w... (an?

:fimfi

. Emma. , w;

.....;..x a
{ .-

L

nu!
<
v

... ,.
4am

 

 

 

52mm. $¢¥§§

a,

&

u.

i“. y,

:2 n.) ’0‘
0. PL.» :01
\

11:364.
1;:th $5.. 9
2 3h .

 

100‘

This is to certify that the

thesis entitled

REACTIONS TO PERFORMANCE FEEDBACK:
THE ROLE OF GOAL ORIENTATION
AND SELF-REGULATORY FOCUS

presented by

Cori A. Davis
has been accepted towards fulfillment

of the requirements for

M.A . degree in Psychology

/fl6/Z

(Major professor

 

Date g/IOIZO/

0-7639 ~ MS U is an Affirmative Action/Equal Opportunity Institution

 

 

LIBRARY

Michigan State
University

 

 

 

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

 

JUNQM Mafia?

 

 

 

 

 

 

 

 

 

 

 

 

 

 

6/01 cJCIRC/DateDuepss-ms

REACTIONS To PERFORMANCE FEEDBACK:
THE ROLE OF GOAL ORIENTATION AND SELF-REGULATORY FOCUS
By

Cori A. Davis

A THESIS

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

MASTER OF ARTS
Department of Psychology

2001

ABSTRACT
REACTIONS To PERFORMANCE FEEDBACK:
THE ROLE OF GOAL ORIENTATION AND SELF-REGULATORY FOCUS
By

Cori A. Davis

Individuals; reactions to performance feedback are poorly understood (Kluger & DeNisi,
1996). It is proposed that goal orientation and self-regulatory focus may help explain
reactions to feedback. Specifically, the research 1) examines the moderating effects of
goal orientation and self-regulatory focus constructs on the relationship between
performance feedback and effort, 2) investigates the relationship between goal orientation
and self-regulatory focus variables, 3) studies the process variables of controllability and
affect, and 4) measures multiple types of effort (time spent on task, self-report, time spent
studying knowledge information, and time spent studying test tips information). Findings
did not support the three proposed sets of relationships. Negative feedback had a strong
main effect such that those who received negative feedback tried harder than those who
received positive feedback regardless of goal orientation or self-regulatory focus.
Exploratory analyses were conducted to further investigate the relationships between the
variables. These analyses revealed that performance-avoid goal orientation tended to
overpower other types of goal orientations as well as self-regulatory focus effects.

Results, limitations, and implications of this study are discussed.

Dedicated to Doug and Janet Davis

Sometimes it’s easy to get lost in the dark,
and become blind to who you are, where you’re from, and what really counts.

Your unconditional love and support have always been my light.

iii

ACKNOWLEDGMENTS

There are many people I would like to recognize and thank. First and foremost,
Dan, I can’t thank you enough for your guidance and support. You gave so willingly of
your time and resources. You were always there for me, willing to meet for hours on end
to mull over ideas, or spend days editing my drafts and giving me feedback. You taught
me to think in different ways and forced the limits of what I thought I could do. This was
an amazing learning experience, because you were truly an amazing mentor.

Rick and Steve, thank you so much for serving on my committee and providing
such insightful comments and guidance. You were always there if I had questions or
needed help, and I appreciated the opportunity to work with you both.

I also want to recognize Jeremy and Matt. Typing math equations and coding
data are pretty much the most tedious tasks imaginable. . .which is why you guys got
stuck with them ©. You really helped me out, especially when I was pressed for time.
Thank you, and good luck to you both.

At the risk of being really sappy, I want to thank my family for hearing me
complain and dealing with my vicious moods at every stage of this process. Mom and
Dad, you were there for me week after week and sadly, year afier year. You never lost
interest in what I was doing or let me dwell on setbacks. Grandma and Grandpa, you
never let me lose perspective by always reminding me of what’s really important and that
I am loved no matter what. Grandma Jackie, I wanted so much to show you something I

could do, but then again, you knew I could do it all along. I miss you.

iv

As for my sisters, Nikki, anytime I was feeling down or getting sick of school,
you would tell me your crazy stories about mean teachers or boring math class and I
would realize that I could never be as sick of school as you already are! You’ve got a
long way to go, and I’m so excited to watch you learn and discover all that lies ahead.
The world is open to you, Kiddo. Embrace it. Mandi, you were there for me even when I
didn’t take advantage of it. You taught me that no matter how busy or far away we
always seem to be, we’ll always be there for each other. Soon, you’ll be teaching many
more things to many more people. . .you’ll surely shine.

Finally, Kevin, thank you for encouraging me when I was completely
overwhelmed and pushing me when I wasn’t. You could usually talk me down during
the rough times, or at the very least talk until I was too tired to think anymore ©. You
were there for me through the whole thing, even as you were going through it yourself.
Always focused and always strong, you are inspiring. Just think, only one more of these

things, and we’re all done! I love you, Honey.

TABLE OF CONTENTS

LIST OF FIGURES .......................................................................................................... x
LIST OF TABLES .......................................................................................................... xi
INTRODUCTION ............................................................................................................ 1
Goal Orientation .................................................................................................... 2
Self-Regulatory Focus .......................................................................................... 3
Contributions of SRF and GO ............................................................................... 5
Complexities of Past GO and SRF Research ........................................................ 7
State/trait differentiation of constructs .............................................................................. 7
Measurement of effort ....................................................................................................... 9
Focus of Research ................................................................................................ 9
REVIEW OF THE LITERATURE ................................................................................ 11
Goal Orientation .................................................................................................. 11
Attribution Theory .............................................................................................. 13
Appropriate outcomes ............................................................................. 14
Self-Regulatory Focus ........................................................................................ 15

Affect .................................................................................................................. 18

The Nature and Measurement of Effort .............................................................. 19
CONCEPTUAL FRAMEWORK AND HY POTHESES ............................................... 21
Purpose ................................................................................................................ 21
Moderation Effects of GO .................................................................................. 21
Hypothesis 1a .......................................................................................... 22

Hypothesis 1b .......................................................................................... 22

Relationship Between GO and SRF .................................................................... 22
Hypothesis 2a .......................................................................................... 23

Hypothesis 2b .......................................................................................... 23

Hypothesis 2c .......................................................................................... 23

The Moderating Effect of SRF ........................................................................... 23
Hypothesis 3a .......................................................................................... 25

Hypothesis 3b .......................................................................................... 25

Attributions ......................................................................................................... 25
Hypothesis 4a .......................................................................................... 25

Hypothesis 4b .......................................................................................... 25

Hypothesis 5 ............................................................................................ 26

Hypothesis 6 ............................................................................................ 26

Affect .................................................................................................................. 26
Hypothesis 7a .......................................................................................... 27

Hypothesis 7b .......................................................................................... 27

Hypothesis 7c .......................................................................................... 27

vi

Hypothesis 8 ............................................................................................ 27

Distinction of Effort ............................................................................................ 27
Hypothesis 9a .......................................................................................... 28
Hypothesis 9b .......................................................................................... 28
Hypothesis 9c .......................................................................................... 28

METHOD ....................................................................................................................... 29

Design ................................................................................................................. 29
Overview ................................................................................................. 29
Task ......................................................................................................... 29

Procedure ............................................................................................................ 3O
Participants .............................................................................................. 3O
Informed consent .................................................................................... 30
Pre-task measures .................................................................................... 31
Test introduction ..................................................................................... 31
Example test questions ............................................................................ 31
Test 1 ....................................................................................................... 31
Feedback ................................................................................................. 32
Goal orientation manipulation 1 ............................................................. 32
Attributions and affect measures ............................................................ 32
Study task introduction/Goal orientation manipulation 2 ....................... 32
Study task ................................................................................................ 33
Goal orientation manipulation check ...................................................... 33
Test 2 ....................................................................................................... 33
Post-task measures .................................................................................. 33
Debrief .................................................................................................... 33

Measures ............................................................................................................. 34
Cognitive ability ...................................................................................... 34
Goal orientation ...................................................................................... 35
Self-regulatory focus ............................................................................... 35

Controllability ..................................................................................................... 37
Affect ...................................................................................................... 38
Effort ....................................................................................................... 38
GO manipulation check .......................................................................... 39

Pilot Studies ........................................................................................................ 4O
Motivation ............................................................................................... 40
Difference between study screens ........................................................... 41
Score believability .................................................................................. 41

RESULTS ....................................................................................................................... 43

Organization of the Results ................................................................................. 43

Descriptive Statistics ........................................................................................... 44

Moderating Effects of GO and SRF .................................................................... 44
Hypotheses 1a-1b .................................................................................... 44
Summary ................................................................................................. 46
Hypotheses 3a-3b .................................................................................... 47

vii

Summary ................................................................................................. 48

Relationship Between GO and SRF .................................................................... 49
Hypotheses 2a-2c .................................................................................... 49

Summary ................................................................................................. 49
Controllability Relationships .............................................................................. 50
Hypotheses 4a-4b .................................................................................... 50

Summary ................................................................................................. 54

Hypothesis 5 ............................................................................................ 54

Hypothesis 6 ............................................................................................ 55

Smnmary ................................................................................................. 55

Affect Relationships ............................................................................................ 55
Hypotheses 7a-7c .................................................................................... 55

Hypothesis 8 ............................................................................................ 56

Distinction Between Forms of Effort Behaviors ................................................ 57
Hypotheses 9a-9c .................................................................................... 57

Other Findings .................................................................................................... 59
Hypothesis 1 ............................................................................................ 59

Hypothesis 3 ............................................................................................ 6O

Hypothesis 9 ............................................................................................ 63

Summary ................................................................................................. 65
DISCUSSION ................................................................................................................. 66
Moderating Effects of GO and SRF .................................................................... 67
Relationship Between GO and SRF .................................................................... 70
Process Variables ................................................................................................ 71
Distinction Between Forms of Effort Behaviors ................................................ 72

Study Limitations ................................................................................................ 73

Task ......................................................................................................... 73

Measures ................................................................................................. 74

Future Research .................................................................................................. 74
Inclusion of other variables ..................................................................... 74

Differences between GO and SRF .......................................................... 75

Conclusion .......................................................................................................... 75
REFERENCES ............................................................................................................... 76
FOOTNOTES ........................................................... 81
FIGURES ........................................................................................................................ 82
TABLES ......................................................................................................................... 99
APPENDIX A: Informed Consent ................................................................................ 117
APPENDD( B: Introduction to the Study ..................................................................... 118

viii

 

 

APPENDIX C: Measures .............................................................................................. 119

Demographics ................................................................................................... 1 19
Goal Orientation Measure ................................................................................. 121
Self-Regulatory Focus Measure ........................................................................ 122
Selves Questionnaire ......................................................................................... 123
Affect Measure .................................................................................................. 126
Causal Dimension Scale ................................................................................... 127
Controllability Measure .................................................................................... 128
Goal Orientation Manipulation Check .............................................................. 129
Motivation Questions ........................................................................................ 130
Difference Between Screens and Score Believability Questions ..................... 131
Performance Test 1 ........................................................................................... 132
Performance Test 2 ........................................................................................... 141
APPENDIX D: Introduction to Test 1 .......................................................................... 151
APPENDD( E: Example Test Questions ...................................................................... 152
Sentence Completion Example Question .......................................................... 152
Analogies Example Question ............................................................................ 153
Quantitative Comparison Example Question............. ....................................... 154
Problem Solving Example Question ................................................................. 155
APPENDIX F: Test 1 Feedback Message .................................................................... 156
APPENDIX G: Goal Orientation Manipulation 1 ........................................................ 157
APPENDIX H: Study Task Instructions ....................................................................... 158
APPENDIX I: Goal Orientation Manipulation... .......................................................... 159
APPENDIX J: Study Task ............................................................................................ 160
Verbal Knowledge Information ........................................................................ 160
Verbal Test Tips Information ............................................................................ 167
Math Knowledge Information ........................................................................... 172
Math Test Tips Information .............................................................................. 183
APPENDIX K: Debriefing Form .................................................................................. 192

ix

Figure 1.

Figure 2.

Figure 3.

Figure 4.
Figure 5.
Figure 6.
Figure 7.
Figure 8.
Figure 9.
Figure 10.
Figure 11.
Figure 12.
Figure 13.
Figure 14.
Figure 15.
Figure 16.

Figure 17.

LIST OF FIGURES

Model based on a strict interpretation of the SRF and GO literatures .......... 82
The conceptual model for this study ............................................................. 83
GO and SRF as moderators of the relationship between performance

feedback and effort ....................................................................................... 84
Model of the procedure ................................................................................. 85
Three-way interaction found in Hypothesis 4 exploratory analyses ............. 86
Two-way interaction found for Hypothesis 4 ............................................... 87
Two-way interaction found in Hypothesis 4 exploratory analyses ............... 88
T wo-way interaction found in Hypothesis 4 exploratory analyses ............... 89
Two-way interaction found in Hypothesis 1 exploratory analyses ............... 90
Two-way interaction found in Hypothesis 1 exploratory analyses ............... 91
Two-way interaction found in Hypothesis 1 exploratory analyses ............... 92
Two-way interaction found in Hypothesis 3 exploratory analyses ............... 93
Three-way interaction found in Hypothesis 3 exploratory analyses ............. 94
Three-way interaction found in Hypothesis 3 exploratory analyses ............. 95
Three-way interaction found in Hypothesis 9 exploratory analyses ............. 96
Two-way interaction found in Hypothesis 9 exploratory analyses ............... 97
Two-way interaction found in Hypothesis 9 exploratory analyses .............. 98

Table 1.

Table 2.

Table 3.

Table 4.

Table 5.

Table 6.

Table 7.

Table 8.

Table 9.

Table 10.

LIST OF TABLES

Summary of Study Hypotheses and Results ................................................... 99

Means, Standard Deviations, Intercorrelations, and Reliabilities for
Variables Included in the Analyses .............................................................. 103

Hypotheses la-lb: Hierarchical Regression Results of the Impact of
Feedback Condition, Trait GO, State GO, and Their Two-Way
Interactions on Time Studied When Controlling for Cognitive Ability ....... 109

Hypotheses la-lb: Hierarchical Regression Results of the Impact of
Feedback Condition, Trait GO, State GO, and Their Two-Way

Interactions on Self-Report Effort When Controlling for

Cognitive Ability .......................................................................................... 110

Hypotheses 3a-3b: Hierarchical Regression Results of the Impact of
Feedback Condition, Self-Regulatory Focus, State GO, and Their
Interactions on Time Studied When Controlling for Cognitive Ability ....... 111

Hypotheses 3a-3b: Hierarchical Regression Results of the Impact of
Feedback Condition, Self-Regulatory Focus, State GO, and Their

Interactions on Self-Report Effort When Controlling for

Cognitive Ability .......................................................................................... 1 12

Hypotheses 4a-4b: Hierarchical Regression Results of the Impact of
Feedback Condition, Trait GO, State GO, and Their Two-Way

Interactions on Outcome Controllability When Controlling for

Cognitive Ability .......................................................................................... 113

Hypotheses 4a-4b: Hierarchical Regression Results of the Impact of
Feedback Condition, Trait GO, State GO, and Their Two-Way

Interactions on Development Controllability When Controlling

for Cognitive Ability .................................................................................... 1 14

Hypotheses 9a: Hierarchical Regression Results of the Impact of

Feedback Condition, Trait GO, State GO, and Their Two-Way

Interactions on Time Spent Studying Knowledge Information When
Controlling for Cognitive Ability ................................................................. 1 15

Hypotheses 9a: Hierarchical Regression Results of the Impact of

Feedback Condition, Self-Regulatory Focus, State GO, and Their

Two-Way Interactions on Time Spent Studying Knowledge Information
When Controlling for Cognitive Ability ...................................................... 116

xi

INTRODUCTION

The importance of giving employees feedback regarding their performance has
long been recognized both in applied and research settings. With recent trends such as
360-degree feedback (Peiperl, 2001) and developmental performance appraisals,
understanding the reactions that employees have to performance feedback is vital. With
these and other types of feedback-based initiatives, it is often assumed that individuals
'will benefit and change their behavior as necessary to improve their performance in
response to feedback. Unfortunately, this assumption is often unreasonable (Ilgen &
Davis, 2000; Ilgen, Fisher, & Taylor, 1979; Kluger & DeNisi, 1996).

It was once thought that the nature of the feedback message itself (e. g. positive or
negative) was a clear predictor of subsequent performance behavior. Kluger and DeNisi
(1996) showed that it is not that simple. Their meta-analysis reported that few consistent
results have been found that verify the direct effect of the nature of feedback on
performance. In other words, just because a person is given praise does not necessarily
mean that he or she will be motivated to “keep up the good work” or “reach for even
higher goals.” Nor is it the case that negative feedback can always be expected to
motivate future performance. One exception to the inconsistency of results has been
found. Results are fairly consistent in showing that motivation is lessened in cases where
negative feedback is given repeatedly (Diener & Dweck, 1978, 1980; Mikulincer, 1994).
Overall, however, research has provided inconsistent evidence for the relationship
between feedback and subsequent motivational reactions (Kluger & DeNisi, 1996).

This inconsistency begs the question: If feedback appears to be necessary to learn

from past behavior, but the nature of the feedback message is not directly responsible for

post-feedback behavior, then what is? Answers to this question have varied widely. In
an attempt to integrate various perspectives on the relationship between feedback and
performance, Kluger and DeNisi (1996) developed the Feedback Intervention Theory
(FIT). FIT integrates concepts from control theory (Carver & Scheier, 1981), goal setting
theory (Locke & Latham, 1990), action theory (Frese & Zapf, 1994), action identification
theory (V allacher & Wegner, 1987), and learned helplessness theory (Mikulincer, 1994),
among others.

The introduction of FIT provided a theoretical framework by which the effects of
cognitive and motivational factors on reactions to feedback can be examined (Kluger &
DeNisi, 1998). Recent trends in industrial psychology have focused on motivational
factors that deal with one’s method of self-regulation. Goal orientation and self-
regulatory focus are two such motivational theories that have received a fair amount of
attention. The first goal of this study was to examine the moderating effects of goal
orientation and self-regulatory focus constructs on the relationship between performance
feedback and effort. In addition, this study investigated the process variables by which
goal orientation and self-regulatory focus have been shown to work: attributions and
affect. The following two sections on goal orientation and self-regulatory focus will
describe the rationale for the research. More elaborate discussions of goal orientation and
self-regulatory focus are provided in the literature review.

Goal Orientation

Goal orientation (GO) is a construct that has been consistently shown to moderate

the relationship between feedback and effort. However, this relationship has really only

been explored in negative feedback situations. Dweck and colleagues (e.g. 1988, 1989)

found that mastery-oriented individuals were more likely to exert effort than
performance-oriented individuals in the face of failure. This finding is generally
explained in the literature via cognitive processes, namely attributions. It has been found
that those with a mastery GO tend to attribute their behavior to controllable causes such
as effort, whereas performance-oriented individuals tend to make attributions to
uncontrollable causes such as ability (Dweck & Leggett, 1988). Therefore, in the face of
failure, mastery-oriented individuals are more likely to believe that they can change their
performance through increased or redirected effort.

Although the dual dimensional taxonomy of G0 is still widely used, more recent
GO studies have found support for adding a new component to the performance side of
G0: the way in which people approach outcomes. Research has shown that within a
performance orientation, individuals may either approach positive outcomes or avoid
negative outcomes (Elliot, 1999; VandeWalle, 1997). These have been termed
performance-prove orientation and performance-avoid orientation, respectively.
Differences in reactions to feedback between these orientations have only recently begun
to be explored.
Self-Regulatog Focus

Others who have addressed the effects of feedback on effort have evoked self-
regulatory focus (SRF) to explain the inconsistencies in the effect of feedback on effort.
SRF theory stems from Higgins’ (1997) attempts to break down the underlying
mechanisms by which individuals approach desired end-states. He argues that people
tend toward certain types of end-states, and that, depending on this tendency, they use

different self-regulatory systems. Higgins divides end-states into what he calls self-

guides. Self-guides can refer to one’s sense of duty and obligation, labeled “ought self-
guides” or to one’s hopes and wishes, labeled “ideal self-guides.” Those with ought self-
guides tend to regulate themselves by avoiding mismatches to desired end-states, and
those with ideal self-guides tend to regulate themselves by approaching matches to
desired end-states. He calls these two self-regulatory systems prevention SRF and
promotion SRF respectively.

Recently, Kluger, Van-Dijk, Kass, Stein, and Lustig (2000) explored Higgins’
(1997) SRF constructs as moderators of the relationship between feedback and effort.
Kluger et a1. (2000) found that individuals who were promotion-focused tended to
increase effort in response to positive feedback, and decrease effort in response to
negative feedback. Conversely, when individuals were prevention-focused, they tended
to increase effort in response to negative feedback, and decrease effort in response to
positive feedback.

In the SRF literature, the moderating effect of SRF on the relationship between
feedback and effort is assumed to occur because of differences in affective states
following feedback as a function of SRF. Higgins (1997) has found that affect leads to
arousal. The higher a person’s arousal, whether positive or negative, the more motivated
a person is. Kluger et a1. (2000) have argued that, in response to positive feedback, a
promotion-focused person would feel elated and a prevention-focused person would feel
satisfied. According to Higgins (1997), elation is an arousing emotion, therefore
motivating, whereas satisfaction is not. In contrast, Kluger et al. (2000) have argued that
when people with a prevention focus fail, they feel anxious, but failure for those with a

promotion focus leads to sadness and depression, not anxiety. Since anxiety, not sadness

or depression, is arousing (Higgins, 1997), Kluger et a1. (2000) claimed that failure
should be motivating primarily to those with a prevention focus.

Kluger et al.’s (2000) findings still account for the likelihood that failure always
results in negative feelings, and success in positive feelings, but the degree of arousal
(excitement) associated with failure or success depends on one’s regulatory focus.
Specifically, failing to meet an outcome (prevention focus goal) or succeeding in
fulfilling an outcome (promotion focus goal) results in emotions characterized by high
arousal (anxiety and elation, respectively). Alternately, meeting an obligation or failing
to fulfill a desire results in low arousal (satisfaction and depression, respectively) (Roney,
Higgins, & Shah, 1995; Strauman & Higgins, 1987).

Contributions of SRF and GO for Understanding the Effects of Feedback on Effort

The preceding discussion shows that both SRF and GO have provided some
explanation for inconsistent reactions to performance feedback. When two conceptual
positions are offered to understand the same relationship, it is common to compare the
two, preferably in a strong inference sense. Yet, in this case, weaknesses in the SRF
measure make strong inferential comparisons between the theoretical constructs of SRF
and GO difficult at the empirical level. SRF measures will be described later, however, it
is important to recognize that all measures used in the past have failed to possess
psychometric properties that are greater than marginal. The result is that any direct,
empirical comparison between SRF and GO that finds stronger relationships for GO
cannot rule out the possibility that the differences are due to the reliabilities of the

measures.

However, the two theoretical positions do make some potentially unique
contributions with regard to the suggested mediators between feedback and effort. As
was mentioned earlier, GO research has only addressed the effects of negative feedback
on effort, and the psychological mechanisms it proposes as the source of effects are
cognitive. Specifically, the mechanisms are attributions. On the other hand, SRF relies
primarily on affective/emotional mediators of the feedback to effort relationship. Taken
together, the two positions suggest a model with attributions and affect as mediators
between feedback and effort as shown in Figure 1.

Although Figure 1 treats SRF and GO separately, conceptually there exists
considerable overlap among the key constructs. This overlap goes unrecognized by each
of the respective literatures, as evidenced by the failure of each literature to refer to the
other. However, both theories are plainly derived fiom how people go about
approaching or avoiding outcomes. For this reason, a third goal of this study was to
investigate the relationship between goal orientation and self-regulatory focus variables.

Both SRF and performance-avoid and perfonnance-prove constructs specify the
ways in which individuals regulate their behavior in order to reach goals. Prevention
focus and performance-avoid G0 are similarly characterized by an avoidance regulation,
while promotion focus and performance-prove GO by an approach regulation. This
comparison is less clear for mastery GO as research has shown that individuals use an
approach regulation in mastery situations (Elliott, 1999). However, the GO literature
conflicts with Kluger et al.’s (2000) findings with regard to reactions to negative

feedback. The G0 literature has shown that mastery-oriented individuals will persist in

the face of failure whereas Kluger et al.’s (2000) found that those with an approach
regulation (promotion focus) will quit in the face of failure.

In addition to mastery not mapping onto promotion focus in a regulation sense, it
also does not completely map on with regard to outcomes. Mastery is typically described
as being concerned with learning and development and promotion is concerned with
approaching ideal needs. Ideal needs refer to advancement, growth, and accomplishment
as opposed to prevention-focused needs for safety, responsibility, and security. Reaching
for ideal needs may involve advancement and accomplishment through learning, similar
to mastery performance, but it is not necessarily so. In fact, SRF studies that manipulate
both promotion and prevention typically do so in performance contexts without any
references to learning or development (Higgins, Shah, & Friedman, 1997).

Given the similarities between GO and SRF, rather than suggest that the two
theories make different predictions, it is more reasonable to suggest that the outcomes
suggested by one theory may also result from the other. However, given that mastery
does not map on as easily as performance-prove and performance-avoid to the SRF
constructs, the hypotheses in this study focused on the similarities between the
performance GO variables to the SRF variables. The hypotheses concerning mastery
GO, then, were considered to be exploratory. This leads us to Figure 2, which shows the
overarching model proposed in this study. The model treats GO and SRF as congruent
constructs, and incorporates the affective and cognitive processes from both literatures.
Complexities of Past GO aid SRF Research

State/trait differentiation of constructs. GO and SRF theories are unique in that

their constructs are considered to be individual traits as well as condition-dependent

states. Therefore, conflicts may exist between one’s trait GO or SRF and the state
created by the context of their environment. In research, there are problems both with
considering state and trait constructs independent of each other, and with considering
them in combination.

With any strictly individual differences (or trait) perspective, elements of
cognitions and situations are ignored. Furthermore, an exclusively individual differences
focus makes application of these theories difficult with regard to practice. In practice,
people perform tasks and get feedback. Once they are on the task there is little control
over the way people are; rather control lies in the environment where people work and, to
some extent, the situations to which they are exposed to. For this reason, theories are of
limited use if they do not address controllable variables. That is not to say that an
individual differences approach is not useful, rather it is bounded in terms of practical
utility.

Of course, similar logic can be applied to a strict state interpretation.
Manipulating GO or SRF within a particular situation may certainly sway people in
particular directions, however, individuals’ trait GO or SRF may still impact their
behaviors. This study will measure traits and manipulate states. Specifically, GO will be
both measured as a trait and manipulated as a state, whereas SRF will be only measured
as a trait. The decision to manipulate G0 was made because such manipulations have
been shown to be robust in the past, especially in comparison to SRF manipulations,
which have been rarely performed. SRF is typically measured as an individual difference
in Higgins’ (1997) work. It is important to keep in mind when reading about the various

theories in this work that GO and SRF refer to both trait and state constructs.

Measurement of effort. Typically, there are two ways in which effort has been
measured in research looking at feedback, SRF, or GO effects. First, it is common for
research involving the assessment of effort to use one-item self-report measures (Kluger
et al., 2000). Not only do one-item scales typically have unknown or low reliabilities, but
self-report measures of effort do not necessarily equate to the actual amount of effort
exerted. Respondents’ own estimates of effort may be biased. It is one thing to report
having tried harder; it is quite another to actually have done so.

Second, both GO and self-regulatory literatures treat performance as the result of
effort where effort is an amorphous behavior without reference to the content of
behaviors in which the effort is directed. Yet the theories direct us toward very different
outcomes that for both themes tend to be of two types — effort to prevent negative
outcomes and protect appearances, and effort to learn and grow. Therefore, if we are to
investigate the theoretical impact of constructs like GO or SRF, we should expect effort
would be invested in somewhat different activities. Therefore, the fourth goal of this
study was to measure multiple types of effort: time spent on task, self-report, time spent
studying knowledge or content information, and time spent studying achievement or
heuristic information).

Focus of Reseaih

In sum, reactions to feedback remain poorly understood. The findings from the
GO literature as well as those produced by Kluger et al. (2000) are steps forward. Both
lines of research bring to bear interesting ideas that should be expanded upon, and
possibly integrated with one another, but both are also limited due to the way in which

key constructs are measured. The research conducted here was aimed at addressing some

of the limitations in a study that incorporated both SRF and GO as explanatory
constructs. Specifically, the research 1) examined the moderating effects of goal
orientation and self-regulatory focus constructs on the relationship between performance
feedback and effort, 2) studied the process variables of controllability and affect, 3)
investigated the relationship between goal orientation and self-regulatory focus variables,
and 4) measured multiple types of effort (time spent on task, self-report, time spent
studying knowledge or content information, and time spent studying achievement or
heuristic information).

The literature review that follows describes the two main theories being
addressed in this research and the processes by which they are presumed to work. A
discussion of effort will then be offered, followed by the specific hypotheses tested in this

study.

10

REVIEW OF THE LITERATURE

@311 Orientation

Goal orientation refers to the reasons individuals approach tasks. According to
Dweck (1989), there are two types of GO, performance and mastery. Performance
orientation is characterized by an emphasis on demonstrating high ability and appearing
competent (J agacinski, 1992). This is contrasted with mastery G0, which emphasizes
improvement, developing skills, and mastering the task (Butler, 1987; Dweck, 1986).
These orientations were once thought to be opposite ends of a single continuum, but were
later shown to be independent constructs (Button, Mathieu, & Zajac, 1996).
Theoretically, then, an individual can exhibit both orientations at once.

Dweck (1989) compared mastery and performance orientations with regard to
performance and found that a performance orientation was often maladaptive. This was
mainly due to the tendency of performance-oriented individuals to quit after failure, as
opposed to mastery-oriented individuals who were more likely to persist after failure.
Dweck argued that the persistence was due to their tendency to try to learn from their
mistakes.

Research has also shown that those with a performance orientation tend to hold an
entity theory about their ability. This means that these individuals view their ability as
being fixed and uncontrollable. In contrast, individuals with a leaming-GO tend to hold
an incremental theory about their ability meaning that they view ability as changeable
(Elliot & Dweck, 1988; VandeWalle, 1997, VandeWalle, Brown, Cron, & Slocum,

1999). This has obvious implications for attributions, particularly for controllability.

ll

This would suggest that those who are mastery-oriented would be more likely than those
who are performance-oriented to attribute their performance to unstable and controllable
factors.

More recent GO research has shown evidence for a further delineation of the
performance orientation construct. Elliot and colleagues (Elliot & Church, 1997; Elliot &
Harackiewicz, 1996) believed there were two dimensions: performance-avoidance and
performance-approach. Similarly, VandeWalle (1997) separated performance-orientation
into performance-avoid and performance-prove. Their separate dimensions of
performance orientation differ with regard to self-regulatory characteristics. According
to VandeWalle, those with a performance-avoid G0 are more likely to exert effort toward
avoiding negative outcomes, whereas those with a performance-prove G0 are likely to
exert effort toward approaching positive outcomes. Research has shown that
performance-avoid individuals tend to be concerned with looking incompetent or
performing poorly. Alternatively, performance-prove individuals tend to be concerned
with looking competent and proving that they can perform better than others (Elliot,
1999; Elliot & Church, 1997; VandeWalle, 1997).

VandeWalle, Cron, and Slocum (in press) investigated the effect of the three
dimensions of G0 on performance after feedback was given on a preliminary task. They
included three mediators: self-report effort, self-efficacy, and self-set goal level.
Interaction analyses for the relationships between GO and feedback were not performed.
However, they did find significant positive relationships between mastery GO and effort,
as well as between performance-prove GO and effort. There was a non-significant,

negative relationship between performance-avoid GO and effort. Although this study did

12

not differentiate between feedback levels, it lends support for further study of the effort
differences between those with a performance-prove and a performance-avoid GO.

In sum, the literature on GO implies that the way in which people respond to
performance feedback, specifically negative performance feedback, may be influenced by
G0. The combination of negative feedback and high performance goals has been shown
to lead to decreased effort even to the point of quitting (Dweck, 1989). One possibility
for this is that the performance is attributed to ability and is seen as beyond the person’s
control. Under a mastery orientation, effort is not expected to drop off with negative
feedback and may even increase. If so, according to the theory, we would expect that
effort would be invested in learning more about the task presumably to perform better in
the future.

As mentioned previously, G0 is influenced by situational conditions and, yet, also
possesses trait-like characteristics. When the situation does not dictate what goals are
favored, the trait goal preferences should have more influence on behavior. Conversely,
if the situation offers strong cues, trait goal preferences may be overridden by the
situational cues (VandeWalle, 1997). The present research created conditions believed to
impact GO and also measured individuals’ pre-study levels of person-specific variance
on the construct.

Attribution Theog

As previously mentioned, the mechanisms thought to be most responsible for
GO’s effect on effort are attributions. Attributions are defined as the causal ascriptions of
events and behaviors. Fiske and Taylor (1991) refer to attributions as being fundamental

to further judgments, emotional reactions, and behavior. Attribution theory addresses the

13

perceptions of individuals with regard to their performance and provides a framework for
understanding the different things that people attribute their performance to. Dimensions
of attributions include locus of causality, stability, and controllability (Russell, 1982;
Weiner, 1980, 1985). Locus of causation refers simply to whether attributions are made
to the self (internal) or to something external. For instance, if someone performs poorly
on an exam, do they blame themselves for not studying enough, or do they blame the
instructor for developing unfair test questions?

Stability refers to how “changeable” the cause of performance is. Two common
examples are ability and effort. Ability is normally considered to be relatively stable and
slow to change, and effort is normally considered to be unstable or changeable.
Therefore, going back to the exam example, one could blame himself or herself for not
having the ability to pass the exam, or for not putting forth enough effort studying to pass
the exam.

Last, controllability refers to whether individuals think that they have control over
the cause of their performance. It is important to keep in mind that controllability is not
dictated by the stability of the cause. For instance, just because people attribute their
performance to an unstable cause does not guarantee that they feel they have control over
changing their performance. Emotions are a good example. If someone is overcome by
anger in a given situation, it does not necessarily mean that this is a stable condition. In a
few hours the person may calm down and no longer be angry. Nonetheless, some would
consider becoming angry an uncontrollable event.

Appropriate Outcomes. A general finding in the literature is that people attribute

success to internal, stable, and controllable factors and failure to external, unstable, and

14

uncontrollable factors (F iske & Taylor, 1991; Taylor, Fisher, & Ilgen, 1984). This is also
known as a self—serving bias because essentially people are taking credit for success and
denying responsibility for failure (e. g. Miller & Ross, 1975).

These three dimensions of attributions all interact with each other to form various
types of attributions. In other words, attributions vary in intemality and also vary in
stability and controllability. The number of combinations is often great. However, with
regard to motivation, the main issue is the belief that the cause of behavior can be
changed through increasing effort (Bandura & Wood, 1989; Martocchio, 1994;
Martocchio & Dulebohn, 1994). If persons are convinced that no amount of trying is
going to change their performance, then chances are they will not be motivated to exert
more effort. This belief that behavior can or cannot be changed is mainly captured in the
controllability dimension, and therefore that dimension was the focus in this study as far
as attributions are concerned.

Self-Regplatog Focus

Higgins’ (1997, 1998) theory of SRF is derived from his attempt to “shed some
light” on the fundamental nature of approach-avoidance motivation. Higgins’ stance is
that it is not enough to explain motivation through the age-old idea that people always
approach pleasure and avoid pain. Psychologists need to understand the underlying
principles involved in why this is so. This led to Higgins’ explorations of the different
ways people approach pleasure and avoid pain.

In order to explore the processes that underlie the operation of approach-
avoidance motivation, Higgins’ looked to the self-regulation literature. Approach-

avoidance is in-and-of itself a form of self-regulation, and Higgins’ goal was to determine

15

if there are other self-regulatory principles that underlie approaching pleasure and
avoiding pain. Self-regulation is generally concerned with both the end-state of behavior,
usually defined in terms of positive or negative, and the direction of regulation, usually
defined in terms of approach or avoidance (Carver & Scheier, 1981, 1982; Higgins, 1987,
1997, 1998; Higgins, Bond, Klein, & Strauman 1986; Kanfer, 1990).

An important contribution from the self-regulation literature is that it expands on
the basic approach-avoidance idea. In addition to the concept that individuals simply
approach pleasure and avoid pain, the self-regulation literature informs us that both
directions of regulation can be used to accomplish both types of end-states. In other
words, one can approach a “match ” with desired outcomes as well as approach a
“mismatch ” to undesired outcomes. Similarly, one can avoid a ”match ” with undesired
outcomes, or avoid a “mismatch ” with desired outcomes. “Match” and “mismatch” in
this context refers to the gap between a person’s behavior and his or her outcomes.

According to self-regulation theories, there are four means by which individuals
approach pleasure and avoid pain, and the predilection for using different types of
regulatory systems depends largely on the valence of the end-state. Further delineating
positive end-states in his self-discrepancy theory, Higgins distinguishes between two
types: ideal self-guides and ought self-guides (Higgins, 1987, 1996; Higgins, et al., 1986;
Higgins, Klein, & Strauman, 1985; Higgins, Roney, Crowe, & Hyrnes, 1994). Ideal self-
guides refer to individuals’ representations of their hopes, wishes, or aspirations. Ought
self-guides refer to individuals’ representations of their duties, obligations, and
responsibilities.

As noted above, the ways individuals approach desired end-states can be

16

differentiated into two systems: either by avoiding a mismatch to the desired end-state, or
by approaching a match to the desired end-state. Higgins (1997, 1998) has labeled these
approaches prevention and promotion SRF (respectively). He describes what type of
focus is relevant with regard to self-guides, concluding that those with ought self-guides
tend toward a prevention SRF and those with ideal self-guides tend toward a promotion
SRF.

Now, what does having a promotion or prevention focus imply? Higgins has
asserted that one’s SRF has implications for performance, strategic tendencies, emotions,
decision-making, and value of incentives (Higgins, 1997, 1998; Higgins et al., 1997). In
general, those with a prevention focus tend to be concerned with protection, safety, and
responsibility and are therefore less likely to engage in risky behaviors or strategies that
promote error. Those with a promotion focus are more concerned with advancement,
growth, and accomplishment, and unlike those with a prevention focus, those with a
promotion focus are more likely to engage in behaviors that involve risk (Higgins, 1997;
Kluger, 2000).

In sum, other models of self-regulation including control theory distinguish
between approaching desired end-states and avoiding undesired end-states (Carver &
Scheier, 1981; Kanfer, 1990). However, they do not go further to assure different types
of end-states fit the construct being regulated as Higgins’ theory does (Higgins, 1997).
More importantly with regard to reactions to feedback, Higgins’ theory suggests when a
positive outcome may lead to more or less effort. Specifically, because those who are
prevention-focused are engaged avoiding a mismatch to the desired end-state, they are

likely to reduce effort once they have reached this end-state due to their concern for

17

protection, safety, and responsibility. Alternatively, those who are promotion-focused are
engaged in approaching a match to their desired end—state and therefore are likely to exert
effort once they reach this end-state due to their concern for advancement, growth, and
accomplishment.

Affept

There is extensive evidence that people feel disappointment or depression-type
emotions when they fail to meet their hopes or ideals, and agitation or anxiety-type
emotions when they fail to meet their obligations or responsibilities (Higgins et al., 1986;
Roseman, 1984; Roseman, Spindel, & Jose, 1990; Scott & O’Hara, 1993; Strauman &
Higgins, 1987). It has also been found that when hopes or ideals are met, people tend to
feel cheerfulness or elated-type emotions. Likewise, when obligations or responsibilities
are met, people tend to feel quiescence or satisfaction-type emotions (Higgins et al.,
1997)

When translated into prevention and promotion terms, prevention-focused
individuals are more likely to be satisfied in the response to positive feedback, and
anxious in response to negative feedback. Promotion-focused individuals are more likely
to be elated or depressed after receiving positive or negative feedback, respectively.

Kluger, et a1. (2000) extended these findings into the performance domain. They
hypothesized that when arousal-type emotions occurred in response to performance
feedback, people would be motivated to put forth effort to improve. Arousal-type
emotions are those that elicit activity as opposed to acquiescence from a person (Kluger,
Lewinsohn, & Aiello, 1994; for a review of the dimensions of affect see Russell, 1980,

1991; for a criticism of the dimensional view of affect see Ortony, Clore, & Collins,

18

1988; Lazarus, 1991). Therefore, Kluger et a1 (2000) proposed that feelings of anxiety
and elation should be motivating. In other words, prevention-focused individuals should
be more motivated by negative feedback, and promotion-focused individuals by positive
feedback.

Affect has been less thoroughly explored in the GO area. Dweck (1986, 1999)
found that performance-oriented individuals were more likely to feel negative affect such
as anxiety in the face of obstacles. Dweck and Leggett (1988) found that mastery-
oriented individuals were generally more optimistic in receipt of negative feedback than
were performance-oriented individuals.

Affect is only just beginning to be explored in connection with the dual
dimensionality of performance GO. With regard to arousal-type emotions, Elliot and
McGregor (1999) found that state test anxiety, or worry, mediated the relationship
between performance-avoid goals and performance. Cron, Slocum, and VandeWalle
(2001) investigated the effects of mastery and performance-avoid G0 on emotional
valence (positive and negative) following negative feedback. They found that a mastery
orientation helped to reduce the intensity of negative emotions that can cause lowered
levels of subsequent self-set goals. Further, the intensity of negative emotional responses
to negative feedback was related to performance-avoid GO. Researchers have yet to
investigate both valence and arousal qualities of affect in relation to G0.

The Nature janp Measurement of Effort

Kluger et al.’s (2000) and VandeWalle, et al.’s (in press) studies are based on a

one-item, self-report measure of effort, yet such measures have a number of drawbacks.

First, single, self-report items are rarely defensible, regardless of their content due to the

19

difficulty in determining their reliabilities. Second, self-report measures of effort are
problematic because people tend to overestimate how much effort they put into work
(Mitchell, 1974). Finally, effort tends to be treated by researchers as an undifferentiated
construct. Effort is effort. Yet, the theories about putting forth effort typically assume
that the effort is invested for some purpose. What the person invests his or her effort in is
really what is important.

With respect to Kluger and many others, effort is expressed in terms of simply
working to perform the task. However, effort can be exerted for different purposes. Two
stand out. One is to learn something, accomplish mastery types of things. The other is to
look good for others, avoid criticism, etc. Therefore, in an attempt to study the
phenomenon of effort expenditure more precisely, this study set up conditions where
effort was assessed independent of the actor’s self-report, and differentiated in terms of
whether it was directed toward mastery and accomplishment or simply toward more time

spent on the task.

20

CONCEPTUAL FRAMEWORK AND HYPOTHESES

Purpose

Performance feedback is a key feature in organizations and all other
performance settings, yet how people respond to it is uncertain. The following
hypotheses address how the motivational theories of GO and SRF potentially add to our
understanding of reactions to feedback. The present research addresses the effects of GO
and SRF on the relation between feedback and effort, the relationship between GO and
SRF themselves, and the impact of controllability and affect in the process. In addition,
some of the limitations inherent in past research with regard to measuring effort and
differentiating between state and trait GO will be explored. The hypotheses are outlined
in Table 1, and the general conceptual models can be seen in Figures 2 and 3.

Moderation Effects of G0 on the Relationship Between Feedback and Effort

As described in the literature review, G0 has been shown to moderate the
relationship between feedback and effort. It has been fairly well established that those
with a mastery orientation are more likely to persist in the face of negative feedback as
compared to those with a performance orientation. Responses to feedback between
performance-prove and performance-avoid oriented individuals is not so clear. However,
one can make some inferences in this regard. Performance-avoid oriented individuals are
concerned with looking bad in front of others. In other words, performance-avoid
individuals tend to focus on avoiding negative outcomes. Conceptually, then,
performance-avoid has similarities to prevention SRF. Given this similarity, one could

predict that performance-avoid individuals will react similarly to those who are

21

prevention-focused in response to feedback. Specifically, performance-avoid individuals
may be more likely to try harder in response to negative feedback than in response to
positive feedback.

The same logic could be followed for performance-prove GO. Those with a
prove GO try to look good compared to others by outperforming them. Therefore, those
with a performance-prove GO attempt to approach positive outcomes. This orientation is
conceptually similar to promotion SRF. Therefore, one can predict that performance-
prove individuals will react similarly to promotion-focused individuals in response to
feedback. Specifically, performance-prove individuals are more likely to try harder in
response to positive feedback as opposed to when responding to negative feedback. The
following hypotheses are not yet making any assumptions about where individuals are
devoting their study time. Rather, it is expected that the total amount of effort exerted
will differ between individuals.

Hypothesis 1a. Individuals with higher levels of performance-avoid GO will
display higher levels of effort when they receive negative feedback and lower levels
when they receive positive feedback.

Hypothesis 1b. Individuals with higher levels of performance—prove GO will
display higher levels of effort when they receive positive feedback and lower levels when
they receive negative feedback.

Relationship Between GO and SRF

Given that GO and SRF have major conceptual similarities, it is expected that

measures of these constructs will be correlated. Specifically, mastery GO, performance-

prove GO, and promotion SRF are all concerned with approaching positive outcomes and

22

are expected to be correlated. Alternatively, performance-avoid GO and prevention SRF
are concerned with avoiding negative outcomes and are expected to be correlated.

Hypothesis 2a. Measures of promotion SRF will be positively correlated with
measures of mastery GO.

Hypothesis 2b. Measures of promotion SRF will be positively correlated with
measures of performance-prove GO.

Hypothesis 2c. Measures of prevention SRF will be positively correlated with
measures of performance-avoid G0.

The Moderating Effect of SRF on the Relationship Between Feedback and Effort

This set of hypotheses deals with the rationale behind Hypotheses la-lb. Kluger
et a1. (2000) used Higgins’ (1997) theory of SRF as a moderator of the relationship
between feedback and effort (Figure 2). Both promotion and prevention-focused persons
prefer positive outcomes to negative ones. But, according to Higgins, they differ in their
approach to the two. Promotion-focused people seek positive outcomes; they direct their
attention to them and strive to get them. Prevention-focused people, by contrast, focus on
avoiding negative consequences and are less likely to devote time energy to gain positive
outcomes. They do the latter only to the extent that by getting positive outcomes they
can avoid negative outcomes.

Kluger et a1. (2000) adapted this approach to feedback assuming positive
feedback is the receipt of positive outcomes and negative feedback is a negative outcome.
If so, promotion-focused persons should be more highly motivated to work hard to
improve their performance following positive feedback (presumably to get more of the

positive outcomes they desire) than those motivated only to prevent poor outcomes.

23

Under negative feedback the reverse Should occur according to Kluger et a1. Here,
prevention-focused persons should feel the pain of poor performance more and work
harder to avoid it in the future.

Kluger et a1. (2000) measured SRF by classifying people according to their
chosen college major. Those pursuing degrees in the liberal arts (theater, history,
languages, religion, etc.) were considered high promotion-focused individuals.
Alternatively, those majoring in business and engineering (accounting, business
administration, economics, etc.) were considered low promotion-focused individuals.
This method of distinguishing between prevention and promotion-focused individuals
treats these constructs as independent but not continuous. This is consistent with the
terminology we have used throughout this work in describing the differences between
self-regulatory foci. There are two reasons for this. One is that this terminology is
generally consistent with Higgins’ own, and two is that research in general has not really
addressed varying levels of prevention and promotion SRF.

Kluger’s measure was indirect. It did not assess the process by which individuals’
chose their majors, it simply classified individuals. This leaves room for the possibility
that other differences between liberal arts and business majors are responsible for the
effect.

Our goal was to replicate Kluger et al.’s (2000) findings. However, because we
are measuring prevention and promotion each as continuous variables, we are also
proposing that the use of measures consistent with Higgins’ conceptualization of the
continuous nature of prevention and promotion focus may help to refine Kluger’s

findings.

24

Hypothesisfi. Individuals with higher levels of prevention SRF will display
higher levels of effort when they receive negative feedback and lower levels when they
receive positive feedback.

Hypothesis 3b. Individuals with higher levels of promotion SRF will display
higher levels of effort when they receive positive feedback and lower levels when they
receive negative feedback.

Attributions

As mentioned earlier, mastery-oriented individuals have been shown to attribute
their performance to more controllable factors than do performance individuals (Dweck
& Leggett, 1988; Elliott & Dweck, 1988). This, however, does not tell us anything about
the difference in attributions between performance-prove and performance-avoid
individuals. However, it is expected that individuals will only exert effort if they believe
that putting forth effort can improve their performance. As hypothesized above,
performance-prove individuals are thought to exert effort when they get positive
feedback. Therefore, it would be consistent to hypothesize that these individuals will
make controllable attributions for their performance, as mastery-oriented individuals
would.

Hypothesis 4a. Individuals with higher levels of mastery GO will make
attributions to more controllable causes in positive and negative feedback conditions.

Hypgthesis 4b. Individuals with higher levels of performance-prove GO will
make attributions to more controllable causes in positive feedback conditions.

In the literature, attributions have been shown to lead directly to affect and effort.

More controllable attributions have been linked to higher levels of positive affect as well

25

as higher levels of effort (Weiner, 1985). In fact, it has been shown that controllability
attributions are a necessary condition for a change in effort. In other words, individuals
must believe that their performance is under their control in order to believe that exerting
effort will change their performance.

Hypothesis 5. As attributions are made to more controllable causes, the level of
positive affect will be higher.

Hypothesis 6. As attributions are made to more controllable causes, the level of
effort will be higher.
A_ff9_c_t

Because affect has been more thoroughly investigated in the SRF literature than in
the GO literature, the inferences regarding the relationships between GO and affect are
mainly drawn from Higgins’ (1997) work. Based on Higgins (1997), Kluger et a1. (2000)
suggested that prevention-focused individuals are more likely to be satisfied in the
response to positive feedback, and anxious in response to negative feedback.
Alternatively, promotion-focused individuals are more likely to be elated or depressed
afier receiving positive or negative feedback, respectively. If we map performance GO
onto this typology, we can then predict that performance-prove individuals would
respond similarly to promotion-focused individuals and that performance-avoid
individuals would respond similarly to prevention-focused individuals.

Also, if the logic of arousal being a motivating factor is correct, then we can also
predict the types of affect mastery-oriented people are likely to exhibit in response to
feedback. Dweck (1989) has shown that compared to performance-oriented individuals,

mastery-oriented individuals have been shown to increase performance in response to

26

positive and negative feedback. Therefore, it is hypothesized that mastery-oriented
individuals will display arousal-type emotions in response to both positive and negative
feedback messages.

Hwothgis 7a. Individuals with higher levels of mastery GO will display higher
levels of elation-type affect when feedback is positive, and higher levels of anxiety-type
affect when feedback is negative.

Hypothesis 7b. Individuals with higher levels of performance-prove GO (and
promotion SRF) will display higher levels of elation-type affect when feedback is
positive, and higher levels of depression-type affect when feedback is negative.

Hymthesis 7c. Individuals with higher levels of performance-avoid GO (and
prevention SRF) will display higher levels of satisfaction-type affect when feedback is
positive, and higher levels of anxiety-type affect when feedback is negative.

It is expected that affect will have a main effect on effort. Higgins work has
shown that arousal type emotions such as anxiety and elation are more likely to lead to
effort as opposed to non-arousal type emotions such as satisfaction and depression
(Higgins, 1987; Kluger et al., 2000).

Hypothesis 8. Arousal (elation/anxiety) will be positively correlated with effort.
Distinption of Effort

Since effort is normally anchored to time spent on task, we thought it would also
be interesting to also look at where effort is spent. Therefore, this set of hypotheses is
concerned with the types of tasks on which individuals choose to exert effort.
Specifically, this study will examine whether effort is extended toward learning or

performance tasks. The G0 literature informs us that different types of people focus on

27

different types of information when studying. Mastery-oriented individuals are more
likely to focus on learning information, and performance-oriented individuals are more
likely to focus on heuristic-type information (Dweck, 1989). Similarly, Elliot &
McGregor (1999) have proposed that a performance-prove G0 is more closely associated
with superficial and instrumental-types of effort as opposed to the more substantial or
comprehensive-types of effort that a mastery G0 is associated with.

Hypothesis 9a. Mastery GO will be positively correlated with the amount of
effort invested in reviewing knowledge information.

Hypothesii 9b. Performance-prove GO will be positively correlated with the
amount of effort invested in reviewing test-taking heuristics.

Hypothesis 9c. Performance-avoid GO will be positively correlated with the

amount of effort invested in reviewing test-taking heuristics.

28

METHODS

Desigp

Overview. Participants performed on two different tests containing actual
Graduate Record Examination (GRE) questions2 with a 15-minute study break between
working on the tests. Before the study break, participants received feedback on their first
test performance and were allowed to choose how to use their break time in preparation
for the second test. The primary behaviors of interest were the ways in which the
participants spent their study time during the break. Individual difference measures were
collected prior to test taking and the tasks were performed under one of four GO
conditions.

The overall desigr used in this study was a 4 (performance-prove/ performance-
avoid/ mastery/ control) x 2 (positive/negative feedback) x A (affect) mixed design where
the first variable was experimentally manipulated (the computer program randomly
assigned and equally divided participants by GO condition) and the last two were
individual difference variables.

ng, There were two types of tasks in this experiment, test-taking and studying.
The tasks, like all other parts of this experiment, were computer-administered. The tests
that participants took consisted of actual verbal and quantitative GRE questions. More
Specifically, verbal question types included analogies and sentence completions.
Quantitative question types included quantitative comparisons and problem-solving. All
of these question types were chosen over other possible types because they had both

knowledge and test taking performance components.

29

The study task involved reviewing information about the above question types.
Two types of information were offered: knowledge and test tips3. Knowledge
information addressed content-based topics such as word meanings in the case of verbal
content and geometry rules in the case of quantitative content. Test tips information
addressed specific test-taking strategies for each question type. The study information
can be seen in Appendix J.
Procedure (See Figure 4 for a time-line of the procedure.)

Participants. A total of 327 undergraduate college students fi'om a'large
midwestem university participated in this study in partial fulfillment of course
requirements. Some participants were dropped fiom the sample for various reasons
resulting in a final sample of 300 participants.4 Participants were mainly white (78%),
18-21 year old (83%), females (71%). Sessions were conducted with groups consisting
of from one to 25 participants. The individuals in these groups were randomly assigned
to one of four GO manipulation conditions until the number of participants in each
condition was over 60 (mastery N=64, perforrnance—prove N=65, and performance-avoid
N=63), then a few more sessions were run for the control group only (N=108). This was
done to increase power, allowing analyses of the trait GO and SRF effects to be run
within the control group only.

Informed consent. When participants arrived at the computer lab for the
experiment, they were asked to enter some identifying information into a computer. At
that point, an informed consent form was displayed on the screen. Before continuing on

with the experiment, all participants were required to read and agree to this consent form.

30

The consent form outlined information regarding the incentives offered for this study (see
Appendix A).

Pre-task measures. After agreeing to participate, individuals were asked to read
some general instructions for the study (see Appendix B) and fill out some general
demographics and two individual difference measures (see Appendix C). They included
questions about participants’ GPAS, ACT/SAT scores, and GRE experience. GRE
experience was assessed to identify individuals who already may have been exposed to
the information in the tasks. The individual difference measures that individuals were
asked to fill out at this time were goal-orientation and self-regulatory focus.

Test introduction Following the questionnaires, participants were given an
introduction to the test (see Appendix D). This introduction included a general
description of the types of test questions and how long they had to take the test.

Example test questions. Participants were given four example test questions and
answers after reading the manipulation statements. There was one example question for
each type of test question (sentence completion, analogies, quantitative comparisons, and
problem solving). These are shown in Appendix E. All participants had to read the
examples before continuing on to the test.

Test 1. Following the test question examples, participants were given test

 

instructions and then Test 1. Participants had 15 minutes to answer as many GRE
questions as they could. Test 1 and Test 2 were composed of all four question types. The
tests were parallel with respect to format. The particular test questions for both Test 1
and Test 2 were chosen for inclusion based on the percentage of GRE examinees who

answered these items correctly between October 1, 1985, and September 30, 1991

31

(approximately 950,000 individuals). We chose questions with percentages ranging from
50 to 70. The average percentage of items answered correctly within each test was 59.
Both tests can be seen in Appendix C.

Feedback. After Test 1, participants were given normative feedback about how
they performed on the task compared to others (see Appendix F). Positive or negative
feedback was presented. Positive feedback was given to participants who answered eight
or more questions correctly, and negative feedback was given to those who answered less
than eight questions correctly. Eight was chosen as the feedback division point because
pilot tests showed that eight was the median number of questions that students answered
correctly. Participants in each experimental manipulation were essentially evenly split
between feedback levels. The overall N’s for each conditioaneedback cell are as
follows: mastery positive/negative=33/3 l , performance-prove positive/negative=3 0/35,
perforrnance—avoid positive/negative=32/31, and control positive/negative=67/41.

Goal orientation manipulation 1. Following feedback, individuals in the first
three conditions were given a verbal manipulation regarding how to interpret their
feedback. Participants in the fourth condition were not given a manipulation (control
condition). These GO manipulation messages can be found in Appendix G.

Attributions and affect measures. Participants filled out attribution and affect
measures after being presented with the GO manipulation (see Appendix C).

Study task introduction/Goal orientation manipulation 2. After answering the
attribution and affect questions, the computer gave participants instrUctions for the study
task (see Appendix H). In addition, a second goal-orientation manipulation was given

here (see Appendix I).

32

Study task. During the first five minutes of study time, participants were forced
to study any or all of 4 study screens: verbal knowledge, verbal test tips, math
knowledge, or math test tips (see Appendix J). For the last ten minutes of study time,
participants were given two additional options: reading information about ethics in
psychology5 and surfing the web. At the beginning of the last ten minutes, a pop-up
screen appeared alerting the participants to their new options. The computer program
recorded how long each participant spent on each study screen.

Goal orientation manipulation check. In order to find out whether or not the GO
manipulation worked, a G0 manipulation check was administered following the study
time (see Appendix C).

M, After the study task, participants were given brief instructions concerning
Test 1, and then took another lS-minute GRE test. This test was similar to Test 1 in
format, however the questions were not identical to those in Test 1 (see Appendix C).

Post-tas_k measures. Even though not necessary for this study, pilot studies
showed that participants wanted feedback after the second test. Therefore, individuals
were told how many questions they got correct afler Test 2. Then, participants answered
three sets of questions. The first set dealt with their general motivation to perform well
on the various components of the task, the second set assessed whether or not individuals
were able to distinguish between the content and strategy information from the study
task, and the third dealt with score believability for both Test 1 and Test 2 (see Appendix
C).

Debrief. Participants were shown the debriefing form as seen in Appendix H.

33

Measures

All measures can be seen in Appendix C. The means, standard deviations, and
reliabilities for each scale can be seen in Table 2. The reliabilities are in bold on the
diagonals of the matrix.

Cpgnitive ability. Individuals’ self-reports of their GPA and ACT/ SAT scores
were used as measures of cognitive ability. Because of the cognitive nature of this task,
cognitive ability was controlled for in the analyses where relevant. Because feedback
was based on true performance in a cognitively based task, it was necessary to control for
cognitive ability in analyses that included feedback. Therefore, in order to control for
ability in hypotheses that included feedback, we chose the index of cognitive ability that
was most highly correlated with the feedback variable. In this case, ACT/SAT scores
was significantly correlated with feedback .57 (p<.01), whereas GPA was not
significantly correlated (r=.l 1). Therefore, ACT/SAT scores were controlled for in
relevant analyses. SAT (mean=1016, SD=226) and ACT (mean=21, SD=4.7) scores
were standardized based on 1999 normative test data.

Using ACT/ SAT scores in the analyses posed one problem. There were 44
participants in this study who did not report their ACT/ SAT scores. Therefore, because
list-wise deletion was used for all of the analyses in this study, the number of participants
used in the analyses was reduced fi'om 300 to 246 when ACT/SAT scores were included.
This lowered the number of people in each conditioaneedback cell to mastery
positive/negative=29/27, performance-prove positive/negative=27/31, performance-avoid

positive/negative=27/26, and control positive/negative=S9/30.

34

QtLlorientation. Goal orientation was assessed using VandeWalle’s (1997)
measure. The measure consists of three scales: mastery, performance-avoid, and
performance-prove GO. All three scales included ratings from 1 to 7 (agree to disagree).
These ratings were converted to a l to 5 range for this study in order to keep all rating
ranges for all measures consistent. Research by Bendig, Komorita, and Matell and
J acoby has shown that the reliability of most measures is not significantly affected by a
change in the number of scale anchor points (as cited in Bass, Cascio, & O’Connor,
1974)

There were four mastery items, five prove items, and four avoid items on
VandeWalle’s (1997) measure. The reliability for each scale is as follows: mastery
(a=.74); performance-avoid (a=.7l), performance-prove (a=.60). These scales were
correlated: mastery/prove=. l 7, mastery/avoid=-.38, prove/avoid=.23. These correlations
are consistent with those found in the GO literature (VandeWalle, 1997).

Self-regulatory focui There were two measures used to assess SRF. First was
the Selves Questionnaire (Higgins et al., 1985). It is a subjective type of survey that asks
individuals to list eight attributes for different self-states. It asks individuals to list
attributes of the type of person they think they ACTUALLY ARE, OUGHT TO BE, and
IDEALLY WOULD LIKE TO BE.

The measure is scored by taking the number of actual-ought matches and
subtracting the number of mismatches. Matches refer to synonym responses on the two
scales, and mismatches refer to antonyrns. The actual-ought discrepancy score has been
used by Higgins to represent an individual’s prevention focus. The same process is used

to determine the actual-ideal discrepancy score, which is representative of a promotion

35

focus (Higgins, 1997).

These measures were coded for each participant by two raters. Initially, responses
fi'om 20 random participants were chosen and independently rated by each scorer. Then,
discrepancies were discussed and scores changed based on mutual agreement. Then, the
first rater scored responses from the remaining participants. The second rater scored a
randomly selected 20% of the remaining responses. When both raters were done, ratings
were compared. Overall, there was a 65% agreement rate for the ought-actual
discrepancies, and a 77% agreement rate for the ideal-actual discrepancies.

Because of its focus on discrepancy scores, the Selves Questionnaire is
conceptually difficult to interpret and apply in relation to other variables. In fact, the
form the Selves Questionnaire used in this study is no longer used in Higgins’ research,
nor was it used in Kluger et al.’s (2000). Higgins’ work now involves a modified version
of the Selves Questionnaire that measures reaction times and asks people to rate the
importance of various ought, actual, and ideal self-states. The older version of the Selves
Questionnaire attempted in this study because of its logistical feasibility relative to the
reaction time measure.

Also, because each discrepancy score is determined by comparing either the ideal
or ought self-guides to the actual self-guides, the discrepancy scores were correlated .70.
This makes it virtually impossible to detect differences between the two on any given
variable. Essentially, it is difficult to determine any differences between the two scores at
all. For these reasons, the Selves Questionnaire was dropped from this study.

Because the Selves Questionnaire was known to be a poor measure, the author

developed a second SRF measure. It originally had 12 items that asked participants why

36

they chose to be enrolled in their current psychology course. Promotion items were
based on ideal self-guides such as wants and hopes, whereas prevention items were based
on ought self—guides such as shoulds and responsibilities. Exploratory factor analyses for
this measure resulted in two factors, however two of the items showed poor factor
loadings (these items are identified in Appendix C). These items were dropped and the
remaining items were combined to form a promotion scale (a=.90) and a prevention scale
(a=.86). These two scales were correlated r=-.18. These scales did not correlate with the
Selves Questionnaire scales. However, given the poor quality of the Selves
Questionnaire, it is uncertain as to the implications of the low correlation for judging the
validity for the constructed measure.

Controllabilig. Russell’s (1982) Causal Dimension Scale was used to assess
attributions. Although this scale measures all three dimensions of attributions, only the
three questions that loaded on the controllability factor were examined for this study (see
Appendix C). The controllability factor had a low reliability of .43 and was therefore not
included in analysis.

The author also developed another measure of controllability. It originally
consisted of six items that asked participants to rate their level of controllability for
improving their test score. Exploratory factor analysis revealed three factors, but one
factor (two items) was dropped due to poor reliability (these items are identified in
Appendix C). This lefi two factors. The first factor was composed of items 1 and 2,
which assessed individuals’ beliefs for improving their scores on the next test. This
factor was named outcome controllability (a=.78). The second factor assessed

individuals’ beliefs that studying could improve their scores (items 5 and 6). This factor

37

was named development controllability (a=.80). The scales were correlated r=.26.

_A_f£ept. The affect measures used were adapted from the State/Trait Anxiety
Index format (STAI: Consulting Psychologists Press Inc., 1968, 1977). The measures
were intended to tap both the valence (positive/negative) and degree (arousal) of affect.
Because theory does not support that those who received positive feedback would feel
negative affect, and that those who received negative feedback would feel positive affect,
the positive affect measure was given only to those who received positive feedback, and
the negative affect measure was given only to those who received negative feedback.
Each measure contained arousal and non-arousal items that factored separately in
exploratory factor analysis. However, some items were found to have poor factor
loadings and were dropped from the measures (these items are identified in Appendix C).

In addition, the arousal and non-arousal factors were highly correlated within each
measure: positive affecF.49 and negative affect-=63. The arousal and non-arousal items
were made into separate scales despite being highly correlated in order to examine any
differences between them. The reliabilities for each scale are as follows: positive non-
arousal (satisfaction)=.85, positive arousal (elated)=.91, negative non-arousal
(depressed)=.92, and negative arousal (anxiety)=.89.

Eflgg Effort was measured in a variety of ways. First, the total time spent
studying was measured in seconds. The total time spent studying was bi-modally
distributed. Most people either stopped studying after five minutes, or continued to study
until the lS-minute study time was over. Second, the amount of time spent studying
knowledge information and the amount of time spent studying test tips information were

calculated. The time spent studying test tips information was positively skewed. Finally,

38

a self-report index of study behavior was collected. The self-report measure consisted of
two items and had an internal reliability index of .77 (see Appendix C). The total time
spent studying and the self-report index of study behavior were highly correlated (r=.54,
p<.01).

GO manipulation checl_(_. Participants were given measures of GO manipulation
checks prior to the second test. These manipulation checks were intended to gauge the
effectiveness of the state GO manipulations. The manipulation check consisted of six
items (two for each type of GO) that were based on the wording used in the GO
manipulations (see Appendix C). Three (two-item) scales were created: mastery (a=.87),
performance-prove (a=.70), and performance-avoid (a=.82).

These scales were highly correlated: mastery/prove=.42, mastery/avoid=.3 7,
prove/avoid =.71. Contrary to hypotheses, all of these correlations were in a positive
direction. It was expected that the mastery and avoid scales would be negatively
correlated. In addition, when all items were entered into an exploratory factor analysis,
the mastery items formed one factor, and the other items formed one factor instead of
two. This could have been due to a problem with the scale, or could possibly reflect the
general performance GO construct. Nevertheless, these scales were retained.

The manipulation check measures were not correlated with their respective
manipulations with the exception of performance-prove (r=.l6, p<.05). Rather, the
manipulation checks were significantly correlated with the measures of trait GO and
SRF. Both the trait GO and SRF measures were generally correlated in expected

directions with the GO manipulation check measures.

39

First, the trait mastery GO measure was significantly correlated with the mastery
GO manipulation check (r=.20, p<.01). Second, the trait performance-prove GO measure
was significantly correlated with both the performance-prove and performance-avoid GO
manipulation checks (r=.30, p<.01 and r=.28, p<.01, respectively). These correlations are
more than likely a result of the performance-prove and avoid GO manipulation measures
being so highly correlated. As mentioned, the factor structure of the manipulation check
measures was not distinct between the two performance dimensions, indicating that the
differentiation between these dimensions is difficult to determine. Finally, the trait
performance-avoid GO measure was negatively correlated with the mastery GO
manipulation check (r=-13, p<.05), and positively correlated with the performance-avoid
GO manipulation check (r=.23, p<.01).

A similar pattern of correlations was found between the SRF measures and the
GO manipulation check measures. The promotion SRF measure was positively
correlated with both the mastery and performance-prove GO manipulation measures
(r=.21, p<.01 and r=.12, p<.05, respectively). The prevention SRF measure was
positively correlated with both the performance-prove and avoid GO manipulation checks
(r=.15, p<.05 and r=.22, p<.01, respectively).

Performance. Actual performance was based on the number of test questions
answered correctly. Performance on the first test was used to determine who would
receive positive or negative feedback. Performance on the second test was used in order
to identify the incentive winners.

Pilot Studies

Three pilot studies (N =1 30) were run to evaluate the content and timing

40

components of the tasks, the framing of the GO and feedback manipulations, the
variability of individual differences, participants’ motivation to perform, distinction
between the types of study information, and score believability. Groups of
undergraduates went through the task, and then answered questions regarding these
issues. This data was analyzed and used to refine the tasks and measures.

Motivation. Participants’ motivation to perform on Test 2 was measured with
three items (see Appendix C). To promote motivation to study, monetary incentives of
$25 were offered to the top two scorers on the second test. Therefore, the first item asked
whether or not the incentive was motivating to people. Most indicated not being
motivated by the incentive (mean=2. 1 8). However, most people indicated wanting to
perform well on the second test (mean=3.70). The latter mean was increased from the
pilot study. Test effort was also assessed. The mean ratings of effort on Test 1 and Test
2 were 3.34 and 3.35, respectively

Difference between study screens. Pilot studies showed that peOple were unable
to tell the difference between the types of information on the study screens (see items 1
and 2 in Appendix C). Adjustments were made to the study screens, and although still
not high, the mean responses from participants of this study were increased as compared
to those from participants in the pilot study: difference between verbal screens=3.50,
difference between quantitative screens=3.60.

Score believab_ility. The mean ratings for Test 1 feedback believability were
similar in the pilot studies (false feedback) and in the current study (true feedback) with
one exception (see items 7-9 in Appendix C). Those who received positive feedback in

the current study were more likely to believe their feedback than those who received

41

positive feedback in the pilot study, or negative feedback in the pilot or current study.
Means were as follows for the current study: positive feedback=3.42 and negative

feedback=2.92.

42

RESULTS

Organization of the Results

The discussion of the results begins with a presentation of descriptive statistics for
all variables and their intercorrelations. This is followed by the presentation of all of the
hypotheses. For the most part, the hypotheses describe the relationship between two
constructs. However, in many cases, each of the constructs is assessed by more than one
measure of the variable representing that construct. For example, consider effort, the
primary dependent variable in this research. Many hypotheses make predictions about
the effects of particular variables on effort. Effort was measured in two different ways,
one self-report and the other an observation of the amount of time devoted to studying
GRE information. Thus, every hypothesis addressing effort will be tested twice -— once
for each of the two effort measures.

Similarly, predictors were often assessed by multiple means. Sometimes these
were with individual differences measures and other times with experimental
manipulations of the construct. The end result is that, for any one hypothesis, there were
often nX2 tests of it where n represented the number of assessments of the independent
variables and 2 the number of effort measures. In the presentation below, the nX2
expressions of each hypothesis are nested under the discussion of the overall hypothesis.

With multiple measures of many variables and several hypotheses that addressed
the same dependent variable, each analysis often addressed more than one hypothesis.
When this was the case, an analysis was first introduced under the discussion of the first

hypothesis to appear in the discussion in which that analysis was needed to address it, and

43

the table in which it was presented identified that subset of the data appropriate for the
hypothesis as well as other ones to be discussed later in the results section.

Finally, by combining many variables for multiple hypotheses in single analysis,
there were times when the analyses yielded information about relationships between
variables that were not of direct interest to this research. In some cases, these were
ignored. However, there were some cases where the pattern of results was of interest to
the issues being raised in this research. Therefore, a final section labeled “Other
Findings” presents these findings.

Descriptive Statistics

Table 2 presents the means, standard deviations, and intercorrelations among all
variables included in the analyses. Where multiple-item scales were used to assess
variables, internal consistency reliabilities (coefficient alphas) are represented in bold in
the diagonal of Table 2. Specific relationships within Table 2 are discussed in the
following section along with the results of additional analyses used to test specific
hypotheses.
Moderation Effects of Goal Orientation and Self-Regulatory Focus

Hypotheses la-lb. The first set of hypotheses addressed the moderating effect of
performance G0 on the relationship between feedback and effort. Given the design of
this study, various methods were used to investigate this effect. First, because the model
addressed the dual nature of performance GO, both performance—avoid GO and
performance-prove GO were considered (V andeWalle, 1997). Hypothesis 1a addressed
the moderating effect of performance-avoid GO and Hypothesis 1b addressed the

moderating of effect of performance-prove GO.

Next, the design of this study allowed for the investigation of both trait and state
effects of the two types of performance GO. GO traits were measured prior to the
beginning of the study, and GO states were manipulated during the study. All
participants completed the trait GO measures and each participant belonged either to a
mastery GO, performance-prove GO, performance-avoid GO, or control (no GO
manipulation) condition.

The types of effort relevant to Hypotheses 1a and lb were self-report and total
seconds spent reviewing the study screens (time studied). The specific forms of
Hypotheses 1a and 1b expressed in the particular variables used are listed below:

0 Trait performance-avoid GO will moderate the relationship between
performance feedback and time spent studying (Hypothesis 1a).

0 State performance-avoid GO will moderate the relationship between
performance feedback and time spent studying (Hypothesis la).

0 Trait performance-avoid GO will moderate the relationship between
performance feedback and self-reported effort (Hypothesis 1a).

0 State performance-avoid GO will moderate the relationship between
performance feedback and self-reported effort (Hypothesis 1a).

- Trait performance-prove GO will moderate the relationship between
performance feedback and time spent studying (Hypothesis 1b).

0 State performance-prove GO will moderate the relationship between
performance feedback and time spent studying (Hypothesis 1b).

0 Trait performance-prove GO will moderate the relationship between
performance feedback and self-reported effort (Hypothesis lb).

0 State performance-prove GO will moderate the relationship between
performance feedback and self-reported effort (Hypothesis 1b).

Regression analyses were used to test Hypotheses 1a and lb. Identical analyses

were used to first regress effort as measured by the time spent studying on the

45

independent variables (Table 3). These analyses were followed by regressing self-
reported effort on the same set of variables (Table 4). In all cases, cognitive ability as
measured by self-reported ACT/ SAT scores was entered in Step 1 as a control variable.
In the same step, the independent variables were added. In the case of Hypotheses la and
1b there were five critical ones: feedback, trait avoid, state avoid, trait prove, and state
prove. Finally, two additional main effect variables were entered in Step 1. Trait and
state mastery variables were entered in order to explore and partial out their effects.

Entered as the second step in the regression analysis were the interactions. In this
case, all two-way interactions were entered. In doing so, a number of these were relevant
to none of the stated hypotheses. The far left column of Table 3 lists the hypotheses for
which each line of the results is relevant, and, when the analysis is relevant for none of
the hypotheses, it is listed as other. Finally, the same set of analyses is repeated using
self-reported effort. The data for these analyses appear in Table 4.

Turning specifically to Hypothesis 1a, there were no significant interactions
between performance-avoid GO and feedback variables as hypothesized for either type of
effort. The analysis for Hypothesis 1b showed similar results. There were no significant
interactions between performance-prove GO and feedback variables as hypothesized.

Some main effects were found. Feedback had a main effect on both time studied
(R2=.07, p<.05; B=-.28, p<.01) and self-report effort (R2=.07, p<.05; B=-.27, p<.01). In
addition, cognitive ability (B=.16, p<.05) and trait performance-avoid GO (B=.l6, p<.05)
had significant main effects on time studied.

Summary. Neither trait nor state avoid or prove performance GO moderated the

relationship between feedback and effort. There was a fairly consistent main effect for

46

feedback such that those who received negative feedback actually tried harder, and
reported trying harder than those who received positive feedback.

Hypotheses 3a-3b. This set of hypotheses addressed the moderating effect of SRF
on the relationship between feedback and effort. It was expected that the dimensions of
SRF would have a similar moderating effect between feedback and effort as the
dimensions of performance GO. Therefore, Hypotheses 3a and 3b mirrored Hypotheses
1a and lb such that prevention and promotion SRF were expected to act like
performance-avoid and performance-prove GO, respectively. The specific hypotheses

are as follows:

0 Prevention SRF will moderate the relationship between performance feedback
and time spent studying (Hypothesis 3a).

0 Prevention SRF will moderate the relationship between performance feedback
and self-reported effort (Hypothesis 3a).

0 Promotion SRF will moderate the relationship between performance feedback
and time spent studying (Hypothesis 3b).

0 Promotion SRF will moderate the relationship between performance feedback
and self-reported effort (Hypothesis 3b).

As with G0, trait measures of SRF were completed by participants prior to
beginning the experiment. Although SRF was not manipulated, the analyses testing
Hypotheses 3a and 3b still included the state GO manipulations in order to partial out any
effect the manipulations may have had on effort. A similar set of hierarchical regressions
was performed for each hypothesis as described for Hypotheses 1a and lb. Identical
analyses were used to first regress effort as measured by the time spent studying on the
independent variables (Table 5). These analyses were followed by regressing self-

reported effort on the same set of variables (Table 6). In all cases, cognitive ability as

47

measured by self-reported ACT/ SAT scores was entered in Step 1 as a control variable.
In the same step, the independent variables were added. In the case of Hypotheses 3a and
3b there were 3 critical ones: feedback, prevention SRF, and promotion SRF.

Entered as the second step in the regression analysis were the interactions. In this
case, all two-way interactions were entered. In addition, three-way interactions were
assessed for those variables showing significant two-way interactions. In doing so, a
number of these were relevant to none of the stated hypotheses. The far left column of
Table 5 lists the hypotheses for which each line of the results is relevant, and, when the
analysis is relevant for none of the hypotheses, it is listed as other. Finally, the same set
of analyses is repeated using self-reported effort. The data for these analyses appear in
Table 6.

Turning specifically to Hypothesis 3a, there were no significant interactions
between prevention SRF and feedback variables as hypothesized for either type of effort.
The analysis for Hypothesis 3b showed similar results. There were no significant
interactions between promotion SRF and feedback variables as hypothesized.

Some main effects were found. Feedback had a main effect on both time studied
(R2=.07, p<.01; B=-.27, p<.01) and self-report effort (R2=.10, p<.01; B=—.28, p<.01). In
addition, cognitive ability (B=.15, p<.05) and promotion SRF (B=.20, p<.01) had
significant main effects on self-reported effort.

Summary. As with G0, SRF failed to moderate the relationship between
feedback and effort. However, promotion SRF had a main effect on self-report effort
such that as participants with higher levels of promotion SRF believed that they put forth

more effort.

48

Relationship Between Goal Orientation and Self-Regulatory Focus

Hypotheses 2a-2c. It was hypothesized that promotion SRF measures would be
positively correlated to mastery and performance-prove GO measures. Further, it was
hypothesized that prevention SRF measure would be positively correlated to
performance-avoid GO measures. In this case, the GO manipulation check measures
were used to represent state GO. The relationships were assessed using correlations that

can be found in Table 2. Specific hypotheses are as follows:

0 Measures of promotion SRF will be positively correlated with measures of
trait mastery GO (Hypothesis 2a).

0 Measures of promotion SRF will be positively correlated with measures of
state mastery GO (Hypothesis 2a).

0 Measures of promotion SRF will be positively correlated with measures of
trait performance-prove GO (Hypothesis 2b).

0 Measures of promotion SRF will be positively correlated with measures of
state performance-prove GO (Hypothesis 2b).

0 Measures of prevention SRF will be positively correlated with measures of
trait performance-avoid GO (Hypothesis 2c).

0 Measures of prevention SRF will be positively correlated with measures of
state performance-avoid GO (Hypothesis 2c).

Of the hypothesized relationships between these variables, promotion SRF was
positively correlated with trait mastery GO (r=.17, p<.01), state mastery GO (r=.21,
p<.01), and state performance-prove GO (r=.12, p<.05). The prevention SRF measure
was positively correlated with both the performance-prove and avoid GO manipulation
checks (r=.15, p<.05 and r=.22, p<.01, respectively).

Summm. The measures of trait mastery GO and promotion SRF were

correlated, but for the most part, the trait performance GO measures were not

49

significantly correlated with the SRF measures. However, the SRF measures were
correlated in the general hypothesized directions with the GO manipulation check
measures.
Controllability Relationships

Hypothesis 4a_-4p. This set of hypotheses addressed the relationships between GO
and controllability. First, mastery G0 was expected to be positively related to
controllability irrespective of feedback level (Hypothesis 4a). Second, perfonnance-
prove G0 was expected to moderate the relationship between feedback and
controllability (Hypothesis 4a). Both state and trait effects 'of GO were examined. In
addition, the two factors of controllability that were discussed in the methods section
were used as DV’s (outcome and development controllability. The specific forms of

Hypotheses 4a and 4b expressed in the particular variables used are listed below:

- Trait mastery GO will be positively related to outcome controllability
(Hypothesis 4a).

0 State mastery GO will be positively related to outcome controllability
(Hypothesis 4a).

0 Trait mastery GO will be positively related to development controllability
(Hypothesis 4a).

0 State mastery GO will be positively related to development controllability
(Hypothesis 4a).

0 Trait performance-prove GO will moderate the relationship between
performance feedback and outcome controllability (Hypothesis 4b).

0 State performance-prove GO will moderate the relationship between
performance feedback and outcome controllability (Hypothesis 4b).

0 Trait performance-prove GO will moderate the relationship between
performance feedback and development controllability (Hypothesis 4b).

50

0 State performance-prove GO will moderate the relationship between
performance feedback and development controllability (Hypothesis 4b).

Regression analyses were used to test Hypotheses 4a and 4b. Analyses were
used to first regress outcome controllability (Table 7), then identical analyses were
performed to regress deveIOpment controllability on the same set of variables (Table 8).
In all cases, cognitive ability as measured by self-reported ACT/SAT scores was entered
in Step 1 as a control variable. In the same step, the independent variables were added.
In the case of Hypotheses 4a and 4b there were five critical ones: feedback, trait mastery,
state mastery, trait prove, and state prove. Finally, two additional main effect variables
were entered in Step 1. Trait and state avoid variables were entered in order to explore
and partial out their effects.

Entered as the second step in the regression analyses were the interactions. In this
case, all two-way interactions were entered. In doing so, a number of these were relevant
to none of the stated hypotheses. The far left columns of Tables 7 and 8 list the
hypotheses for which each line of the results is relevant, and, when the analysis is
relevant for none of the hypotheses, it is listed as other.

Turning specifically to Hypothesis 4a, there were no significant main effects
between trait or state mastery GO and outcome controllability. There was, however, a
significant standardized regression coefficient for the relationship between trait mastery
GO and development controllability (B=.16, p<.05), although the overall R2 for this
model was not significant.

The analysis for Hypothesis 4b showed mixed results. There was no interaction
between feedback and either trait or state performance-prove GO for outcome

controllability. However, there was a significant main effect for trait performance-prove

51

GO (R2=.07, p<.01; B=-.21, p<.01). Although the model was not significant, exploratory
analyses showed a three-way interaction between feedback, trait performance-prove GO
and state performance-avoid G0 on outcome controllability (B=1.74, p<.05; see Figure
5). The results of this interaction suggest participants who were given a performance-
avoid goal reported increasing levels of controllability as trait performance-prove
increased -a trend that was consistent across feedback conditions. Conversely, feedback
appeared to play an important role in reports of controllability across trait levels of
performance-prove GO. When positive feedback was given, participants in the control
condition reported similar levels of controllability regardless of their level of
performance-prove GO. However, when given negative feedback, participants high in
performance-prove GO reported higher levels of outcome controllability. It appears that
negative feedback interacted with trait levels of performance-prove GO, causing
participants who were high in the trait to report greater levels of control over their test
score.

With regard to development controllability, there was a significant interaction
between feedback and state performance-prove GO (R2=.l7, p<.01; AR2= .12, p<.01; B:-
.22, p<.05). However, the interaction was in the opposite direction hypothesized (see
Figure 6). Those in the performance-prove condition actually expressed lower, not
higher, ratings of development controllability than those in the control group following
positive feedback. Figure 6 shows the two-way interaction between feedback and state
performance-prove GO for development controllability. As the graph indicates,
participants in the control condition reported slightly higher levels of development

controllability when given positive feedback than when given negative feedback.

52

Perhaps positive feedback, in the absence of an external goal, acted as a motivator,
causing participants to report high levels of development controllability. Conversely,
when given a performance-prove goal, participants reported higher levels of development
controllability when given negative feedback than when given positive feedback. The
graph seems to indicate that positive feedback combined with a performance-prove goal
caused participants to report less control over their ability to learn. Perhaps participants
stopped trying because they achieved their goal or felt that additional study effort would
have minimal impact as they already achieved top performance.

Exploratory analyses showed two other significant interactions in the same model.
There was a feedbackXtrait performance-avoid interaction (B=.91, p<.01) and a
feedbacsztate performance-avoid interaction (B=-.33, p<.01). These interactions are
difficult to interpret, however they seem to suggest that performance-avoid GO may have
had more of an effect on development controllability than did performance-prove GO
(see Figures 7 and 8). The graph in Figure 7 depicts the two-way interaction between
feedback and trait performance-avoid GO for development controllability. The results
are similar to those shown in Figure 6. Participants who were low in trait performance-
avoid GO reported less development controllability when given positive feedback
compared to negative feedback, similar to participants who were given perfonnance-
prove goals. Once again, positive feedback may have caused participants to stop trying
or to see little room for improvement. However, the results were opposite for participants
who were high in trait levels of performance-avoid GO. Positive feedback resulted in
high levels of reported development controllability than did negative feedback. Positive

feedback may have caused these participants to see a link between studying and their

53

ability to learn the information, while negative feedback may have caused them to see
studying as having little impact on learning.

The resulted graphed in Figure 8 appear contrary to those in Figure 7. When
given a performance-avoid goal, positive feedback caused participants to report lower
levels of development controllability than when given negative feedback, while
participants in the control condition reported slightly higher levels of development
controllability when given positive feedback. These seemingly contradictory findings
could be attributable to differences in state versus trait GO. When given a state goal,
participants may have stopped trying when they achieved the external goal, seeing further
learning or development as unnecessary, while participants graphed in Figure 7 took
positive feedback as a sign that they could learn and develop.

Summary. Trait mastery GO and developmental controllability were positively
related. There was also a significant interaction between feedback and state
performance-prove G0 on development controllability, however, it was in the opposite
direction of that hypothesized.

Hypothesis 5. Hypothesis 5 addressed the relationship between controllability
and affect. Specifically, it was expected that positive affect would increase as
controllability increased. In other words, it was expected that people would be more
satisfied and elated when ratings of controllability were higher in positive feedback
conditions, and less depressed and anxious when ratings of controllability were higher in
negative feedback conditions. According to the correlations in Table 2, there were no
significant correlations found between any type of affect and either factor of

controllability.

54

Hypothesis 6. It was expected that higher levels of controllability would be
related to higher levels of effort. Both types of total effort were examined: time studied
and self-report. In addition, both factors of controllability were assessed: outcome and
development. These relationships were all investigated using correlations from Table 2.

There was a significant correlation development controllability and self-report
effort (r=.2l, p<.01). Apparently, participants who indicated that studying and practice
could improve their score also reported that they put forth more effort during the study
time. However, the results indicate that these participants did not necessarily invest more
time studying than those who had low development controllability scores.

Summm. GO did not consistently predict participants’ levels of controllability,
and in turn, controllability failed to predict affect or effort.

Affect Relationships

Hygrthesia 7a-7c. This set of hypotheses dealt with the relationships between GO
and SRF with different types of affect. Specifically, in positive feedback situations it was
expected that mastery GO, performance-prove GO, and promotion SRF would be related
to elation. Also, it was expected that performance-avoid GO and prevention SRF would
be related to satisfaction.

In negative feedback situations, it was hypothesized that mastery GO,
performance-avoid GO, and prevention SRF would be related to anxiety. Also, it was
expected that performance-approach GO and promotion SRF would be related to
depressive-type affect.

Inherent in these hypotheses are multiple relationships. Both trait and state

(condition) GO’s are relevant, as well as all four types of affect (satisfaction, elation,

55

depression, and anxiety). It is important to recall that each of these affect-types were
fixed to feedback condition. In other words, only those who received positive feedback
completed the satisfaction and elation scales. Similarly, only those who received
negative feedback completed the depression and anxiety scales. Because scales were not
crossed by feedback level, this set of hypotheses was simply assessed using the
correlation matrix found in Table 2.

There were no significant correlations between affect and mastery or
performance-prove GO. However, those in the performance-avoid GO condition
experienced higher levels of negative affect (depression r=.27, p<.05; and anxiety F30,
p<.05) than those in the control condition. Trait performance-avoid G0 was also related
to depression (r=l 8, p<.05). In addition, trait performance—avoid G0 was negatively
correlated with elation (r=-.23, p<.01). Given the high intercorrelations of the affect
measures, it is difficult to distinguish differences between arousal and non-arousal
emotions. However, the findings with regard to performance-avoid G0 are consistent
with those of Cron, Slocum, and VandeWalle (2001).

Additionally, there were significant correlations with regard to SRF. These
correlations are less clear. It seems that both promotion and prevention SRF were
positively correlated with negative affect (depression r=.25 and .24, p<.01; and anxiety
r=.l7 and .18, p<.05). Conceptually, it does not make sense that both of these constructs
would be related in the same direction with depression and anxiety. This was likely due
to the high correlation between the arousal and non-arousal measures.

Hymthesis 8. Hypothesis 8 dealt with the relationship between arousal-type

affect (elation and anxiety) and effort. Consistent with Higgins (1997) theory, it was

56

expected that higher levels of elation and anxiety would be associated with higher levels
of effort. Contrary to what was expected, correlations from Table 2 indicated that
depression was the only type of affect that correlated significantly with effort.
Specifically, depression was positively correlated with both self-report effort and time
studied (r=.29, p<.01 and r=.30, p<.01, respectively). Therefore, Hypothesis 8 was not
supported.
Distinction Between Forms of Effort Behaviors

Hypothesis 9a-c. It was expected that participants’ goal orientations would be
related to the types of information they chose to study. There were two types of study
information: knowledge and test tips. It was hypothesized that mastery GO would be
positively related to the amount of time spent studying knowledge information, and that
avoid and prove performance orientations would be positively related to the amount of
time spent studying test tips information. Both trait and state GO's were examined.
Specific hypotheses are as follows:

0 Trait mastery GO will be positively related to time spent studying knowledge
information (Hypothesis 9a).

0 Participants in the mastery GO condition will study knowledge information
more than participants in the control condition (Hypothesis 9a).

0 Trait performance-prove GO will be positively related to time spent studying
test tips information (Hypothesis 9b).

0 Participants in the performance-prove GO condition will study test tips
information more than participants in the control condition (Hypothesis 9b).

0 Trait performance-avoid GO will be positively related to time spent studying
test tips information (Hypothesis 9c).

0 Participants in the performance-avoid GO condition will study test tips
information more than participants in the control condition (Hypothesis 9c).

57

Similar hierarchical regressions as used to test Hypotheses la and lb were used
for this set of hypotheses. Identical analyses were used to first regress effort as measured
by the time spent studying knowledge information on the independent variables (Table
9). These analyses were followed by regressing time spent studying test tips information
on the same set of variables. In all cases, cognitive ability as measured by self-reported
ACT/SAT scores and feedback were entered in Step 1 as control variables. In the same
step, the independent variables were added. In the case of Hypotheses 9a-9c, there were
six critical ones: trait mastery, state mastery, trait prove, state prove, trait avoid, and state
avoid.

Entered as the second step in the regression analysis were the interactions. In this
case, all two-way interactions were entered. In doing so, a number of these were relevant
to none of the stated hypotheses. The far left column of Table 9 lists the hypotheses for
which each line of the results is relevant, and, when the analysis is relevant for none of
the hypotheses, it is listed as other.

With regard to time spent studying knowledge information, there were weak
results. There were some significant main effects for cognitive ability (B=.15, p<.05) and
feedback (B=-.22, p<.01), however the overall model of the first regression was not
significant (see Table 9). No overall model effects or significant standardized regression
coefficients were found with regard to time spent studying test tips information.
Therefore, no relationships were found between type of GO and time spent studying test
tips.

Exploratory analysis was performed to investigate the effects of SRF on choice

of study information. The two regressions used for Hypotheses 3a and 3b were used with

58

time spent studying knowledge information and time spent studying test tips information
as the dependent variables. Again, no effects were found for time spent studying test tips
information, and the overall model for knowledge information was not significant (see
Table 10). However, feedback had a main effect on time spent studying knowledge
information (B=-.22, p<.01).
Other Findings

Hypothesis 1. Exploratory analyses for Hypothesis 1 showed some significant
interactions between trait and state GO's on both types of effort (see Tables 3 and 4). The
models in which these interactions were found were not significant, however, the
interactions showed some interesting trends. First, Table 3 shows that for total time
studied, there was a significant interaction between trait mastery GO and state
performance-prove GO (B=-l .23, p<.05; see Figure 9). When the trait measure of
mastery G0 was low, study time was slightly less for the prove condition when compared
to the control condition. Individuals with a prove GO tended to be more focused on
performance than leaming. However, when the trait measure of mastery G0 was high,
individuals in the prove condition spent more time studying than those in the control
condition. The interaction implies that mastery-oriented individuals spent more time
studying when given a performance-prove oriented goal than when given no goal (i.e.,
control condition). It appears that the performance-prove manipulation enhanced
participants’ goal orientation trait, when that trait was strong. In this interaction,
mastery-oriented participants tended to study more when given the performance-prove

goal.

59

Second, Table 3 shows an interaction between trait performance-avoid GO and
state performance-prove GO (B=-.95, p<.05; see Figure 10) on total time studied. In this
interaction, participants who were low in the performance-avoid trait tended to study
more when given a performance-prove goal than those who were given no goal (i.e.,
control condition). However, participants high in the performance-avoid trait spent less
time studying when given a performance-prove goal than those who were given no goal.
As with Figure 9, the prove manipulation appeared to enhance participants’ natural
tendencies when the GO trait was strong. Thus, high performance-avoid participants
studied less when given a performance-prove goal, as one would expect given their
tendency to exert just enough effort in order to avoid performing poorly.

Finally, Table 4 shows a significant interaction between trait mastery GO and
state performance-prove GO ([3=-1.15, p<.05; see Figure 11) on effort. The control
condition was consistent with what one might expect: as mastery GO increased, so did
study effort. However, the most critical portion of the graph appears to be the differences
in self-reported for the performance-prove condition. When given a performance-prove
goal, participants who were low in mastery-orientation reported more study effort than
participants who were high in mastery GO. It is possible that participants who were low
in mastery GO truly studied harder to score higher on the tests, however, with a self-
reported measure, it is difficult to rule out impression management as another plausible
explanation for the findings.

Hypothesis 3. Table 5 shows the exploratory analyses performed for Hypothesis
3. The second step of this regression was significant (R2=.l4, p<.01; AR2= .08, p<.05)

and showed a significant interaction between promotion SRF and state mastery G0 on

60

total time studied (B=-l.l6, p<.01; see Figure 12). The graph of this interaction is
contrary to what one would expect. When given a mastery-oriented goal, participants
tended to spend more time studying if they were low in promotion SRF than when high in
promotion SRF. However, the opposite would be expected: promotion SRF should
combine with mastery GO to enhance study time further. This contrary finding was more
pronounced when compared to the control condition. In the absence of a goal, high
promotion participants studied considerably more than high promotion participants who
were given a mastery goal. One explanation is that by telling high promotion-focused
participants not to focus on performance, the mastery GO manipulation actually increased
attention on performance, thus reducing study time.

The third model in Table 5 was also significant, but not incrementally above the
second model (R2=.17, p<.01). It contained a significant three-way interaction between
feedback, promotion SRF, and state performance-avoid GO (B=-l .26, p<.05; see Figure
13). This three-way interaction is consistent with G0 and SRF theories. When given
positive feedback, low promotion-focused participants in the control condition reported
less study effort than those in the performance-avoid condition. However, participants
who were high in promotion focus reported greater study time in the control condition but
less study time in the performance-avoid condition. This interaction indicates that in the
absence of goal manipulation, high promotion-focused participants exerted more study
effort when given positive feedback, consistent with G0 theory. However, when given a
performance-avoid goal and positive feedback, high promotion-focused participants

reported less study effort than those in the control condition.

61

In the negative feedback situation, regardless of condition, participants reported
greater study time as levels of promotion SRF increased. However, participants reported
greater study time when given a performance-avoid goal than when given no goal at all.
Consistent with the research, this finding supports the notion that performance-avoid
participants tend to exert more effort when given negative feedback and this trend is
enhanced when the participant has a promotion SRF.

Table 6 also shows a significant three-way interaction. For this regression, the
third model was significant (R2=.l9, p<.01), however did not add significantly to the
variance explained in the first and second models. The three-way interaction in the third
model mimicked the three-way interaction found with regard to study time (see Figure
14). Feedback, promotion SRF, and state performance-avoid GO formed a three-way
interaction on self-report effort (R2=.19, p<.01; B=-l .21, p<.05). Therefore, the three-
way interaction graphed in Figure 14 is consistent with the interaction gaphed in Figure
13. When given positive feedback, high promotion-focused participants spent more time
studying than low promotion-focused participants and participants in the performance-
avoid condition. There appeared to be little change in the study time of participants in
the performance-avoid condition, despite the level of promotion SRF. As with Figure 13,
participants in the performance-avoid condition spent more time studying than
participants in the control condition as indicated by the parallel lines. Within each
condition, the amount of study time also increased for high promotion-focused
participants in the presence of negative feedback. Once again, this finding is consistent
with the theory. When no goal manipulation was given, high promotion-focused

participants spent more time studying, which is similar to what one would expect from

62

participants high in mastery GO. However, for participants in the performance-avoid
condition, negative feedback resulted in more study time when compared to the control
condition. Study time also increased for participants who were high in promotion SRF.
Hypothesis 9. Although none of the steps are significant in Table 9. There was a
consistent pattern of interactions between trait mastery and state prove on time spent
studying knowledge material. There was a significant two-way interaction between trait
mastery and state performance-prove GO (B=-1.23, p<.05), however, this relationship
seemed to be due to a three-way interaction between feedback, trait mastery, and state
prove (B=-1.83, p<.05). Figure 15 depicts the three-way interaction between feedback,
mastery GO, and the performance-prove condition. The interaction is consistent with G0
research. When given positive feedback, participants high in mastery GO spent more
time studying knowledge materials, (i.e., focused on learning), than those low in mastery
GO and participants in the performance-prove condition. Thus, positive feedback
appeared to motivate mastery-oriented participants to spend more time learning when no
external performance goal is given. Conversely, when given negative feedback and a
performance-prove goal, participants spent less time studying knowledge information
than participants in the control condition. However, this difference was more pronounced
for participants high in mastery-orientation. When given negative feedback and a
performance-prove goal, participants high in mastery GO tended to spend less time
studying knowledge material than participants low in mastery-orientation and even less
time than participants in the control condition. It appears that the performance-prove
manipulation re-directed the study efforts of high mastery-oriented participants away

from knowledge information.

63

Table 10 shows a significant interaction between feedback and state performance-
avoid GO ([3=-.23, p<.05; see Figure 16). However, again, it is important to recognize
that neither of the models in the regression shown in Table 10 were significant. Figure
16 depicts the two-way interaction between feedback and performance-avoid condition.
When no goal manipulation was given, participants who received positive feedback
typically spent less time studying knowledge information than participants who were
given negative feedback. However, when participants were given a performance-avoid
goal the results were more pronounced. Negative feedback resulted in slightly more time
spent studying knowledge materials when compared to the control condition, but
considerably less time compared to the control condition when positive feedback was
given. These data are consistent with the conclusion that positive feedback combined
with a performance-avoid goal caused participants to spend less time studying when their
performance reached a minimum criterion. Thus the performance-avoid goal may have
been de-motivating in the presence of positive feedback.

Finally, Table 10 also showed a significant interaction between promotion SRF
and state performance-prove GO (B=-.97, p<.05; see Figure 17). Consistent with Figure
15, Figure 17 shows that giving a performance-prove goal caused high promotion-
focused participants to spend less time studying knowledge information than low
promotion-focused participants and less time than high and low promotion-focused
participants in the control condition. Once again, it would appear that giving a
performance-prove goal re-directed attention away from knowledge information or

learning.

64

Summgy. As a whole, the interactions were consistent with the notion that
performance GO had a stronger impact on effort than mastery GO, regardless of which
was the state and which was the trait. Furthermore, when compared across performance
orientations, performance-avoid appeared to have a stronger impact on effort than did
performance-prove or promotion SRF regardless of which was the state and which was

the trait. These effects were firrther enhanced in the presence of feedback.

65

DISCUSSION

In close to every aspect of our lives, we are inundated by feedback. Feedback
may come in various forms, and we may choose to ignore it or accept it, but the goal
inherent in almost all types of feedback messages is to maintain or change behavior. In
the workplace, feedback canies special significance and is often linked to important
internal and external outcomes. In some cases, negative feedback can indicate that one is
performing poorly on the job and is unlikely to be promoted, and/or it can result in
lowered self-efficacy and negative affect. In other cases, negative feedback can motivate
people to improve their behaviors and face tough challenges. Similarly, positive
feedback can indicate that one’s work will be rewarded, and/or it can result in increased
self-efficacy and positive affect. However, positive feedback may also decrease one’s
motivation move past an already successfully reached goal.

If the goal in giving feedback is to change one’s behavior, it is crucial that we
understand how feedback works. Unfortunately, there are no clear answers here (Kluger
& DeNisi, 1996). Reactions to feedback are impacted by individual differences,
cognitions, how the feedback is framed, and of course by the environmental conditions
under which it is delivered.

Research has shown GO and SRF to be related to reactions to feedback. These
constructs are unique in that they are considered to be both individual differences as well
as environmentally induced states. This study was set out to further our understanding of
reactions to performance feedback. The hypotheses emerged from a combination of the

GO and SRF literatures.

66

This study had four goals. First, we wanted to replicate the moderation effects of
GO and SRF on the relationship between feedback and effort. Second, we wanted to
explore the relationship between GO and SRF themselves. Third, we incorporated the
process variables of controllability and affect in order to gauge their impact on reactions
to feedback. Last, we set out to address some of the limitations inherent in past research
with regard to measuring effort and differentiating between state and trait G0. The
following section addresses the findings for each goal, limitations of the research in
attaining each goal, and implications for future research.

Moderating Effects of GOand SRF

Hypotheses 1 and 3 addressed the moderating effects of GO and SRF,
respectively, on the relationship between feedback and effort. There was no support for
either of these hypotheses. There are at least three possible reasons for this. First, the
negative feedback manipulation in this study had a strong main effect on effort, which
may have left little room for more subtle effects of GO or SRF. Second, both the trait
and state performance-avoid constructs appeared to have a dominant effect relative to the
other GO and SRF constructs. Third, the task itself provided a strong situation that was
heavily performance oriented and familiar to participants.

Pervasive throughout the analyses was negative feedback’s effect on various types
of effort. Despite Kluger and DeNisi’s (1996) claim that the feedback message itself is
not solely responsible for changes in behavior, it was very consistent in this study.
Participants in every condition, regardless of levels of trait GO and SRF, tried harder in
response to negative feedback. Reasons for negative feedback’s strong effect are likely

due to the nature of this task. First, this was a cognitively-based, achievement-oriented

67

testing situation. Given that college students are likely used to receiving scores (or
grades) in their classes, and usually put a premium on high grades, they may have
differentially focused on this aspect of the task.

Another reason for negative feedback’s strong effect could be that the feedback
itself contained an implicit goal. The feedback given to participants was normative, and
therefore lent itself to natural goal-setting. Even those in the control condition were
likely to evaluate where they stood relative to others and subsequently set personal goals
for themselves, particularly in cases where they were told that they performed less well
than others. Individuals in the other conditions were given explicit goals. However,
these implicit, self-set goals may have over-ridden any effects that the manipulation-
stated goals may have had on these individuals. Particularly considering that this was a
performance-based task, mastery GO and promotion SRF had little effect on participants’
effort levels. This lends support to the theory that perhaps people were setting their own
goals in response to the feedback.

The exploratory analyses revealed an interesting finding that could have also had
implications for the lack of GO and SRF moderation effects. Although most of the
interactions analyzed in the exploratory analyses were not significant, they showed a
trend with regard to performance-avoid GO: Performance-avoid GO had a pervasive
effect on effort. Whether individuals naturally tended toward a performance-avoid GO,
or are put in a performance-avoid condition, they exhibited behaviors typical of that
construct. In other words, performance-avoid individuals tried harder in response to

feedback, particularly negative feedback, than did other individuals.

68

This domination effect of performance-avoid GO may have been partly due to
the nature of the task. The GRE task was quite performance oriented for two reasons.
First, the task was a test taking exercise. Although effort was measured during the study
time, the participants were likely influenced to be thinking about performance given that
they received feedback on tests.

Second, the GRE task in this study was performance oriented because individuals
in all of the conditions were given normative feedback. Therefore, even participants in
the mastery condition were told how they performed relative to other participants.
Normative comparisons are common in performance GO manipulations. Such feedback
was chosen for this study in order to enhance the meaningfulness of the feedback and
maintain consistency of feedback messages across the conditions.

In addition to the task being performance oriented, it was also a familiar one for
most participants. Given that all of the participants were college students, most took
either the ACT or SAT to gain admission to college. These tests are similar to the GRE.
Given past experience on such tests, this nature of this task could easily have primed
students to consider how they have prepared for and performed on these tests in the past.
It is fairly well recognized that studying for 15 minutes is not going to substantially raise
someone’s GRE score. This perspective could easily have overridden any orientation
individuals had toward learning or performance. In a sense, past experience with these
types of exams could have served as a barrier for the manipulations if the participants
believed that they could neither improve their performance nor learn from the study

SCI'CCI‘IS.

69

Unfortunately, the implications of the nature of the task do not fully explain the
results. Given that the task and feedback were performance oriented, we might expect
this to lead to a strengthening of the hypothesized relationships affected by the
performance GO manipulations. However, this did not prove to be the case, the
hypothesized relationships were not significant. In addition, the GO manipulation check
appeared to correlate more highly with the trait GO items than with the manipulations.
_Re_lationship Between GO and SRF

Hypothesis 2 addressed the relationship between GO and SRF variables. As
discussed in the results, trait mastery GO and promotion SRF were correlated. However,
measures between the rest of the constructs were not correlated. This is likely a factor of
the quality of the SRF measures. As described in the methods section, a SRF measure
was constructed specifically for this study and was not related to an established SRF
measure (the Selves Questionnaire).

Despite the lack of significant results for the direct test of the hypotheses,
exploratory analysis, although largely not significant, did present some interesting trends.
For instance, Figures 15 and 17 show that individuals who are high on both trait mastery
GO and promotion SRF responded to performance-avoid goal situations in similar ways.
These individuals studied knowledge information less time when given a performance-
prove goal than those who were not given a performance-avoid goal (control condition).
Although this trend was not consistent, it lends support for firrther study of the
relationships between these variables, particularly considering the limitations presented

by the quality of the scales used in this study.

70

Process Variables

The two process variables under consideration in this study were controllability
and affect. There were few significant findings that corresponded with the hypotheses for
these variables. The affect variables did not relate as expected with G0 or effort. This
may have been partly due to the quality of the affect measures. As discussed in the
methods section, the arousal and non-arousal components of the affect scales were highly
correlated.

Despite the lack of relationships shown for Hypotheses 4 through 7, past research
has shown that this is a fruitful line of inquiry. It is fairly well established that
attributions are a firndamental part of response to events (F iske & Taylor, 1991; Weiner,
1986). In addition, researchers have found that negative discrepancies between current
performance and self-set goals create emotional responses that can serve as an incentive
for increased effort (V andeWalle, 2001; Bandura, 1997).

Given feedback, individuals are bound to feel something about the feedback, and
are likely to make attributions for their performance. It comes as no surprise that
individuals will be less satisfied with negative feedback than with positive feedback.
With respect to attributions, research has shown that in receipt of positive feedback,
individuals will attribute their behavior to themselves and conditions they control rather
than factors not under their control. Likewise, negative feedback is more likely to
produce performance attributions to external factors or, if internal, to things over which
the person has no control. Therefore, it is still worth exploring how to increase feelings
of controllability in feedback situations, particularly when considering reactions to

negative feedback (Ilgen & Davis, 2000).

71

Distinction Between Forms of Effort Behaviors

The first distinction of effort we examined was actual study time versus self-
reported study effort. As discussed in the methods section, the two measures were highly
correlated (r=.54, p<.01). Overall, the findings from analyses using study time as the DV
were similar to the findings using self-reported effort as the DV. It appears that
individuals were fairly honest and accurate in reporting their effort levels.

The second distinction of effort we examined was time spent studying knowledge
information versus test taking heuristics. Hypothesis 9 proposed that mastery-oriented
individuals would be more likely to spend time studying knowledge information, and that
performance-oriented individuals would be more likely to study test tips information.
However, Hypothesis 9 was not supported. There are at least two possible reasons for
this. First, it seems that participants had a difficult time telling the difference between the
types of study screens. The mean responses for questions pertaining to the differences
between screens were 3.50 (verbal) and 3.60 (quantitative). Although these means were
higher than those obtained from pilot studies, they were still less than ideal.

In addition, individuals spent much more time on the knowledge screens as
opposed to the test tips screens overall (knowledge mean $19.79 seconds, test tips
mean=268.42 seconds). The test tips screens were also rated as being less helpful than
the knowledge screens for learning (test tips mean=2.84, knowledge mean=3.51) and
improving scores (test tips mean=2.98, knowledge mean=3. 14). This was likely due to
the format and types of information in the tests. It could be that these participants were
already familiar with the test format and types of test questions. If so, then they more

than likely did not need to focus on test taking heuristics. Alternatively, the test content

72

may have required a need for “brushing up” on past knowledge. A prime example of this
would be the geometry questions. Individuals may have felt that if they reviewed the
content information in geometry, they would have a better chance of improving their
performance.
Study Limitations

Given the implications and limitation of the results, it is also important to
consider some of the broader limitations that may influence the generalizability of this
study.

T_asi<; The achievement nature of the task itself may have overridden the
likelihood that individuals would exhibit behaviors associated with a mastery GO or
promotion SRF. In addition, participants found this task to be dull, and therefore, were
less likely to feel motivated to perform. This was confirmed by the moderate to low
means on the motivation measures. In addition, when given options to engage in the task,
surf the net, or watch the clock count down the end of the study time, the experimenter
observed many of the participants choose to simply watch the timer.

In addition, it seems that people were not likely to believe their feedback,
particularly in negative feedback situations. It is a common finding that individuals are
less likely to accept negative feedback, however, feedback in this study was based on true
scores in order to alleviate the effect of low believability in negative feedback situations.
This could be due to the fact that even though participants’ feedback conditions were
based on their true performance, individuals’ feedback messages were not accurate as to
their true performance. Participants who answered eight or more questions correctly

received the positive feedback message, and those who answered seven or less questions

73

correctly received the negative feedback message. Therefore, the feedback messages for
the first test were not individualized based on true score.

Measures. As discussed in the methods section, a number of crucial measures in
this study had to be constructed. Measures were created for SRF, controllability, and all
forms of affect. Using non-established measures led to uncertainty about the quality and
validity of the measurements. In addition, the manipulation appeared to have little effect
on the outcome variables within each hypothesis. However, problems arose when
attempting to determine the effectiveness of the manipulations because the manipulation
check measure had poor structure. The manipulation check items were highly
intercorrelated and did not factor well into the expected categories. Therefore, it was
difficult to ascertain whether the manipulations did not work, or worked but failed to
produce the hypothesized effects.

Future Research

Inclusion of other variables. As with any one study, there were important
variables left out of the experiment. Variables such as self-efficacy, are likely to play a
role in reactions to feedback. Self-efficacy has been shown to be related to both
attributions and affect (Bandura, 1997). It could possibly help further explain the
relationships between affect and attributions and controllability. In this study, two types
of controllability examined may have tapped into similar constructs as self-efficacy.
They both examined individuals’ beliefs about their ability to change their behaviors.
Essentially, individuals were asked about whether or not they thought they could control
their performance on the second test, and if they thought that they could improve their

performance through studying. These constructs approach the types of issues that self-

74

efficacy might tap. However, self-efficacy may play a role earlier in the process, as well
as at many points throughout the process.

Differences between GO and SRF. Given that GO and SRF were largely found to
be unrelated in this study, this leaves room for further research to investigate the
relationships between these constructs. As noted in some of the exploratory analyses,
trait and state GO did have interactions with promotion SRF. Perhaps research designed
specifically for the purpose of isolating these interactions would shed some light on the
distinction between these variables.

Conclusion

Despite the lack of significant findings that emerged from this study, there still
remains an interesting question with regard to the relationship between two seemingly
parallel constructs. Theoretically, GO and SRF have many similarities, and their
respective literatures could possibly benefit from exploring their relationships with each
other, and with various process variables such as attributions and affect. Some areas that
need to be improved for future research include better measures of SRF and affect. In
addition, given the complexities of separating out the various state and trait components

in this study, it might be beneficial to investigate these various effects individually.

75

 

REFERENCES

Bandura, A. (1997). Self-efficacy: The exercise of control. New York: W. H.
Freeman & Company.

Bandura, A., & Wood, R. (1989). Effect of perceived controllability and
performance standards on self-regulation of complex decision making. Journal of

Personalig and Social Psychology, i6, 805-814.

Bass, B. M., Cascio, W. F ., & O’Connor, E. J. (1974). Magnitude estimations of
expressions of frequency and amount. Journal of Applied Psychology, 5_9_, 313-320.

Butler, R. (1987). Task-involving and ego-involving properties of evaluation:
Effects of different feedback conditions on motivational perceptions, interests, and
performance. Journal of Educational Psychology, B, 474-482.

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, 6_7, 26-48.

 

Carver, C. S., & Scheier, M. F. (1981). Attention and self-regulation: A control
theog; approach to human behavior. New York: Springer-Verlag.

Carver, C.S., & Scheier, M. F. (1982). Control theory: A useful conceptual
framework for personality-social, clinical, and health psychology. Psychological
Bulletin, 2, 111-135.

Cron, W.L., Slocum, J.W., & VandeWalle, D. (2001, February). Opal
disconfirmation and self-setfll level: The role of emotional reactions and go_al
orientation. Paper presented at the 16th annual conference of the Society for Industrial
and Organizational Psychology, San Diego, CA.

Diener, C. 1., & Dweck, C. S. (1978). An analysis of learned helplessness:
Continuous changes in performance strategy and achievement cognitions after failure.
Journal of Personalig: and Social Psychology, &, 451-462.

Diener, C. 1., & Dweck, C. S. (1980). An analysis of learned helplessness: II.
The processing of success. Journal of Personality and Social Psycholpgy, 32, 940-952.

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

Dweck, C. S. (1989). Motivation. In A. Lesgold and R. Glaser (Eds),
F ounfirtimra for a psychology of education. Hillsdale, NJ: Erlbaurn.

76

Dweck, C. S. (1999). Self-theories: Their role in motivation, personality, and
development. Philadelphia, PA: Psychology Press.

Dweck, C. S., & Leggett, E. L. (1988). A social-cognitive approach to
motivation and personality. Psychological Review. _9_5, 256-273.

Elliot, A. J. (1999). Approach and avoidance motivation and achievement goals.
Educational Psychologist, 351, 169-189.

Elliot, A. J., & Church, M. A. (1997). A hierarchical model of approach and
avoidance achievement motivation. Journal of Personality and Social Psychology, Q,
218-232.

Elliot, A. J., & Harackiewicz, J. M. (1996). Approach and avoidance
achievement goals and intrinsic motivation: A mediational analysis. Journal of

Personalig and Social Psychology, 19, 461-475.

Elliot, A. J., & McGregor, H. A. (1999). Test anxiety and the hierarchical model
of approach and avoidance achievement motivation. Journal of Personality and Social
Psychology, 16, 628-644.

Elliott, E. S., & Dweck, C. S. (1988). Goals: An approach to motivation and
achievement. 101ml of Personality and Social Psychology, fl, 5-12.

Fiske, S. T., & Taylor, S. E. (1991). Social cognition. New York, NY: McGraw-
Hill.

Frese, M., & Zapf, D. (1994). Action as the core of work psychology: A German
approach. In H. C. Triandis, M. D. Dunnette, & L. M. Hough (Eds), Handbook of
industrial and organizational psychology, (2nd ed., Vol. 4, pp. 271-340). Palo Alto, CA:
Consulting Psychologists Press.

Higgins, E. T. (1987). Self-discrepancy: A theory relating self and affect.
Psychological Review, 9_4, 319-340.

Higgins, E. T. (1996). The “self digest”: Self-knowledge serving self-regulatory
functions. Journal of Personality and Social Psycholog, ll, 1062-1083.

Higgins, E. T. (1997). Beyond pleasure and pain. American Psychologist, 5_2,
1280 —1300.

Higgins, E. T. (1998). Promotion and prevention: Regulatory focus as a
motivational principle. Advances in Experimental Social Psychology, 30, 1—46.

Higgins, E. T., Bond, R. N., Klein, R., & Strauman, T. (1986). Self-
discrepancies and emotional vulnerability: How magnitude, accessibility, and type of

77

discrepancy influence affect. Journal of Personalig and Social Psycholcgy, 5_l_, 5-15.

Higgins, E. T., Klein, R., & Strauman, T. (1985). A psychological model for
distinguishing among different aspects of depression and anxiety. Social Cognition. 3,
51-76.

Higgins, E. T., Roney, C. J. R., Crowe, E., & Hymes, C. (1994). Ideal versus
ought predilections for approach and avoidance: Distinct self-regulatory systems. Journal
of Personality and Social Psycholgjgy, fl, 276-286.

 

Higgins, E. T., Shah, J ., & Friedman, R. (1997). Emotional responses to goal
attainment: Strength of regulatory focus as moderator. Journal of Personality and Social
Psychology, 2, 515-525.

Ilgen, D. R., & Davis, C. A. (2000). Bearing bad news: Reactions to negative
performance feedback. Applied Psychology: An International Review. fl, 550-565.

Ilgen, D. R., Fisher, C. D., & Taylor, M. S. (1979). Consequences of individual
feedback on behavior in organizations. J oumal of Applied Psychology, 95, 349-371.

J agacinski, C. M. (1992). The effects of task involvement and ego involvement
on achievement-related cognitions and behaviors. In D. H. Schunk, J. L. Meece, et al.
(Eds), Studentaperceptions in the classroom (pp. 307-326). Hillsdale, NJ: Erlbaum.

Kanfer, R. (1990). Motivation theory and industrial and organizational
psychology. In M. D. Dunnette & L. M. Hough (Eds), Handbook of industrial and
organizational psychology, (2"d ed., Vol. 1, pp. 75-170). Palo Alto, CA: Consulting
Psychologists Press.

Kluger, A. N. (2000, February). Needs. self-regulation. and risl_< preference?
Paper presented at the 15th annual conference of the Society for Industrial and
Organizational Psychology, New Orleans, LA.

Kluger, A. N., & DeNisi, A. (1996). The effects of feedback interventions on
performance: A historical review, a meta-analysis, and a preliminary feedback
intervention theory. Psychological Bulletin. 119. 254-284.

Kluger, A. N., & DeNisi, A. (1998). Feedback interventions: Toward the
understanding of double-edged sword. Current Directions in Psychological Science, 1,
67-72.

Kluger, A. N., Lewinsohn, S., & Aiello, J. R. (1994). The influence of feedback
on mood: Linear effects on pleasantness and curvilinear effects on arousal.

Organizational Behavior and Human Decision Processes. @, 276-299.

Kluger, A. N., Van-Dijk, D., Kass, R., Stein, E. Z., & Lustig, H. (2000,

78

February). Positive (negative) feedback: Encouragement or discouragement? Paper

presented at the 15"I annual conference of the Society for Industrial and Organizational
Psychology, New Orleans, LA.

 

Lazarus, R. S. (1991). Progress on a cognitive-motivational-relational theory of
emotion. American Psychologiat, aa, 819-834.

Locke, E. A., & Latham, G. P. (1990). A theoryof goal setting and task
performance. Englewood Cliffs, NJ: Prentice Hall.

Martocchio, J. J. (1994). Effects of conceptions of ability on anxiety, self-
effrcacy, and learning in training. Journal of Applied Psychology. 3, 819-825.

Martocchio, J. J ., & Dulebohn, J. (1994). Performance feedback effects in
training: The role of perceived controllability. Personnel Psychology. 111, 357-373.

Mikulincer, M. (1994). Human learned helplessness: A coping perspective. New
York: Plenum Press.

Miller, D. T., & Ross, M. (1975). Self-serving biases in the attribution of
causality: Fact or fiction? Psychological Bulletin, 8_2_, 213-225.

Mitchell, T. R. (1974). Expectancy models of job satisfaction, occupational
preference, and effort: A theoretical, methodological, and empirical appraisal.
Psychological Bulletin, 8_l, 1053-1077.

Ortony, A., Clore, G. L., & Collins, A. (1988). The cognitive structure of
emotions. Cambridge: Cambridge University Press.

Peiperl, M. A. (2001). Getting 360-degree right. Harvard Business Review,
January 2001, 142-147.

Roney, C. J. R., Higgins, E. T., & Shah, J. (1995). Goals and flaming: How
outcome focus influences motivation and emotion. Personality aid Social Psychology
Bulletin, ;1_, 1151-1160.

Roseman, I. J. (1984). Cognitive determinants of emotion: A structural theory.
Review of Personality and Social Psychology. 5, 11-36.

Roseman, I. J ., Spindel, M. S., & Jose, P. E. (1990). Appraisals of emotion-
soliciting events: Testing a theory of discrete emotions. Journal of Persomrlity and Social
Psychology, 52, 899-915.

Russell, D. (1982). The causal dimension scale: A measure of how individuals
perceive causes. Journal of Personality and ScLial Psychology, 5;, 1137-1145.

79

Russell, J. A. (1980). A circumplex model of affect. Journal of Personality and
Social Psychology, &, 1152-1168.

Russell, J. A. (1991). Categories of emotions. Psychological Bulletin, 110, 426-

 

450.

Scott, L., & O’Hara, M. W. (1993). Self-discrepancies in clinically anxious and
depressed university students. Journal of Abnormal Psychology, m2, 282-287.

Strauman, T. J., & Higgins, E. T. (1987). Automatic activation of self-
discrepancies and emotional syndromes: When cognitive structures influence affect.
Journal of Personality and Social Psychology, 53, 1004-1014.

Taylor, M. S., Fisher, C. D., & Ilgen, D. R. (1984). Individuals’ reactions to
performance feedback in organizations: A control theory perspective. In K. Rowland &

J. Ferris (Eds), Research in personnel and human resource management (Vol. 2, pp. 81-
124). Greenwich, CT: JAI Press.

VandeWalle, D. (1997). Development and validation of a work domain goal
orientation instrument. Educational and Psychologigrl Measurement. 52, 995-1015.

VandeWalle, D., Brown, S.P., Cron, W.L., & Slocum, J .W. (1999). The
influence of goal orientation and self-regulation tactics on sales performance: A
longitudinal field test. Journal of Applied Psycholagy, 8_4, 249-259.

VandeWalle, D., Cron, W.L., & Slocum, J .W. (in press). The role of goal
orientation following performance feedback. Journal of Applied Psychology.

Vallacher, R. R., & Wegner, D. M. (1987). What do people think they’re doing?
Action identification and human behavior. Psychologgrl Review. _9_4_, 3-15.

~Weiner, B. (1980). Human motivation. New York: Holt, Rinehart & Winston.

Weiner, B. (1985). An attributional theory of achievement, motivation, and
emotion. Psychological Review. E, 548-573.

Weiner, B. (1986). An attributional theory of motivation and emotion. New
York: Springer-Verlag.

80

ENDNOTES

' Because technically participants will have no other measure of their
performance except for the feedback they receive, performance and feedback will be
treated synonymously here.

2 All of the test questions used in this study were taken from the following
source, reprinted by permission of the Educational Testing Service, the copyright owner.

GRE Practicing to Take the General Test #9, 9th Edition, 1994

3 All of the study materials used in this study were taken from the following
sources, reprinted by permission of the Stanley H. Kaplan Educational Center, the

copyright owner.

Lesson Book: GRE Preview, Classroom, Review, 1996

Verbal Home Study: GRE Vocabulary Review List and Word Root List, 1995
Mathematics Home Study: GRE Problem Solving, 1995

Mathematics Home Study: GRE Quantitative Comparisons and Graphs, 1995

4 Twenty-seven participants were dropped from the study for various reasons.
The computer program failed to collect data for some of the participants, and some
participants failed to follow experimenter instructions.

5 Ethics in psychology information was taken fi'om portions of the American
Psychological Association Ethics Code (version 1992). The APA Ethics code can be
seen on the APA website (www.apa.org). Inquiries concerning the substance or
interpretation of the APA Ethics Code should be addressed to the Director, Office of
Ethics, American Psychological Association, 750 First Street, NE, Washington, DC
20002-4242.

81

@2388: 00 use mam 2: mo 5382825 firm a co wows: Eve—2 ._ 83w?—

    
      

mDOOu
>m0h<430wm
-mmmw

¥0<mome

mZO_.rDm_m.E.<

ZO_._.<._.zm_mO
4(00

 

82

 

wI

mI

wZO_._.Dm_mt.<

55% $5 8% 388 33388 2E. .m oSwE

98:

\

9.2.:

 

 

 

$300“.

 

o-mmI

>mOH<JDmedjmw
\ZO_._.<._.Zm=mO .200

 

 

 

¥0<mommm

83

4
00

.toto cam 6.0330.“ cough—etc; 28253 3:30:22 2: .3 B88262: mm mam use 00 .m Rama

¥0<mommu

 

9mm:
9m 2..

 

wDOOu
>m0h<430mmi4ww
\ZO.._.<._.Zw_mO 4(00

 

 

 

 

 

$32.23 28w
teem :eauxéom
c2325:

mumswém—z

 

 

 

 

EEO:

 

E... :
Ream

Seq—cocoa 05 we Eco—z .v oSwE

 

 

 

 

 

 

 

 

 

 

 

.30 2 5:80 v88”.
A > > >
03% Pmmz. v.m<.—. >QDPm
¥U<mowmm N ._.mm._. xomIU ZOHSDEEE .00

 

 

 

 

 

 

mZO Emmi.

 

 

 

 

 

.mdm
.06 :8...
8298950

mmm3w<w2

 

 

 

 

 

mZO_.—.mmDO mam—245nm

 

 

 

¥U<mommm _ HmmE.

 

 

 

 

 

_ ZOHan=Z<Z .0.0

 

 

 

 

 

BmmmSJOthOU

 

 

 

 

 

N ZOFSDEZ<2 .Od

 

 

85

Hypothesis 4 Other Findngs: Tine-Way Interaction Between Feedback, Trait Performance-
Prove GO. and State Performance-Avoid GO on 011ch Controllability

POSITIVE FEEDBACK

4.0

O
’ * l
I
|

 

 

:0: "stag marriage (if;
a°__.0_°i‘_‘.'3'299"9“90_

 

Outcorm Corlrolabilly
.0
8

.w
8
i

 

 

 

 

 

 

 

 

 

 

 

2.50 ' m_- —-_._———- _-_
2.00 -__ ,
Low High
Trait Performance-Prove GO
Hypothesis 4 Other Findngs: Three-Way Interaction BaMeen Feedback. Trait Performance-
Prava GO, and State Performance-Avoid GO on Outcome Controllability
NEGATIVE FEEDBACK
500 ~— _ __
4.50 l —— . __-_fia
. . o 4.38

 

 

 

—.—‘ __ _ ' m—S—taHTeF-a—‘i‘o‘nirrI-nc—etAVORt-GOT
._. ’.---CP."§9LCfl‘9‘_'°9_.,

 

Ortcomo Controlahily
. 9’

0|

0

 

 

 

2.50 T - ——--- —-

 

2.00 i——— _ L,,.___-,., _ a“, a
Low High
Trait Performance-Prove GO

 

Figge 5. Three-way interaction found in Hypothesis 4 exploratory analyses.

86

Hypothesis 4 Other Findings: Tine-Way Inbraction Between Feedback, Trait Performance-
Prove GO, and Stab Performance-Avoid GO on Oracome Controllability

POSITIVE FEEDBACK

5.00

 

:‘

8
l”
I

l

 

 

 

 

:0: Sbte Patormance-Avoid G‘O—
L‘_.C‘_’"_"9'_99"£‘_‘2L ‘

 

 

Outcome Controlabtky
9°
8
L.

.0
o
o

i
l

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

2.50 i
2.00 ‘ . - »~-~—
Low High
Trait Performance-Prove GO
Hypothesis 4 Other Flnclnga: Three-Way Inbractlon Between Feedback, Trait Performance-
Prove GO, and Stab Performance-Avoid GO on Outcome Controllability
NEGATIVE FEEDBACK
5.00 —
i
.5. i. __ - u-
; ' , . o 4.38
g . 00 +— _ —_
z i
g i 3.75
g 3 50 ‘ :Sbb Patormanoe-Avoid GO“:
0 . i I 0 Control Condtion ’
.5. !
5 3.00 - -
i
2.50 f -—
i
2.00 -——- __ ,
Low High

Trai Performance-Prove GO

Figme 5. Three-way interaction found in Hypothesis 4 exploratory analyses.

86

Hypothesis 4b: Interaction Between Feedback and Stab Performance-Prove 60 on
Development Controllability

 

5.00 ~—~-*

 

4.50 7

 

 

 

_ I—O—Sbb P86nmnoe-ErozejGO
;'_‘_‘_"_._021'L°'.E£"_°1'92_

 

 

s»

8
!L_
I

 

 

 

 

Dovcbpmmrcmouaunty
(a)
8‘
I
I
i
I
I
I

2.50

 

 

 

 

Negative Posrtwe
Feedback

Figge 6. Two-way interaction found for Hypothesis 4b.

 

87

Hypothesls4 Other Findings: Inbraction Between Feedback and Trait Performance-Avoid GO
on Development Controllability

5.00 .._.__- — w —— __ I-

 

 

 

 

4.50 r— v -——— A__ _-__._ _

.0-
8

 

 

 

_ . __ _ :‘fifi‘fiifpafifiaadé‘i
.'_'3_';.L_°W Traitfmmemsiflig <20_

é
I

 

Developmmt Control labilly
"E
I
I
I
I
I

 

 

Negative Positive
Feedback

F igare 7. Two-way interaction found in Hypothesis 4 exploratory analyses.

88

 

Hypothesis 4 Other Findings: Interaction Between Feedback and Sbte Pertomrance-Avold
GO on Development Cortroliabllity

5.00 ~ — . __..

 

 

4.50 I ~-

 

a.
8
I
I
a
S

 

 

 

 

 

thvelopmmt Corlrol labilly
g .
O
I

 

 

 

 

 

3 3-72 isms Putormance-Avoid 60‘.
I f - 9 - ‘ Control Condtion
I
I
3.00 ; —w— _
2.50 I - 7 a- 7
2.00 L
Negaive Positive

Feedback

Figge 8. Two-way interaction found in Hypothesis 4 exploratory analyses.

89

Hypothesis 1 Other Findings: Interaction Between Trait Mastery GO and State Perforrnance-
Prove GO on Total Time Studied

 

 

 

7_——e—“ gramaaeefia
.- - '0- ' °Control Condition

L__.__.._. -_-._.

 

 

Total Time Studied (in seconds)
E
8
I
I
I
I
|
I
I
I

 

 

500.00 ~- , - -—k A.
Low High
Trail Mastery GO

Figpre 9. Two-way interaction found in Hypothesis 1 exploratory analyses.

90

Hypothesis 1 Other Findings: Interaction Between Trait Performance-Avoid GO and State
Perfonnance-Prove GO on Total Time Studied

30000 -~——-———~

 

 

 

 

 

 

 

 

 

 

 

’5?
§
E - __
E I—O—State Performance-Prove GO I
g . I- - O- - ~Control Condition_ I
. g
.E I
P I
£0 600.00 P -
o
.—
I
55000 ' __ _ _
500.00 .

Low High
Trait Performance-Avoid GO

Figare 10. Two-way interaction found in Hypothesis 1 exploratory analyses.

91

Hypothesis 1 Other Findings: Interaction Between Trait Mastery GO and State Performance-
Prove GO on Self-Reported Study Effort

5.00 -——~——-——u ...- — .- _.___ .. __________

 

 

 

8

 

+ State Performance-Prtgao I
_ - - O- - -Control Condition

 

 

 

Self-Reported Study Effort
U
8
I
l
I

8

 

 

 

 

200-—

 

Low High
Trait Mastery GO

E_igure_11. Two-way interaction found in Hypothesis 1 exploratory analyses.

92

Hypothesis 3 Other Findings: Interaction Between Promotion SRF and State Mastery G0 on
Total Time Studied

 

 

80000 ~ ~-»~—- - —.

' 9 765.57

 

750.00 — ,-

 

70000 -

F—e—_‘*Séiéii§sié}§ ‘00

650.00 I I h
|- - Q- - -Control Condition

 

 

 

§
84 ALL—.222

 

Total Time Studied (in seconds)

 

55000 '

 

 

Low High
Promotion SRF

Figge 12. Two-way interaction found in Hypothesis 3 exploratory analyses.

93

 

Hypotheels 3 Other Findings: Three-Way lrteraction Between Feedback, Promotion SRF, and
State Performance-Avoid 60 on Self-Reported Study Effort

POSITIVE FEEDBACK

 

5.00 .—u_—W. -*_-.__ . m... A #

 

 

 

 

7 Iz—Sbb Pdormanoe-Avoid GO I
If 9 - - Control Condtion

 

 

 

 

§

 

 

Self-Reported sudy Start
3
O
-L— ._._._.__- +_
I
I
I

 

 

2.50 ‘-

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

I
i
2.00 L s -m
Lou High
Promotion SRF
Hypothesis 3 Other Finding: Thee-Way Inbraction Between Feedieck, Promotion SRF, and
State Performance-Avoid GO on Self-Reported Study Effort
NEGATIVE FEEDBACK
5.00 I
4.50 .‘ - ——————~-——r——
i
e I
g 4.00 I __- *
g I .
3 50 I_ __ - _.__. _ E—O—State Pdorrnanoe-Avoid GO I
3 ’ I 332 i"9°'Gontrol Condtion '
E I e 3.29
I _ . . .
- 3,00 + " ___
v8 .
f _, .
2.77
2.50 -
2.00 +— V - - -
Low High
Promotion SRF

F igp_re l3. Three-way interaction found in Hypothesis 3 exploratory analyses.

94

Hypothesis 3 Other Findings: Three-Way Inbraction Between Feedback. Promotion SRF, and
Stab Performance-Avoid GO on Total Time Studied

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

POSITIVE FEEDBACK
800.00 .—7fi—~_-._~___.--_c_____ __#2
750.00 » a
é _,.o 739m
E I '
§ 700.00 I __
. i
s I __ __ _‘
3 650 00 I g—State Patormance-Avoid GO
3 ' ; - - o - "Control Condtion
'3 i
E I
5 600.00 ‘ _-_ _
3. 1
550.00 I ~— —- —— i r a“
500.00 I—
Low High
Promotion SRF
Hypothesis 3 Other Findings: Three-Way Inbraction Between Feedback, Promotion SRF, and
Stab Performance-Avoid GO on Tobi Time Studied
NEGATIVE FEEDBACK
900.00 I _,
850.00 ~—
.. I 829.71
a
§ 800.00 .
e I 796.74
5 I _ _
3 I __ __ +8tab Patormance-Avoid GO
8 75000 I - - o - 'Control Condtion I
7’ I
. I
E l
3 700.00 . -
I .
I O
" I sans
850.00 I u -
I
800.00 > ._
Low High

Promotion SRF

Figare 14. Three-way interaction found in Hypothesis 3 exploratory analyses.

95

Hypothesis 9 Other Findings: Three-Way Interaction Between Feed back, Trait Mastery GO,
and State Performance-Prove GO on Time Spent Studyhg Knowledge Information

POSITIVE FEED BAC K

600.00 M —-—— ._,...___ — _.__.__~_ ~———-——_——. .——— WA,

550 00 . ~ — ~ A __-. ._ _n ___-__ .__,nn,.

 

500.00 ‘ — 7 4—»——— —- —

 

 

 

 

 

I
I . M, . ”Wines“;
I ." -

450.00 I- __ ..;_ _» . +State Patormanoe Prove GO .
I . ' ' 438.00 1 0 Control Condtion

 

 

 

400.00 .

 

350.00 3 _ -,.

 

Time amid Knowieche Normationfln second)

 

 

 

 

 

 

 

 

 

300.00 f —~ _, — ~
Low High
Trek Mastery GO
Hypothesis 9 Other Findings: Three-Way interaction Bebueen Feedback, Trdt Mashry GO.
and State Performance-Prove GO on Time Spent Studyhg Knowledge Information
NEGATIVE FEEDBACK

600.00 I — a 7

I
550.00 5 m ——
500.00 ‘ — -_ ~——————

i 452 00 .... 459.00

:‘mefi'figfifiébi
_’, _, QOnIroI Condtion .

 

 

450.00: 94 -_. ' * ._...__,___

 

 

400.00 I—— ——

 

 

350.00 ~

Tine auded Knowledge hfonnetioniin seconds)

300.00 ¥ G
Low High

Tran Mastery GO

 

Figg‘e 15. Three-way interaction found in Hypothesis 9 exploratory analyses.

96

Hypothesis 9 Other Findings: Inbraction Between Feedback and State Performance-Avoid
GO on Time Spem Studying Knowledge Information

 

600.00 - - -

I
I
I
I
I
I
I
I
I
I

a 4 4

___§_97,-99 __ __

 

8

 

3—e— State Paforrnanoe-Avoid G0 I
.. - 6 - - Control Condlion

8

 

 

 

' ‘ "0410.12

 

Tine amid Knowledge information (In seconds)
&
8

 

 

 

400.00 I
I
35000 f— 7,? __ _n _,__
I \ 318.59
300.00 *— —— —
Negative Positive

Feediedr

FigLLI'C l6. Two-way interaction found in Hypothesis 9 exploratory analyses.

97

Hypothesis 9 Other Findings: Imencdon Buween Promotion SRF md State Performance-
Prove GO on Time Spam Studying Knowledge Information

600.00 .

 

550.00 ' — -. _____--__-,,--_-_.--_____-_.-_____ —‘— “7+

 

 

500.00 ———--- —. -- - —w.-, __ _. Au,h._.-__#_

 

 

 

1 State Paformenoe—Prove GO I

450.00 , .
' I_- ' 9 ° ' Contol Condtion

 

 

§
8

 

 

 

Tine auded Knowledge Informationun seconds)

350.00 I M .

321.54

300.00 c __ _
Low High

Pl'omotlon SRF

 

Figure 17. Two-way interaction found in Hypothesis 9 exploratory analyses.

98

 

.oo 223

 

 

 

 

 

 

 

 

 

cocoa—mam xoosu .952 Box/sorted -ooaascotem CO 3582: 53» Bum—oboe 3.2568

3393 Eo><utom 52% ”am 55555 on 53 ”am :28?GC .«o 3.382 ”om mmme5o§m

.00 055

3:896 0325 9:22 0.55.5.8; oozes—cot?“ mo 8:5on 53» 5822.50 £2568

33th— o>oiutom :9? ”am 55555 on EB “Em 985088“. mo 3:532 5N m_mo5o§m

x830 .352 beam—z .00 Dean:— mo 8:68.: 5? 5828.80 b25568

cocoon—am b07432 :EH “Em 550605 on E? “Em 550805 mo gunman—z ”mm £85253

Joan—00$ gamma: 0388 >05 55> 20>»:

OO 30.5%3 832 58 535005 3588 0388 >05 5:3 tobo

votaazm team Conny—flow 88m 00 265%05 38% mo £96— 832 chm—.55 EB 00 035-85.52.53

.oZ 08? 52m 555.80 xomnwoom .3 225. him:— 55, 335355 x: £85233

£23505 “Eamon 0388 >05 5:? £96—

00 Eo><utom 88m .832 use c.8555 gamma: 0382 .35 5:? Echo

votaasm totm “Sonya—om OO Eo><utom :89 mo £32 8&2 the—ammo 53 00 Regencmscotoq

82 08¢ 53m 55550 0.8nt we £96— conwE 53> £9535:— 22 $3509»:
5.8m A503ata> Eat—.23: Anya—auteur “no—E235:— $85.5»:

 

 

 

$38M cam 88508,: 53w mo anus—am

N 55

99

 

 

 

 

 

 

00 00.00.0000 00055000 #0050000 032000 E 08000 032—00000

0000096 023500000 000000—050 003m 00 005.000 Ear—i 00000 00 00050050 00—000 EB 00 00000000005000

002 >EEm=00000 0000800 0200000 0.0000000 00 m~0>0_ 005$: 5_B 2000355 ”50 m_m0500>$

0005:0000

0605005 0>50w00 000 0220.00 0m 00300 032—00000

0000096 3250205000 “000000—050 00 ©0502 003m 0008 00 00050050 00:00 EB 00 >050:—

>=£tmm 3252—00000 0000300 00 b00002 50,—. 00 20>»: 000mm: 5_B 2002200— 03 mmm0500>3

£03005 0>50w00 030000 >05 00:B

m_0>0_ 00B2 000 0.005.000 0>Em00 030000 >05 002B

0000005 tobm 00035.3 "Em 00508000 00,000 00 m_0>0_ 0.32 3305 EB "Em 005000000

002 000% >006 0005000 #0000000 00 £00. 02E: 5MB $002305 5m mmm0500>m

10050000 020000 030000 >05 000B

000— 00B2 000 0030000 0>=aw00 030000 >05 02B

00000006 000m 00035—3 "Em 00500500 00.00 .00 m_0>0_ 000mm: >235 EB mam 00500000

002 000,—. >005 0056000 0305000..“— ‘00 m_0>0_ 0&2 .EB 2002300— am 0605252
0.0001 @0330; 00000000G 303025» «00002—000— £00500»:

 

 

 

 

 

 

08:80:80 _ 0000

100

 

.0>50w00 mm 20055000 005B 500.50
00>?>00_x00 00 m20>02 005mm: 500 635000 2 20005000

 

 

 

 

 

 

 

 

 

55 00500500 005B 5000.5 003005050500 00 m_0>0_ 005$: >20m5
50500005 >505a< 00 553300 0005 EB 2055 00500500 505 00 55500000000000
002 005000.055 00 5_0><..t00 0500‘ .50 m20>02 00:95 50B 20053502 ”05 2350030
.0>50w00 mm 20055000 005B 5000.5 00b
005300005 00 m_0>0_ 005mm: 500 .0>50m00 2 20055000
.55 005000000 005B 500.50 00300520 00 232 00:25 >20m5
5000005 005000000 00 05.00.0000 EB 0.35 005000000 5000 00 03000000005000
502 0052m— 0505 00 0>000.0000 500k 00 205. .0525 EB 020053502 8K 200.50%:
.0>50w00 mm 20055000 005B 500.50 00>5>505~00
00 296. 00525 500 “95600 2 20055000 005B 500.50
50000005 >50mx0< 00 E05002 055 00300520 00 20>02 005mm: >20m5 EB 00 E05020
502 00520 00 E05002 500‘ 00 205. 0053: 55B 20053500 an 20050030
205000 0008002050
50500005 00000 500000.505 205000 00000500 00555 00 EB 00000 .50 20>02 05 £00000 032205000
b22000 0050. >505 >55255000 000.0 05 0520 000 000505550 m< ”5 20.050030
@0592
000800000 20.0000 500600.030 00:35
5000005 0052m— 205000 00000000 05 EB 500000 95600 00 20>02 05 £00000 0202205000
002 00505205 0552205000 0.008 05 05000 08 000505550 m< 5 20050030
«.0001 3035...; .00500m0n 30320.5 50050000500 02005003—

 

 

 

330050050 >505 00 >03

32.5880 _ 03¢

101

 

00520000 500022-000 w00B00>00 0m 500m0>00

 

 

 

 

 

5000005 00 55370000 0005 00000 00 0000.00 05 50B 50020000 >_0>50000
sz 0% .30 80:0 25 00 062.000 05 B :03 00 2398505030 ”no 080200
0052000; 5522-000 w00B00>00 00 50000000
5000005 00 05000000 00000 00 0000.00 05 00B 50020000 >20>5m00
002 $05 0000 50505 000% 0005 00 05000000 500‘ 00 EB 00 05000000000000 ”50 000050030
005000005 0550305. w00B00>00
5000005 00 00022 0005 00 500050 00000 00 0000000 05 50B 50020000
002 05503055 50505 00F 00 000022 0000. >_0>5_m00 00 EB 00 000002 ”00 2850030
5000005 00000 00000005 005000 .0000 EB 50020000 >_0>50000
002 000% >505 00520 00 EB 900582005200 50002 5 2850030
5.000: 30.00005» 0.0500030 30—50005» 0.0500350— 02005003.—

 

 

 

 

 

 

920000 500 0085 E 505 00 .0005

62.5080 0 2.3

102

A0005 w0_00_En00000w 0050 :0 500 .ou0000w .00000 ._n0000wv 009% .00000 00. 00 0005

05 500 000% 05 00 000:. 05 000500 0000-50-25 00000000.. >05 00000. 0000—0000 0 0_ 0B000 000B 05 000.000 5 5050.05 «00 000B 00300?
0000.0"; .Su00000w 00000 :0 ._u0000wv 50050.00 000.00% 0050 05 00 :0 00 000E 0.0 00 5200800 000% .05 .00 000.0 05 000500 000200-05
05 00000000.. 005 00000. 00.00.0000 0 E 0B000 000B 00>0B00 000—000 05 E 5000 00500 000005 .0300 05 000B 0000-030 :0.V0..:.. .00.V0...

 

 

E 80 m: 80 m: 80 80 000 z
3.- 8.- 00.- 3.- 8.- S. 8. 3. 00. 30 0020023200
E 80 m: 80 N: 80 80 000 z
:8...- ......0m.- :80- :00; :80- ...am: .2. 8. 00.. 00. .0200000.050.9020 .0
E E N: z
- *2: - - -- -- :.- :0.- 00 R. 22.80.000.52 :20 .0
80 80 80 03 z
.. - :8.- -- :8.- 8.- 8.- 3. a. 0.000.052 00.3. .0-
80 m: 0: z
-- :80 -- - .00- 00.- 3. 0m. 002.50.00.05: 02:0 .0
80 80 03 z
-- -- :2.- 0o.- 8.. 3. mm. 0.000.005: 02:0 .0
N: N: 0: z
-- 80 2.- 3.- 0.. R. 28.50.00.052 02:: .0
80 000 z
-- 8.- S.- S. a. 02.3.0.5: 02:2 .0
000 z
- :00. 00. 0.0. 500.80 00308... .N
-- :0: s: 0:00.500 000.? 200.05 .—
000 00
0 b 0 0 .- 0 N _ 00 :32 030...;

 

 

 

00030-0040 05 0_ 5050—000 003000> 000 000230000 500 00002000000000 000003000 5005000m 00002

N 030%

103

5V0: .mOVQ...

 

 

:— oom «.5 com N: cam can 5mm Z
3.- we: 3. 0o. wor oo.- 0o. :8. ow. m5.m 3:52—03:00 00.00.00 .0—
K— com 30 Gem N: com com 5mm Z
3.- 00.- Sr 3. S. we. vo. S. mo. 3.»- 5=50=0h000 .m—
EZ oom m: com m: com com 5mm Z
mo. 5. No. Nov 2. we. no: N_.- 50— mod 01m 00300300 .5
0: com 30 com NS com gm 0mm Z
no. mo. mo. 5. co. mo: 00. 3. in N06 Em 00:00.0...— .5—
0: con m: com N: com com 53 Z
mo. 00: 2. no. m 0. 00. oo. o_. :._ 00m 10000 .0000: 5m0>< .3
.5 com m: com a: con com 0mm Z
2. 8. 0.5—. o.- wo. oo. no. 3. 3: NZ“ 00000 0.002 0000.5 .3
:0 com m2 com N: ooM com 53 Z
8. S. 5. Sr mo. 8. 0.5—.- ...m_.- 2: cod 0.0009 .0000: 0.0.002 .2
0: 00m m: com m: 03 com 03 Z
mo: cor No. co. m_. ... _ _. mo. co.- 3. Cum 90.5 20.04 .2
:— oom 3.— COM N: com com 03 Z
00.- 8.- oo.- cor vo. mo. 0mm. 1.0m. mm. mm.m 30.5 0000.5 .3
a b e m v m N 0 Gm 0002 0305.3,

 

 

 

 

 

0003000. 05 E 5050—000 003%; 000 000250200 500 00000—000885 00000500Q 5005000m .00002

825053 N 00.0.

104

.mwcooom ... Em moEmfig 3:58 0...: o... .8 8.283% $86.88 was 8.82 ”802
.26. x2338 .3 Ex... 225 3.3.3:. .oob<nu ”EVE... .mo.vm_..

 

 

E 88 E 88 E 888 88 88 z
8.. 8.. 8. 3. :.- 8..- 8.- 3.- 8.88 8.88 .2... .8: 3:58 o...:- ...N
E 88 8. com .8. 88 8... 88 z
3.- 8.- 8.- 8.- 8. 8. .m..- 8. 3.23 8.8:. ...-5.8.55.8 38:8 2::- .8
E 88 E 88 E 88 8..- 88 z
3. 8. 8.- 8. 8.- 8.- ...-8.- 8.- .38 ...-:88 :53. 38:8 2...... .3
E 888 E com E 88 com 88 z
.... 8. ... 8. 8. 8.- ......E- .8. m . .. 88 fie-=2... .85. 8:8 .8
8 8. 8. 8. 8 ...... .... 7.

8m. ...... .... 8. .... 8. m..- m... 88 89.5.... .3
8 8. 8. 8. 8 8. .... z

.8. m .. .m. 8. 8.. 8. .-- 3. m . .. 8.8 5.32%: .8
8 8. 8 8. 8. 8. 8. z
8. 8. 8.- 8.- 8.- 8.- 8. 8. 38 8.8... .8
8 8. 8 8. 8. 8. 8. z
8. 8. . ..- . ..- 8. 8. .-- 8. 8. 88m 8.3888 ..N
E 88 8. 88 E 8..- 8m 88 z
8.- 8..- 8.- 8.- ... .. .. 8.- .8. 8. 88 8.82.288 825.225 .8

8 b 8 m w m N . am :38 ”88......»

 

 

 

 

88:83 05 ... 39:05 moBmtm> .8 mouzfiflwzom v.8 .w...o_§otoo.ofi_ 88.8.59 ©8985 .9802

82.588. m 0......-

105

.22. .... 8.. 85:58.3. 882

 

 

8!... .82..
88. 88. 88. 88m 88. 88. 88. 88. 88m
8.. 3. 8. 8.. ...... ...m. 8.- ...m. ...N. 3. 8.82.285...
88. 88m 88. 88. 88. 88m 88m 88m
8... ......- ..8. .2. 8.. 8. 8. 8.- 8.- .......-..~.m 5858......
88. 88m 88. 88. 88. 88m 88m
8. 8. .8. ...N. 8.- 8. ..E. 8.- 82:58 8832......
88. 88. 88m 88m 88m 88.
8. .... ..8. ..8. ....N. 8.- 1... .85 .88.: .8... .2
88m 88. 88. 88m 88m
.... ..8. 8. ...... 8. 2.. ..85....§2 $2....
88. 88. 888 888
8. .2; 8. ...... 8.- 826 .88.... @288 .2
88m 88m 88.
... ..8. ......- 3.- 8.5. .83. .N.
88m 88m
8. ... .. 8. £2: .8... ...
88
3. 8. 8...... has... ....
.. 2 2 2 ... 2 N. .. ... 8 2......»

 

 

$2.89... 0... ... cons—o... 838.8.» 8.. 3.55.2.3. ...... .mco..m_otoo.o.:.discugofl magnum .882).

888.88. m 2......

106

.8.v.._..... .882...

 

 

888 888 888 888 888 888 888 888 888 888 z
88. 88. 88. 88. 8.. 88.- 8.. .8. 88.- .8.- .2... .8.. 88.838 2:; .88
888 888 888 888 888 888 888 888 888 888 z
.8. .8. 88. 88. 88. :88. 88.- 88.- 88. 88. $888.55.. 88.838 85.8 .88
888 888 888 888 888 888 888 888 888 888 z
..8. 88. 8.. 8.. ...... :88. .8. 88.- 88. .8. .38.. 88.838 85.8 .88
888 888 888 888 888 888 888 888 888 888 z

1.8.. ..8. ...... 2.8.. :88. :8... 88.- .8. 88. 88.- €882.28. 28...... 8.5.8 .88
88. , 88. 88. 88. 88. 88. 88. 88. 88. 88. z
.... .8.. 8.8.. :88. .8.. ...88. 88. 88.- 88.- ...:- 8.8.8.2 ...8
88. 88. 88. 88. 88. 88. 88. 88. 88. 88. z
8.. ...-.88. :88. :8... ...... .... :88. .8.. ... 88. .88.- 88.88.82. .88
88. 88. 88. 88. 88. 88. 88. 88. 88. 88. z
88. 88.- 8.. 88. .8.. ...8 .. ......88.- 88.- 8.. 88. 88.3... .88
88. 88. 88. 88. 88. 88. 88. 88. 88. 88. z
88. 8..- 88. 8..- .8.- 88. 88.- 88.- 88. 88. 8888888888 ..8
888 888 888 888 888 888 888 888 888 888 z

:88. 88. 88. 88. 8 .. ...... . 8. .8. 8 .. 1.8.. .8. 85.88.3580 32:88.85: .88
888 888 888 888 888 888 888 888 888 888 z

:88. 88. ..8. 8 .. ...... . 8. ......8.. 88.- :88. 1.8.. 88. 88:88:85.... 2.335 .8.

8. 8. 8. 8. ... 8. 8. .. 8. 8 838...;

 

 

 

 

 

8882898. 2.. ... 3.5.0:. 8058...... > 80.. 80.....fl..o~. 8%.8co..m.oboo.8£ duos-8.5m. ......ESm 8:802

€38.28. 8 288.-

107

....oa ... 28 88....38..o~. .207.

 

 

88.38.88
on. 85.8 8:26.288 8882.. 8:25:80 403.88.. 2:88 8... 2.5.3 .6588... .0: 9.83 88.8.8.8? omen-Hug m.o.v8...:.. Kev...
888 888 888 88. 88. 88. 88. 888 888 z
:8...- _...._88. :88. 88. ... .8.- 88.- 88. 88. .8.... .88.. 88.838 28.8 .88
888 888 88. 88. 88. 88. 888 888 z
-- :88. :88. 88. 8.. 8.. .8.- 88. 88.- 6888.385: 88.858 8...; ...8
888 88. 88. 88. 88. 888 888 z
-- :88. 8.. :88. 8.. 88.- 88. 88.- 88.8.. 88.858 88.-8 .88
88. 88. 88. 88. 888 888 z
E. 88. :88. ..8. 88.- :.8. 88. €882.88. :88... 8.5.8 .88
88. 88. 88. z
88. :88. 8.. 8.. 8.8.8.2 ...8
88. 88. z
88. 8.. 88. 88.88.8.8... .88
88. 88. 88. z
.8. :8... 88. 88. 88...“... .88
88. 88. z
88. .8. 88. 88.38.8888 ..8
888 z
88. :88. 88.32.8880 38588.2... .88
8... 88.32.8885 88.8.5 .8.
88 88 88 ..8 88 88 .8 88 8. 8.88....»

 

 

88838.58. 0... ... 3.822.. 88.8.8...8> ..o. 88.558823. .8.... 8.5.3.8588... 85:88.38. 6.8.8.85 .8..on

.8.-5:888. 8 8.88.-

108

Table 3

Hypotheses la-lb: Hierarchical Regression Results of the Impact of Feedback Condition, .
Trait GO. State GO. and Their Two-Way Interactions on Time Studied When Controlling

for Cognitive Abilig

 

 

Hyp Variable N p R2 AR2
STEP 1: 256 .07*
Cognitive Ability (ACT/SAT Scores) .16*
Feedback (1=positive, 0=negative) -.28**
Trait Avoid .16*
State Avoid (dummy coded vs. control) .04
Trait Prove -.03
State Prove (dummy coded vs. control) -.06
Trait Mastery .10
State Mastery (dummy coded vs. control) -.10
STEP 2: 256 .13 .06
1a Feedback x Trait Avoid -.02
1a Feedback x State Avoid -.02
1b Feedback x Trait Prove -.14
lb Feedback x State Prove -.05
Other Feedback x Trait Mastery .26
Other Feedback x State Mastery .03
Other Trait Avoid x State Avoid -.16
Other Trait Avoid x State Prove -.95*
Other Trait Avoid x State Mastery .50
Other Trait Prove x State Avoid -.31
Other Trait Prove x State Prove -.15
Other Trait Prove x State Mastery -.81
Other Trait Mastery x State Avoid .25
Other Trait Mastery x State Prove -1.23*
Other Trait Mastery x State Mastery .20

 

Note: [3 is the standardized regression coefficient associated with each step of the
hierarchical regression. Increments for variables entered at the AR2 significance levels
are based on the F tests for that step. *p<.05, **p<.01

109

Table 4
Hypotheses la-lb: Hierarchical Regression Results of the Impact of Feedback Condition.

Trait GO. State GO. and Their Two-Way Interactions on Self-Report Effort When

Controlling for Cognitive Abilitv

 

 

Hui Variable N [3 R2 AR2
STEP 1: 256 .07*
Cognitive Ability (ACT/SAT Scores) .14
Feedback (l=positive, 0=negative) -.27**
Trait Avoid .02
State Avoid (dummy coded vs. control) .10
Trait Prove .06
State Prove (dummy coded vs. control) .05
Trait Mastery .06
State Mastery (dummy coded vs. control) -.02
STEP 2: 256 .14 .07
la Feedback x Trait Avoid -.15
1a Feedback x State Avoid -.19
1b Feedback x Trait Prove -.25
1b Feedback x State Prove -.04
Other Feedback x Trait Mastery -.54
Other Feedback x State Mastery -.02
Other Trait Avoid x State Avoid -.39
Other Trait Avoid x State Prove -.67
Other Trait Avoid x State Mastery .14
Other Trait Prove x State Avoid .74
Other Trait Prove x State Prove .70
Other Trait Prove x State Mastery -.27
Other Trait Mastery x State Avoid .40
Other Trait Mastery x State Prove -1.15*
Other Trait Mastery x State Mastery .17

 

Note: [3 is the standardized regression coefficient associated with each step of the

hierarchical regression. Increments for variables entered at the AR2 significance levels

are based on the F tests for that step. *p<.05, **p<.01

110

Table 5
Hypotheses 3a-3b: Hierarchical Regression Results of the Impact of Feedback Condition-
Self-Regplatory Focus. State GO. and Their Interactions on Time Studied When

Controlling for Cognitive Ability

 

 

Hyp Variable N p R2 AR2
STEP 1: 256 .O7**
Cognitive Ability (ACT/SAT Scores) ' .14
Feedback (l=positive, 0=negative) -.27**
Prevention SRF .06
Promotion SRF .12
State Mastery (dummy coded vs. control) -.09
State Prove (dummy coded vs. control) -.05
State Avoid (dummy coded vs. control) .02
STEP 2: 256 .14** .08*
3a Feedback x Prevention SRF .02
3b Feedback x Promotion SRF -.65
Other Feedback x State Mastery .05
Other Feedback x State Prove -.11
Other Feedback x State Avoid -.08
Other Prevention SRF x State Mastery .37
Other Prevention SRF x State Prove -.03
Other Prevention SRF x State Avoid .18
Other Promotion SRF x State Mastery -1.16**
Other Promotion SRF x State Prove -.80
Other Promotion SRF x State Avoid -.23
STEP 3: 256 .17** .03
Other Feedback x Promotion SRF x State Mastery -.67
Other Feedback x Promotion SRF x State Prove -1.11
Other Feedback x Promotion SRF x State Avoid -1.26*

 

Note: [3 is the standardized regression coefficient associated with each step of the
hierarchical regression. Increments for variables entered at the AR2 significance levels
are based on the F tests for that step. *p<.05, **p<.01

111

Table 6
Hypotheses 3a-3b: Hierarchical Regression Results of the Impact of Feedback Condition.
Self-Regplatorv Focus, State GO. and Their Interactions on Self-Report Effort When

Controllipg for Cognitive Ability

 

 

Hyp Variable N p R2 AR2
STEP 1: 256 .10“
Cognitive Ability (ACT/SAT Scores) .15*
Feedback (l=positive, 0=negative) -.28**
Prevention SRF .08
Promotion SRF .20**
State Mastery (dummy coded vs. control) -.03
State Prove (dummy coded vs. control) .04
State Avoid (dummy coded vs. control) .08
STEP 2: 256 .15** .05
3a Feedback x Prevention SRF .09
3b Feedback x Promotion SRF -.67
Other Feedback x State Mastery -.01
Other Feedback x State Prove -.05
Other Feedback x State Avoid -.124
Other Promotion SRF x State Mastery -.55
Other Promotion SRF x State Prove .62
Other Promotion SRF x State Avoid .09
Other Prevention SRF x State Mastery .23
Other Prevention SRF x State Prove .12
Other Prevention SRF x State Avoid -.11
STEP 3: 256 .19** .04
Other Feedback x Promotion SRF x State -.17
Mastery
Other Feedback x Promotion SRF x State Prove -.24
Other Feedback x Promotion SRF x State Avoid -1.21*

 

Note: B is the standardized regression coefficient associated with each step of the
hierarchical regression. Increments for variables entered at the AR2 significance levels
are based on the F tests for that step. *p<.05, **p<.01

112

Table 7
Hypotheses 4a-4b: Hierarchical Regression Results of the Impact of Feedback Condition-
mu GO. State GO. and Their Two-Wav Interactions on Outcome Controllability When

Controllingfor Cognitive Ability

 

 

Hyp Variable N B R1 AR2
STEP 1: 256 .O7**
Cognitive Ability (ACT/SAT Scores) .16*
Feedback (l=positive, 0=negative) -.03
4a Trait Mastery .05
4a State Mastery (dummy coded vs. control) -.05
Trait Prove .21**
State Prove (dummy coded vs. control) .03
Trait Avoid -.13
State Avoid (dummy coded vs. control) -.01
STEP 2: 256 .13" .06
4b Feedback x Trait Prove .02
4b Feedback x State Prove .04
Other Feedback x Trait Mastery .19
Other Feedback x State Mastery .06
Other Feedback x Trait Avoid .57
Other Feedback x State Avoid -.05
Other Trait Prove x State Prove -.28
Other Trait Prove x State Mastery .02
Other Trait Prove x State Avoid .31
Other Trait Mastery x State Prove -.28
Other Trait Mastery x State Mastery .84
Other Trait Mastery x State Avoid -.46
Other Trait Avoid x State Prove .68
Other Trait Avoid x State Mastery .56
Other Trait Avoid x State Avoid -.01
STEP 3: 256 .15** .02
Other Feedback x Trait Prove x State Prove .75
Other Feedback x Trait Prove x State Mastery -.15
Other Feedback x Trait Prove x State Avoid 1.74*

 

Note: B is the standardized regression coefficient associated with each step of the
hierarchical regression. Increments for variables entered at the AR2 significance levels
are based on the F tests for that step. *p<.05, **p<.01

113

Table 8

Hypotheses 4a-4b: Hierarchical Rgzression Results of the Impact of Feedbegc Condition.

T_rait GO. State GO. and Their Two-Way Interactions on Development Controllability

When Controlling for Cognitive Ability

 

 

Hyp Variable N [3 R2 AR2
STEP 1: 256 .05
Cognitive Ability (ACT/SAT Scores) .00
Feedback (l=positive, 0=negative) -.04
4a Trait Mastery . 16*
4a State Mastery (dummy coded vs. control) .09
Trait Prove .07
State Prove (dummy coded vs. control) -.02
Trait Avoid .03
State Avoid (dummy coded vs. control) -.05
STEP 2: 256 .17** .12"
4b Feedback x Trait Prove -.46
4b Feedback x State Prove -.22*
Other Feedback x Trait Mastery -.35
Other Feedback x State Mastery -.19
Other Feedback x Trait Avoid .91 **
Other Feedback x State Avoid -.33**
Other Trait Prove x State Prove -.86
Other Trait Prove x State Mastery .66
Other Trait Prove x State Avoid .37
Other Trait Mastery x State Prove -.27
Other Trait Mastery x State Mastery -.56
Other Trait Mastery x State Avoid -.89
Other Trait Avoid x State Prove .63
Other Trait Avoid x State Mastery -.02
Other Trait Avoid x State Avoid -.41

 

Note: B is the standardized regression coefficient associated with each step of the

hierarchical regression. Increments for variables entered at the AR2 significance levels

are based on the F tests for that step. *p<.05, **p<.01

114

Table 9
Hypotheses 9a: Hierarchical Regression Results of the Impact of Feedback Condition.
flait GO. State GO. and Their Two-Way Interactions on Time Spent Studying

Knowledge Information When Controlling for Cognitive Ability

 

 

H);- Variable N [3 R2 ARZ
STEP 1: 256 .04
Cognitive Ability (ACT/SAT Scores) .15*
Feedback (l=positive, 0=negative) -.22**
9a Trait Mastery .09
9a State Mastery (dummy coded vs. control) .01
Trait Prove -.10
State Prove (dummy coded vs. control) -.09
Trait Avoid .07
State Avoid (dummy coded vs. control) -.02
STEP 2: 256 .12 .07
Other Feedback x Trait Mastery .80
Other Feedback x State Mastery .00
Other Feedback x Trait Prove -.48
Other Feedback x State Prove .02
Other Feedback x Trait Avoid -.10
Other Feedback x State Avoid -.18
Other Trait Mastery x State Mastery -.50
Other Trait Mastery x State Prove -1.23*
Other Trait Mastery x State Avoid .25
Other Trait Prove x State Mastery -.34
Other Trait Prove x State Prove -.18
Other Trait Prove x State Avoid -.38
Other Trait Avoid x State Mastery .38
Other Trait Avoid x State Prove -.34
Other Trait Avoid x State Avoid .01
STEP 3: .14 .02
Other Feedback x Trait Mastery x State Mastery -.27
Other Feedback x Trait Mastery x State Prove -1.83*
Other Feedback x Trait Mastery x State Avoid -.21

 

Note: B is the standardized regression coefficient associated with each step of the
hierarchical regression. Increments for variables entered at the AR2 significance levels
are based on the F tests for that step. *p<.05, **p<.01

115

Table 10

Hypotheses 9a: Hierarchical Regression Results of the Impact of Feedback Condition,

Self-Regplatory Focus. State GO. and Their Two-Way Intergtions on Time Spent

Studying Knowledge Information When Controlling for Cogiitive Ability

 

 

Hyp Variable N B R2 AR2
STEP 1: 256 .04

Cognitive Ability (ACT/ SAT Scores) .15
Feedback (l=positive, 0=negative) -.22**

9a Promotion SRF .07
Prevention SRF .04
State Mastery (dummy coded vs. control) .01
State Prove (dummy coded vs. control) -.09
State Avoid (dummy coded vs. control) -.03

STEP 2: 256 .10 .06

Other Feedback x Promotion SRF -.04

Other Feedback x Prevention SRF .00

Other Feedback x State Mastery .01

Other Feedback x State Prove .01

Other Feedback x State Avoid -.23*

Other Promotion SRF x State Mastery -.59

Other Promotion SRF x State Prove -.97*

Other Promotion SRF x State Avoid .29

Other Prevention SRF x State Mastery .31

Other Prevention SRF x State Prove .12

Other Prevention SRF x State Avoid .24

 

Note: B is the standardized regression coefficient associated with each step of the

hierarchical regression. Increments for variables entered at the AR2 significance levels

are based on the F tests for that step. *p<.05, **p<.01

116

APPENDIX A
Informed Consent

The study in which you are about to participate investigates your performance on
a simulated Graduate Record Examination (GRE) task. This task is a shortened version
of the paper and pencil GRE. This experiment consists of three major stages. First, you
will be asked to complete a brief (15 minute) test containing questions like those found
on the GRE. You will be given feedback about how you performed on the test. Next,
you will have an opportunity to study material intended to help you improve your
performance. Finally, you will take a second test.

In addition to completing the above tasks, you will be asked some general
questions about how you approach tasks and why you think you performed the way you
did. You will also be asked to report some basic demographic information including
your ACT or SAT scores. Your ACT or SAT scores may be verified against scores on
file in the registrars office. All of your responses will be kept confidential and your
privacy will be protected to maximum extent allowable by law, so we ask that you
answer all of the questions carefully and honestly.

The entire experiment will take approximately one and a half hours. The top two
performers on the second test will get $25 at the end of this semester. Incentive winners
will be notified by e-mail. Please keep in mind that this simulation is NOT a real GRE.
Also, please remember that your scores on these two tests are true for these tests only,
and not indicative of how you would perform on the real GRE.

Participation in this study is voluntary. By clicking on “CONTINUE” below, you
are indicating that you agree to participate in this study. You are also indicating that you
were flee to refuse to participate in this project or any part of the project. You may
refuse to answer some of the questions and/or may discontinue your participation at any
time without penalty by typing the keys “CTRL + F12.”

If you have any concerns or questions about your participation in this project, or
would like a copy of the results, you can contact Cori Davis at 517-355-2171
(daviscor@msu.edu) or Dan Ilgen at 517-355-7503 (ilgen@msu.edu).

If you have any questions or concerns regarding your rights and involvement in
this research you may contact:

The University Committee on Research Involving Human Subjects (U CRII-IS)
246 Administration Building, Michigan State University

East Lansing, M148824-1046 1

Phone: (517)355-2180 Fax: (517)353-2976

Email: UCRIHS@msu.edu Web Site: http/www.msu.edu/unit/vprgs/ucrihs

117

APPENDIX B
Introduction to the Study
Thank you for agreeing to participate in the GRE Experiment. During this experiment,
you will be presented with various screens. Some have questions for you to answer, and

others just contain information for you to read. You will need a pen/pencil and scratch
paper for this study. Please keep the following instructions in mind:

1) You must answer every question before the computer will allow you to continue to the
next screen. Also, if you make a mistake answering a question, finish answering the rest
of the questions on that screen, and THEN go back to correct your answer. Please answer
each question honestly.

2) Once you have hit “CONTINUE,” you will not be able to go back to the previous
screen.

3) You may use scratch paper during this study, however, please do not use a calculator.

4) Please do not talk, use other computer programs, do homework, or leave the room
during this study unless instructed otherwise.

Now we would like to ask you some questions, please hit “CONTINUE.”

118

APPENDIX C

Measures

Demogr_aphics

Please answer the following questions. We need information about who you are, so that
we can keep track of you data and contact you if you win an award. Also, if you are a
college freshman, please enter you high school GPA when prompted. Thanks.

1. Please enter your pilot login ID.
2. What is your gender?

1 = Male

2 = Female

3. What is your age?

l=lessthan18
2:18-19
3=20-21
4:22-23

5 = greater than 23

4. What is your year in college?

1 = F irst-Year
2 = Sophomore
3 = Junior

' 4 = Senior
5 = Other

5. What is your race?
1 = African-American

2 = Asian
3 = Hispanic/Latino (N on-white)
4 = White
5 = Other

6. What is your major, or area of interest?

119

Demogzaphics (continued )

7. What was your grade point average (GPA) as of the end of your last full semester in
college (either at MSU or elsewhere)?
1 = less than 2.50

2 = 2.50 to 2.79
3 = 2.80 to 3.19
4 = 3.20 to 3.59

5 = 3.60 or greater

8. What was your ACT or SAT total test score? If you took both exams, please enter
your score on the ACT exam.

9. Have you taken or prepared for the Graduate Record Examination (GRE)?

1=Yes
2=No

10. Have your participated in this study before?

1=Yes
2=No

120

Goal-Orientation Measure

Next, we need you to answer the following questions about yourself using the 5-point
scale below.

9‘5”???)1'"

N

9.

10.
11.
12.

13.

 

l 1- % l l

l 2 3 4 5
Strongly Disagree Neutral Agree Strongly
Disagree Agree

I prefer to work on projects where I can prove my abilities to others. (prove)

I often look for opportunities to develop new skills and knowledge. (mastery)
I’m concerned with showing that I can perform better than others. (prove)

I prefer to avoid situations where I might perform poorly. (avoid)

For me, development of my ability is important enough to take risks. (mastery)
Avoiding a show of low ability is more important to me than learning a new skill.
(avoid)

I am willing to accept a challenging assignment that I can learn a lot from.
(mastery)

I prefer to work in situations that require a high level of ability and talent.
(prove)*

I would avoid taking on a new task if there was a chance that I would appear
incompetent to others. (avoid)

I try to figure out what it takes to prove my ability to others. (prove)

I enjoy challenging and difficult tasks where I’ll learn new skills. (mastery)

I’m concerned about taking on a task if my performance would reveal that I had
low ability. (avoid)

I enjoy it when others at work are aware of how well I am doing. (prove)

*This item loaded with the mastery items in an exploratory factor analysis. However, for
the purposes of this study it was kept as part of the prove scale in order to maintain
consistency with other studies using this scale.

Self-Regplatory Focus Measure

We are interested in why you decided to take the psychology course(s) you are enrolled
in. Please use the following scale to indicate your reasons for choosing this course.

1 l l l l

l I I l I

I 2 3 4 5
Strongly Disagree Neutral Agree Strongly
Disagree Agree

It is a requirement for me. (prevention)

I thought it would be fun. (promotion)

I thought it would be interesting. (promotion)

I was told I should take it. (prevention)

I need it because it is a prerequisite for other classes. (prevention)
I thought it would be challenging. (promotion)*
I wanted to take this course. (promotion)

I thought it would be easy. (prevention)*

I had to take this course. (prevention)

10. I thought I would enjoy this course. (promotion)
1 1.1 like psychology. (promotion)

12. I had to take a psychology course. (prevention)

PWNQMPPNT‘

*Dropped fiom the final analyses due to poor factor loading and/or internal consistency.

122

Selves Qpestionnaire

In the following three screens, you will be asked to list the attributes of the type of person
you think you ACTUALLY, IDEALLY, and OUGH T to be. Please read and answer the
following questions carefully.

Your ACTUAL attributes refer to the characteristics you believe you ACTUALLY
possess. Please list 8 words that describe your ACTUAL attributes.

1)
2)

 

3)
4)
5)
6)
7)

8)

123

Selves Questionnaire (continued)
Your OUGHT attributes refer to the characteristics you believe you SHOULD or
OUGHT to possess. These represent your normative rules or prescriptions for yourself.

Please list 8 words that describe your OUGHT attributes. (It is okay to use words from
the last screen if they apply here.)

1)
2)
3)
4)
5)
6)
7)

3)

124

Selves Ouestiogaire (continued)
Your IDEAL attributes refer to the characteristics you would IDEALLY like to possess.

They represent the ultimate goals for yourself. Please list 8 words that describe your
IDEAL attributes. (It is okay to use words fi'om the last 2 screens if they apply here.)

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

7)

 

8)

125

Affect Measure
The next two screens deal with how you feel about your performance on the first test.

If your score was ABOVE the 65th percentile. . .please put 1’s in the blanks for the first

screen (questions 1-12), and then answer questions on the second screen (questions 13-
24) honestly.

If your score was BELOW the 65th percentile. . .please answer questions on the first
screen (questions 1-12) honestly, and put 1’s in the blanks for the second screen
(questions 13-24).

A number of statements people have used to describe themselves are given below. Read
each statement and fill in the appropriate box to the left of the statement to indicate how
you feel about how you performed on the test. There are no right or wrong answers. Do
not spend too much time on any one statement but give the answer that seems to describe
your feelings the best.

1 l l I I

l r I l I

1 2 3 4 5

Not to a small somewhat moderately very

at all extent much
Negative feedback: Positive feedback:
1. I feel agitated (arousal)* 13. I feel eager (arousal)
2. I feel disappointed (non-arousal) 14. I feel cheerful (arousal)
3. I feel frustrated (non-arousal) 15. I feel self-confident (non-arousal)*
4. I feel anxious (arousal) 16. I feel satisfied (non-arousal)*
5. I feel tense (arousal) 17. I feel happy (arousal)
6. I feel on edge (arousal) 18. I feel excited (arousal)
7. I feel disheartened (non-arousal) 19. I feel secure (non-arousal)
8. I feel discontent (non-arousal) 20. I feel calm (non-arousal)
9. I feel uneasy (arousal)* 21. I feel enthusiastic (arousal)
10. I feel discouraged (non-arousal) 22. I feel delighted (arousal)
11. I feel stressed (arousal) 23. I feel relaxed (non-arousal)
12. I feel dissatisfied (non-arousal) 24. I feel content (non-arousal)

I"Dropped from the final analyses due to poor factor loading and/or internal consistency.

126

Causal Dimension Scale

The following questions are asking about the cause of your performance on this test.
Please answer these questions on a scale of one to five:

1) Is the cause(s) something that reflects an aspect of: (locus — reverse coded)

YOURSELF 1 2 3 4 5 THE SITUATION

2) Is the cause(s) something that is: (stability — reverse coded)

PERMANENT 1 2 3 4 5 TEMPORARY

3) Is the cause(s) something that is: (locus)

OUTSIDE 1 2 3 4 5 INSIDE
of you of you
4) Is the cause(s) something that is: (stability)

VARIABLE l 2 3 4 5 STABLE
over time over time
5) Is the cause(s) something about: (locus — reverse coded)

YOU 1 2 3 4 5 OTHER PEOPLE

6) Is the cause(s) something that is: (stability)

CHANGEABLE l 2 3 4 5 UNCHANGING

7) Is the cause(s): (controllability — reverse coded)

CONTROLLABLE l 2 3 4 5 UNCONTROLLABLE
by you or other people by you or other people

8) Is the cause(s) something: (controllability — reverse coded)

INTENDED l 2 3 4 5 UNINTENDED

by you or other people by you or other people
9) Is the cause(s) something for which: (controllability)

NO ONE 1 2 3 4 5 SOMEONE
is responsible is responsible

127

Controllability Measure

Please answer the following questions using the 5-point scale below

1 l l l l

I I I I T

1 2 3 4 5
Strongly Disagree Neutral Agree Strongly
Disagree Agree

I can influence my score on the next test. (outcome factor)

I can do better on the next test. (outcome factor)

I typically don’t do well on these kinds of tests. (reverse coded)*

My score will be the same on the next test no matter what I do. (reverse coded)*
Studying can improve my score. (development factor)

I can improve my score with practice. (development factor)

95"?pr

*Dropped from the final analyses due to poor factor loading and/or internal consistency.

128

Goal Orieriation Manipulation Check

How important to you are the following things?

99:59PM"

1 l l l

I I I I

1 2 3 4
Not to a small somewhat moderately
at all extent

Getting a high score on Test 2 (prove)

Learning math and verbal information (mastery)
Doing as well as other students (avoid)
Improving verbal and math skills (mastery)

Not looking bad compared to others (avoid)
Scoring better than others on Test 2 (prove)

129

Man—

very
much

Motivation Questions

Please answer the following questions using the 5-point scale below.

J l l I I

 

I I T I I

I 2 3 4 5
Strongly Disagree Neutral Agree Strongly
Disagree Agree

I was motivated to perform well on the second test by the chance to win money.
I wanted to perform well on the second test.

I tried to be one of the top performers on the second test.

I put forth my best effort when taking Test 1.

I put forth my best effort when taking Test 2.

If I would have been given more time to study, I would have studied longer.*

I put forth my best effort while studying.*

NP‘S‘PP‘NT‘

*These two items made up the self-report study effort scale.

130

Difference Between Screens and Score Believabilig Questions

Please answer the following questions using the 5-point scale below.

PP°>'.°‘.V‘

l l l l

l
l I I l I

I 2 3 4 5
Strongly Disagree Neutral Agree Strongly
Disagree Agree

. I could tell the difference between the types of information on the VERBAL

KNOWLEDGE study screen and the VERBAL TEST TIPS study screen.
I could tell the difference between the types of information on the MATH
KNOWLEDGE study screen and the MATH TEST TIPS study screen.

. The verbal and math KNOWLEDGE screens were best for learning math and

verbal information.

The verbal and math TEST TIPS screens were best for learning math and verbal
information.

The verbal and math KNOWLEDGE screens were best for improving your score.
The verbal and math TEST TIPS screens were best for improving your score.

I think the score I received for TEST 1 accurately described my performance.

I believe the score I received for TEST 1 was correct.

My percentile score for TEST 1 (i.e., my score compared to others) was
believable.

10. I think the score I received for TEST 2 accurately described my performance.
11. I believe the score I received for TEST 2 was correct.

131

Performance Test 1

POW?

Column A Column B
333 (1/2)3
666

if the quantity in Column A is greater;

if the quantity in Column B is greater;

if the two quantities are equal;

if the relationship cannot be determined from the information given.

- . :-
(...
.\ '4 I

\y

 

The circle has center 0 and radius 1.

Column A Column B

The area of the shaded region n/2

POP?

if the quantity in Column A is greater;

if the quantity in Column B is greater;

if the two quantities are the same;

if the relationship cannot be determined from the information given.

3. DOORzROOM ::

WPOW?

rudderzanchor
boatzship
patio:terrace
hatchzhold
basementzattic

4. CHOREOGRAPHYzDANCE ::

590??

ceremonyzsermon
agendazadvertisement
poetry:recitation
instrumentationzconductor
plot:story

132

20

 

9 16

What is the value of x in the figure above?

12
12.5
15

9J3
18

WPOP?

. If the circumference of a circle is less than 107:, which of the following could be the
area of the circle?

201:
251i
361:
817:
100a

WPOW?

. It is puzzling to observe that Jones’s novel has recently been criticized for its -------
structure, since commentators have traditionally argued that its most obvious ------- is
its relentlessly rigid, indeed schematic, framework.

attention to..preoccupation
speculation about..characteristic
parody of..disparity

Violation of..contradiction

lack of..flaw

FPOP?

. The painting was larger than it appeared to be, for, hanging in a darkened recess of
the chapel, it was ------- by the perspective.

improved

aggrandized
embellished
je0pardized
diminished

FPOW?

133

9.

10.

11.

A student has test scores of 85, x, and y, respectively, and an average (arithmetic
mean) score of 95 on the three tests.

Column A Column B
The average (arithmatic mean) of x and y 100

if the quantity in Column A is greater;

if the quantity in Column B is greater;

if the two quantities are equal;

if the relationship cannot be determined from the information given.

POP”?

x and y are positive integers
x > 1
y < 2

Column A Column B

x 2y

if the quantity in Column A is greater;

if the quantity in Column B is greater;

if the two quantities are equal:

if the relationship cannot be determined from the information given.

POP”?

POTTERY2SHARD ::
syrnphonyzmusician
bread:crumb
wallzbrick
shoezheel
buildingzarchitect

even?

12. ENZYMEzCATALYST ::

vaccinezallergy
bacterium:microbe
glandzmuscle
veinzorgan
neuronzcorpuscle

P153095?

134

13.

14.

15.

16.

If 3 < x < 8 and 5 < y < 11, which of the following represents all the possible values
of xy?

 

 

 

 

A. 3<xy<11
B. 8<xy<l9
C. 15<xy<88
D. 24<xy<55
E. 33<xy<40
Ifx¢0,then x+7_l=
7x x
A. x+6
6x
B. x+6
7x
C. —6x+7
7x
1). l
7
E. -1
7

After thirty years of television, people have become “speed watchers”; consequently,
if the camera lingers, the interest of the audience ------- .

broadens
begins
varies
flags
clears

51.5.09“?

Biologists ------- isolated oceanic islands like the Galapagos, because in such small,
laboratory-like settings, the rich burly-burly of continental plant and animal
communities is reduced to a scientifically ------- complexity.

explore..diverse
desert..manageable
exploit..intimidating
reject..intricate
prize..tactable

£159.09”?

135

17.

The diameter of the circle is 10.

Column A Column B

The area of the region enclosed by quadrilateral ABCD 4O

POW?

18.

POP?

if the quantity in Column A is greater;
if the quantity in Column B is greater;
if the two quantities are equal;

if the relationship cannot be determined from the information given.

Column A Column B
ST + TR RS

if the quantity in Column A is greater;

if the quantity in Column B is greater;

if the two quantities are equal;

if the relationship cannot be determined from the information given.

19. ABACUS:CALCULATE ::

590W?

organzworship
patentzinvent
caliperszregulate
manuscriptzedit
sextantmavigate

136

20.

21.

22.

23.

24.

ESCAPEzCAPTURE ::
wamzdanger
immersezdampness
feintzthrust
dodgezblow
investzbankruptcy

91.5.05”?

If the average (arithmetic mean) of two numbers is 20 and one of the numbers is x,
what is the other number in terms of x?

A. 40—x
B. 40-2x
C. 20+x
D. 20—x
E. 20—2x

Ifx+y=n,thenx2+2.xy+y2=

A. 2n

B. n2

C. n(x — y)

D. n2 + 2y(n — y)
E. n + xn —

Crosby’s colleagues have never learned, at least not in time to avoid embarrassing
themselves, that her occasional ------- air of befuddlement ------- a display of her
formidable intelligence.

genuine..dominates
alert..contradicts
acute..precludes .
bogus..presages
painstaking..succeeds

51.5.0.5”?

That many of the important laws of science were discovered during experiments
designed to ------- other phenomena suggests that experimental results are the -------
of inevitable natural forces rather than of planning.

analyze..foundations
disprove..predecessors
alter..adjuncts
illuminate..consequence
verify..essence

apnea

137

25. Column A Column B

4+2J2 2+4J2

if the quantity in Column A is greater;

if the quantity in Column B is greater;

if the two quantities are equal;

if the relationship cannot be determined from the information given.

page

26.
3x = 4y

xy at 0
Column A Column B
The ratio of x to y The ratio of y to x

if the quantity in Column A is greater;

if the quantity in Column B is greater;

if the two quantities are equal;

if the relationship cannot be determined from the information given.

page?»

27. MIMICRY:CAMOUFLAGE ::
photosynthesiszpollination
territorialityzmigration
hibernation: generation
mutationzvariation
digestion:rumination

ever?

28. WAFTzPLUMMET ::
skim: glide
dreamzcaptivate
tosszcatch
flailzassault
meander:dash

£159.09"?

29. If a+b=10,then (a+%)+(b+%)=

A. 5
B. 10
C. 15
D. 20
E. 25

138

30.

31.

32.

IO miles

A F C

8 miles

 

According to the figure above, traveling directly from point A to point B, rather than
from point A to point C and then from point C to point B, would save approximately
how many miles?

590??
Unth—d

Compared mathematically to smoking and driving, almost everything else seems
relatively risk-free, ------- almost nothing seems worth regulating.

w
since
so

even though
as long as

mp???

It was not only the ------- of geologists that ------- earlier development of the
revolutionary idea that the Earth’s continents were moving plates; classical physicists,
who could not then explain the mechanism, had declared continental movement
impossible.

indecisiveness..challenged
radicalism..deterred
conservatism..hindered
assumptions..hastened
resistance..mandated

WPCP?

139

 

_l L
+wm—+

we L

l.- .-

{_ 25fi——->

 

 

 

 

 

The figure represents the floor of a certain room.

Column A Column B
The area of the floor 350 square feet

if the quantity in Column A is greater;

if the quantity in Column B is greater;

if the two quantities are equal;

if the relationship cannot be determined from the information given.

Page

. If 3x + 1 represents an odd integer, which of the following represents the next larger
odd integer?

3(x + l)
3(x + 2)
3(x + 3)

3x + 2

3(x + 2) + l

FUDGE”?

140

Performance Test 2

30°
A C

 

Column A Column B
AB BC

if the quantity in Column A is greater;
if the quantity in Column B is greater;
if the two quantities are equal;

if the relationship cannot be determined from the information given.

POW?

2. 750 < n < 1,500

Column A Column B

1,500 -— n n —— 750
if the quantity in Column A is greater;
if the quantity in Column B is greater;

if the two quantities are equal;
if the relationship cannot be determined from the information given.

POW?

3. HYMNzPRAISE ::
A. waltzzjoy
B. liturgyzrite
C. lullaby:child
D. dirge:grief
E. prayerzcongregation

4. STRAYzGROUP ::
miscalculatezsolution
improvisezsuggestion
slurzpronunciation
deletezchange
digresszsubject

WPPW?

141

5. If x can have only the values —3, 0, and 2, and y can have only the values —4, 2, and 3,
what is the greatest possible value for 2x + y?

A. 13
B. 15
C. 16
D. 20
E. 22
6- i P(2,7)
o

 

 

Q (6,0)
The area of A OPQ in the figure above is

A. 6
B. 12
C. 14
D. 21
E. 42

7. Politeness is not a ------- attribute of human behavior, but rather a central virtue, one
whose very existence is increasingly being ------- by the faddish requirement to
“speak one’s mind.”

superficial..threatened
pervasive..undercut
worthless..forestalled
precious..repudiated
trivial..affected

WPPW?

142

8.

10.

People frequently denigrate books about recent catastrophes as morally -------
attempts to profit from misfortune, but in my View our desire for such books, together
with the venerable tradition to which they belong, ------- them.

inopportune..encourages
fortuitous..fosters
treacherous..safeguards
despicable-.1egitimizes
corrupt. generates

WPPW?

 

 

 

 

On a turntable, a record of radius 6 inches is rotating at the rate of 45 revolutions per
minute.
Column A Column B
The number of inches traveled per The number of inches traveled per
minute by a point on the minute by a point on the record five
circumference of the record inches from the center of the record
A. if the quantity in Column A is greater;
B. if the quantity in Column B is greater;
C. if the two quantities are equal;
D. if the relationship cannot be determined from the information given.
8
X Y
4
W Z
Column A Column B
The area of a square 36

region with a perimeter
equal to the perimeter
of rectangular region
WXYZ

if the quantity in Column A is greater;

if the quantity in Column B is greater;

if the two quantities are equal;

if the relationship cannot be determined with the information given.

PWW?

143

11. LOGIC:REASONING ::
sensitivityzmorality
arrogance:leadership
ethicszbehavior
creativityzenthusiasm
braveryzcharisma

WPOW?

12. LIEN:CLAIM ::
briefzinvestigation
mortgagezinterest
foreclosurezpleading
gamishment:presumption
subpoena:command

WPPW?

13. If x is positive and y is 1 less than the square of x, which of the following expresses x
in terms of y?

A. x=y2—1
B. x=y2+l
C.x= y+1
D. x= l—y
E = l+y
1 2 3
14.I+—3-+§—

2 3 4
A. l

9
B. E

12
c. 33

12
D. 8
E. 9

144

15. Du Bois’ foreign trips were the highlight, not the ------- , of his travels; he was
habitually on the go across and around the United States.

16.

17.

18.

WPOW?

idiosyncrasy
result
precursor
culmination
totality

Nurturing the Royal Ballet’s artistic growth while preserving its institutional stability
has been difficult, because the claims of the latter seem inescapably to -------
development; apparently, attaining artistic success is simpler than ------- it.

WPOW?

ensure..promoting
inhibit..perpetuating
undermine..resurrecting
modify..appreciating
supplement..confining

Triangular regions T1 and T2 have equal areas and have heights h] and h,
respectively.
Column A Column B

POW?

POW?

The area of T1 The area of T3
h] h;

if the quantity in Column A is greater;

if the quantity in Column B is greater;

if the two quantities are equal;

if the relationship cannot be determined from the information given.

x2y>0
xf<0

Column A Column B
X y
if the quantity in Column A is greater;
if the quantity in Column B is greater;

if the two quantities are equal;
if the relationship cannot be determined from the information given.

145

19. TANGOzDANCE ::
arabesqueztheme
tonality:instrumentation
rhyme2pattem
stanzazline

elegy:poem

WPOW?

20. PLANT:SOIL ::
germzbacterica
organism:medium
sample: grth
nutrientzliquid
tree:root

WPOW?

21. On the number line, 1.4 is halfway between which of the following pairs of numbers?

—1.4 and 2.4
-1 and 2
—0.3 and 3.1
0.15 and 1.55
0.4 and l

WPOW?

22. What is the circumference of a circle with radius 8?

8/1:
16/1t
81t
161:
641:

WPOW?

23. Prudery actually draws attention to the Vice it is supposed to ------- ; the very act that
forbids speech or prohibits sight ------- what is hidden.

condemn..distorts
monitor..signals
repress..dramatizes
obviate..fosters
divulge..conceals

WPOW?

146

24. Although often extremely critical of the medical profession as a whole, people are
rarely willing to treat their personal doctors with equal ------- .

impetuosity
sarcasm
mockery
contempt
condescension

WPOW?

25. B

/\

A
The circle with center 0 has a radius of 5.
Column A Column B

The perimeter of AABC 24
if the quantity in Column A is greater;
if the quantity in Column B is greater;
if the two quantities are equal;
if the relationship cannot be determined from the information given.

POW?

26.
Column A Column B
1 1
4 + #1 2 + —1—1
3+— 3+—
2 4

if the quantity in Column A is greater;

if the quantity in Column B is greater;

if the two quantities are equal;

if the relationship cannot be determined from the information given.

POW?

147

27. FIDGETzNERVOUSNESS ::
cringe:dread
stallzfrustration
regale:amusement
doubtzconsternation
nagzannoyance

meow?

28. DORMANTleACTIVITY ::
starkzornateness
malleable:placticity
pronezuprightness
infuriatingztedium
slackzexcess

WPOW?

29. If a and b are both positive even integers, which of the following must be even?

I. ab
11. (a + l)”
111. a“ + 1’

I only

11 only

I and 11 only

I and 111 only
I, II, and III

30.
A

“I F

apnea

 

 

 

What is the perimeter of the pentagon above?

21
26
28
31
41

WPOW?

148

31. Despite claims that his philosophy can be traced to ------- source, the philosophy in
fact draws liberally on several traditions and methodologies and so could justifiably
be termed ------- .

a particular..consistent

a schematic..multifaceted
a dominant..cogent

an authoritative..derivative
a single..eclectic

WPOW?

32. The startling finding that variations in the rate of the Earth’s rotation depend to an
degree on the weather has necessitated a complete of the world’s time-keeping
methods.

A. unexpected..overhaul
B. anticipated..recalibration
C. indeterminate..rejection
D. unobservable..review

E. estimated..acceptance

33.

Nlr—tlfl Ip—I
Uilw

Column A Column B

r t
t r

if the quantity in Column A is greater;

if the quantity in Column B is greater;

if the two quantities are equal;

if the relationship cannot be determined from the information given.

POW?

149

34.

II I

 

III IV

 

Points (x,-3) and (-2,y), not shown in the figure above, are in quadrants IV and II,
respectively. If xy at 0, in which quadrant is point (x,y)?

I

II

III

IV

It cannot be determined from the information given.

WPOW?

150

APPENDIX D
Introduction to Test 1

In a few minutes, you will be taking the first test. Here is some general information
about the test you are about to take.

1) You will have 15 minutes to answer as many questions as you can. The questions are
taken from the verbal and quantitative (math) sections of past GRE exams.

2) There are two types of verbal questions: sentence completions and analogies. There
are also two types of quantitative questions: problem solving and quantitative
comparisons.

3) You are not expected to be familiar with the GRE, simply follow the instructions for
answering each question. You are allowed to use scratch paper, however, PLEASE DO
NOT USE CALCULATORS.

Before you begin the test, please review the following example questions. There are four
examples, one for each type of question. Please click CONTINUE to see the example
questions.

151

 

APPENDIX E
Example Test Questions
Sentence Completion Example Question

Directions: Each sentence below has one or two blanks, each blank indicating that
something has been omitted. Beneath the sentence are five lettered works or sets of
words. Choose the word or set of words for each blank that best fits the meaning of the
sentence as a whole.

Because it is ----- to ----- all the business costs related to employee discontent, an accurate
estimate of the magnitude of these costs is not easily calculated.

A) difficult...measure
B) impossible...justify
C) improper. . .overlook
D) useless...discover
E) necessary...pinpoint

Answer to Sentence Completion Example Ouestion

Because it is ----- to ----- all the business costs related to employee discontent, an accurate
estimate of the magnitude of these costs is not easily calculated.

A) difficult. . .measure — the correct answer is A
B) impossible...justify

C) improper...overlook

D) useless...discover

E) necessary...pinpoint

152

Analogies Example Question

Directions: In each of the following questions, a related pair of words or phrases is
followed by five lettered pairs of words or phrases. Select the lettered pair that best
expresses a relationship similar to that expressed in the original pair.

APPREHENSION:TERROR::
A) interest: conspiracy

B) affection: adoration

C) indifference: animosity
D) reluctance: termination

E) anxiety: faith

Answer to Analogies Example Question

APPREHENSION:TERROR::

F) interest: conspiracy

G) affection: adoration — the correct answer is B
H) indifference: animosity

I) reluctance: termination

J) anxiety: faith

153

Quantitative Comparison Example Question

Directions: Each question consists of two quantities, one in Column A and one in Column
B. You are to compare the two quantities and choose:

if the quantity in Column A is greater;

if the quantity in Column B is greater;

if the two quantities are equal;

if the relationship cannot be determined from the information given.

POW?

Ma: Since there are only four choices, NEVER MARK “E "
Example:
Column A Column B
2%percentof1,120 22.7

Answer to Quantitative Comparison Example Question

 

Column A Column B
1

ZEpement of1,120 22.7

- the correct answer is C

154

Problem Solving Example Question

Directions: The following question has five answer choices. Select the best of the answer
choices given.

9ft

 

 

6ft

 

 

 

 

 

 

The rectangular rug shown in the figure above has a floral border 1 foot wide on all sides.
What is the area, in square feet, of that portion of the rug that excludes the border?

A) 28
B) 40
C) 45
D) 48
E) 53

Afiwer to Problem Solving Example Question

9ft

 

 

:1: 6R

 

 

 

 

 

 

The rectangular rug shown in the figure above has a floral border 1 foot wide on all sides.
What is the area, in square feet, of that portion of the rug that excludes the border?

F) 28 — the correct answer is A
G) 40
H) 45
I) 48
J) 53

155

APPENDIX F
Test 1 Feedback Message
Positive feedback

Compared with scores from people who have answered these questions on past GRE
exams, your score is: ABOVE THE 80th PERCENTILE.

Negative feedback

Compared with scores from people who have answered these questions on past GRE
exams, your score is: BELOW THE 50th PERCENTILE.

156

APPENDIX G
Goal Orientation Manipulation 1
Mastery Manipulation 1
IMPORTANT !!! IMPORTANT !!! IMPORTANT !!!

Your score reflects your knowledge of math and verbal concepts. Your goal in this study
should be to LEARN information in order to IMPROVE YOUR MATH AND VERBAL
SKILLS. Your score is provided as a gauge for you to measure subsequent learning.
Your score should not be used to compare your performance to others’.

Performance-Prove Manipulation 1
IMPORTANT! !! IMPORTANT! ! ! IMPORTANT ! ! !

Your score reflects your performance relative to others’. Your goal in this study should
be to SCORE BETTER THAN THE OTHER STUDENTS by getting the HIGHEST
SCORE you can. Using your performance on Test 1 as a gauge, focus on maximizing
your score on Test 2.

Performance-Avoid Manipulation 1

IMPORTANT !!! IMPORTANT !!! IMPORTANT !!!
Your score reflects your performance relative to others’. Your goal in this study should
be to score AT LEAST AS WELL AS OTHER STUDENTS participating in this

experiment (at the 65th percentile). Using your performance on Test 1 as a gauge, focus
on scoring AT LEAST IN THE 65th PERCENTILE on Test 2.

157

APPENDIX H
Study Task Instructions

Now that you have received feedback about how you performed, you will be given an
opportunity to study before taking another test. You will have 15 minutes to study:

*F or the first 5 minutes, you will have the option to study different types of information.
You can study verbal and math KNOWLEDGE material, or verbal and math TEST
TAKING TIPS. KNOWLEDGE screens contain information about word definitions and
math rules commonly used in the GRE. TEST TAKING TIPS screens contain
information about the types of test questions found in the GRE.

*After the first 5 minutes, a screen will pop-up letting you know that you have 10
minutes left. At that point, you will have two more options in addition to studying. One
new Option will be to read information on Ethics in Psychology. Another new option will
be watch a timer that counts down the minutes to Test 2. If you chose the timer option,
you are welcome to surf the intemet. HOWEVER, PLEASE KEEP YOUR EYE ON
THE TIMER so that you are aware when your time is up. . .and you can continue with the
experiment at the proper time. PLEASE DO NOT ACCESS THE INTERNET UNLESS
YOU HAVE CHOSEN THE TIMER OPTION!

158

APPENDIX I

Goal Orientation Manipulation 2

Maste_ry Manipulation 2

IMPORTANT !!! IMPORTANT !!! IMPORTANT !!!

Remember that your goal in this study is to LEARN. Studying math and verbal
information can help you accomplish this goal. During the study time, focus on
information that will help you further DEVELOP YOUR MATH AND VERBAL
SKILLS.

You now have 15 minutes to study for the next test. Remember, you must study for at
least the first 5 minutes. Then, during the last 10 minutes, you will have the option of
studying, reading about Ethics in Psychology, or watching the timer and surfing the net.

Performance-Prove Manipulation 2
IMPORTANT! ! ! IMPORTANT! ! ! IMPORTANT ! ! !

Remember that your goal in this study is to get the HIGHEST SCORE. Studying math
and verbal information can help you accomplish this goal. During the study time, focus
on information that will help you PERFORM BETTER THAN OTHER STUDENTS
and get the highest score on Test 2.

You now have 15 minutes to study for the next test. Remember, you must study for at
least the first 5 minutes. Then, during the last 10 minutes, you will have the option of
studying, reading about Ethics in Psychology, or watching the timer and surfing the net.

Performance-Avoid Manipulation 2
IMPORTANT !!! IMPORTANT !!! IMPORTANT !!!

Remember that your goal in this study is to PERFORM AT LEAST IN THE 65th
PERCENTILE. Studying math and verbal information can help you accomplish this
goal. During the study time, focus on information that will help you PERFORM AT
LEAST AS WELL AS OTHER STUDENTS. '

You now have 15 minutes to study for the next test. Remember, you must study for at
least the first 5 minutes. Then, during the last 10 minutes, you will have the option of
studying, reading about Ethics in Psychology, or watching the timer and surfing the net.

159

APPENDIX J

Study Task

Verbal Knowledge lnforrnation

The VERBAL KNOWLEDGE screens contain definitions of the types of words likely to be
found on the GRE.

The following categories of words commonly found on the GRE. The general meaning of
the words in each list are in ALL CAPS.

DIFFICULT TO UNDERSTAND
Abstruse
Arcane
Enigmatic
Esoteric
Inscrutable
Obscure
Opaque
Rarefied
Recondite
Turbid

CRITICIZE/CRITICISM
Aspersion

Belittle

Berate

Calumny

Casti gate

Diatribe

Tirade

DEBAUCHED/DEBAUCHERY
Depraved

Dissipated

Salacious

Sordid
Turpitude

F UNERAL/DEATH
Bereave

160

F ALSEHOOD
Chicanery
Dirge
Dissemble
Equivocate
Feint

Feigned

Guile

BITING (AS IN WIT OR TEMPERAMENT)
Caustic

PRAISE
Aggrandize
Eulogize

Fawn
Laud/Laudatory
Venerate/Veneration

HARMFUL/TO HARM
Baneful

Perfidious

J eopardize

TIMID
Diffident
Trepidation

BORING
Banal
Hackneyed
Mundane
Tediurn
Trite

RENDER USELESS/WEAKEN
Envervate

Inhibit

Obviate

OVERB LOWN/WORDY
Bombastic

Garrulous

Loquacious

Verbose

161

IMPULSIVE
Irnpetuous
Rash

HOSTILE
Malevolent
Misanthropic
Vindictive

STUBBORN
Obdurate
Obstinate
Vexing

GENEROUS/KIN D
Altruistic
Magnanimous
Philanthropic

GREEDY
Avaricious
Miserly
Penurious

TERSE
Laconic
Tactiturn

LAZY / S LUGGISH
Indolent

Languid

Lethargic
Phlegrnatic
Quiescent

Slothful

PACIFY
Mitigate
Placate
Propitiate

POOR
Destitute
Disingenuous
Dissemble
Indigent

162

ECCENTRIC/DISSIMILAR
Anomalous

Eclectic

Iconoclast

Radicalism

BACK OUT, OFF
Abate

Abjure

Recant

Rescind

HIGHEST POINT
Climax
Culmination

BEGINNING
Engender
Indigenous
Nascent

LAW
Codicil
Culpable
Exculpate
Litigation

POLITICS & GOVERNMENT
Abdicate

Anarchy

Bourgeois

Matriarchy

Oligarchy

The following is a list of words and definitions commonly used in GRE questions.

Abase (v): cast down, bring low, humble, degrade, disgrace. abassare (to lower)

 

Abash (v): dismay, confound, disconcert, discomfit, make ashamed. esbahir (to make
ashamed)

Abdicate (v): renounce, abandon or relinquish a position, right, power, or trust. a_b_
(fiom, away) + dicare (to proclaim)

Aberration (n): wandering, straying from the right way; deviation, esp. from what is

163

right, normal or natural; mental disorder. Q (from, away) + we (wander)

Acumen (n): acuteness of mind, judgment, sagacity, perspicacity, shrewdness,
discernment

Admonish (v): (a) warn, caution, reprove, rebuke, reprimand
(b) inform or remind by way of a warning

(c) advise, exhort. ad (to) + monere (to warn)

Baleful (adj): sinister, harmful, evil, calamitous, deadly, pernicious, destructive. bealu
(an evil), fiom balw (evil) + _fu_ll_

Bane (n): that which causes death, destruction; the cause of injury, mischief or
destruction. bana (murderer)

Beneficient (adj): bringing about or doing acts of kindness or charity; resulting in
benefit; charitable, munificent, kindly. bene (well) + facere (to do)

Bombast (n): high-sounding but silly or meaningless language; pomposity,
grandiloquence; an inflated style; fustian. From bombasium, a cotton doublet
(orig. meaning = cotton stuffing, padding)

Cacophony (n): janing or disagreeable combination of sounds, words, tones, or musical
notes; any harsh, jarring sound; discord. kakos (bad) + phone (voice)

Cajole (v): flatter, coax, persuade, wheedle, deceive by flattery. cajoler (to coax)

Chicanery (n): trickery, especially legal; deceit, guile, unfair artifice used in contest or
discussion.

Condescension (n): a patronizing manner or behavior
Constemation (11): great fear or shock that makes one feel helpless or bewildered
Divulge (V): to make known; disclose; reveal

Fortuitous (adj): accidental, by chance. fortis (chance)

 

Irnmerge (v): to plunge or disappear, as in a liquid
Meander (v): wander aimlessly, ramble, peregrinate.
Perpetuate (V): make perpetual, preserve, cause to endure, save from oblivion.

Repudiate (v): reject or disown as having no validity

164

Below is a list of word roots. Learning these can help you to determine the meaning of
words you are unfamiliar with.

ABLE, IBLE: CAPABLE OF, WORTHY OF. changeable, durable, laudable,
indubitable, inevitable, infallible, irreducible, tolerable, variable.

AD, A: TO. adapt, adequate, adumbrate, advocate, accede, adduce, affiliate, aggregate,
allocate, annunciation, appall, arrest, assiduous, attract.

ANTI, ANT: AGAINST, OPPOSITE. anticlimax, antidote, antipathy, antiphony,
antipodes, antithesis, antagonism.

AUTO, AUT: SELF. autobiography, autocracy, autograph, automation, autonomous,
autopsy, autism.

BEL, BELL: BEAUTIFUL. belle, embellish.

CO, COM, CON: WI'I'H, TOGETHER. coeducation, coefficient, coincide,
communicate, communist, compare, concert, concubine, conflict, cooperate,
correspond.

CONTRA, CONTRO, COUNTER: AGAINST. contradict, contrary, controversy,
counter, counteract, counterattack, counterfeit, countennand, counterpart,
counterpoint, encounter.

CULP: FAULT, BLAME. culpable, culprit, inculpate, exculpate.

DI, DIS: AWAY, APART. disagreeable, discard, discern, disdain, dismay, dismiss,
distant, diverge.

E, EX, EC: OUT. eliminate, emanate, eradicate, erase, evade, evict, evince, exact,
excavate, except, excerpt, excise, excite, exclusive, excommunicate, exile, exit.

FUG: FLEE. centrifuge, fugitive, fugue, refuge, refugee, subterfuge.

GEN: BIRTH, CLASS, KIN. gender, gene, general, generation, generosity, genesis,
genetics, genial, genital, genius, gentle, gentile, gentility, gentry, congenital,
degenerate, engender, eugenics, ingenious, ingenuity, ingenuous, progeny,
progenitor, regenerate.

HEMI: HALF. hemiptera, hemisphere, hemistich

IM, IN: NOT. immature, immutable, imperfect, improvident, indigestible, inhospitable,
innocuous, intolerant.

165

IN, IM, 1: IN, ON. (often N is dropped and the first letter to which I is prefixed is
doubled) illuminate, incantation, induct, infer, imbibe, immigrate, impact,
irrigate.

JUNCT, JUG: JOIN. adjunct, conjugal, conjunction, injunction, junction, junta,
subjugate, subjunctive.

LUX, LUC: LIGHT. elucidate, lucid, lucubrate, luster, pellucid, translucent.

MAL: BAD. maladroit, malady, malediction, malefactor, malevolence, malice,
malinger.

MUT: CHANGE. commute, immutable, mutability, mutation, mutual, permutation,
transmute.

OMNI: ALL. omnibus, omnipresent, omnipotent, omniscient, omnivorous.

ONER: BURDEN. exonerate, onerous, onus.

POST: BEHIND, AFTER. posterior, posterity, postem, posthumous, postrneridian, post-
mortem, postpone.

RE: BACK, AGAIN. recline, refer, regain, remain, reorganize, repent, request.

SEX: SIX. sextuplets, sextile, sextet

SOPH: WISDOM. philosopher, sophism, sophist, sophisticated, sophistry, sophomore.
TACIT: SILENT. reticent, tacit, tacitum.

UN: NOT. unaccustomed, unruly, unseen, untold, unusual.

166

Verbal Test Tips Information

The verbal TEST TIPS screens contain information about the types of GRE verbal
questions.

SENTENCE COMPLETIONS

Make sure you memorize the instructions for sentence completions:

Each sentence below has one or two blanks, each blank indicating that something has
been omitted. Beneath the sentence are five lettered words or sets of words. Choose the
word or set of words for each blank that @fits the meaning of the sentence as a whole.

Here are some strategies for answering sentence completions:

- Read the sentence strategically, using you knowledge of scope and structure to see
where the sentence is heading.

0 In your own words, anticipate the answer.

0 Look for answers close in meaning to yours. (Eliminate tempting wrong answers
using the clues.)

o Read your choice back into the sentence to be sure it fits.

There are elements in the sentence that limit the possible answers. Finding these
elements will guide you to the correct answer. Here are some important “roadsigns” to
look for:

“Straight Ahead” sigps are used to make one part of the sentence support or elaborat_e
another part.

A. The winning argument was ------- and persuasive. (cogent, flawed)
and thus since
similarly because consequently
in addition likewise also
therefore

167

“Detour” signs ch_ange the direction of the sentence.

B. The winning argument was ------- but persuasive. (cogent, flawed)
but despite unless
yet however rather
although while unfortunately
nonetheless

On the GRE a semicolon by itselealways connects two closely related clauses.

 

C. The play’s script lacked depth and maturity; likewise, the acting was altogether ------
-. (sublime, amateurish)

In two-blank sentences. look for the roadsigps for each blank. Notice that sometimes the
second blank is easier to workrwith.

5. Unfortunately, there are some among us who equate tolerance with immorality, they
feel that the ------- of moral values in a permissive society is not only likely, but -------.

decline. . .possible

upsurge. . .predictable
disappearance. . .desirable
improvement. . .commendable
deterioration. . .inevitable

WPOW?

TECHNIQUES FOR THE HARDEST QUESTIONS

0 For the hardest problems, always rephrase the sentence in your own terms to get the
gist.

0 Eliminate any choice in which one word doesn’t fit the sentence.

0 Eliminate any choice in which words don’t fit with each other.

STRATEGIES FOR AN ALOGIES

Make sure you memorize the instructions for analogies:

168

In each of the following questions, a related pair of words or phrases is followed by five
lettered pairs of words or phrases. Select the lettered pair that best expresses a
relationship similar to that expressed in the original pair.

Every Analogy question consists of two words which are separated by a colon. Below
them are five answer choices.

There wilLallways be a logical and necessary relationship between the words in the stem
pair. You express this relationship by making a short sentence called the bridge. Always

t_ry to make a bridge before looking at the answer choices.
STRONG AND WEAK BRIDGES

A weak bridge expresses a relationship that isn’t necessary or direct. Some signs of weak
bridges are: “usually,” “can,” “seldom,” or “sometimes.”

Weak bridges:

A. Some MAPS are put in ATLASES.

B. A MAP is usually smaller than an ATLAS.

C. MAPS and ATLASES have to do with geography.
D. A MAP is a page in an ATLAS.

A strong bridge expresses a logical and necessary relationship.

The sign of a strong bridge: It contains (or can contain) a definite word like “always,”
“never,” or “must.”

The best bridge is a strong bridge that fits exactly one answer choice.
Strong bridges:

E. MAPS are what an ATLAS contains.

F. MAPS are the unit of reference in an ATLAS.

G. An ATLAS is an organized collection of MAPS.

Never fall for Same Subject Analogies-pairs of words that are not related to each other,
but only to a third word.

Same-subject Analogies:
H. Necklace : Bracelet

1. Temperature : Humidity
J. Relay : Marathon

169

THE FOUR-STEP SOLUTION TO AN ALOGIES

Find a strong bridge between the stem words.

Plug the answer choices into the bridge

Adjust the bridge as necessary

If stuck, make bridges out of the answer choices and work backwards.

PPNI‘

Eliminate choices with weak bridges, or two choices that have the same bridge.
Look at the answer choices to determine the parts of speech.

Try matching positive or negative connotations in the stem words with those in the
answer choices.

Even when you don’t know all the words, elimination and intelligent guessing will earn
you more points.

FIVE CLASSIC BRIDGES

ETS uses some kinds of bridges over and over on the GRE. You don’t have to memorize
these, but exposing yourself to them will give you a feel for the sort of bridge that will
get you the right answer.

1. Definition (“is always” or “is never”)

13. PLATITUDE : TRITE ::
riddle : unsolvable
axiom : geometric
omen : portentous
syllogism : wise
canard : concise

WPOW?

2. Function / Purpose

14. AIRPLANE : HANGAR ::
music : orchestra
money: vault

finger : hand

tree : farm

insect : ecosystem

WPOW?

3. Lack
15. LUCID : OBSCURITY ::

A. ambiguous : doubt
B. provident : planning

170

C. furtive:legality
D. economical : extravagance
E. secure : violence

. Degree (sometimes to the point of excess)

16. ATTENTIVE : RAPT ::
loyal : unscrupulous
critical : derisive
inventive : innovative
jealous : envious
kind : considerate

WPOW?

. Characteristic Actions / Items

17. SCALPEL : SURGEON ::
palette : artist

stage : dancer
dictionary: poet

lock : burglar

chisel : sculptor

WPOW?

171

Math Knowledge Information

The AJAT H KNOWLEDGE screens contain information about the types of math rules
likely to be needed for answering GRE questions.

NUMBER OPERATIONS
I. ORDER OF OPERATIONS

PEMDAS = Please Excuse My Dear Aunt Sally — this mnemonic will help you remember
the order of operations

P = Parentheses

E = Exponents

M = Multiplication (in order fiom left to right)

D = Division (in order from left to right)

A = Addition (in order from left to right)
S = Subtraction (in order from left to right)

II. FRACTIONS

o Canceling and Reducing Fractions

Generally speaking, when you work with fractions on the GRE, you’ll need to put
them in lowest terms. That means that the numerator and the denominator are not

divisible by any common integer greater than 1. For example, the fraction % is in

lowest terms, but the fraction % is not, since 3 and six are both divisible by 3. The

method we use to take such a fraction and put it in lowest terms is called reducing.
That simply means to divide out any common multiples from both the numerator and
denominator. This process is also commonly called canceling.

Ex: Reduce :17: to lowest terms.

First, determine the largest common factor of the numerator and
denominator. Then, divide the top and bottom by that number to reduce.

15_3x5_3XS-:—5_3
35 7x5 7x5+5 7

 

o Multiplication of Fractions

E 19... 2,. _8_
9 4 15
. . . 2 10) '3 82
First, reduce (cancel) diagonally and vertically. — x 7 x 1—5—
3 l 3

172

Then multiply numerators together and denominators together.
2X1x2 4

3x1x3 -9

 

Division of Fractions
Dividing is the same as multiplying by the reciprocal of the divisor. To get the
reciprocal of a fraction, just invert it by interchanging the numerator and the

denominator. For example, the reciprocal of the fraction 3 is g.

.44

To divide, invert the second term (the divisor), and then multiply as above.

4 4 4 9 ‘4 9’ 1x3

_+_:—x—=——X——=———=3

3 9 3 4 ,3 4, 1X1
MixedNumbers

Mixed Numbers are numbers consisting of an integer and a fraction. For

example, 3211-, 12% and 5:;- are all mixed numbers. Fractions whose numerators

are greater than their denominators may be converted into mixed numbers, and
vice-versa.

Ex: Convert ~223— to a mixed number.

23 20 3 3
_=_+_=5_
4 4 4 4

Ex: Convert 2% to a fraction.

23=2+3=14+

_3_u
7 7777

III. POWERS AND ROOTS

In the term 3x2 , 3 is the coefficient, x is the base, and 2 is the exponent. The exponent

refers to the number of times the base is multiplied by itself, or how many times the base

is a factor. For instance, in 43, there are 3 factors of 4: 43 = 4 o 4 o 4 = 64.

A number multiplied by itself twice is called the square of that number, e. g. x2 is x
squared.

A number multiplied by itself three times is called the cube of that number, e. g. 43 is 4
cubed.

173

To multiply two terms with the same base, keep the base and add the exponents.

Ex=2’-2’=(2-2)(2-2-2) or 22.23:?”
=(202020202) :25
:25

4 7 4+7 1
Exzx 0x =x =xl

To divide two terms with the same base, keep the base and subtract the exponent of the
denominator from the exponent of the numerator.

O
404

 

Ex: pr¢0, %—=p3-2 8p! =p

2

To raise a power to another power, multiply the exponents.

EX: (32)4 ___ (3 . 3)4 or (32)4 = 32 X4
= (2 ° 3)(3 ' 3)(3 ' 3)(3 ' 3) = 38
= 3

Any non zero number raised to the zero power is equal to l. a0 = 1, if a at O. 00 is
undefined.

A negative exponent indicates a reciprocal. To arrive at an equivalent expression, take
the reciprocal of the base and change the sign of the exponent.

a'“= J? or (1)"
a a
l 1 1
Ex: 2'3 = — 3 =——=—
(2) 23 8

A fractional exponent indicates a root.
1

(a); = Kl; (read “the nth root of a.” If no “n” is present, the radical sign means square
root.)

1
Ex: 83 = 3 8 = 2
On the GRE, you will probably only see the square root. The square root of a non-
negative number x is equal to the number which when multiplied by itself gives you x.
Every positive number has two square roots, one positive and one negative. The positive
square root of 25 is 5; and the negative square root of 25 is —5, since (-5)2 = 25 also.

174

Other types of roots have appeared on the test (cube root, or 1]— , is an example), but they
tend to be extremely rare.

NEE: In the expression 3x2, only the x is being squared, not the 3. In other words,
3x2: 3(x ). If we wanted to square the 3 as well, we would write (3x)2.
(Remember in the order of operations we raise to a power before we
multiply, so in 3x2 we square x and then multiply by 3.)

IV. RULES OF OPERATIONS WITH ROOTS

By convention, the symbol J— (radical) means the positive square root only.

Ex: J5=+3;— 9 =-3
Even though there are two different numbers whose square is 9 (both 3
and —3) we say that J9 is the positive number 3 only.

When it comes to the four basic arithmetic operations, we treat radicals in much the same
way we would treat variables.

0 Addition and Subtraction of Roots

Only like radicals can be added to or subtracted from one another.

Ex: 2./§+4./§-./§—3./§=(4./§—./§)+(2./§-3./§) [Notez J5=1J§
=3J§+(-\/§)
=3JE-J3

o Multiplication and Division of Roots

To multiply or divide one radical by another, multiply or divide the numbers outside
the radical signs, then the numbers inside the radical signs.

Ex: (6J5)x(2./§)=(6x2)-(./§x./§)=12./3x5 =12.f1§

Ex: 12./Sew?=(12+2)-(./E+./§)=6(\/§)=6s/§

4J7?
xz—T=(% )<—1—-‘/_68=) XEF 2f

If the number inside6 the radical 15 a multiple of a perfect square, the expression can be
simplified by factoring out the perfect square.

Ex: fi=xl36x2=J36xJ2=6J2

175

V. BASIC ALGEBRA
OPERATIONS WITH POLYNOMIALS

All of the laws of arithmetic operations, such as the associative, distributive, and
commutative laws, are also applicable to polynomials.

o Communitative Law

2x+5y=5y+2x
5ax3b=3bx5a =15ab

o Associative Law

2x—3x+5y+2y=(2x—3x)+(5y+2y)=—x+7y

(—2x)(% x><3y)(—2y) = (—x2 )(—6y2) = 6x’y’

- Both Laws

2x + 3x2 — 6x + 4x2 = 2x - 6x + 3x2 + 4x2 (commutative law)
= (2x — 6x) + (3x2 + 4x2) (associative
law)

= —4x + 7x2
Note: This process of simplifying an expression by subtracting or adding together those
terms with the same variable component is called combining like terms.

0 Distributive Law

3a(2b - 5c) = (3a x 2b) — (3a x Sc) = 6ab — 1 Sac

Note: The product of two binomials can be calculated by applying the distributive law
twice.

Ex: (x+5)(x-2)=x0(x—2)+50(x—2)
=xox—x-2+5-x—502
=x2—2x+5x—10

= x2 + 3x -10
LINEAR EQUATIONS
An gguation is an algebraic sentence that says that two expressions are equal to each

other. The two expressions consist of numbers, variables, and arithmetic Operations to be
performed on these numbers and variables. To solve for some variable we can

176

manipulate the equation until we have isolated that variable on one side of the equal sign,
leaving any numbers or other variables on the other side. Of course, we must be careful
to manipulate the equation only in accordance with the equality postulate: whenever we
perform an operation on one side of the equation we must perform the same operation on
the other side. Otherwise, the two sides of the equation will no longer be equal.

A linear or first-degree equation is an equation in which all the variables are raised to the .
first power (there are no squares or cubes). In order to solve such an equation, we’ll
perform operations on both sides of the equation in order to get the variable we’re solving
for all alone on one side. The operations we can perform without upsetting the balance of
the equation, are addition and subtraction, and multiplication or division by a number
other than 0. Typically, at each step in the process, we’ll need to use the reverse of the
operation that’s being applied to the variable in order to isolate the variable. In the
equation n + 6 = 10, 6 is being added to n on the left side. To isolate the n, we’ll need to
perform the reverse operation, that is, to subtract 6 from both sides. That gives us n + 6 —
6=10-6,orn=4.

Let’s look at another example.

Ex: If 4x-7=2x+5,whatisx?

1. Get all the terms with the 4x — 7 = 2x + 5
variableononesideofthe 4x—2x—7=2x-2x+5
equation. Combine the terms. 2 7 5
x -— =
2. Get all constant terms on the 2x — 7 + 7 = 5 + 7
other side of the equation. 2x :12
3. Isolate the variable by 2x _ 12
dividing both sides by its 3— ' 3—
coefficient. x = 6

We can easily check our work when solving this kind of equation. The answer we arrive
at represents the value for the variable in the original equation. If the equation holds true,
we’ve found the correct answer. In the above example, we got a value of 6 for x.
Replacing x with 6 in our original equation gives us 4(6) — 7 = 2(6) + 5, or 17 = 17.
That’s clearly true, so our answer is indeed correct.

TRIANGLES

GENERAL TRIANGLES

A triangle is a closed figure with three angles and three straight sides. An acute triangle
is a triangle with three acute angles. An obtuse triangle is a triangle with an obtuse angle.

A right triangle is a triangle with a right angle. In a right triangle, we call the side
opposite the right angle the hypotenuse and the other two sides the Leg.

177

A \

Acute Obtuse Right

THE SUM OF THE INTERIOR ANGLES
OF ANY TRIANGLE IS 180 DEGREES.

Each interior angle is supplementary to an adjacent exterior angle. The degree measure
of an exterior angle is equal to the sum of the measures of the two non-adjacent (remote)
interior angles, or 180° minus the measure of the interior angle.

In the figure below, a, b, and c are interior angles. Therefore a + b + c =180. In addition,
d is supplementary to c; therefore d + c = 180. And since d is an exterior angle, it is
equal to the sum of the two remote interior angles: a and b.

 

 

The altitude (or height) of a triangle is the perpendicular distance from a vertex to the
side opposite the vertex. The altitude can fall inside the triangle, outside the triangle, or
on one of the sides.

 

 

 

   

Altitude=AD Altitude=EH
Altitude=AC

0 Sides and Angles Of Triangles

The length of any side of a triangle is less than the sum of the lengths of the other
two sides, and greater than the difference of the lengths of the other two sides.

b+c>a>b-c
a+b>c>a-b b C
a+c>b>a-c c B

178

If the lengths of two sides of a triangle are unequal, the greater angle lies opposite the
longer side and vice versa. In the figure above, if [A > AB > 1C, then a > b > c.

Area of a Triangle

The area of a triangle is

% base x height.

Remember that the height (or altitude) is perpendicular to the base. Therefore, when
two sides of a triangle are perpendicular to each other, the area is easy to find. In a
right triangle, we call one leg the base and the other leg the height. Then the area is
one half the product of the legs, or

A=lbh
2
1

=-—ll><l2

2

Perimeter of a Triangle

The perimeter of a triangle is the distance around the triangle. In other words, the
perimeter is equal to the sum of the lengths of the sides.

Isosceles Triangles

An isosceles triangle; is a triangle that has two sides of equal length. The two equal
sides are called Leg and the third side is called the bag.

Since the two legs have the same length, the two angles opposite the legs must have
the same measure. In the figure at right, PQ = PR, and 1Q: AR.

 

Equilateral Triangles

179

An equilateral triangle has three sides of equal length and three 60° angles.

...; .-

LM=MN=LN
AL= 4M=4N=60 "
0 Right Triangles

As we stated in the previous section, the longest side (which lies Opposite the right
angle, the largest angle) of a right triangle is called the hypotenuse. The other two
sides are called the Leg.

 

 

PYTHAGOREAN THEOREM
B
Hypotenuse
Leg. a C
H
C b A

Leg;
(Leg [)2 + (Leg2)2 = (Hypotenuse)2 or a2 + b2 = c2

The Pythagorean theorem holds for all right triangles, and states that the square of
the hypotenuse is equal to the sum of the squares for the legs.

Some sets of integers happen to stisfy the Pythagorean theorem. These sets of
integers are commonly referred to as “Pythagorean triplets.” One very common set
that you might remember is 3, 4, and 5. Since 32 + 42 = 52, you can have a right
triangle with legs of lengths 3 and 4, and hypotenuse of length 5. This is probably the
most common kind of right triangle on the GRE. You should be familiar with the
numbers, so that whenever you see a right triangle with legs of 3 and 4, you will
immediately know the hypotenuse must have length 5. In addition, any multiple of
these lengths makes another Pythagorean triplet; for instance, 62 + 82 = 102, so 6, 8,
and 10 also make a right triangle. One other triplet that appears occasionally is 5, 12,
and 13.

180

The Pythagorean theorem is very useful whenever you’re given the lengths of two
sides of a right triangle; you can find the length of the third side with the Pythagorean
theorem.

CIRCLES

 

CIRCLE
The set of all points in a plane at the same distance from a certain point. This point is
called the center of the circle.

A circle is labeled by its center point: circle 0 means the circle with center point 0. Two
circles of different size with the same center are called concentric.

DIAMETER
A line segment that connects two points on the circle and passes through the center of the
circle. In circle 0, AB is a diameter.

RADIUS
A line segment from the center of the circle to any point on the circle. The radius of a
circle is one-half the length of the diameter. In circle 0, 0A, OB, and GT are radii.

CIRCUMFERENCE AND ARC LENGTH

The distance around a circle is called the circumference. The number 11 (“pi”) is the ratio
of a circle’s circumference to it’s diameter. The value of rt is 3.1415926. . ., usually
approximated 3.14. For the GRE, it is usually sufficient to remember that 1t is a little
more than 3.

Since 1! equals the ratio of the circumference to the diameter, a formula for the

circumference is A
B
C = 1rd
or
C = 21tr
X

An ar_c is a portion of the circumference of a circle. In the figure to the right, AB is an arc
of the circle, with the same degree measure as central angle A08. The shorter distance

181

between A and B along the circle is called the minor arc; the longer distance AXB is the
major arc. An are which is exactly half the circumference of the circle is called a
semicircle (in other words, half a circle).

The length Of an arc is the same fraction of a circle’s circumference as its degree measure

is of the degree measure of the circle (360°). For an arc with a central angle measuring n
degrees,

= (-—n—)(circumference)

 

Arc length 360
n
= —— x 2717‘
360
AREA OF A CIRCLE A
The area of a circle is given by the formula A
B
Area = 1tr2

A sector is a portion of the circle, bounded by two radii and an arc. In the circle to the
right with center 0, OAB is a sector. To determine the area of a sector of a circle, use the
same method we used to find the length of an arc. Determine what fraction of 360° is in
the degree measure of the central angle of the sector, and multiply that fraction by the
area of the circle. In a sector whose central angle measures 11 degrees,

Area of sector = (fi) x (Area of circle)

n
:——x7zr2

360

182

Math Test Tips Information

The MATH TEST TIP screens contain information about types of GRE math questions

PROBLEM SQLVING STRATEGIES

FIVE STEPS FOR SQLVING MATH PROBLEMS

1.

Read the problem.

By this, we mean read THE WHOLE PROBLEM, including the answer choices.
Some people work on the problem one piece at a time before reading all the way
through, and end up doing extra calculations or solving for variables they never
needed to know. This is apt an efficient way to proceed. Approach math problem
solving like reading comprehension (although perhaps not with the same dread):
since you don’t stop after each detail in a reading passage to digest what you’ve
read, you shouldn’t do so with a math problem either. Read through the whole
problem, so you get the “big picture.” Also, keep in mind the difficulty level of
the problem: if it’s one of the first few multiple choice questions, it will probably
be rather straightforward. If, on the other hand, it’s one of the last questions on
the exam, then it’s a good bet that the phrasing is intentionally confusing or the
problem takes an unexpected turn.

“Fine,” you say, “that makes sense, but why should I read the answer choices?

I’ll only need those at the end.” Actually, if you wait until you’ve finished
solving the problem to look at the answer choices, you’ll have missed some of the
most important clues the GRE gives you. Look at the answer choices and ask
yourself:

OAre the answer choices close together or far apart? Can I approximate?

OWhat form are the answer choices in? Are they variables? Fractions? Radicals?
Exponents?

OCan I plug the answer choices back into the problem?

OCan I eliminate some answer choices using common sense?

Looking at the answer choices and asking yourself these questions can save time.
Determine what the question is.

This isn’t as easy (or as obvious) as it sounds. You must know what the question
specifically asks you to find. Sometimes, a question is set up to lead you to think
they’re going to ask for the value of some particular quantity, but instead they ask

for something different. Here’s an example:

Every week, John spends 20 percent of his
paycheck on groceries, 40 percent on

183

 

clothing, and saves the remainder. If in one
week he spends $35 on groceries, how much
does he spend on clothing?

If you don’t pay attention to what this question is asking, you may assume you
need to find his total income for the week-and chances are that that w be 3
answer choice! In fact, you don’t need to find his total income anywhere in the
solution: since the 40 percent he spends on clothing is twice the 20 percent he
spends on groceries, he’ll spend twice $35, or $70 a week on clothing.

Determine what you’re given

What you’re given can be stated explicitly in the problem or in an accompanying
diagram. Often this will be all you need, especially in the easier problems. But
also look for information that is stated IMPLICITLY. For instance, if you’re
given one side of an isosceles right triangle, you can find the length of any other
side. (Now we’re not saying that you should as a matter of course figure out the
other side lengths; wait and see whether you need to find them. But keep in mind
that you 8L6 given the lengths.) Chances are that this implicitly offered
information will prove usefill in solving many problems. At the same time, be
careful. Don’t get caught up in every little detail; occasionally, some are designed
to throw you off.

Identify a path for getting from what you are given to the answer.

Often one path will appear to you immediately. It may not be the fastest way,
however, and it may be worth spending a few seconds looking for another method
if it will save you a minute of test time. Sometimes you must invest time to save
time. If your method looks like it will take five minutes Of calculations, then you
KNOW that there has to be a shortcut.

Also remember: you are looking for the ANSWER, not the SOLUTION. There’s
a big difference. The GRE is a multiple-choice test, so the ANSWER is either A,
B, C, D, or E, (except for quantitative comparison questions where the answer is
A, B, C, or D). If a shortcut lets you pick out the answer without going through
all the math, all the better. At times you don’t need to solve anything; you can

0 Plug in the answer choices;
0 Pick numbers to stand in for variables;
0 Eliminate answer choices using common-sense approximation, and the like.

Select an answer choice.
You may find at this point that the path you have chosen does not work.

Unpleasent, but we must face reality. You will have several choices: you can try
a different method, you can leave the question until later, or you can guess. Your

184

 

decision should depend on how much time you’ve already spent on the question,
how far you are in the section, how much time you have left, and your own “gut”
instinct.

ALTERNATIVE METHODS AND GUESSING

There are a number of common techniques that are useful in solving many types
of questions. That is why we introduce them here, rather than in any particular
section. Remember, these techniques are often faster and simpler than solving the
question using traditional math, so bear them in mind when trying to find a path to
solve a problem. They will be covered in more detail throughout the text,
particularly in the discussion of question types that lend themselves to these
alternative techniques. And, of course, alternative methods will be discussed
occasionally in the solutions to homework problems. The examples used here are
strictly for demonstrative purposes; these questions are fairly straightforward and
can be best answered in the conventional way. Focus you attention on the method
introduced, not the specific problem.

ALTERNATIVE METHODS
Backsolvirg

This technique takes advantage of an important feature of the GRE: you’re always
given the correct answer. The task at hand is merely deciding which answer of
the five choices is the correct one. One way to do that is simply to try each
possibility. In other words, use the answer choices to work back through the
question stem. What makes this technique of backsolving expecially useful is
that, in general, you only need to check at most two choices to find the answer.

Numerical answer choices on the GRE are nearly always listed in ascending or
descending order. Therefore, when you’re checking numerical choices, you
should always start with answer choice C, the middle choice. Then if choice C is
not correct, move to either answer choice B or D, depending on whether the result
you got with choice C was higher or lower than what you wanted. And, generally
speaking, those are all the choices you’ll need to check.

Picking Numbers

In the algebra section, we’ll review how algebra is tested on the GRE. Problems
that involve variables can sometimes be tricky and time-consuming. One way
around that is to pick simple numbers to stand in for variables. Substitute a
number for a variable into the question stem and find that value the question is
asking for. Then, substitute the same number for that variable in the answer
choices, and the answer choice which gives the same result will be correct. If two
answer choices give the same result, try another number just on those two
choices.

185

 

When picking numbers, it’s important to check every answer choice, and if two

choices give the desired result, pick an additional number to narrow things down
fiirther.

This method is generally most useful on problems with variables and complicated
algebraic expressions. Problems that involve percents, particularly percent word
problems, and be simplified greatly by picking number.

Guessing

You should never leave anything blank on the GRE. Even if you have absolutely
no time to attempt a problem, you should still select an answer choice. In some
instances, you will find that either you don’t have enough time to complete a
problem or you’re unable to find a problem’s solution. In either case, the natural
Option is to make a guess. In both cases, though, how you guess can make a
difference. It’s often the case that , even if you don’t have the time or the
knowledge to find the answer, you can still use logic and common sense to
eliminate some answer choices and guess among what remains. And, of course,
the more wrong choices you’re able to eliminate, the better your chances of
guessing correctly.

Logical Elimination

Sometimes you can eliminate answer choices by logic; you might discover that
the correct answer must lie in a certain range, or it must be even, or it must be
negative, etc. Use any information like this to eliminate choices which don’t fit
the bill and guess among what remains.

Obvious answer choices

On the hardest questions the obvious answer choice is almost never correct. This
simple statement raises two important questions: 1. Which are the hardest
questions? And 2. What is an obvious answer choice? First, the hardest questions
are the ones that the testmakers expect the vast majority of people to get wrong.
Specifically, the last few Quantitative Comparisons and the last couple of multiple
choice questions. Second, the obvious answer choice is the one that leaps out at
you as correct-the answer choice that seems immediately to be the answer, with
virtually no use of math at all. So why is that choice generally wrong on the
hardest questions? Because if it were correct, then most people would pick it (it’s
obvious, right?) and so most people would get the question right. But the
testrnaker knows ahead of time most people will get that question wrong (since
it’s a hard one), and hence the obvious answer choice is not correct.

186

QUANTITATIVE COMPARISON STRATEGIES

Problem solving questions are pretty straightforward and self-explanatory — you’re given
a question and five answer choices. Quantitative comparisons, on the other hand, can
seem perplexing at first. It’s not immediately apparent what you’re supposed to do —
you’re given two columns, with numbers or variables in the columns.

A Quantitative Comparison question presents two numbers, expressions, variables,
whatever, in two columns: one labeled Column A and one labeled Column B. Your job
is to determine which column has the greater amount. QC’s are an example of what is
called a “fixed-format” question; the answer choices are the same for all QC’s. You
should answer:

A. if the quantity in Column A is greater

B. if the quantity in Column B is greater

C. if the two quantities are equal;

D. if the relationship cannot be determined from the information given.

QC’s are meant to be done about twice as fast as problem solving questions; this is
because you do not actually have to solve anything for most QC’s, you only need to make
a comparison.

DO I HAVE TO FIND ACTUAL VALUES?

NO. What’s very important to understand about these questions is that being asked to
compare amounts doesn’t mean you have to know exactly what the amounts are. You
only have to know their relative values. This has some important implications, the most
important being that

QC’s reguire insight

It is crucial that you understand this for a number of reasons. First, as we said, it means
that QC questions should take less time to answer than a regular problem solving
question. You should not spend more than about thirty seconds on a QC problem on your
first run through the section. Second, your objective on these questions is to find a way
to make a comparison that doesn’t require a lot of work. With time’s winged chariot
galloping at the heels of even the best student, it is unlikely that anybody will see the best
method for every problem immediately. If you dOn’t see a fast method immediately, and
it seems like the problem will take forever, skip it, and come back to it later.

Avoid computation: look for the s_hortcut
As we just said, QC questions rely on insight, not on computation or brute force. They
have to be designed that way, if they expect you to do them in about 30 seconds each.

Many QC questions will look at first as if you have to multiply, divide, whatever; but if
they look that way to you, then you are missing something. They’re not interested in

187

 

seeing whether you know how to multiply two numbers, they want to see if you
understand the basic principles of arithmetic.

Ex. 5.
Column A Column B

24x56x39 12x112x35

A lot of people’s first thoughts with this problem would be to multiply the numbers out.
If you did that here, though, you probably would not have that much time left for the rest
of the section — it could take a couple of minutes to multiply these numbers out. You
have to use your time efliciently on the GRE. It is a timed test, and your goal is to answer
as many questions as you can (correctly, of course) in the time allotted. The clever way
to do this problem, and the way you are supposed to do it is to realize that you can factor
the numbers in Column A, and rewrite them this way.

24x56><39 = 12x2x56x39 = 12x112x39

Now we’ve got two of the same factors in both columns. We can cancel then from each,
and just compare 35 to 39. Since 39 is bigger, the answer is A.

Do the same thing to both sides

Usually, if you do the same thing to both sides, you won’t change the answer, but you can
make you job much easier. Always remember: you’re not interested in the actual values
of the columns, only in the relative values — how they compare to each other. We used
this principle in our last example: we cancelled 12 x 112 fi'om both columns, and just
compared what was left. Look at this example:

Ex. 6.
Column A Column B
_6_ 2.7.
13 65

We can’t make the comparison easily this way, so let’s make the fractions look more
alike. Multiply the numerator and the denominator in column A by 5, and we end up

with :6—(5). Since this is larger than column B, A must be the answer.

Many QC questions will have more than one term in a column. Our advice here is

Compare one piece at a time

188

 

Ex. 11.
Column A Column B

mam/53 6+7

What you can’t do here is just add the numbers under the square roots in column A, and
decide that x/ZI- + J59 = 4100. Math isn’t quite so easy. But you can compare each
piece in column A to each piece in column B. How does m compare to 6? Well, 36 is

62, so 41 must be bigger than 6. Similarly, since 49 = 72, J5 must be bigger than 7.
Since we’re adding two larger terms in column A, the answer is A.

Make the columns look similar

What we mean by this somewhat ineffable statement is to look at the form of the columns
— sometimes that can give you an idea of how to proceed. If one column has a square
root, and the other doesn’t, try squaring both columns to eliminate the square root. If one
is a percent, and the other is a fraction, try converting.

Sometimes it can be very difficult (or time-consuming) to compare the columns directly,
but each can be compared to a third number. That’s our next tip:

Compare to a thircmlue (such as O or 1)

It can be very easy on QC questions to make foolish assumptions, for instance, assuming
that a variable represents an integer, or assuming that when they say “number,” they
don’t really mean zero. That’s our next tip:

Never make assumptions

Column A Column B
John’s weight Bob’s weight

Ex. 14. John is taller than Bob.

Very often, one of the first few questions on a GRE QC section will look like this. Here,
they obviously hope that you will think “Well, if John is taller, then he must weigh
more.” That’s a foolish assumption. There’s no absolute correlation between height and
weight. Fortunately, problems like this can be relatively easy to spot.

You usually can’t find the values of the columns in algebra questions, but this doesn’t

necessarily mean that you can’t make a comparison. One tip that can work very well
with algebra QC’s is

189

 

Try picking numbers

Remember that what we’re interested in is the relative values of the columns. Picking
values for variables can give us some idea of that. Picking numbers will not immediately
tell you what the answer is: after all, if you pick a number and find that column A is

bi gger, that doesn’t mean that column A will always be bigger. For that reason, it’s
usually best to pick a couple of different numbers. Of course if you get different
relationships depending on what you pick, you know that D must be the answer. When
you pick numbers, however, you must keep in mind what limitations have been given for
the variables. Must they be negative? Positive? Can one of them be zero? If you do
pick values, try unusual values: negative numbers, fractions, negative fractions, zero.
Often, especially with the later questions, one relationship will be obvious, and you
should try to think of a number that will change the relationship, to prove that the answer
is choice D. '

 

If a geometry problem doesn’t have a diagram. draw one

This tip applies everywhere on the GRE math sections, not just OC’s. It can be
expecially helpful on QC’s, however. Now we don’t mean that you should carry a ruler
and compass into the test with you — you’re not allowed to anyway. All we mean is that
drawing a quick sketch can make an otherwise esoteric question concrete.

Tgy setting the columns egual

Many QC’s will involve comparing a variable, or something vague, to a number. At
times, the only concrete thing you’ll have will be one of the columns — so why not start
with that?

Ex. 17. g of the people in the room are men. There are 68 men in the room.

Column A Column B
Number of people in the 105
room

Working backwards from the numbers in the centered information can be time-
consuming. What is much easier here is go start with the value in column B. Suppose
there were 105 people in the room. Then how many men would there be? If there were

105 people, and g of them were men, then gx105 or 60 people are men. Since this is

less than the actual number of men, we know that the total number of people must be
greater than 105. Therefore, column A is bigger, the answer is A.

GUESSING ON QC QUESTIONS

190

 

As we’ve said before, even if you have no idea what the answer to a GRE question is,
guess something.

0 It can’t hurt, and it might help. The good thing about guessing on QC questions is
that it is so easy.

0 You can almost always eliminate at least one choice on a QC. For instance, look at
the following:

Ex. 18. x=2

A. Column A B. Column B

x56 x55 + x55

Even if you have no idea of what to do here, you know that the answer can ’t be D. After
all, we have a definite value for x; we could always figure out the values of the columns;
raising 2 to the 55th power. We’re not suggesting that you should do this, or even that
you should consider multiplying 255 out, but you should realize that it is possible to make
the comparison. Therefore, D can’t be the answer, and you can guess among A, B, and
C. The answer to this question, by the way, is C; we can rewrite column B as 2(255) or
256, and see that it is equal to column A.

In certain cases, you will be able to eliminate D from consideration on a QC, but what is

more common are questions where you can eliminate everything but D and another
choice. This can be one of your most powerful strategies on this section.

191

 

APPENDIX K
Debriefing Form
Thank you for participating in the GRE Experiment!

If you are an incentive winner, you will be notified by e-mail. If your e—mail address
changes before the end of the semester, please contact one of the investigators below.

If you have any concerns or questions about your participation in this project, or would
like a copy of the results, you can contact Cori Davis at 517-3 55-2171
(daviscor@msu.edu) or Dan Ilgen at 517-355-7503 (ilgen@msu.edu).

If you have any questions or concerns regarding your rights and involvement in this
research you may contact:

The University Committee on Research Involving Human Subjects (U CRIHS)
246 Administration Building, Michigan State University

East Lansing, MI 48824-1046

Phone: (517)355-2180 Fax: (517)353-2976

Email: UCRIHS@msu.edu Web Site: http/www.msu.edu/unit/vprgs/ucrihs

192