA GFI CFI Chi-square
BES .05 .90 .94 361.35**
General .05 .91 .95 320.37**
Global .05 .91 .95 334.79**

 

 

 

 

 

Note: ** = p<.01, degrees of freedom for Chi-square = 144

82

 

In order to check the assumption that the links between variables were
appropriately constrained, a revised model was tested in which none of these paths were
constrained. In other words, the path coefficients were allowed to vary across ratee. A
Chi-squared test was run to determine whether the models were significantly different.
When the BES scores were used as the performance measure, the original and revised
model were not significantly different, indicating that the relationship among constructs
did not change over ratee. However, when the global performance rating and the general
performance rating were used, the unconstrained model fit the data significantly better
than the constrained model, suggesting that at least one relationship was different across
ratee.

By constraining each path individually in turn, it was determined that the path that
differed was the path between liking and performance rating. In other words, for all other
paths, constraining the path to be the same across ratee did not lead to a significantly
different Chi-square value. A model in which the liking to performance path was
constrained functioned the same as the original model. That is, it was significantly
different from the revised model, but not significantly different from the original model.

Closer inspection revealed that for both the global performance rating and the
general performance measure, the liking to performance path was weaker when the ratee
was highly performance oriented. This is consistent with the point made in reference to
Table 15, in which relationships related to performance orientation were generally weaker

than those related to mastery orientation.

83

 

In sum, the supplementary analyses indicate that the links between the underlying
processes remain intact. That is, when one includes perceived similarity and liking in the
model, the connections between actual similarity and performance ratings can be
determined. Actual similarity leads to perceived similarity. Perceived similarity leads to
attributions of understanding and reinforcement. These attributions lead to liking, which

leads to increased performance ratings. This process holds for all three ratees.

Discussion

On the whole, the data support some of the stated hypotheses. Hypothesis 1,
predicting a relationship between actual similarity and performance ratings was not
supported by the data. Thus, the match between rater and ratee on mastery and
performance orientation did not have a significant effect on the rater’s evaluation of the
ratee’s performance. This failure will be discussed further with respect to the
supplementary analyses incorporating perceived similarity and liking.

Similarly, hypothesis 2, predicting a perceived match between ratee and job was
not supported. Raters were not influenced by the appropriateness of mastery and
Performance strategies for the two jobs examined. One possible explanation for this is
that the raters did not utilize any sort of matching judgment between job and employee.
In other words, they were using a general model rather than taking the particular job into
account. Another possibility is that there was a disagreement over the appropriate

strategies for the particular jobs in question. In other words, the raters may not have

84

believed that the mastery oriented behaviors were more appropriate for the research job
and that the performance oriented behaviors were more appropriate for the assembly line

job. While the pilot subjects indicated that the jobs were better suited for their respective
approaches (i.e. a mastery approach for the research job and a performance approach for
the assembly job), it is important to note that the pilot subjects had received training on
mastery and performance orientation while the actual subjects had not. This training may
have cued them towards that type of judgment.

The interaction proposed in hypothesis 5 between actual similarity and self-esteem
also failed. Ratees were not differentially susceptible to the effect of actual similarity on
performance rating. One possibility for this failure is that competing forces were at work.
Subjects with low self-esteem may have been more motivated to see similar ratees
positively, but high self-esteem subjects may also have had a reason to give high ratings.
High self-esteem subjects, by definition, like themselves. If a ratee is similar to the high
self-esteem rater, the rater may also find them likable. Another possibility is that this

effect is contingent upon the rater’s making the connection between actual and perceived
similarity. The latter possibility is addressed in the supplementary analyses and will also
be discussed further.

As reported in previous literature, a correlation was found between mastery
orientation and self-esteem, but the counterpart relationship between performance
orientation and self-esteem was not found. Highly mastery oriented subjects had higher
selfresteem than those low on mastery orientation. The theoretical explanation for this

effect presupposes that mastery orientation causes an increase in self-esteem, but it is

85

important to bear in mind that this study did not examine causal order. Thus, it may be
the case that having high self-esteem allows a person to focus on self-worth and therefore
want to increase their abilities (i.e. learn).

The effect of actual similarity on the attributions was mixed. Byme’s hypothesis
that similarity leads to liking because rating a similar person highly is reinforcing to the
rater is not well supported by this particular data set when actual similarity is used. While
a significant relationship was found between this attribution and the performance rating
given (H9), the corresponding relationship between actual similarity and the attribution
was not found (H7). Thus, while subjects who reported that rating a ratee highly would
be reinforcing did, in fact, rate that ratee higher, this reinforcement was not affected by
how objectively similar the ratee was to the rater. As will be discussed below, this failure
disappears when perceived similarity is included.

A similar problem was found when examining the additional explanation
hypothesized here that it is an increased understanding of the ratee by the rater that
mediates the similarity-attraction relationship. This attribution was found to be related to
the performance rating (H8), but only partial support was found for the hypothesis that
this attribution was influenced by the actual similarity between rater and ratee (H6).
When raters reported that they understood the ratee, they gave the ratee a higher
Performance rating, but understanding was only partially related to how objectively
Similar the ratee and rater were.

The supplementary analyses, incorporating perceived similarity, may provide

ClueS as to why many of these hypotheses failed. By exploring the processes in terms of

86

perceived similarity, a better understanding may be obtained for their failure. The re-
examination of hypothesis 5, substituting perceived similarity for actual similarity did not
clearly account for the original failure of the self—esteem by actual similarity interaction.
Although there appeared to be a tendency in the hypothesized direction, this effect did not
reach significance in two of the three analyses and therefore cannot be used to completely
explain the failure. The fact that these two analyses indicate that the effect bordered on
significant and that the third one did, in fact, reach significance, suggests that the effect

was simply too small to reach significance. The effect sizes for this interaction were, in

fact, small (7]2 = .01).

The additional variables proved more helpful, though, when exploring the failure
of hypotheses 1a and b. The analyses incorporating perceived similarity and liking
indicated that these variables were related to the performance ratings given. Further, each
of the variables examined (See Figure 3) was significantly related to the theoretically
adjoining variables on either side. There was partial support that the subjects were
sensitive to the actual similarity manipulation, as indicated by the link between actual and
perceived similarity and the ANOVA analyses detailed in Table 8 and by the significance
of these links in the structural equation model. Perceived similarity was significantly
linked to the understanding and reinforcement attributions, and these attributions were
linked to liking. Liking was significantly related to performance ratings.

It is likely, then, that the lack of a relationship between actual similarity and
performance rating can be attributed to two sources. First, there was a less than perfect

1‘81ationship between actual and perceived similarity. Subjects appeared to be mostly

87

sensitive to the mastery orientation and performance orientation manipulations made, but
in two of the six cases, were not. Second, the relationships between the variables appear
not to have been strong enough to carry from the beginning of the model to the end
significantly, as indicated by the failed interaction for hypothesis 1.

Some implications of this study are clear. First, including perceived similarity
and liking in the model allows a better understanding of the connection between actual
similarity and performance ratings. Actual similarity leads to perceived similarity, which
leads to the understanding and reinforcement attributions. The attributions lead to
increased liking of the ratee which leads to increased performance ratings. Thus the
processes remain intact, regardless of how easily they can be detected.

There are many implications stemming from the inclusion of attributional
processes in the model. First, perceived similarity was directly related to the attributions
made. Raters who perceived the ratees to be similar to themselves reported that they
understood those ratee and that rating them would be reinforcing. These attributions, in
turn, were directly related to liking. When the raters understood the ratees and felt that
rating them highly would be reinforcing, they liked them better. Also, liking was directly
related to performance ratings. Raters gave higher ratings to those ratees who were liked.

These were not, however, the only influences on the performance rating. One can safely
presume that performance ratings are affected by other, more obviously relevant factors
Such as the quality of work produced.

It is unfortunate that the actual similarity manipulation used in this study, while

mostly successful in affecting perceived similarity, did not prove to be strong enough to

88

 

have a practical impact on performance ratings. According to these analyses,
manipulating the actual similarity between rater and ratee would not have a perceivable
effect on the performance rating given. It is possible that this is due, in part, to the fact
that the raters were presented with ratees who were fictional and with whom the ratees
had no previous or future contact. The raters in this case were given very limited
information and that may have proved to not be enough in terms of forming lasting
impressions.

Whatever the reason, the implications of actual similarity in this study will have to
be confined to its theoretical effects on the other variables studied rather than on
immediate, practical results. The raters in this study were sensitive to the manipulation of
goal orientation. In general, they were able to examine the work related behaviors of the
ratees and determine the similarity of these behaviors to their own work style, as
indicated by the rater by ratee interaction on perceived similarity. Thus, subjects were

willing to use goal orientation as the basis for these judgments when given the

opportunity to do so.

The attributions made by the rater provide more information about the underlying
processes of the similarity-attraction phenomenon. Raters made specific attributions

about ratees when they felt the ratees were similar to themselves. The data support both

the reinforcement theory and the fact that perceived similarity led to increased

Understanding of the ratee. These attributions, then, were significantly related to

Performance ratings.

89

Underlying processes have been mostly ignored in recent literature on the
similarity-attraction phenomenon. Most researchers have been content to accept the
phenomenon as a whole and examine its practical implications without exploring the
processes involved (e.g. Golightly, Huffman and Byme, 1972; Good and Good, 1974;
Griffitt and Jackson, 1970). The exception to this is Byme and his colleagues’
explanation that the similarity-attraction phenomenon stems from reinforcement (Byme
& Clore, 1967; Byme & Griffitt, 1969; Byme & Nelson, 1965).

The current study indicates that the underlying processes may be more complex
than originally thought. While the reinforcement hypothesis supported by Byme and his

colleagues accounts for part of the link between perceived similarity and liking,
additional variance is accounted for by the rater’s increased understanding of the ratee.
Thus, this study is consistent with Byme’s previous research supporting reinforcement,
but adds to by providing an additional mechanism through which similarity affects liking
and performance.

One implication of this is that ratees who are perceived as dissimilar to the ratee
may be at a disadvantage because the rater does not understand the ratee’s actions. This
suggests that the link between perceived similarity and liking may be reduced if the ratee
can provide the rater with some sort of explanation or rationale for his behavior.

There are several limitations of this study. First, the subjects were not actual
managers rating actual employees in a business setting. While they were asked to play
that role, they may or may not have been successful in emulating real managers. Thus,

While the general tendency towards these processes may be present, managers may have

90

 

characteristics that allow them to overcome them. Second, the model explored in the
supplementary analyses indicates that there is some room for improvement. Other
processes, besides those explored here, may further explain the relationships between
these variables.

Future research in this area should be extended to a business setting, where real
supervisors and subordinates can be used. It is possible that in a real setting, where rater
and ratee have more prolonged interaction, the rater may know the ratee’s working style
better, and this style may be a more critical component of the evaluation process.
Conversely, other factors may serve to suppress these processes.

Also, future research may attempt to examine the attributions made in a less direct
manner. For example, if perceived similarity causes increased understanding, subjects
might indicate more confidence in predicting future actions of the ratee.

Finally, some attention may also be paid to the appropriateness of the attributions

made. It is reasonable that a supervisor’s understanding and agreement with a
subordinate’s choice of work behaviors may in some cases be rightfully linked to good
performance. A supervisor should, hopefully, recognize productive behaviors when he
sees them. It should also be recognized, though, that these attributions might be subject
to influence by irrelevant factors. If this is the case, then these attributions should not be
seen as undue influence, but may perhaps be used to improve and refine performance

measures.

91

APPENDICES

92

APPENDIX A

Job Performance Descriptions

High pgrformance, low masteg:
Terry picked this job because he thought he could do well at it.

On the job,

He does everything possible to make sure finished products contain no flaws.

He asks how well others have done at tasks.

He cares too much about how well he does things when other people are watching.
He always wants to know how well other people are performing on the same job.
He approaches the job in the manner in which he thinks he will perform it the best.
He is pleased when he performs well.

He believes that his failures are a result of low ability.

He does not ask very many questions.

He enjoys finding out that he is performing better than other people.

He worries about what other people will think.

He tries to avoid mistakes at all costs.

He tries to gain favorable judgments of his work.

He seeks out alternative tasks on which to work when the current task is not going well.
He is not interested in learning about other aspects of the company.

He gets frustrated when he performs badly.

When given the opportunity to choose a project, he chooses one on which he is confident
he can perform well.

High mastery. low performance:

John picked this job because it would give him the opportunity to learn a lot.

On the job,

He enjoys taking on new/different tasks.
He concentrates on improving.

He Wastes time trying to understand things when they are not necessary to complete the
job.

He approaches the job in the manner he thinks will make him learn the most.
He asks a lot of questions centered around understanding the job.

93

(APPENDD( A Cont’d)

He is pleased when he works hard and learns a lot.

He changes strategies when he is failing.

He wants to understand the jobs of the people around him (even if he doesn’t need to in
order to do his job.)

He seeks challenges.

He is not bothered when his performance is less than perfect.

He tries to understand the job in detail.

He picks projects based on how much can be learned from them rather than how useful
they would be or how important they would be.

He is persistent.

He is satisfied if a goal is partially attained if it means that he has increased his skill level.

He does not compare his performance with anyone else’s.
He frequently interrupts his coworkers to ask them questions.

When given the opportunity to choose a project, he chooses one on which he can learn.

H'gh mastery, high performance:

Frank picked this job because he thought he could do well at it, and it would give him the
opportunity to learn a lot.

On the job,

He works hard to ensure that finished products contain no flaws.

He enjoys taking on new and/or different tasks.

He asks how well others have done at tasks.

He concentrates on improving.

He cares too much about how well he does things when other people are watching.

He wastes time trying to understand things when they are not necessary to complete the
job.

He wants to know how well other people are performing on the same job.

He asks a lot of questions centered around understanding the job and tries to understand
the job in detail.

He approaches the job in the manner in which he thinks he will perform very well while
learning a lot.

He is pleased when he works hard, learns a lot and performs well.

He believes that any of his failures are partially a result of low ability, but combats this by
Changing the strategies he is using.

He enjoys finding out that he is performing better than other people.

He wants to understand the jobs of the people around him (even if he doesn’t need to to
do his job.)

He worries about what other people will think.
He seeks challenges.

94

E: 3
:

 

 

(APPENDD( A Cont’d)
He tries to gain favorable judgments of his work and to avoid mistakes.

When given an opportunity to choose a project or task, choose one from which he can
both learn and perform well.

Sometimes when the current task is not going well, he seeks out alternative tasks, and
sometimes he persists in the one he’s working on.

Distractor 1:
Bill picked this job because it paid well.

On the job,

He concentrates on living up to his responsibilities.

He does everything possible to make sure projects are completed on time.
He believes that his failures indicate that he didn’t try hard enough.
He asks the advice of his co-workers when things are not going well.
He is pleased when he finishes quickly.

He approaches the job in a tested and proven manner.

He looks for promotion opportunities.

He plans out each project carefully.

He approaches tasks carefully.

He attempts to complete tasks quickly.

He tries to be consistent.

He is occasionally unfriendly.

When given an opportunity to choose a project or task, chooses the one that seems the
most important.

Distractor 2:

Chris picked this job because it interested him.

On the job,

He changes strategies periodically.

He enjoys finding out that his work makes a difference.

He wastes time talking to co-workers.

He gets worried when things are not going well.

He is pleased when he completes a task without any help.

He approaches the job in the way most other people do.

He is helpful to his co-workers.

He uses many different resources when gathering information.

95

(APPENDIX A Cont’d)

He approaches tasks confidently.

He attempts to make the best use of his resources.
He tries to balance his various responsibilities.
He takes a long time to make decisions.

When given an opportunity to choose a project or task, he chooses the one that interests
him.

96

 

APPENDIX B

Job Descriptions

Manufacturing Job

This employee works in an automobile manufacturing plant. Various workers work on an
assembly line. Each person on the assembly line has a different job that he or she repeats
over and over. This person’s job is to assemble the brakes to be installed in cars. He is
responsible for making sure that there are no defects in the braking systems. There are
several people who also do this job, but they work on other shifts. Thus, they do not

check or duplicate his work.

Research Job

This employee works for a chemical company. One division of the company designs and
manufactures different types of plastics. For example, some products require plastics that
need to be more durable than others. Some require very hard plastics while others require
a more malleable plastic. This person’s job is to create and develop new types of plastics
to be used for a variety of products. This means that he must discover ways in which
plastics can be created that combine all the desired characteristics required by particular
products. There are several people who also do this job, but each of these people works
independently of the others. Thus, they do not check or duplicate his work.

97

 

APPENDIX C

Goal Orientation Measures

Iprefer to do things that I can do well rather than things that I do poorly.

I’m happiest at work when I perform tasks on which I know that I won’t make any errors.
The things I enjoy the most are the things I do the best.

The opinions others have about how well I can do certain things are important to me.

Ifeel smart when I do something without making any mistakes.
Ilike to be fairly confident that I can successfully perform a task before I attempt it.

Ilike to work on tasks that I have done well on in the past.
I feel smart when I can do something better than most other people.

The opportunity to do challenging work is important to me.

When I fail to complete a difficult task, I plan to try harder the next time I work on it.
Iprefer to work on tasks that force me to learn new things.

The opportunity to learn new things is important to me.

I do my best when I’m working on a fairly difficult task.

I try hard to improve on my past performance.
The opportunity to extend the range of my abilities is important to me.
When I have difficulty solving a problem, I enjoy trying different approaches to see which

one will work.

98

 

 

 

APPENDIX D

Self-Esteem Measure

Ifeel that I’m a person of worth, at least on an equal plane with others.
Ifeel that I have a number of good qualities.

lam able to do things as well as most other people.
Ifeel I do not have much to be proud of.

Itake a positive attitude toward myself.

At times I think I am no good at all.

All in all, I am inclined to feel that I am a failure.
0n the whole, lam satisfied with myself.

I wish I could have more respect for myself.
Icertainly feel useless at times.

99

 

 

 

I. Ivr'ii

APPENDIX E

Performance Measure for Brake Manufacturing Job

Inspection and Maintenance

Inspects machines on a regular basis and performs routine maintenance (i.e., oiling,
sharpening, changing molds, replacing old or worn parts) using various tools and

measuring devices to ensure safety and optimum operation.

 

The statements to
the right are
examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Makes sure a failed part gets reordered and
records all maintenance done including all
relevant information. Does work in an area
without leaving any kind of a mess or slowing
down the work of others. Does work thoroughly
so that it lasts for a long time.

 

 

 

The statements to
the right are
examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Sometimes gets distracted from work, but usually
does good work. Follows maintenance schedule
but doesn’t go beyond.

 

 

 

The statements to
the right are
examples of
behavior that one
would eXpect of
individuals usually
rated “low” on this
dimension,

 

Typical Low Behavior
Fails to notify others of dangerous situations.
Fails to notify next shift of what remains to be
done. Fails to record work done, thus causing the
work to be done twice.

 

 

 

 

(APPENDIX E Cont’

Diagnosis

Inspects product for errors in meeting specifications such as width, smoothness, and
functionality using both observation and measuring devices and machine gauges to ensure

d)

machine is operating in a way that produces quality product.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Assesses quality of machine functioning quickly
and accurately. Tries to improve quality even
though specifications are being met. Careful to
take accurate readings of machine calibration and

functioning. Feels personally responsible for error
detection.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior

Does product inspections, but occasionally fails to
gauge or measure product. Usually attends to
detail, but sometimes misses problems.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

Typical Low Behavior

Doesn’t check after completion to be sure machine
is working properly, just installs a part and leaves.
Relies on others to catch errors. Doesn’t use all
sources of information.

 

 

101

 

 

 

 

I l

(APPENDIX E Cont’d)

Calibration

Reads production orders to determine quantity and type of product needed and then sets
up/calibrates machines to make different products (i.e. different in size, or shape) using

various hand tools, knowledge of specifications, and operator’s manuals.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Checks schedules of maintenance at beginning of
shift, then plans schedule to get important things
completed. Knows the particular problems to be
addressed and makes sure appropriate parts and
tools are available before beginning.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Verifies part availability before beginning work on
a machine. Neglects repair on a machine because
he/she doesn’t think it will be needed on that shift.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

Typical Low Behavior
Jumps in and starts with little or no planning or
attention to when the machine is needed for
production. Fails to make sure all needed parts are
in stock before beginning.

 

 

102

 

 

 

 

 

.\e‘

If

 

In!

V

 

 

(APPENDIX E Cont’d)
Communication
Records maintenance done on equipment using standard forms to ensure that a record of

problems and maintenance performed is available when later inspections are done or
problems develop. Records defective product and routes to scrap as appropriate.

 

 

 

 

 

 

 

   
   

The statements to the Typical High Behavior
right are examples of In addition to putting date and operation on the
behavior that one legend, also puts any other information that might
would expect of be useful. Detailed and precise in filling out 5
individuals usually records. Communicates problems to the next shift
rated “high” on this both verbally and in written form when necessary.
dimension.

4
The statements to the Typical Average Behavior
right are examples of Doesn’t pay attention to the entire work process.
behavior that one Focuses only on present task. Records only the 3
would expect of major work that is done or only the date at which
individuals usually the equipment was inspected.
rated “average” on
this dimension.

2
The statements to the Typical Low Behavior
right are examples of Wastes the time of others by keeping records on
behavior that one their own tool cart or in other inconvenient places. 1

would expect of Fails to record the work that is done. Fails to put
individuals usually parts back where they belong.

rated “low” on this
dimension.

 

 

 

 

 

 

103

 

i.-

 
 

 

2.!

‘lgs

 
 

\

 

(APPENDD( E Cont’d)

Preventative Maintenance

Monitors equipment and performs necessary maintenance on and makes adjustments to
equipment using knowledge of the equipment and its malfunctions, necessary tools and

diagnostics, and operator’s manuals.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Knows and is aware of entire systems instead of
focussing only on the specific part that is being
worked on. Does things that aren’t necessarily
his/her job. Prioritizes work in terms of criticality.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Only uses operator’s manual when in trouble
instead of using it to learn about the equipment.
Uses diagnostics only when needed. Pays some
attention to detail, but doesn’t go out of way to see
that a machine is running maximally.

 

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

Typical Low Behavior
Puts more pride in tools than in machine being
worked on. Does not ask question when he/she

doesn’t know something.

 

 

104

 

 

 

 
 
 
 

 

(APPENDIX E Cont’

Housekeeping

Cleans equipment or tools using solvents and scrubbing materials as part of regular
maintenance. Cleans work area and/or parts according to guidelines to maintain a safe

(0

and orderly work area and clean parts.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Leaves an area just as clean as it was before work if
not cleaner. Removes all useless materials from
area. Knows proper materials for cleaning.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Doesn’t always communicate problems or dangers
to others. Uses appropriate cleaning equipment
and materials. Doesn’t pay that much attention to
prioritizing work in terms of importance.

 

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

Typical Low Behavior
Does no more than is explicitly state. Ignores
cleaning problems. Leaves grease and oil on
vehicles.

 

 

 

 

105

t“t

is
c-179- .

 

“In 11.5» «Mia... JV. —.a .. u.m.. wry. .I. .ln. ut- ..-._ _-.h min hrs. .1 . ,V .4... n - ~\l;l. . .. oil. ,I. m._. are ....l.u\

 

(APPENDIX E Cont’d)

Maintaining a Safe Work Environment

Works in potentially dangerous situations using established safety procedures to ensure

safety.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Always uses fire and safety procedures regardless
of time constraints. Sees potential dangers and
informs supervisor. Tests systems and looks for
problems that might produce dangerous situations.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Sometimes overreacts to dangers. Usually
communicates safety problems to supervisor.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

 

Typical Low Behavior
Fails to read safety instructions, thus creating
dangerous situations. Uses safety procedures as an
excuse to avoid work. Uses much more detergent
than is necessary to get the job done.

 

 

 

106

“‘L. 1*?! .

.’.'9?11
.1
. W

4
1 .
9L» w my.

 

(APPENDIX E Cont’d)

Maintains A Good Work Relationship

Discusses work-related issues with supervisor and other work group members using
knowledge gained from work experience to improve work unit efficiency or quality, or to

solve work-related problems.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Exercises tact in bringing up problems when group
members are affected. Takes initiative to see that
suggestions get canied out. Uses all available
evidence to convince people that he/she has a better
way to do things.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Develops new work procedures, but doesn’t share
them. Takes initiative only if it is beneficial to
himself/herself. Sometimes fails to address
problem directly, but makes some effort to
communicate problems and solutions.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

 

Typical Low Behavior
Disrupts meetings by changing topics or by
horseplay. Doesn’t ask questions even if unsure of
procedures. Upsets people instead of trying to
help. Complains about work but doesn’t do
anything to change it. Leaves area in a mess so that
accidents are more likely to happen.

 

 

107

 

(APPENDD( E Cont’d)
Performing Mechanical Work
Uses hands and various tools to perform many basic mechanical operations (e. g.

hammering, tightening/loosening nuts, drilling, soldering, etc.). Displays mechanical
ability in performing these operations. Displays an understanding of how various

machines work.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Tightens/loosens nuts, screws, etc. faster than most
people. Knows proper single-ended wrench size
just by looking at nut. Uses hands and tools
skillfully and effortlessly. Understands how
various machines and tools work with minimum or
no explanation or by reading available manuals.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Tightens/loosens nuts, screws, etc. with average
speed. Displays slight awkwardness in using hands
and tools. Can be taught how to repair new
machines or use new tools and procedures.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

Typical Low Behavior
Tightens/loosens nuts, screws, etc. relatively
slowly. Has difficulty learning how to repair new
machines even with instruction. Occasionally has
accidents or makes a mistake in repairing a
machine.

 

 

108

 

 

APPENDD( F

Performance Measure for Research Engineering Job

Technical Communication

Transmits written and oral information related to technical projects and assignments.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Can sell, orally and in writing, a technical
improvement to management which is initially
opposed to change. Is able to instruct other
engineers in new technology. Tailors written and
oral technical presentations to fit the audience.
Documents difficult technical material effectively.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Reports only the core information pertinent to the
problem at hand. Communicates well only with
engineers within his or her specific technical
discipline.

 

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

Typical Low Behavior
Writes technical reports which are too wordy.
Rarely contributes information to engineering staff
discussions of technical problems. Frequently has
a proposed project rejected by a manager because
of a poor presentation. Has difficulty explaining
technical results.

 

109

 

 

(APPENDIX F Cont’d)

Gathering Technical lnforrnation from Others

Seeks appropriate others in the business for guidance, advice and reaction to one’s own

approach to dealing with technical problems.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Asks questions of technical experts to obtain the
appropriate and needed information. Willingly
asks others for help, but only on significant or
difficult problems. Appropriately expands network
of people who can be contacted for advice. Knows
the appropriate sources of technical information.

 

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Occasionally doesn’t know where to go for help or
what to request. Occasionally has to be encouraged
to seek advice of others. Seeks advice of others but
sometimes ignores their good suggestions. Limits
efforts for obtaining advice to immediate
supervisor and close co-workers.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

Typical Low Behavior
Seldom asks others for advice; often has to be told
to do so. Often assumes that the help of others
cannot be obtained or that they can be of little help
even if asked. Almost always asks for help on all
technical problems.

 

110

 

 

(APPENDD( F Cont’d)

Scientific and Technical Knowledge

Possesses fundamental scientific, mathematical and engineering knowledge necessary for

adequate completion of a project or assignment.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Is considered the technical expert in the department
or component of the organization. Uses the latest
technical principles, rather than cookbook
formulas, to perform tasks. Performs assignments
with minimal technical assistance.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Has a good working knowledge of the applicable
technology. Has fundamental grasp of engineering
principles. Is aware of relevant technology
advances.

 

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

Typical Low Behavior
Is unfamiliar with the precise definitions of many
technical terms. Often needs technical help from
colleagues in order to complete an assignment.
Has difficulty understanding basic engineering
designs.

 

111

 

 

(APPENDIX F Cont’d)

Organization and Planning

Manages projects and assignments including establishing priorities, meeting deadlines,

and attending to details.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Is able to establish or appropriately change project
priorities without the specific guidance of the
supervisor. Prepares schedules identifying project
milestones as well as contingency plans. Offers
ideas developed from current projects as proposals
for possible future projects.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Usually can reorganize a project due to schedule or
technical specification changes. Is able to use
systematic scheduling procedures. Sometimes
spends too much time on details.

 

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

Typical Low Behavior
Frequently misses project deadlines. Does not
follow up through implementation after the major
components of a project are completed. Drags out
each assignment to the maximum. Uses excessive
manpower and equipment resources due to poor
project management.

 

112

 

 

(APPENDIX F Cont’d)

Problem Recognition and Definition

Understand the cause(s) of a problem.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Is able to recognize quickly the existence of a
technical problem before all the negative symptoms
are apparent. Is able to distinguish between
symptoms and causes of a problem. Is able to
identify a specific problem as being an example of
a general class of problems which has certain
possible solutions. Goes to the location of a
problem to get direct information about it.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Usually determines the cause of a problem as the
project progresses. Considers other people’s
opinions about the cause of a problem. Is often
unwilling to offer a tentative diagnosis of a
problem based on observable symptoms.

 

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

Typical Low Behavior
Often attacks the first symptoms of a problem,
rather than looking for its real causes. Often
misses one or two important factors in a problem.
Usually is not able to see which problem symptoms
are related to each other and treats each symptom
as if it were a separate problem.

 

113

 

 

(APPENDIX F Cont’d)

Development of Alternative Solutions

Creates several technically feasible solutions to a problem.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Appropriately looks for better ways to do a job.
Presents to management several alternate solutions
to a technical problem and justifies the
recommended alternative. Creates imaginative
solutions to long-term problems.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Sometimes offers several solutions to a technical
problem for management to choose from.
Develops another approach to a problem only when
the current approach fails. Occasionally requires
prompting by the supervisor to look for more than
one possible solution to a problem.

 

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

Typical Low Behavior
Doesn’t consider challenging the “status quo” of a
traditional approach to an engineering problem.
Will propose and defend the first solution to come
to mind.

 

114

 

 

(APPENDIX F Cont’d)

Evaluation of Alternative Solutions

Uses theoretical, analytical, and empirical methods to determine the likely consequences

of alternative solutions.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Selects a solution based upon well-documented and
thorough analysis. Quantifies all known pro’s and
con’s associated with the possible consequences of
proposed solutions. Quickly finds the strong and
weak points of alternatives.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Does not prejudge any possible solution before the
evaluation data are complete. Evaluation of
alternatives is limited to obvious or conventional
considerations.

 

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

Typical Low Behavior
Sometimes cannot point out the comparative
advantages and disadvantages of two alternatives.
Occasionally ignores or fails to consider some
significant data when evaluating alternative
solutions.

 

 

 

115

..,..., traces-ts. -r_

 

w .445... can

(APPENDD( F Cont’d)

Implementation of Chosen Alternative

Is able to make an alternative operational by fitting the solution to the particular situation.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Anticipates implementation problems and plans for
their solution. Accepts minor changes in a problem
solution in order to gain its implementation without
compromising the design or business objectives.
Usually overcomes small obstacles to the
implementation of a solution

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Makes an idea operational although it may not
function at rated capacity. Forces the chosen
alternative solution into operation, compromising
some of the desired goals of the project.

 

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

Typical Low Behavior
Expects every solution to work as smoothly as
possible upon implementation. Tries to implement
a new procedure before being sure that operating
management fully understands it. Sticks with the
original solution longer than its performance
justifies. Rigidly adheres to textbook solutions

 

without considering the specific situation.

 

116

 

 

 

(APPENDIX F Cont’d)

Responding to Change

Accepts and actively participates in changes in the way work is performed.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Actively seeks available information to understand
the change. Anticipates and works to minimize the
disruptive aspects of the change. Advocates the
benefits of the change to others. One of the first to
participate in or use the changed approach/method.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Accepts given explanation of reason for change.
Tolerates and occasionally tries to overcome
disruptive effects of change. Is uninterested in
discussing the pro’s and con’s of the change.
Participates in the changes as it becomes necessary.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

Typical Low Behavior
Does not accept the given explanation for the
change. Disrupts implementation of change by
trying to cling to old approach/method. Complains
to others about the disadvantages of the change.
Participates in the change only if told to do so.

 

117

 

 

 

(APPENDIX F Cont’d)

Professional Activities

Participates in professional activities.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Seeks leadership roles in professional societies.
Teaches a technical refresher course for the local
professional society chapter. Frequently presents a
paper at regional or national technical society
meeting.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Attends most chapter meetings of the technical
society. Has never submitted a paper for
presentation at a technical society meeting.

 

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

Typical Low Behavior
Joins professional societies only if encouraged to
do so. Does not attend professional society
functions.

 

118

 

 

(APPENDIX F Cont’d)

Continuing Education Activities

Uses learning procedures to maintain or obtain up-to-date technical skills.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Has made definite plans for self-development in
technical areas. Devotes a substantial portion of
spare time to reading technical publications and
taking technical courses. Completes university
courses on advanced technical topics. Completes
as many relevant company-sponsored technical
seminars and short courses as possible.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Completes most technical courses or seminars held
outside the company. Completes relevant technical
courses only if encouraged by others to do so.

 

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

Typical Low Behavior
Expects the organization and its management to
initiate continuing education efforts. Frequently
content to rely upon co—workers for learning about
new techniques. Never attends an in-house
technical seminar. Reads technical literature only
when told to by the supervisor.

 

119

 

 

(APPENDIX F Cont’d)

Work Assignments Sought

Desires and pursues job activities.

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “high” on this
dimension.

Typical High Behavior
Prefers assignments which involve several
technical disciplines. Is willing to accept an
assignment which has an uncertain chance of
success. Tries to get assignments which focus on
different applications of a specific technical
interest.

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “average” on
this dimension.

Typical Average Behavior
Rarely expresses reluctance to accept an
assignment. Tends to remain with assignments in
which he feels comfortable.

 

 

 

The statements to the
right are examples of
behavior that one
would expect of
individuals usually
rated “low” on this
dimension.

 

Typical Low Behavior
Desires assignments which are more administrative
than technical. Is content to perform in current
assignment for an indefinite amount of time. Tries
to avoid assignments in unfamiliar technical areas.
Prefers to work on rather routine and mundane
assignments.

 

120

 

 

(APPENDIX F Cont’d)
Technical Interest and Curiosity

Shows interest and curiosity regarding recent developments in science and technology

 

 

 

 

 

 

The statements to the Typical High Behavior
right are examples of Seeks information about all technical areas. Seeks
behavior that one involvement in relevant technical developments.
would expect of Works extra hours on own initiative to learn about 5
individuals usually new developments.
rated “high” on this
dimension.

4
The statements to the Typical Average Behavior
right are examples of Occasionally reads journals in related technical
behavior that one areas. Interest in new technology is usually limited 3
would expect of to own area only. Sometimes displays a negative
individuals usually attitude toward new ideas.
rated “average” on
this dimension.

2
The statements to the Typical Low Behavior
right are examples of Is pessimistic and cynical about new technical
behavior that one developments. Has little curiosity about 1
would expect of technologies related to own specific area. Adopts
individuals usually an attitude of “if it’s important, someone will tell
rated “low” on this me about it” toward developments.
dimension.

 

 

 

 

121

APPENDIX G

General Performance Items

On the job, this person is trustworthy.
On the job, this person is responsible.
On the job, this person is honest.

On the job, this person is hardworking.
On the job, this person is safe.

122

 

APPENDD( H

Perceived Similarity and Attraction Measures
Perceived Similarity:
This person is similar to me.
This person probably shares my attitudes, values and beliefs.
This person thinks about his work the way I do.
Attraction:
I would like this person.

I would enjoy working with this person.
I would enjoy being around this person.

123

 

REFERENCES

124

REFERENCES

Bandura, A. (1986). Social foundations of thought and action. Englewood Cliffs,
NJ: Prentice Hall.

Baskett, G. D. (1973). Interview decisions as determined by competency and
attitude similarity. Journal of Applied Psychology, 57, 343-345.

Bemardin, H. J. (1978). Effects of rater training on leniency and halo errors in
student ratings of instructors. Journal of Applied Psychology, 63, 301 - 308.

Borman, W. C. (1975). Effects of instructions to avoid halo error on reliability
and validity of performance evaluation ratings. Journal of Applied Psychology, 60, 556-
560.

Buffardi, L. & Gibson, J. F. (1980). Relation between raters’ characteristics and
halo error. Perceptual and Motor Skills, 51, 1003-1011.

Button, S. B., Mathieu, J. E., & Zajac, D. M. (1996). Goal orientation in
organizational research: A conceptual and empirical foundation. Organizational Behavior
and Human Decision Processes, 67, 26-48.

Byme, D (1971). The Attraction Paradigm. New York: Academic Press.

Byme, D., & Clore, G. L. (1967). Effectance arousal and attraction. Journal of
Personality and Social Psychology, 6, 1-18.

Byme, D., & Griffitt, W. (1969). Similarity and awareness of similarity of
personality characteristics as determinants of attraction. Journal of Experimental
Research in Personality, 3, 179-186.

Byme, D., & Nelson, D. (1965). Attraction as a linear function of proportion of
positive reinforcements. Journal of Personality and Social Psychology, 6, 659-663.

Cleveland, J. N., Murphy, K. R., & Williams, R. E. (1989). Multiple uses of

performance appraisal: Prevalence and correlates. Journal of Applied Psychology, 74,
130-135.

Cohen, J. (1988). Statistical power analysis for the behavior sciences. Hillsdale,
NJ: Lawrence Erlbaum Associates.

Cohen (1992). A power primer. Psychological Bulletin, 112, 155-159.

125

Drory, A., & Ben-Porat, A. (1980). Leadership style and leniency bias in
evaluation of employees’ performance. Psychological Reports, 46, 735-739.

Dweck, C. S. (1995). Implicit theories as organizers of goals and behavior. In
Sources and Contents of Action Goals

Dweck, C. S. (1986). Motivational processes affecting learning. American
Psychologist, 41, 1040-1048.

Dweck, C. S. (1992). The study of goals in psychology. Psychological Science, 3,
165-167.

Dweck, C. S., & Elliot, E. S. (1983). Achievement motivation In P. Mussen, gen.
ed., and E. M. Hetherington, vol. ed., Handbook of Child Psychology, Vol 4 (pp. 643-
691). New York: Wiley.

Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to motivation
and personality. Psychological Review, 2, 256-273.

Elliot, E. S., & Dweck, C. S. (1988). Goals: An approach to motivation and
achievement. Journal of Personality and Social Psychology, 54, 5-12.

Feldman, J. M. (1981). Beyond attribution theory: Cognitive processes in
performance appraisal. Journal of Applied Psychology, 66, 127-148.

Feldman, J. M. (1986). A note on the statistical correction of halo error. Journal
of Applied Psycholog)’. 71, 173-176.

Fiedler, K. (1993). Constructive processes in person cognition. British Journal of
Social Psychology. 32, 349-364.

Golightly, C., Huffman, D. M., & Byme, D. (1972). Liking and loaning. Journal
of Applied Psychology. 56, 521-523.

Good, L. R. & Good, K. C. (1972). Attitude similarity and evaluation of potential
counselors. Psychological Reports, 31, 963-966.

Good, K. C. & Good, L. R. (1974). Attitude similarity and liking for a supervisor.
The Journal of Psychology, 313-316.

Griffitt, W. & Jackson, T. (1970). Influence of information about ability and non-
ability on personnel selection decisions. Psychological Reports, 27, 959-962.

126

Hakel, M. D., Hollmann, T. D., & Dunnette, M. D. (1970). Accuracy of
interviewers, certified public accountants, and students in identifying the interests of
accountants. Journal of Applied Psychology, 54, 115-119.

Hamilton, D. L. & Fallot, R. D. (1974). Information salience as a weighting factor
in impression formation. Journal of Personality and Social Psychology, 74, 444-448.

Hartel, C. E. J. (1993). Rating format research revisited: Format effectiveness and
acceptability depend on rater characteristics. Journal of Applied Psychology, 78, 212-
217.

Havenstein, N. M. A. (1992). An information-processing approach to leniency in
performance judgments. Journal of Applied Psychology, 77, 485-493.

Ilgen, Barnes-Farrell, & McKellin. (1993). Performance appraisal process
research in the 1980’s: What has it contributed to appraisals in use? Organizational
Behavior and Human Decision Processes, 54, 321-368.

Jackson, D. N., Peakcock, A. C., & Smith, J. P. (1980). Impressions of
personality in the employment interview. Journal of Personality and Social Psychology,
39, 294-307.

Keller, R. T (1983). Predicting absenteeism from prior absenteeism, attitudinal
factors and non-attitudinal factors. Journal of Applied Psychology, 68, 36-540.

Krahe, B. (1983). Self-serving biases in perceived similarity and causal
attributions of other people’s performance. Social Psychology Quarterly, 46, 318-329.

Levy, D. A., Kaler, S. R., & Schall, M. S. (1988). An empirical investigation of
role schemata: Occupations and personality characteristics. Psychological Reports, 63, 3-
14.

London, M. & Poplowski, J. R. (1976). Effects of information on stereotype

development in performance appraisal and interview contexts, Journal of Applied
Psychology, 61 , 199-205.

Lydon, J. E., Jamieson, D. W., & Zanna, M. P. (1988). Interpersonal similarity
and the social and intellectual dimensions of first impressions. Social Cognition, 6, 269-
286.

Maurer, T. J., & Taylor, M. A., (1994). Is sex by itself enough? An exploration of

gender bias issues in performance appraisal. Organizational Behavior and Human
Decision Processes, 60, 231-251.

127

Medin, D. L., Goldstone, R. L., & Gentner, D. (1993). Respects for similarity.
Psychological Review, 100, 254-278.

Mount, M. K., & Thompson, D. E. (1987). Cognitive categorization and quality of
performance ratings. Journal of Applied Psychology, 72, 240-246.

Murphy, K. R., & Balzer, W. K. (1986). Systematic distortions in memory-based

behavior ratings and performance evaluations: Consequences for rating accuracy. Journal
of Applied Psychology, 71, 39-44.

Murphy, K. R., Philbin, T. A., & Adams, S. R. (1989). Effect of purpose of
observation on accuracy of immediate and delayed performance ratings. Organizational
Behavior and Human Decision Processes, 43, 336-354.

Paunonen, S. V., & Jackson, D. N. (1987). Accuracy of interviewers and students
in identifying the personality characteristics of personnel managers and computer
programmers. Journal of Vocational Behavior, 31, 26-36.

Paunonen, S. V., Jackson, D. N., & Oberman, S. M. (1987). Personnel selection
decisions: Effects of applicant personality and the letter of reference. Organizational
Behavior and Human Decision Processes, 40, 96-1 14.

Pulakos, E. D & Wéxley, K. N. (1983). The relationship among perceptual
similarity, sex and performance ratings in manager-subordinate dyads. Academy of
Management Journal, 26, 129-139.

Pulakos, E. D., White, L.A., Oppler, S. H., & Borman, W. C. (1989). Examination
of race and sex effects on performance ratings. Journal of Applied Psychology, 74, 770-
780.

Roedel, T. D., Schraw, G., Plake, B. S. (1994) Validation of a measure of learning
and performance goal orientations. Educational and Psychological Measurement, 54,
1013-1021.

Rosenberg, M., Schooler, C., & Schoenbach, C. (1989). Self-esteem and
adolescent problems: Modeling and reciprocal effects. American Sociological Review, 53,
1004-1018.

Rosenberg, M., Schooler, C., Schoenbach, C., & Rosenberg, F. (1995). Global

self-esteem and specific self-esteem: Different concepts, different outcomes, American
Sociological Review, 60, 141-156.

128

Sackett, P. R., DuBois, C. L. Z., & Noe, A. W. (1991). Tokenism in performance
evaluation: The effects of work group representation on male-female and white-black
differences in performance ratings. Journal of Applied Psychology, 76, 263-267.

Schneier, C. E. (1977). Operational utility and psychometric characteristics of
behavioral expectation scales: a cognitive reinterpretation. Journal of Applied
Psychology, 62, 541-548.

Schmitt, N., & Hill, T. E. (1977). Sex and race composition of assessment center
groups as a determinant of peer and assessor ratings. Journal of Applied Psychology, 62,
261-264.

Senger, J. ( 1971). Managers perceptions of subordinates’ Competence as a
function of personal value orientations. Academy of Management Journal, 415-423

Smith, R. E., Meadow, B. L., & Sisk, T. K. (1970). Attitude similarity,
interpersonal attraction, and evaluative social perception. Psychonomic Science, 18, 226-
227.

Tan, D. T. Y, & Singh, R. (1995). Attitudes and attraction: a developmental study

of the similarity-attraction and dissimilarity-repulsion hypothesis. Personality and Social
Psychology Bulletin, 21, 975-986.

Turban, D. B. & Jones, A. P. (1988). Supervisor-subordinate similarity: Types,
effects and mechanisms. Journal of Applied Psychology, 73, 228-234.

van Lange, P. A. M., Liebrand, W. B. G., & Kuhlman, D. M. (1990). Causal
attribution of choice behavior in three n-person prisoner’s dilemmas. Journal of
Experimental Social Psychology, 26, 34-48.

Varma, A., Denisi, A. S., & Peters, L. H. (1996). Interpersonal affect and
performance appraisal: A field study. Personnel Psychology, 49, 341-360.

Weiss, H. M. (1978). Social learning of work values in organizations. Journal of
Applied Psychology, 63, 711-718.

Wetzel, C. G. & Insko, c. A. (1982). The similarity-attraction relationship: Is
there an ideal one? Journal of Experimental Social Psychology, 18, 253-276.

Wexley, K. N. & Youtz, M. A. (1985). Rater beliefs about others: their effects on
rating errors and rater accuracy. Joumal of Occupational Psychology, 58, 265-275.

129

Williams, K. J ., DeNisi, A. S., Blencoe, A. G., & Cafferty, T. P. (1985). The role
of appraisal purpose: Effects of purpose on information acquisition and utilization.
Organizational Behavior and Human Decision Processes, 35, 314-339.

130

“‘ttttllliiltilt