<%BANNER%>

The Effects of Theories of Intelligence and Task Outcome on Avoidance of Performance Feedback

Permanent Link: http://ufdc.ufl.edu/UFE0044473/00001

Material Information

Title: The Effects of Theories of Intelligence and Task Outcome on Avoidance of Performance Feedback
Physical Description: 1 online resource (90 p.)
Language: english
Creator: Novell, Corinne
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: academic -- avoidance -- failure -- feedback -- information -- performance -- toi
Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Three studies examined whether incremental and entity theories of intelligence differentially prompt avoidance of performance feedback. Studies 1 and 2a supported the hypothesis that an entity theory of intelligence prompts greater avoidance of performance feedback, and Study 2a showed that this relationship occurs primarily when entity theorists experience failure. Mediation analyses revealed that participants tended to avoid feedback when they felt less able to cope with the feedback and when they perceived the feedback as less useful. The pattern of feedback avoidance maps onto Dweck’s model of Mastery-oriented vs. Helpless Response patterns to failure and supports the contention that feedback avoidance may be a previously unexamined part of the Helpless Response pattern that entity theorists exhibit when they fail. The findings also suggest an alternative explanation for the achievement gap between incremental and entity theorists in school.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Corinne Novell.
Thesis: Thesis (Ph.D.)--University of Florida, 2012.
Local: Adviser: Shepperd, James A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044473:00001

Permanent Link: http://ufdc.ufl.edu/UFE0044473/00001

Material Information

Title: The Effects of Theories of Intelligence and Task Outcome on Avoidance of Performance Feedback
Physical Description: 1 online resource (90 p.)
Language: english
Creator: Novell, Corinne
Publisher: University of Florida
Place of Publication: Gainesville, Fla.
Publication Date: 2012

Subjects

Subjects / Keywords: academic -- avoidance -- failure -- feedback -- information -- performance -- toi
Psychology -- Dissertations, Academic -- UF
Genre: Psychology thesis, Ph.D.
bibliography   ( marcgt )
theses   ( marcgt )
government publication (state, provincial, terriorial, dependent)   ( marcgt )
born-digital   ( sobekcm )
Electronic Thesis or Dissertation

Notes

Abstract: Three studies examined whether incremental and entity theories of intelligence differentially prompt avoidance of performance feedback. Studies 1 and 2a supported the hypothesis that an entity theory of intelligence prompts greater avoidance of performance feedback, and Study 2a showed that this relationship occurs primarily when entity theorists experience failure. Mediation analyses revealed that participants tended to avoid feedback when they felt less able to cope with the feedback and when they perceived the feedback as less useful. The pattern of feedback avoidance maps onto Dweck’s model of Mastery-oriented vs. Helpless Response patterns to failure and supports the contention that feedback avoidance may be a previously unexamined part of the Helpless Response pattern that entity theorists exhibit when they fail. The findings also suggest an alternative explanation for the achievement gap between incremental and entity theorists in school.
General Note: In the series University of Florida Digital Collections.
General Note: Includes vita.
Bibliography: Includes bibliographical references.
Source of Description: Description based on online resource; title from PDF title page.
Source of Description: This bibliographic record is available under the Creative Commons CC0 public domain dedication. The University of Florida Libraries, as creator of this bibliographic record, has waived all rights to it worldwide under copyright law, including all related and neighboring rights, to the extent allowed by law.
Statement of Responsibility: by Corinne Novell.
Thesis: Thesis (Ph.D.)--University of Florida, 2012.
Local: Adviser: Shepperd, James A.

Record Information

Source Institution: UFRGP
Rights Management: Applicable rights reserved.
Classification: lcc - LD1780 2012
System ID: UFE0044473:00001


This item has the following downloads:


Full Text

PAGE 1

1 THE EFFECTS OF THEORIES OF INTELLIGENCE AND TASK OUTCOME ON AVOIDANCE OF PERFORMANCE FEEDBACK By CORINNE A NOVELL A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE R EQUIREMENTS FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2012

PAGE 2

2 2012 Corinne A Novell

PAGE 3

3 To my Mom Christine and my twin sist er Jen the mo st amazing women I know for the support they have given me through out my life journey s

PAGE 4

4 ACKNOWLEDGEMENTS I would like to thank the members of my committee for their time and efforts toward this project. I would especially like to thank my advisor Dr. James Shepperd for his guidance through my graduate years. Finally, I would like to acknow their assistance with the formatting of my dissertation

PAGE 5

5 TABLE OF CONTENTS page ACKNOWLEDGEMENTS ................................ ................................ ................................ ............. 4 LIST OF TABLES ................................ ................................ ................................ ........................... 8 LIST OF FIGURES ................................ ................................ ................................ ......................... 9 LIST OF ABBREVIATIONS ................................ ................................ ................................ ........ 10 ABSTRACT ................................ ................................ ................................ ................................ ... 11 CHAPTER 1 INTRODUCTION ................................ ................................ ................................ .................. 12 Entity vs. Incremental Theories of Intelligence ................................ ................................ ...... 12 Attributions for Success and Failure ................................ ................................ ............... 13 Approachi ng Tasks with Performance vs. Learning Goals ................................ ............. 14 Helpless vs. Mastery Oriented Response Patterns to Failure ................................ .......... 16 Does the Helpless Response Include Information Avoidance? ................................ ....... 20 Information Avoidance ................................ ................................ ................................ ........... 20 Theories of Intelligence and Information Avoidance ................................ ............................. 27 Overview of the Current Research ................................ ................................ ......................... 29 2 STUDY 1: ASSOCIATIONS BETWEEN THEORIES OF INTELLIGENCE, GOAL PREFERENCES, AND ACADEMIC INFORMATION AVOIDANCE .............................. 31 Method ................................ ................................ ................................ ................................ .... 31 Participants ................................ ................................ ................................ ...................... 31 Materials ................................ ................................ ................................ .......................... 31 Procedure ................................ ................................ ................................ ......................... 32 Results ................................ ................................ ................................ ................................ ..... 32 Theories of Intellige nce and the Inclination to Avoid Performance Feedback ............... 33 Theories of Intelligence and Goal Preferences ................................ ................................ 33 Goal Preferences and the Inclination to Avoid Performance Feedback .......................... 35 3 STUDY 2A: INDUCED THEORIES OF INTELLIGENCE, TASK OUTCOME, AND FEEDBACK AVOIDANCE ................................ ................................ ................................ .. 37 Method ................................ ................................ ................................ ................................ .... 38 Participants and Design ................................ ................................ ................................ ... 38 Materials and Procedure ................................ ................................ ................................ .. 38 Results ................................ ................................ ................................ ................................ ..... 42 Manipulation Check ................................ ................................ ................................ ........ 42 Preliminary Analyses ................................ ................................ ................................ ....... 42

PAGE 6

6 Avoidance of Performance Feedback ................................ ................................ .............. 44 Hypothesis 1 (Effec ts of ToI) ................................ ................................ .......................... 46 Hypothesis 2 (Effects of Task Outcome) ................................ ................................ ........ 47 Hypothesis 3 (ToI x Task Outcome Interactions) ................................ ........................... 48 Hypothesis 4 (Mediation Analyses) ................................ ................................ ................ 53 Hypothesis 5 (Associations among ToI, Goal Preferences, and the Inclination to Avoid Performance Feedback ) ................................ ................................ .................... 56 Theories of intelligence and inclination to avoid performance feedback ................ 56 Theories of intelligence and goal preferences ................................ .......................... 56 Goal preferences and inclination to avoid performance feedback ........................... 57 4 STUDY 2B: MEASURED (IMPLICIT) THEORIES OF INTELLIGENCE AND RESPONSES TO FEEDBACK ................................ ................................ .............................. 59 Method ................................ ................................ ................................ ................................ .... 59 Participants and Design ................................ ................................ ................................ ... 59 Materials and Procedure ................................ ................................ ................................ .. 59 Results ................................ ................................ ................................ ................................ ..... 60 Preliminary Analyses ................................ ................................ ................................ ....... 60 Avoidance of Performance Feedback ................................ ................................ .............. 62 Hypothesis 1 (Effects of ToI) ................................ ................................ .......................... 63 Hypothesis 2 (Effects of Task Outcome) ................................ ................................ ........ 63 Hypothesis 3 (ToI x Task Outcome Interactions) ................................ ........................... 65 Profiles of Seekers vs. Avoiders ................................ ................................ ...................... 65 Hypothesis 5 (Associations between ToI, Goal Preferences, and the Informa tion Avoidance Scale) ................................ ................................ ................................ ......... 65 Theories of intelligence and inclination to avoid performance feedback ................ 65 Theories of intelligence and goal preferences ................................ .......................... 65 Goal preferences and inclination to avoid performance feedback ........................... 66 5 DISCUSSION ................................ ................................ ................................ ......................... 67 General Discussion ................................ ................................ ................................ ................. 67 Implications ................................ ................................ ................................ ............................ 69 Alternative Explanation and Limitations ................................ ................................ ................ 71 Future Directions ................................ ................................ ................................ .................... 74 Conclusion ................................ ................................ ................................ .............................. 74 APPENDIX A THEORIES OF INTELLIGENCE ADULT SCALE ................................ ............................. 76 B INFORMATION AVOIDANCE SCALE ................................ ................................ .............. 77 C GOAL PREFERENCES SCALE ................................ ................................ ........................... 78 D CONSENT FORM ................................ ................................ ................................ .................. 79

PAGE 7

7 E ADDITIONAL QUESTIONNAIRE ITEMS ................................ ................................ ......... 80 F DEBRIEFING FORM ................................ ................................ ................................ ............ 84 LIST OF REFERENCES ................................ ................................ ................................ ............... 86 BIOGRAPHICAL SKETCH ................................ ................................ ................................ ......... 90

PAGE 8

8 L IST OF TABLES Table page 1 1 ................................ ................................ ............................. 17 1 2 Characteristics of the two response patterns ................................ ................................ ...... 17 2 1 Correlations among variables ................................ ................................ ............................ 33 3 1 Zero order correlations among variables ................................ ................................ ........... 43 3 2 Means for outcome measures ................................ ................................ ............................. 44 3 3 Hypothesis 1 results: main effects of ToI ................................ ................................ .......... 46 3 4 Hypothesis 2 results: main effects of task outcome ................................ ........................... 48 3 5 Hypothesis 3 results: tests of interactions ................................ ................................ .......... 49 4 1 Zero order correlations among variables ................................ ................................ ........... 61 4 2 Means for outcome measures ................................ ................................ ............................. 62 4 3 Hypothesis 2 results: effects of task outcome ................................ ................................ .... 64 4 4 Hypothesis 3 results: ToI x feedback interactions ................................ ............................. 64

PAGE 9

9 LIST OF FIGURES Figure page 2 1 Percentages of incremental and entity theorists preferri ng learning vs. performance goals.. ................................ ................................ ................................ ................................ 34 3 1 theories of intelligence and task outcome. ................................ ................................ ......... 45 3 2 Anticipated negative affect values as a function of theories of intelligence and task outcome.. ................................ ................................ ................................ ............................ 51 3 3 Anticipated difficulty coping with the feedbac k as a function of theories of intelligence and task outcome.. ................................ ................................ .......................... 51 3 4 Perceived threat from the feedback as a function of theories of intelligence and task outcome.. ................................ ................................ ................................ ............................ 52 3 5 Perceived utility of the feedback as a function of theories of intelligence and task outcome.. ................................ ................................ ................................ ............................ 52 3 6 Partial correlations between feedback condition and performance feedback avoidance as mediated by anticipated difficulty coping.. ................................ .................. 54 3 7 Partial correlations between task outcome and performance feedback avoidance as mediated by perceived ut ility of the feedback. ................................ ................................ .. 55 3 8 Partial correlations between task outcome and performance feedback avoidanc e: multple mediation by coping and perceived utility. ................................ ........................... 55 3 9 Percentages of incremental and entity theorists preferring learning vs. performance goals. ................................ ................................ ................................ ................................ .. 57

PAGE 10

10 LIST OF ABBREVIATION S IA Information avoidance the deliberate delay or avoidance of potenti ally unwanted information. TOI Theories of Intelligence refer to different beliefs about the nature of intelligence.

PAGE 11

11 Abstract of Dissertation Presented to the Graduate School of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy THE EFFECTS OF THEORIES OF INTELLIGENCE AND TASK OUTCOME ON AVOIDANCE OF PERFORMANCE FEEDBACK By Corinne A. Novell August 2012 Chair: James A. Shepperd Major: Psychology Three studies examined wheth er incremental and entity theories of intelligence differentially prompt avoidance of performance feedback. Studies 1 and 2a supported the hypothesis that an entity theory of intelligence prompts greater avoidance of performance feedback, and Study 2a show ed that this relationship occurs primarily when entity theorists experience failure. Mediation analyses revealed that participants tended to avoid feedback when they felt less able to cope with the feedback and when they perceived the feedback as less usef ul. oriented vs. Helpless Response patterns to failure and supports the contention that feedback avoidance may be a previously unexamined part of the Helpless Response pattern that entity theorists exhibit when they fail. The findings also suggest an alternative explanation for the achievement gap between incremental and entity theorists in school.

PAGE 12

12 CHAPTER 1 INTRODUCTION Entity vs. Incremental Theories of Intelligence Although most peop le would agree that intelligence is a good thing to have, they often disagree on how 1988). People who hold an entity or fixed ToI believe that intelligence is a trait that a person is born with. Accordingly, they view intelligence as relatively uncontrollable and unchangeable and incremental or malleable ToI believe that intelligence is something that a person acquires in life and reflects the amount of effo rt invested. Accordingly, they assume that people can increase Interestingly, when it comes to performance or scores on in telligence tests, people with different theories tend not to differ in cognitive ability or in confidence in their abilities they simply hold different beliefs about why they scored how they did (Dweck, 2000; Dweck, Hong, & Chiu, 1995; Levy et al., 1998). Entity theorists believe their score reflects their innate ability, for better or for worse. Incremental theorists believe their score reflects the effort they put into the task. It is also important to note that although entity and incremental theorists differ in their implicit theories are domain specific and do not extend to perceptions of changeability of other attributes, such as personality (Dweck, 2000). In addition, the domain specificity of implicit theories makes theories of intelligence distinct from more general constructs such as locus of

PAGE 13

13 control (Rotter, 1954). Despite this level of specificity, it remains unclear whether theories of intelligence may further be subject specific such that a person may have difference beliefs regarding the fixedness of math, language abilities, etc. Entity and incremental ToI differ in several important, consequential ways. The beliefs underlying the theories create dif different attributions for success and failure, adoption of different goals when approaching tasks, and different response patterns to failure (Dweck, 2000; Dweck & Leggett, 1998). Although perhaps already apparent, it is important to note that the primary interest of this research is situations in which people fail rather than succeed. Attributions for Success and Failure The contrasting attributions made by entity and incremental theorists lead them to form different interpretations and draw different inferences about failure and, consequently, to assume different implications for self evaluation following a poor performance (Dweck, 1999). Because entity theorists believe that intelligence is a f ixed trait, they tend to draw dispositional inferences from their performance on intelligence tests. That is, entity theorists believe that performances measure and are diagnostic of the amount of intelligence they permanently possess. Entity theorists int erpret failure or even the need to put forth effort (i.e. in a challenging or difficult situation) as evidence that they lack natural ability. Thus failure reveals the limit of their intelligence. The implications for self evaluation are serious: Failure i s a direct threat to their belief that they are intelligent and to their self esteem (Dweck & Bempechat, 1983). In fact, worth is contingent upon success and failure (Dweck, 2000; Dykman, 1998). Incremental theorists, on the other ha nd, are less quick to fuse an intelligence test outcome success or failure to the self. Rather, incremental theorists attribute a performance to the adequacy of effort or the appropriateness of the strategy with which they approached a task.

PAGE 14

14 According to a study by Hong, Chiu, Dweck, and Lin (as cited in Dweck, 2000), i ncremental theorists interpret failure as temporary and the investment of effort as a natural part of the learning process and, iron ically, part of getting smarter And, as suggested by a stu dy by Mueller and Dweck (as cited in Dweck, 2000), t he implications of failure on self evaluation are not as serious for incremental theorists because performance (and failure) is not directly tied to the self, and their self worth is not contingent upon p er formance The attributions made by entity and incremental theorists manifest in their responses when they encounter difficulty. A study that examined the effects of academic difficulty and failure revealed different self relevant thoughts among 5 th gra ders who held either entity or incremental theories (Diener & Dweck, 1978). Whereas over 50% of entity theorists derogated their abilities following failure, attributing their failure to disposition (i.e., not being smart), not a single incremental theoris t attributed their failure to disposition. Instead, incremental theorists attributed failure to external factors such as the difficulty of the task. In fact, incremental theorists responded to harder, more challenging problems positively, stating things su Other research by Zhao, Dweck, and Mueller (as cited in Dweck, 2000) f inds that simply imagining failure led to denigration of abilities among entity but not incremental theorist s Approachi ng Tasks with Performance vs. Learning Goals approach tasks and influence the types of tasks they prefer to engage in. Entity theorists believe that every occasion f or evaluation has potential to unravel their positive self beliefs. Thus, they approach performance tasks with two motives: They are motivated to show others and themselves that they are smart and at the same time are motivated to conceal demonstrations of failure; that is, they want to look competent at all costs (Dweck & Elliott, 1983; Dweck &

PAGE 15

15 Elliott Moskwa, 2010; Hong, Dweck, Lin, & Wan, 1999). A study by Bandura and Dweck (as cited in Dweck, 2000), found that t hese two motives lead entity theorists to prefer easy tasks over difficult ones because easy tasks are more likely to yield a positive judgment .Yet, entity theorists 2000, p. 3), entity theorists also show motivation toward positive evaluation and may weigh the ir odds of positive evaluation when they choose one task over another. Incremental theorists, on the other hand, do not view such occasions with as high stakes as do the entity theorists; that is, a single performance is not going to shake them. As such, incremental theorists approach tasks with learning goals: They strive to develop abilities, even if doing so necessitates encountering some failures along the wa y. Incremental theorists tend to prefer challenging, difficult tasks over easy tasks because challenges provide an opportunity for growth whereas easy tasks are relatively unhelpful for growth. According to a study by Mueller and Dweck (as cited by Dweck, 2000), e respective beliefs that intelligence is fixed vs. m alleable It is important to note that entity theorists do not exclusivel y value performance goals and incremental theorists do not exclusively value learning goals each type of theorist generally values both learning and documenting capability. However, many situations in life demand a choice between one goal and the other. Th is choice could arise, for example, when student s register for classes. Indeed, the Mueller and Dweck study (as cited in Dweck, 2000) that asked college students to choose between getting a good grade and being challenged in a class found that entity theor ists tended to choose getting a good grade whereas incremental theorists tended to choose bei ng challenged In such cases,

PAGE 16

16 entity theorists and incremental theorists reveal their goal preferences by which one they are willing to sacrifice. Several studies demonstrate the relationship between theories of intelligence and specific goals. For instance, one study examined the effect of implicit theories of intelligence on task choices in school. Researchers presented 8 th graders with three options for an acade mic activity that they would soon undertake. The options included two that reflected a performance goal and a third that reflected a learning goal. The task descriptions for the performance goal choices were 1) a very easy task in which people were unlikel y to err, 2) a medium difficult task that was difficult enough that it made the task taker look good, and the description for the third, learning goal choice was 3) a difficult task in which mistakes were inevitable but task takers would learn useful thing s. Whereas only 20% of the entity theorists chose the learning goal, more than 60% of incremental theorists chose it. Conversely, more than half of the entity theorists opted for the very easy task that represented a performance goal. A second study replic ated this effect of goals on task choice, but after manipulating theories of intelligence (Dweck & Leggett, 1988). Participants who read a passage that promoted an entity theory tended to choose one of the tasks that reflected a performance goal, whereas p articipants who read a passage that promoted an incremental theory tended to choose the task that reflected a learning goal. This second study supports the idea that the relationship between theories of intelligence and their respective goals are not just correlational, but causal. Helpless vs. Mastery Oriented Response Patterns to Failure Both theories of intelligence and preferred goals prompt distinct affective, emotional, and behavioral response patterns to failure and setbacks (Diener & Dweck, 1978, 1 980; Dweck, 1975; Elliott & Dweck, 1988). Specifically, entity theorists tend to respond to failure or a prospect thereof in a Helpless Response pattern that is marked by negativity, defensiveness, and

PAGE 17

17 avoidance. In contrast, incremental theorists respond to failure or the prospect thereof in an adaptive, Mastery oriented Response pattern marked by persistence (Dweck, 2000). Table 1 1 summarizes the relationships between ToI, goal preferences, and the response patterns to success and failure as adapted fr om Dweck & Leggett (1988) and Baird, Scott, Dearing, & Hamill (2009). Table 1 2 summarizes the affective, cognitive, and behavioral characteristics of the Mastery oriented and Helpless Response patterns. As is evident, the response patterns of entity and i ncremental theorists differ only when they perceive low ability ; that is, when they experience or anticipate failure. Table 1 1. odel Theory of Intelligence Goal Preference Perceived Ability Response Pattern Entity Performance High Mastery oriented Low Helpless Incremental Learning High Mastery oriented Low Mastery oriented Table 1 2. Characteristics of the two response patterns Helpless Response Mastery oriented Response Cognitions Loss of belief in efficacy of effort Continued belief in efficacy of effort Meaning of Effort Effort confirms low ability Effort level is consistent with task requirements Attention Divided between goal and task Undivided attention on serve goal Affect Negative affect interferes with task and can prompt withdrawal Affect channeled into task Challenge Seeking Avoid challenging tasks Seek out challenge Performance after obstacles Deterioration of performance Persistence after failure Several studies document significant differences betwee n entity and incremental theorists in their affect following failure (Diener & Dweck, 1978; Henderson & Dweck, 1990; Robins & Pals, unpublished, as cited in Dweck, 2000 ; Sorich & Dweck, 2000; Zhao et al., unpublished, as cited in Dweck, 2000 ). For example, one study demonstrated not only the well established

PAGE 18

18 finding that entity theorists exhibit more negative affect than do incremental theorists when they fail, but also found that failure made non depressed entity theorists (but not non depressed incrementa l theorists) lapse into a depressive state similar to that of a different sample of currently depressed participants (as measured by the Beck Depression Inventory) (Beck, Ward, Mendelson, Mock, & Erbaugh, 1961; Zhao et al., unpublished, as cited in Dweck, 2000 ). Other research that examined predictors of dysphoria following failure found that pursuit of a performance like validation goal (i.e. entity theorists) was positively related to anxiety, self esteem loss, and other negative emotions, whereas pursuit of a learning like growth goal was negatively related to these same measures (Dykman, 1998). Helpless and Mastery oriented Response patterns produce different cognitions following failure (Diener & Dweck, 1978). A study by Zhao, Dweck, and Mueller (as ci ted in Dweck, 2000) found that n ot only do entity theorists make negative self evaluations after experiencing failure but they also have thoughts of quitting or escaping from the domai n of failure ose of continued self belief and a commitment to try harder the next time. That is, they do not see failure as dead end; rather, it is detour along the journey to success. Still other evidence reveals that entity and incremental theorists differ in the way s that they react to failure. Whereas entity theorists respond to negative feedback in defensive, avoidant ways, incremental theorists respond with positive confrontation (Dweck & Elliott Moskwa, 2010). For instance, a study by Hong, Chiu, Dweck, and Lin (as cited in Dweck, 2000) conducted at the University of Hong Kong where classes are conducted in English found that international students who learned that they scored low on English proficiency were less interest in taking a remedial English course if th ey were entity theorists than if they were increment al

PAGE 19

19 theorists That is, when presented with a real world opportunity that could help them but also initially expose deficiencies, entity but not incremental theorists chose to avoid that learning opportuni ty. Instead, entity theorists opted for the choice that offered the short term benefit of Other research in the field finds similar helpless and mastery oriented patterns in response to setbacks (Henderson & Dweck, 1990). The researchers tr acked the grades of students with either dispositional entity or incremental theories as they transitioned to a more difficult grades improved. This pattern w as due, in part, to entity theorists withdrawing effort when they encountered increased challenges a form of self handicapping that is clearly maladaptive for success in school. Incremental theorists, on the other hand, increased their efforts when they en countered more difficult tasks. Indeed, self handicapping is another type of helpless, avoidant behavior that entity theorists may use to distance themselves from thoughts that they are not smart enough to conquer the task. Importantly, research suggest s that people do not even need to experience real failure to exhibit helpless or mastery oriented response patterns (Sorich & Dweck, 2000; Zhao et al., unpublished, as cited in Dweck, 2000 ). One study had participants imagine failing a school subject found that entity theorists reported a desire to avoid that particular subject in the future, presumable so as to not have to deal with more failure. Incremental theorists, on the other hand, reported a desire to exert greater effort on the subject in the futur e (Sorich & Dweck, 2000). Another study that examined responses to hypothetical failure found that incremental theorists coped well, where they planned for the future to turn failure into success. Entity theorists, in contrast, coped poorly, and wanted to remove themselves from the situation in which they

PAGE 20

20 imagined failure (Dweck, 2000; Zhao et al., unpublished, as cited in Dweck, 2000 ). Collectively these findings suggest that, whereas entity theorists exhibit an avoidant, Helpless Response pattern to (the prospect of) failure, incremental theorists exhibit a proactive, Mastery oriented Response pattern. Does the Helpless Response Include Information Avoidance? Despite the array of avoidant, helpless responses that the ToI literature documents, an important type of avoidant behavior remains unexamined. I suspect that entity and incremental theorists may choose differently if presented a dichotomous choice between learning vs. not learning feedback about how they performed on an important intellectual task. P resumably choosing to learn performance feedback would represent a mastery oriented response because it involves learning and improving on an academic outcome even if failure is initially likely. In contrast, avoiding performance feedback would represent a helpless response because it involves withdrawing from an opportunity to learn and improve on an academic outcome. Table 1 1 suggests that valence of the outcome is of paramount importance in predicting a Helpless vs. Mastery orientation Response pattern. Thus, according to Table 1 1 avoidance of performance feedback should occur only when participants perceive low ability (i.e. they experience failure). That is, when faced with a poor task performance, entity theorists may be more inclined to sacrifice l earning opportunities (i.e. avoid performance feedback) in their effort to avoid negative feedback (Elliott & Dweck, 1988; Hong et al., unpublished, as cited in Dweck, 2000 ). Information Avoidance Support for this general hypothesis comes from research on information avoidance. unwanted information (Sweeny, Melnyk, M iller & Shepperd, 2010). The beliefs, goals, and response patterns of entity theorists and i ncremental theorists differ in several ways and these

PAGE 21

21 differences suggest several reasons why entity theorists would be more likely than incremental theorists to avoid information (i.e. performance feedback) when they experience failure. First, entity the orists should be more likely than incremental theorists to avoid information because failure feedback would be more threatening to their self view. In general, people are motivated to maintain desired self views (Cai, Wu, & Brown, 2009; de la Ronde & Swann 1993; Kunda, 1990; Swann, de la Ronde, & Hixon, 1994). One way to maintain a desired self view is to avoid the information that might potentially threaten that self view (Sweeny et al., 2010). Research on selective exposure and motivated searches finds t hat people prefer information that supports and confirms their self views over information that challenges or disconfirms their self views ( Hart, Albarracin, Eagly, Brechan, Lindberg, & Merrill, 2009; Smith, Fabrigar, & Norris, 2008). As one might imagine, information that poses a high (serious) threat to important self views is likely to prompt greater avoidance than is information that poses a low threat. Indeed, people are motivated to avoid negative self views (Baumeister, Bratslavsky, Finkenauer, & Voh s, 2001; Hoorens, 1996). One study that manipulated severity of threat found that people were more willing to learn how raters rated their attractiveness if these raters attended a different university than their own (low threat) than if these raters atten ded their own university (high threat) 1 Because entity theorists make more dispositional inferences about the self from failure than do incremental theorists, failure should pose a greater threat to self views for entity theorists. As such, entity theor ists should be more likely than incremental theorists to avoid feedback when they experience failure. 1 Malone, W., Shepperd, J. A., & Sweeny, K. (2009). Unpublished manuscript, University of Florida, Gainesv ille.

PAGE 22

22 Second, entity theorists should be more likely than incremental theorists to avoid information because they view performance on intelligence tests as unco ntrollable. People tend to display greater information avoidance about outcomes that they suspect are unfavorable when they perceive the outcome as uncontrollable or unchangeable rather than controllable or cha ngeable (Melnyk & Shepperd, 2012 ; Yaniv, Benad or, & Sagi, 2004). Support for this idea comes from various domains of research. In the medical domain, a set of studies that surveyed onset disease found that people more often wanted know their di agnostic genetic standing when the disease was described as treatable than when it was described as untreatable. Indeed, the most common reason cited for avoiding learning genetic standing was the untreatable nature of the disease (Yaniv, Benador, & Sagi, 2004). Another study found that women were more likely to avoid learning their risk for breast cancer when the predictors of breast cancer were described as uncontrollable rather than contr ollable (Melnyk & Shepperd, 2012 ). Evidence that people are more l ikely to avoid information that they suspect yields unfavorable results (e.g. failure) if it is uncontrollable rather than controllable also comes from the social domain (Trope, Gervey, & Bolger, 2003). Specifically, several studies show that perceived cha positive and negative feedback about those abilities. Amidst a three way interaction between feedback valence (positive vs. negative), perceived controllability of social abil ity (controllable vs. uncontrollable), and perceived ability (low vs. high), the only participants who preferred learning positive over negative feedback were those participants who perceived that they had low social ability and that this ability was uncon trollable; that is, the combination of low ability and an uncontrollable outcome led to defensiveness of unfavorable feedback. However, when people

PAGE 23

23 perceived that they had low social ability but also perceived this ability as controllable, or when people p erceived their ability to be high, regardless of controllability of the outcome, they seemed to be less defensive and tended to prefer negative feedback over positive. Because entity theorists perceive intelligence as uncontrollable whereas incremental t heorists perceive intelligence as controllable, only entity theorists should tend to avoid information, and only when they perceive low ability (i.e. experience failure). The predicted discrepancies in information avoidance between entity and incremental theorists may depend on their expectations for positive rather than negative feedback. Research on information avoidance cites mixed results about whether people are more likely to seek or avoid information when they anticipate bad news. On the one hand, p eople may be more likely to seek information only to the extent that this information is positive or favorable. In the studies by Trope et al. (2003), perceptions of ability (low vs. high) about which feedback was available arning positive and negative feedback about this ability. Specifically, people who expected less favorable results (i.e. people who perceived their ability to be low), were less interested in receiving negative feedback than were people who expected more f avorable results (i.e. those who perceived their ability to be high). This finding should not be too surprising: important theories that propose a motive to seek positive rather than negative information include motivated searches in memory (Kunda, 1990) a nd the Theory of Motivated Information Management (Afifi et al., 2006). Additional support that people avoid information decisions to seek or avoid diagnostic testing for type II diabetes. Participants who received high

PAGE 24

24 preliminary risk feedback for diabetes were more likely than participants who received low preliminary risk feedback to avoid taking a free diagnostic test at the end of the study 2 On the other hand, expe cting a negative outcome may be the very reason why people seek information rather than avoid it (Brewer et al., 2007; Cappelli et al., 2001; Katapodi, Lee, Faci one, & Dodd, 2004) 3 People may be motivated to learn unfavorable information for several reaso ns: if the outcome is controllable they might want to use it to change their behavior and reduce the likelihood of a later, more serious negative outcome; if the outcome is uncontrollable they might still seek the information if they think it will help the m prepare for the future in other ways (e.g. helping family cope and get their affairs in order, etc.); or people might have low would be the lesser of the two evils. The effect of expectation on information avoidance may best be understood in concert with other factors (Sweeny et al., 2010). For instance, the effect may depend on the type of domain (e.g. health or social). For the current research, however, the relevant interaction may involve Expectation (positive vs. negative) x Perceived Controllability (low vs. high), where people exhibit avoidance when the expectation for negative feedback is high and controllability of the outcome is low (Sweeny et al., 20 10; Trope et al., 2003). As applied to theories of intelligence, because entity theorists perceive low control over the outcome, they should be more likely to avoid information, but only when they expect negative feedback (associated with a failure perform ance). 2 Novell, C. A., & Shepperd, J. A. (2012). Hints of bad news and avoidance of health information. Unpublished manuscript, University of Florida, Gainesville. 3 Miller, W. (2010). The effects of magnitude and likelihood on information avoidance (Unp ublished doctoral dissertation). University of Florida, Gainesville.

PAGE 25

25 Third, entity theorists should be more likely than incremental theorists to avoid information because entity theorists cope poorly with failure. People tend to avoid information when they report insufficient coping resources, because having suffic ient coping resources/ability helps people deal with the aftermath of learning unfavorable info rmation (Melnyk & Shepperd, 2012 ). 4 Ability to cope with a current situation (e.g. receive potentially threatening information) likely involves two factors: t he coping resources that people currently have to use and the demands of the situation itself. Theoretically, if situational demands are held constant, people with better coping ability will be less likely to avoid information. Likewise, if coping resource s are held constant, situations that demand greater coping resources (i.e. presumably due to high severity and/or low controllability of the outcome) will likely prompt greater avoidance. If people lack the necessary resources to deal with the current dema nds because of either a personal deficiency or excessive situational demands, or both, avoiding information (at least temporarily) Because entity theorists view failure as a threat to their self views and as low in controllability, the resources needed to cope with failure should greater for entity theorists than for incremental theoris ts. As such, entity theorists should be most likely to avoid information when the situational coping demands are high (i.e. experience failure). Fourth, entity theorists should be more likely than incremental theorists to avoid information because they ex perience greater negative affect when they fail. People are motivated to minimize negative experiences and emotions such as fear, embarrassment, and disappointment 4 Novell, C. A., & Shepperd, J. A. (2012). Hints of bad news and avoidance of health information. Unpublished manuscript, University of Florida, Gainesville.

PAGE 26

26 (Kirsch, 1985). Empirical findings support this motive. For example, in a study that invited gay men to learn the results of an HIV antibody test, 48% of those who declined to learn their results reported that a primary reason for avoiding the results was fear of a positive test result ( Lyter, Valdiserri, Kingsley, Amoroso, & Rinaldo 1987 ). In a ddition, research in the domain of consumer decisions shows that negative emotions while making a decision prompted avoidance. Interestingly, the decision to avoid predicted less negative emotion afterward than did the decision to seek (Luce, 1998). Togeth er these findings suggest that not only does negative emotion prompt avoidance but avoidance may be part of an emotional regulatory process to attenuate negative emotions. Because entity theorists but not incremental theorists display a host of negative e motions when they fail (or simply imagine failing), including anxiety and depression, entity theorists should be more likely to avoid information. Obviously because these negative emotions manifest only when considering failure, entity theorists should be more likely to avoid feedback only when they anticipate failure. Fifth, entity theorists should be more likely than incremental theorists to avoid information because they should anticipated greater regret to seek and less regret to avoid the information. Complex cost benefit analyses often precede many decisions, and the decision to learn or not learn information about the self is likely no exception. Cost benefit analyses may reflect anticipated regret values for selecting any of the possible choices ove r the others (Lee, 1971). Because regret is such an aversive experience, people may factor in anticipated regret when making a decision, selecting the choice likely to yield the lowest regret (Loomes, Sugden, 1982; Roese & Olson, 1995). Empirical studies s upport this possibility. For instance, one study found specific antigens (PSA) tests for past patients

PAGE 27

27 correlated highly with their present decisions to order PSAs for hypothetical patients (Sorum et al., 2004). Other research found that women were more likely to learn their risk for breast cancer when they anticipated greater regret about seeking than about avoiding the risk info rmation (Melnyk & Shepperd, 2012 ). The prospect of anticipated regret appli es to theories of intelligence. Although choosing to learn information may yield benefits for both entity and incremental theorists, choosing the learn yields differential repercussions for entity theorists than for incremental theorists. For instance, ent ity theorists risk indictment of their intelligence whereas incremental theorists do not (Dweck, 2000). Entity theorists likely risk (and may already begin feeling) extreme negative affect whereas incremental theorists likely do not. Entity theorists may c ope more poorly with failure than do incremental theorists. All of these difference and others may lead to different anticipated regret values for entity and incremental theorists facing the choice of learning or not learning feedback about their performan ce on an academic task. Theories of Intelligence and Information Avoidance To date, the theory of intelligence literature has focused on a host of Helpless responses to failure, including self handicapping, preference for easy tasks that will garner pos itive feedback, feelings of depression, and preference for flattering feedback over accurate feedback (Dweck & Elliott Moskwa, 2010). When it comes to the effect of ToI and success vs. failure on decision making and choosing tasks, however, the Helpless re sponse patterns documented represent preferences between two types of known information or tasks, indicative of selective exposure findings. No previous studies operationalize the Helpless vs. Mastery oriented patterns as the choice made when presented wit pattern includes complete information avoidance. Although no prior studies have examined

PAGE 28

28 whether entity theorists would be more likely than would incremental theorists to avoid test performance feedback via a dichotomous choice, a few studies suggest evidence that they would for feedback about ability as a function of the perceived changeability of that ability. Results found that when ability was perceived as uncontrollable, people reported a stronger preference for positive feedback over negative feedback. However, when abil ity was perceived as controllable, people did not report differential preference for positive and negative feedback. Similarly, Trope et al. (2003) found that whereas participants did not report a preference for positive over negative feedback when ability was perceived as uncontrollable and high nor when ability was perceived as controllable, they reported a preference for positive over negative feedback when ability was perceived as uncontrollable and low. Although these studies highlight the importance o f perceived controllability of the ability, neither study examined whether theories of intelligence differentially predict information avoidance from a dichotomous (i.e. seek or avoid overall feedb ack) perspective. In addition, the dependent variable in th e Dunning (2005) studies was preference ratings for specified types of feedback (i.e. positive or negative) and was based on feedback and measured on a Likert t ype scale. My dissertation directly examines the question of whether theories of intelligence, which differ in perceived control of the intelligence, differentially prompt complete avoidance of ostensibly available performance feedback following either suc cess or failure on that performance. To date, literature on information avoidance highlights situational and motivational factors influencing information avoidance (perceived controllability, anticipated regret, etc.), but no published studies examine the effects of individual differences on information avoidance.

PAGE 29

29 Sweeny et al. (2010) noted that individual differences in coping style may play a role in avoidance. Because entity and incremental theorists cope with failure differently and because they differ on several other dimensions that predict avoidance, theories of intelligence seem an ideal candidate to examine the effects of individual difference on information avoidance. Overview of the Current Research The current research has two primary goals. T he first goal is to replicate past associations of ToI, including goal preferences as well as various responses to failure: That is, to show that entity theorists but not incremental theorists display a host of helpless responses to failure, including anti cipated negative affect, perceived threat to self views, difficulty coping, etc. A second goal is to extend past research by examining the possibility that ToI (entity vs. incremental) and task outcome (success vs. failure) will differentially prompt avoid ance of ostensibly available performance feedback and thus making an argument that information avoidance may be a behavioral component of the Helpless Response pattern. I conducted three studies that examine the relationship between ToI and avoidance of p erformance feedback. Study 1 is correlational and examines the association between implicit ToI and the inclination to avoid academic performance feedback (e.g., tes t scores). Study 1 also examines the association between implicit ToI and goal preferences, and between inclination to avoid information and goal preferences. Studies 2a and 2b experimentally examine the relationship between theories of intelligence and information avoidance by manipulating (Studies 2a) and measuring (Studies 2b) ToI and manipul ating task performance (success or failure). Two key differences between the proposed research and traditional theories of intelligence studies are 1) Dweck and colleagues typically use within subjects designs; that is, both entity and incremental theori sts first experience success on a given task, and later, both types of theorists experience failure on a task. The proposed research uses a between subjects design in which

PAGE 30

30 participants experience either success (which will lead to positive feedback expect ations) or failure (which will lead to negative feedback expectations). For obvious reasons, within subjects designs do not lend themselves well to studies in which information avoidance is the dependent measure; and 2) Dweck recommends excluding participa nts with average scores in the middle of recommendation, I reana lyzed the data after excluding participants whose composite scores were between 3.01 and 3. 99 and classifying participants with a theory of intelligence composite score ranging from 1 3 as an incremental theorist and classifying participants with a composite score ranging from 4 6 as an entity theorist. Across all analyses, dichotomizing theorie s of intelligence produced similar, though slightly weaker results. Any analyses I present that dichotomize the ToI scale (relating to goal preferences) are for illustrative purposes only.

PAGE 31

31 CHAPTER 2 STUDY 1: ASSOCIATION S BETWEEN THEORIES O F INTELLIGENCE, GOAL PREFERENCES, AND ACA DEMIC INFORMATION AV OIDANCE Study 1 examined the hypothesis (H1) that entity theorists are more inclined than are incremental theorists to avoid information about their academic performance. Study 1 also examined the relationship between theories of intelligence and goal preferences, and goal preferences and inclination to avoid information about academic performance. Method Participants Participants were 1259 ninth graders from the state of Florida participating in an unrelated, larger survey of adolescent religiousness. The three measures relevant to the present study were included in the larger survey. Participants received a gift card to Wal Mart for completing the larger survey. Materials The first scale was a 3 item scale ad apted from the Theories of Intelligence Adult Scale (Dweck, 2000; Dweck et al., 1995; Mueller & Dw eck, unpublished, as cited in Dweck, 2000 ). s trongly disagree ; 6 = strongly agree ToI items is .85 in other research (e.g. Baird et al., 2009). In the current study the alpha was .90 ( M = 2.8, SD = 1.53). d information about academic performance (Information Avoidance Scale) and consisted of eight items created for this study. Example items At times I avoid information about s foolish for me to avoid information about how I did on a point

PAGE 32

32 Likert type scale ( 1 = strongly disagree ; 5 = strongly agree ) To create a composite measure of inclinat ion to avoid academic information, I averaged responses to the eight items, after reverse M = 2.1, SD = .83) The third measure was the 4 item Goal Choice scale (Dweck, 2000; Dweck et al., 1995; Mueller & Dweck, unpublish ed, as cited in Dweck, 2000 poin t Likert type scale (1 = strongly disagree ; 6 = strongly agree) (Appendix C) except for a single had to choose between getting a good grade and being challenged tomous ( getting a good grade or being challenged ). Researchers have examined the four items individually and have also combined the four items to form a single measure of goal preference (Baird et al., 2009; Mueller & Dweck, unpublished, as cited in Dweck, 2000 ). In the present study, the alpha that emerged wh analyzed the four items individually. Procedure Participants received a web address and password via email for the larger survey. They then completed the survey, including the measures relevant to the present study. Results I excluded from relevant analyses participants who chose not to answer an item within either the theories of intelligence or academic information avoidance composites, or chose not to answer t he relevant goal preference items.

PAGE 33

33 Theories of Intelligence and the Inclination to Avoid Performance Feedback Table 2 1 presents the correlations between all measures. Consistent with Hypothesis 1, entity theorists were more inclined than were incremental theorists to avoid information about their academic performance. Table 2 1. Correlations among v ariables Item ToI Composite Dichotomized ToI Info Avoidance Scale GP1: Dichotomous Item GP2: Avoid tasks when failure likely GP3: Rather do well than learn ToI Composite Dichotomized ToI .87** Info Avoidance Scale .28** .23** GP1: Dichotomous item .19** .14** .20** GP2: Avoid tasks when failure likely .37** .34** .28** .28** GP3: Rather do well than learn .17** 15** .08* .23* .31** GP4: Learning trumps grades .01 .03 .03 .19** .07* .18** Note. N s range from 1036 1109 depending on the number of participants excluded from the analyses. Correlations reported are Pearson moment co rrelations. ** p < .01. p < .05. Theories of Intelligence and Goal Preferences I first examined the dichotomous goal preference item (GP1) in which goal preferences are directly pitted against one another. Consistent with expectations, the more strongly participants endorsed an entity theory of intelligence, the stronger was their preference for a performance goal, replicating past findings ( Dweck & Legg ett, 1988; Mueller & Dweck, unpublished, as cited in Dweck, 2000 ). As evident in Figure 2 1, when exami ning the percentages alone, participants who were classified as incremental theorists were equally likely to adopt a performance (49.7%)

PAGE 34

34 or learning goal (50.3%), whereas participants who were classified as entity theorists were more likely to adopt a perf ormance goal (64.6%) over a learning goal (35.4%), 2 (1) = 19.95, p < This finding suggests that the relationship between theories of intelligence and Figure 2 1. Percentages of incremental and entity theorists preferring learning vs. performance goals. Whereas no goal preference difference was found for incremental theorists, entity theorists preferred a performance goal over a learning goal. The a nalysis in this figure is based on dichotomizing theories of intelligence (Dweck, 2000). However, this classification of incremental and entity theorists is for illustrative purposes only and omits participants from the middle of the distribution. The othe r goal preference items generally showed a pattern similar to the pattern of the forced choice item. Consistent with Hypothesis 1, the more strongly participants endorsed an finding is noteworthy because it suggests that entity theorists are inclined to avoid situations perhaps opportunities for feedback that present a high likelihood of failure. Likewise, the more 0 10 20 30 40 50 60 70 Incremental Entity % Preferring Each Goal Implicit Theory of Intelligence Learning Goal Performance Goal

PAGE 35

35 strongly participants endorsed an entity theory of intelligence, the more they agreed with the finding is not surprising. After all, as suggested in a study by Mueller and Dweck (as cited in Dweck, 2000), both entity and incremental theorists tend to hold both performance and learning goals goal preferences emerge most clearly when the goals directly conflict Goal Preferences and the Inclination to Avoid Performance Feedba ck I tested the hypothesis that people who prefer different goals (performance vs. learning) differ in their inclination to avoid information about their academic performance. An independent sample t test for the forced choice goal preference item found th at students reported a greater tendency of avoiding academic information if they said they preferred a good grade (performance goal), n = 559, M = 2.2, SD = .84, than if they said they preferred to be challenged (learning goal), n = 489, M = 1.9, SD = .78, in a class, t (1151) = 6.31, p = .0001, d = .37. The other three goal preference items generally showed similar patterns. Participants reported stronger inclination to avoid academic performance feedback the more strongly they endorsed GP2 and GP3. Again academic performance avoidance was unrelated to GP4. In sum, Study 1 found that entity theorists were more inclined than were incremental theorists to avoid academic performance information. In addition, replicating past findings, entity theorists were more likely than were incremental theorists to prefer a performance goal over a learning goal. Finally, preference for a performance goal corresponded with greater avoidance of academic information. Although these data support the proposed hypotheses, they are correlational. The remaining studies extend this study by experimentally manipulating theories of

PAGE 36

36 intelligence and/task outcome to examine their relationship with avoidance of actual performance feedback.

PAGE 37

37 CHAPTER 3 STUDY 2A: INDUCED TH EORIES OF INTEL LIGENCE, TASK OUTCOM E, AND FEEDBACK AVOIDANCE Study 2a manipulated whether participants endorsed an entity vs. incremental theory of intelligence then provided participants with a success or failure experience on an academic writing test and tested five hy potheses. H1: Compared with participants who endorse an incremental ToI, participants who endorse an entity ToI will display greater avoidance of detailed performance feedback, lower perceived control over their writing score, greater current and anticipa ted negative affect, greater anticipated difficulty coping with the feedback, greater regret to seek feedback, greater perceived threat posed by the feedback, lower regret to avoid feedback, and lower perceived utility of the feedback. H2: Compared with participants who experience a success outcome on the task, participants who experience a failure outcome on the task will display greater avoidance of detailed performance feedback, greater current and anticipated negative affect, greater difficulty copin g with the feedback, greater regret to seek feedback, greater perceived threat by the feedback, and lower regret to avoid feedback. H3: The effect of task outcome on many of the outcomes will depend on theories of intelligence. Among participants who endo rse an entity ToI, there will be a directional effect of task outcome on these measures (see H2 for these directional effects); among participants who endorse an incremental ToI, however, the effect of task outcome on these measures will be in one of the f ollowing three categories: a) in the same direction but closer to zero, b) zero, or c) in the opposite direction of the effect predicted for entity participants. I have no compelling reason for predicting different effects of task outcome among entity vs. incremental participants on perceived control or for perceived utility. For instance, because belief about the controllability of intelligence is independent of actual ability (high or low) (Dweck, 2000), perceived control of the academic outcome should be independent of success or failure. Similarly, if feedback is uncontrollable, participants may also see that feedback as less useful; I therefore predict that perceived utility will show a pattern similar to that of perceived control and be independent of success or failure. Nonetheless, I conducted exploratory analyses to test for possible differences in effects on perceived control and perceived utility of the feedback. H4: Perceived control, anticipated affect, coping resources, anticipated regret, perc eived threat, and perceived utility of the feedback will separately mediate the relationship between task outcome and decision to seek or avoid performance feedback in the entity condition but not in the incremental condition. H5: Compared with participa nts who endorse an incremental ToI, participants who endorse an entity ToI will display greater preference for a performance goal than a learning goal and greater inclination to avoid academic information (on the Information Avoidance

PAGE 38

38 Scale). In addition, compared with participants who prefer a learning goal, participants who prefer a performance goal will be more inclined to avoid academic information. These findings would replicate Study 1. Method Participants and Design Participants were 203 people (84 males, 118 females, age M = 34.0, SD = 13.17) recruited conditions in a 2 (ToI: entity vs. incremental) x 2 (Task Outcome: success vs. failure) between subjects fa ctorial design. Materials and Procedure Mturk is a site hosted by Amazon that facilitates efficient data collection and allows people to sign up and get paid for completing surveys listed by researchers. Mturk provides a more diverse sample than do tradit ional data collection methods yet yields data as reliable as collected using traditional data collection methods (Buhrmester, Kwang, & Gosling, 2011). People who accessed the study via Mturk opened the survey link and completed an online consent form (App endix D). They then read a background paragraph about the survey that said: success in graduate school. The GRE typically comprises writing, mathematical, and verbal tasks. The a dministrators at the Educational Testing Service (ETS), who are in charge of developing the GRE, are pilot testing new critical writing topics to determine if they are appropriate for future GRE exams. We have enlisted an Mturk sample to help pilot these e ssay topics. In this session, you will write an essay, receive feedback on your essay from one of our online GRE analysts, and complete some additional items. All participants learned that the GRE writing task that they would see entailed analyzing an issu e and would be graded on a 1 6 scale in half point increments, where 1.0 was the lowest possible score and 6.0 was the highest possible score. Participants in the entity essay condition

PAGE 39

39 were instructed to write an essay supporting a statement that endorsed an entity view of intelligence: Scientists of all fields including neurologists, biologists, and developmental psychologists agree that intelligence is mostly genetically inherited. People are born with a certain potential regarding analytical, mathemati cal, and writing abilities. When people encounter challenge and/or failure, the limit of their potential is revealed. Participants in the incremental essay condition were instructed to write an essay supporting a statement that endorsed an incremental vie w of intelligence: Scientists of all fields including neurologists, biologists, and developmental psychologists agree that intelligence can dramatically increase with practice and re limitless as long as people continue to challenge themselves and are resilient in the face of failure. Participants in both essay conditions received instructions to write 1 2 paragraphs in which they provide examples that support the observation and t o include at least one policy that schools should adopt in light of this observation. After the essay task, all participants completed the Theories of Intelligence Adult Scale (the same items from Study 1) as a manipulation check that the essay task succes sfully induced the desired ToI (entity or incremental). They then completed a distracter task while the online GRE analyst ostensibly assessed their essay. Participants were told that a CONTINUE button would appear on the screen when the online GRE analyst had finished assessing their essay. Clicking the CONTINUE button would allow them to view their essay grade. I programmed the CONTINUE button to appear after an 8 minute delay to facilitate the cover story. Clicking the CONTINUE button introduced the tas k outcome manipulation. In the failure condition, participants received a low essay score (2.0 out of 6). In the success condition, participants received a high essay score (5.5 out of 6). The computer screen again displayed the scoring system (1.0 = the l owest possible score; 6.0 = the highest possible score) below their

PAGE 40

40 score to ensure that participants knew that they scored poorly or well and to induce either a positive or negative expectation for more detailed feedback. After viewing their score, partic ipants completed a series of items. First they completed the primary dependent measure. Participants received a message on the screen stating that in addition to the numeric grade they saw, the GRE analyst had also provided detailed comments about their es a) b) c) No thanks, I do not want To enhance believability, the email option included a space to type an email address. Participants then wrote why they chose the option they chose. Next, participants then completed several measures about t he task (Appendix E). Three difficult), comparative performance (H to other participants who have completed the same task?; 1 = much worse than others; 7 = much task?; 1 = no effort, 7 = a lot of effort). The next five items were adjectives (calm, nervous, anxious, relaxed, and tense) that assessed how participants felt about their performance on the essay task (1 = not at all; 4 = very much). I combined the fiv e affect items (after reverse coding calm and relaxed) into a Current me feel bad [depressed, sad, etc.]; 1 = not at all; 7 = very much), items assessing anticipated

PAGE 41

41 difficulty coping (e.g. pretty well ; 1 = strongly disagree; 7 = strongly agree [reverse cod ed]). I combined the two coping items (after reverse coding the items so that higher numbers would reflect greater items assessing anticipated regret to seek or avoid the feedback (e.g. Imagine that you chose to regardless of which option you actually chose. How much do you anticipate regretting that decision later? ; 1 = not at all; 5 = very much), perceived control o ver the outcome (e.g. There are things I can do to improve my results should I learn that I did not perform well on the Analyze an Issue task; 1 = strongly disagree; 7 = strongly agree), perceived utility of the feedback (e.g. It would be useful to see the about my essay; 1 = strongly disagree; 7 = strongly agree), and four items assessing perceived my essay might threaten an important belief a bout myself such as that I am intelligent or competent; 1 = strongly disagree; 7 = strongly agree). I combined the four threat items into a forced choice goal prefere nce item (If I had to choose between getting a good grade and being Finally, participants reported their SAT scores (if they remembered them) and the same individual differe nce measure of inclination to avoid performance feedback used in Study 1 2.3, SD = 1.19). Participants were then thanked, read a debriefing that described the stud y and its hypotheses (Appendix F), and were paid for their time.

PAGE 42

42 Results Manipulation Check To ensure that the entity and incremental essay writing task successfully oriented participants to the desired ToI (entity or incremental), I combined the three ToI items to create a M = 3.3, SD = 1.77). Participants reported higher scores on the composite index if they wrote an essay supporting an entity theory of intelligence (the entity condition, M = 4.1, SD = 1.78) than if they wrot e an essay supporting an incremental theory of intelligence (the incremental condition, M = 2.6, SD = 1.44), t (201) = 6.55, p < .0001, d = .93. Because the essay task generally oriented participants to the desired ToI, the manipulation was successful 1 Pr eliminary Analyses Regarding information avoidance, i n response to the opportunity to see the detailed feedback provided by the online GRE analyst, most participants opted to see detailed feedback about their essay; specifically,161 (79.3%) participants se lected Option 1 (see feedback now), 17 chose Option 2 (receive feedback later via email), and 25 chose Option 3 (decline to receive feedback). Because information avoidance can be either temporary (delay) or permanent (avoid), participants who chose Option 2 or 3 were combined and classified as avoiders. The zero order correlations among all variables appear in Table 3 1 and the cell means of all variables appear in Table 3 2 1 As noted earlier, Dweck typ ically categorizes people into three groups based on their ToI composite score. The incremental group consists of people with composite scores less than or equal to 3.0, the entity group consists of people with composite scores greater than to equal to 4.0 and the unclassified group consists of people with condition fell into the three categories as follows: 35.4% incremental theorists; 7.1% uncla ssified; and 57.6% entity incremental theorists; 11.5% unclassified; and 20.2% entity theorists.

PAGE 43

43 Table 3 1 Zero order correlations among v ariables ToI Manipulation Task Out come Manipulation Information Avoidance Decision Info Avoidance Scale Goal Preference Perceived Control Current Negative affect Anticipated Negative Affect Difficulty Coping Regret to Seek Regret to Avoid Perceived Threat ToI Manipulation Task Outcome Manipulation .08 Information Avoidance Decision .14* .24** Info Avoidance Scale .11 .05 .31** Goal Preference .15* .10 .01 .21** Perceived Control .22** .04 .12 .29** .02 Curre nt Negative Affect .08 .29** .04 .03 .03 .22** Anticipated Negative Affect .17* .35** .19** .26** .10 .08 .37** Difficulty Coping .22** .18** .24** .45** .12 .25** .17* .35** Regret to Seek .13 .17* .21** .43** .14* .13 .09 .24** 45** Regret to Avoid .09 .01 .25** .25** .03 .24** .08 .01 .07 .02 Perceived Threat .16* .16* .13 .33** .14* .03 .20** .61** .50** .40** .11 Perceived Utility .12 .14* .32** .32** .03 .50** .14* .03 .29** .29** .41** .02 Note. N = 2 03. Correlations reported are Pearson moment correlations. ** p < .01. p < .05.

PAGE 44

44 Table 3 2 Means for outcome measures Induced Incremental ToI Induced Entity ToI Success n = 56 Failure n = 48 Success n = 45 Failure n = 54 M SD M SD M SD M SD Information Avoidance Scale 2.1 (1.06) 2.2 (1.20) 2.3 (1.41) 2.5 (1.11) Perceived Control of Performance 5.4 (1.76) 5.6 (1.47) 4.6 (1.98) 4.9 (1.66) Current Negative Affect 1.5 (.60) 1.9 (.69) 1.6 (.53) 2.0 (.79) Anticipated Negative Affect 1. 9 (1.09) 2.6 (1.77) 1.9 (1.32) 3.5 (1.83) Anticipated Difficulty Coping 1.8 (1.16) 1.8 (.87) 1.9 (1.03) 2.7 (1.36) Anticipated Regret to Seek feedback 1.3 (.59) 1.5 (1.22) 1.4 (1.18) 1.9 (1.09) Anticipated Regret to Avoid Feedback 4.1 (2.80) 3.7 (2.49) 3.1 (2.00) 3.7 (2.63) Perceived Threat of Feedback 1.8 (1.00) 1.8 (1.17) 1.8 (.96) 2.5 (1.21) Perceived Utility of Feedback 5.5 (1.77) 5.5 (1.47) 5.6 (1.50) 4.7 (1.93) Note N = 203. Avoidance of Performance Feedback Although I gener ally organize the results around the hypotheses, I first present the results for academic information avoidance because a) it is the dependent variable of primary interest, and b) academic information avoidance is dichotomous and the analyses performed on this variable differ from those performed on the other dependent variables. Figure 3 1 displays the percentage of participants in each condition that opted to avoid square. Consist ent with Hypothesis 1, I found a marginally significant trend for more participants to avoid seeing the 2 (1) = 3.57, p 2, more participants opted to avoid

PAGE 45

45 2 (1) = 11.36, p influence avoidance of feedback only among participants in the entity condition (H3), I conducted separate chi square analyses for each ToI condition. As predicted, more participants in in the 2 (1) = 8.08, p 2 (1) = 2.81, p Figure 3 1 theories of intelligence and task outcome. Whereas task outcome did not significantly affect avoidance in the incremental condition, experie ncing a failure outcome led to greater avoidance in the entity condition. One additional notable finding was a significant correlation between the Information r = .31, p < 001. Specifically, avoiders reported higher scores ( M = 3.0 SD = 1.33) on the Information Avoidance Scale than did seekers ( M = 2.1 SD = 1.08), t (201) = 4.62, p < .001, d = .79. In 0 10 20 30 40 Incremental Entity % Avoiding Feedback Induced Theory of Intelligence Success Failure

PAGE 46

46 addition, these two variables showed similar patterns of correlations with the other variables. These findings suggest that responses to the scale may serve as a proxy for actual information avoidance. Together, the results support Hypotheses 1 through 3 with regard to information avoidance and show that participants were more l ikely to avoid feedback about an academic task they had failed, particularly if they had an entity ToI. Hypothesis 1 (Effects of ToI) I tested Hypothesis 1 by conducting a series of t tests to examine mean differences between the induced entity and incre mental conditions for perceived control, current negative affect, anticipated negative affect, anticipated difficulty coping, regret to seek the feedback, regret to avoid the feedback, perceived threat posed by the feedback, and perceived utility of the fe edbac k. The results appear in Table 3 3 Table 3 3 Hypothesis 1 results: main effects of ToI Induced Incremental ToI n = 104 Induced Entity ToI n = 99 M SD M SD T p d Perceived Control (of performance) 5.5 (1.63) 4.7 (1.81) 3.12 < .01 .47 C urrent Negative Affect 1.7 (.67) 1.8 (.71) 1.11 ns .15 Anticipated Negative Affect 2.2 (1.48) 2.8 (1.78) 2.46 < .02 .37 Anticipated Difficulty Coping 1.8 (1.03) 2.3 (1.27) 3.21 < .01 .43 Anticipated Regret to Seek Feedback 1.4 (.94) 1.7 (1.15) 1 .79 < .08 .29 Anticipated Regret to Avoid Feedback 3.9 (2.65) 3.5 (2.37) 1.23 ns .16 Perceived Threat (of the feedback) 1.8 (1.08) 2.2 (1.16) 2.35 .02 .36 Perceived Utility (of the feedback) 5.5 (1.63) 5.1 (1.79) 1.66 ns .23 Note N = 203.

PAGE 47

47 As pre dicted, compared with participants who endorsed an incremental ToI, participants who endorsed an entity ToI reported lower perceived control of their writing ability, greater anticipated negative affect, greater anticipated difficulty coping with the detai led feedback, greater anticipated regret to see the feedback (marginally significant), and greater perceived threat posed by the feedback. However, analyses for anticipated regret to avoid the feedback and perceived utility failed to support Hypothesis 1. Coupled with the data for information avoidance, the data overall support the hypotheses that link an entity theory of intelligence to greater information avoidance and to the helpless response entity theorists often report (Dweck, 2000). Hypothesis 2 (Ef fects of Task Outcome) I tested Hypothesis 2 by conducting a series of t tests to examine mean differences between the success and failure conditions for perceived control, current negative affect, anticipated negative affect, anticipated difficulty coping regret to seek the feedback, regret to avoid the feedback, perceived threat posed by the feedback, and perceived utility of the feedback. The results of t hese analyses appear in Table 3 4 As predicted, compared with participants who failed the task, par ticipants who succeeded on the task reported greater current negative affect, greater anticipated negative affect, greater anticipated difficulty coping with the detailed feedback, greater anticipated regret to see the feedback, and greater perceived threa t posed by the feedback. However, analysis revealed that task outcome did not affect anticipated regret to avoid feedback. Exploratory analyses revealed no main effect of feedback on perceived control of performance. Interestingly, exploratory analyses rev ealed a significant effect of task outcome on perceived utility such that participants who failed reported the feedback would be less useful than did participants who succeeded. This finding is rather counterintuitive because people who

PAGE 48

48 do not perform well have objectively more room to improve and in theory more benefit to gain from feedback than do people who perform well. Coupled with the data for information avoidance, the data overall support the hypotheses that link failure to greater avoidance of deta iled feedback. Table 3 4 Hypothesis 2 r esults: m ain e ffects of t ask o utcome Success n = 101 Failure n = 102 M SD M SD T p d Perceived Control (of performance) 5.0 (1.61) 5.2 (1.90) < 1 ns .11 Current Negative Affect 1.6 (.74) 2.0 (.57) 4.2 1 < .001 .61 Anticipated Negative Affect 1.9 (1.85) 3.1 (1.19) 5.35 < .001 .77 Anticipated Difficulty Coping 1.8 (1.23) 2.3 (1.10) 2.62 < .01 .43 Anticipated Regret to Seek Feedback 1.3 (1.16) 1.7 (.90) 2.40 < .02 .34 Anticipated Regret to Avoid Feedback 3.7 (2.55) 3.7 (2.51) < 1 N s .00 Perceived Threat (of the feedback) 1.8 (1.24) 2.2 (.98) 2.29 < .03 .36 Perceived Utility (of the feedback) 5.5 (1.77) 5.1 (1.63) 1.90 .05 .24 Note N = 203. Hypothesis 3 (ToI x Task Outcome Interactions) T o test Hypothesis 3 I conducted a series of ANOVAs to examine ToI x Task Outcome interactions on perceived control, current negative affect, anticipated negative affect, anticipated difficulty coping, regret to seek the feedback, regret to avoid the feedba ck, perceived threat posed by the feedback, and perceived utility of the feedback. The results of these analyses appear in Table 3 5

PAGE 49

49 Table 3 5 Hypothesis 3 results: tests of i nteractions F p 2 Perceived Control (of performance) < 1 ns .00 Current Negative Affect < 1 ns .00 Anticipated Negative Affect 3.29 .07 .02 Anticipated Difficulty Coping 5.68 .02 .03 Anticipated Regret to Seek Feedback < 1 ns .003 Anticipated Regret to Avoid Feedba ck 1.86 ns .009 Perceived Threat (of the feedback) 6.88 .01 .03 Perceived Utility (of the feedback) 3.00 .08 .02 As predicted, analyses revealed a marginally significant ToI x Feedback interactions for anticipated negative affect, anticipated difficul ty coping with the detailed feedback, and perceived threat posed by the feedback (see also Figures 3 2 through 3 4 ). As evident in Figure 3 2 simple effects tests revealed that theories of intelligence did not influence anticipated negative affect in the success condition (entity M = 1.9, SD = 1.16; incremental M = 1.9, SD = .87), t (99) < 1, d = .00. However, in the failure condition, participants reported greater anticipated negative affect in the entity condition ( M = 3.5, SD = 1.83) than in the increme ntal condition ( M = 2.6, SD = 1.77), t (100) = 2.40, p < .02, d = .88. This interaction differed from the hypothesized interaction (that compared with experiencing success, experiencing failure would lead to greater anticipated negative affect only in the e ntity condition [H3]); instead, compared with participants who endorsed an incremental ToI, participants who endorsed an entity ToI reported greater anticipated negative affect only in the failure condition. However, the pattern reveals a ToI x Task Outc ome interaction nonetheless and is thus important in it s own right. As seen in Figure 3 3 simple effects tests revealed that task outcome did not influence coping in the incremental condition (success M = 1.8, SD = 1.16; failure M = 1.8, SD = .87), t (102 ) < 1, d =

PAGE 50

50 .00. However, in the entity condition, participants reported greater difficulty coping in the failure condition ( M = 2.7, SD = 1.36) than in the success condition ( M = 1.9, SD = 1.03), t (97) = 3.15, p < .01, d = .66. As evident in Figure 3 4 s imple effects tests revealed that task outcome did not influence perceived threat in the incremental essay condition (success M = 1.8, SD = 1.00; failure M = 1.8, SD = 1.17), t (102) = .29, p = ns, d = .00. However, in the entity essay condition, participa nts reported that the feedback posed greater threat in the failure condition ( M = 2.5, SD = 1.21) than in the success condition ( M = 1.8, SD = .96), t (97) = 3.34, p < .01, d = .64. Interestingly, an exploratory analysis revealed a marginally significant interaction on perceived utility of the feedback. As seen in Figure 3 5 simple effects tests revealed that task outcome did not influence perceived utility of the feedback in the incremental condition (success M = 5.5, SD = 1.77; failure M = 5.5, SD = 1.4 7), t (102) < 1, d = .01. However, in the entity condition, participants reported the feedback would be less useful in the failure condition ( M = 4.7, SD = 1.93) than in the success condition ( M = 5.6, SD = 1.50), t (97) = 2.41, p < .02, d = .52. The four significant interactions revealed that participants in the entity condition who failed displayed a pattern characteristic of the Helpless Response to failure (Dweck, 2000). Specifically, they exhibited greater anticipated negative affect, greater difficul ty coping with the available feedback, greater perceived threat, and lower perceived utility of the feedback. Because this finding was not predicted, it should be interpreted cautiously. The exploratory analysis for as not significant. Inconsistent with predictions, analyses revealed no significant interaction for either of the regret items (regret to seek and regret to avoid feedback). Combined with the data for information avoidance, the data generally support Hypot hesis 3 and also the argument that information avoidance aligns with the Helpless

PAGE 51

51 Response pattern typically exhibited by entity theorists when they experience failure (Dweck, 2000). Figure 3 2 Anticipated negative affect values as a function of theorie s of intelligence and task outcome. Whereas anticipated negative affect did not differ as a function of theories of intelligence in the success condition, anticipated negative affect did differ as a function of theories of intelligence in the failure condi tion. Figure 3 3 Anticipated difficulty coping with the feedback as a function of theories of intelligence and task outcome. Whereas anticipated difficulty coping did not differ as a function of task outcome in the incremental condition, anticipated dif ficulty coping did differ as a function of task outcome in the entity condition. 1.0 2.0 3.0 4.0 Incremental Entity Anticipated Negative Affect Induced Theory of Intelligence Success Failure 1.0 1.5 2.0 2.5 3.0 Incremental Entity Anticipated Difficulty Coping Induced Theory of Intelligence Success Failure

PAGE 52

52 Figure 3 4 Perceived threat from the feedback as a function of theories of intelligence and task outcome. Whereas perceived threat did not differ as a function of task outc ome in the incremental condition, perceived threat did differ as a function of task outcome in the entity condition. Figure 3 5 Perceived utility of the feedback as a function of theories of intelligence and task outcome. Whereas perceived utility did n ot differ as a function of task outcome in the incremental condition, perceived utility did differ as a function of task outcome in the entity condition. 1.0 1.5 2.0 2.5 3.0 Incremental Entity Perceived Threat Induced Theory of Intelligence Success Failure 1.0 2.0 3.0 4.0 5.0 6.0 7.0 Incremental Entity Perceived Utility Induced Theory of Intelligence Success Failure

PAGE 53

53 Hypothesis 4 (Mediation Analyses) Because chi square analyses revealed an overall moderation of inform ation avoidance, where task outcome influenced avoidance of detailed feedback only in the entity condition and not the incremental condition, I tested for mediation of the effect of task outcome on information avoidance separately for each ToI group using traditional steps for mediation testing (Baron & Kenny, 1986). Given the overall moderation of performance feedback avoidance (i.e. the logistic regressions for the entity condition would always be significant [ b = 1.42, SE = .52, p < .01, OR = .24], and the logistic regressions for the incremental condition would always nonsignificant [ b = .91, SE = .55, p = ns, OR = .40]), I ran the logistic regression for the incremental essay condition once and then focused on mediation analyses of information avoidan ce for only the entity condition. I treated any variables for which either task outcome produced a significant or marginally significant effect (main effect or interaction) as a potential mediator. Thus, potential mediators included current negative affect anticipated negative affect, difficulty coping, regret to seek, perceived threat, and perceived utility of the feedback. Logistic regression analyses revealed that current negative affect, anticipated negative affect, regret to seek, and perceived threat did not mediate the relationship between feedback expectations and information avoidance. That is, although task outcome influenced current negative affect ( b = .37, SE = .14, p < .01, R 2 = .07), anticipated negative affect ( b = 1.53, SE = .33, p < .001, R 2 = .18), regret to seek the feedback ( b = .45, SE = .23, p = .05, R 2 = .04), and perceived threat ( b = .75, SE = .22, p < .01, R 2 = .10) in the entity condition, none of these potential mediators significantly affected the outcome variable when controlli ng for task outcome ( b = .10, SE = .33, p = ns, OR = .90), ( b = .11, SE = .14, p = ns, OR = 1.11), ( b = .30, SE = .19, p = ns, OR = 1.35), and ( b = .25, SE = .21, p = ns, OR = 1.28), respectively.

PAGE 54

54 However, analyses showed that both anticipated difficulty coping and perceived utility of the feedback partially mediated the re lationship. As shown in Figure 3 6 feedback significantly influenced anticipated difficulty coping ( b = .78, SE = .25, p < .01, R 2 = .09) and difficulty coping predicted information av oidance when controlling for task outcome ( b = .40, SE = .19, p < .04, OR = 1.49). However, feedback remained a significant predictor ( b = 1.15, SE = .54, p < .04, OR = .32), indicating only partial mediation. As shown in Figure 3 7 task outcome signific antly influenced perceived utility of the feedback ( b = .85, SE = .35, p < .02, R 2 = .06) and perceived utility predicted information avoidance when controlling for task outcome ( b = .44, SE = .14, p < .01, OR = .65). Again, task outcome remained a signi ficant predictor ( b = 1.15, SE = .55, p < .04, OR = .32), indicating only part ial mediation. However, Figure 3 8 shows that when perceived utility and anticipated difficulty coping were entered simultaneously when controlling for task outcome, perceived u tility remained significant ( b = .43, SE = .15, p < .01, OR = .65) but neither anticipated difficulty coping ( b = .37, SE = .21, p = ns, OR = 1.44) nor task outcome ( b = .89, SE = .57, p = ns, OR = .41) was significant. Figure 3 6 Partial correlations for the relationship between feedback condition and performance feedback avoidance as mediated by anticipated difficulty coping. The partial correlation between feedback condition and performance feedback avoidance controlling for anticipated difficulty coping is in parentheses. ** p < .01. p < .05. .22* .31** Task Outcome Performance Feedback Avoidance Anticipated Difficulty Coping .29** (.22*)

PAGE 55

55 Figure 3 7 Partial correlations for the relationship between task outcome and performance feedback avoidance as mediated by perceived utility of the feedback. The partial correlation betwee n task outcome and performance feedback avoidance controlling for perceived utility is in parentheses. ** p < .01. p < .05. Figure 3 8 Partial correlations for the relationship between task outcome and performance feedback avoidanc e as simultaneously mediated by anticipated difficulty coping and perceived utility of the feedback. The partial correlation between task outcome and performance feedback avoidance controlling for anticipated feedback and for perceived utility is in parent heses. ** p < .01. p < .05. .32** .19 .24* .31** Task Outcome Performance Feedback Avoidance Anticipated Difficulty Coping Perceived Utility .29** (.16) .34** .24* Task Outcome Performance Feedback Avoidance Perceived Utility .29** (.22*)

PAGE 56

56 Hypothesis 5 (Associations among ToI, Goal Preferences, and the Inclination to Avoid Performance Feedback) Theories of intelligence and inclination to avoid performance feedback As shown in Table 3 1 and inconsistent with Hyp othesis 5, induced ToI was not significantly related to inclination to avoid performance feedback on the Information Avoidance Scale. Similarly, students reported no greater inclination to avoid performance feedback in the entity condition than in the incr emental condition. Of note, ToI did significantly correlate with the actual avoidance decision. Theories of intelligence and goal pr eferences Consistent with expectations and Study 1 and replicating past research that ToI can cause goal preferences (Dwec k & Leggett, 1988), an entity ToI was associated with a preference for a performance goal (see Table 3 1 ) Also, as shown in Figure 3 9 when examining the percentages alone, participants who endorsed an incremental ToI were more likely to adopt a learning goal (58.7%) than a performance goal (41.3%), whereas participants who endorsed an entity ToI were more likely to adopt a performance goal (56.6%) than a learning goal (43.4%), 2 (1) = 4.70, p = This finding suggests that the effect of theories of intelligence on goal preference is preference for learning goals. This patte rn differs from the pattern found in Study 1, where the preference for a performance goal.

PAGE 57

57 Figure 3 9 Percentages of incremental and entity theorists prefer ring learning vs. performance goals. Whereas incremental theorists preferred a learning goal over a performance goal, entity theorists preferred a performance goal over a learning goal. Goal preferences and inclination to avoid performance feedback As s hown in Table 3 1 and consistent with Hypothesis 5 and Study 1, goal preference significantly correlated with the IA Scale such that preference for a good grade (performance goal) was associated with greater inclination to avoid performance feedback. Of no te, goal preference did not significantly correlate with the actual avoidance decision. In summary, Study 2a found that avoidance of detailed feedback was higher in the entity condition than in the incremental condition, but only when participants failed the writing task. In addition, participants reported greater anticipated negative affect, difficulty coping with feedback, and perceived threat of the feedback, and lower perceived utility of the feedback if they were induced to have an entity ToI and then failed the task. In addition, main effects revealed that participants reported lower perceived control over the feedback and greater regret to seek the feedback in the entity condition than in the incremental condition. Participants also 0 10 20 30 40 50 60 70 Incremental Entity % Preferring Each Goal Induced Theory of Intelligence Learning Goal Performance Goal

PAGE 58

58 reported greater current negative affect and greater regret to seek the feedback in the failure condition than in the success condition. Last, mediation analyses revealed that the effect of task outcome on avoidance in the entity condition was explained by entity participa nts anticipating that they would cope worse and see the feedback as less useful in the failure than in the success condition. Importantly, Study 2a replicated the finding from Study 1 that ToI is related to goal preferences and that goal preferences are re lated to inclination to avoid performance feedback. However, Study 2a failed to replicate the relationship between ToI and inclination to avoid performance feedback. This failure to replicate could be due in part to lower power in Study 2a, or to the fact ToI and this measure should not be as strong.

PAGE 59

59 CHAPTER 4 STUD Y 2B: MEASURED (IMPL ICIT) THEORIES OF IN TELLIGENCE AND RESPO NSES TO FEEDBACK Study 2b tested the same hypotheses tested in Study 2a, except that Study 2b measured rather than manipulated theories of intelligence. Method Participants and Design Participant s were 168 people (72 males, 96 females, age M = 32.9, SD = 12.59) recruited participants were randomly assigned to success vs. failure task outcomes in a between subjects fa ctorial design. Materials and Procedure The materials and procedure for Study 2b were identical to those in Study 2a with one exception. Because theories of intelligence were measured rather than manipulated, participants in all conditions completed the to beliefs about intelligence (one that actually appears on practice GRE exams, see http://www .ets.org/gre/revised_general/prepare/analytical_writing/issue/sample_task ). Consequently, after learning about piloting the GRE essay, I asked participants to write an essay supporting the following statement: As people rely more and more on technology t o solve problems, the ability of humans to think for them selves will surely deteriorate. ToI ( N M = 3.1, SD = 1.54). As in Study 2a, I also combined relevant items to form composites for the Information Avoidance scale (reverse coding when appropriate) ( = .86, M = 2.2, SD = 1.10), for Current Negative affect (after reverse coding c alm and relaxed ) (

PAGE 60

60 = .88, M = 1.7, SD = .72) for Anticipated Difficulty Coping ( M = 2.0, SD = 1.38), and for Perceived Threat posed by the feedback ( M = 1.9, SD = 1.06). Results Preliminary Analyses Information Avoidance Decision. Most participants opted to see detailed feedback about their essay provided by the online GRE analyst. Specifically, 132 participants (78.6%) selected Option 1 (see feedback now), 18 chose Option 2 (receive feedback later via email), and 18 chose Option 3 (dec line to receive feedback). As in Study 2a participants who chose Option 2 or 3 were combined and classified as avoiders. The zero order correlations among all variables appear in Table 4 1 and the cell means of all variables appear in Table 4 2

PAGE 61

61 Table 4 1. Zero order correlations among variables ToI Composite Dichotomized ToI Task Outcome Manipulation Info Avoidance Decision Info Avoidance Scale Goal Preference Perceived Control Current Negative affect Anticipated Negative Affect Anticipated Di fficulty Coping Regret to Seek Regret to Avoid Perceived Threat ToI Composite Dichotomized ToI .90** Task Outcome Manipulation .04 .00 Info Avoidance Decision .01 .03 .02 Info Avoidance Scale .11 .12 .04 .18* Goal Preference .06 .04 .14 .02 .19* Perceived Control .17* .18* .08 .04 .21** .01 Current Negative Affect .08 .01 .21** .01 .18* .03 .06 Anticipated Negative Affect .23** .24** .32** .06 .24** .03 .05 .37** Anticipated Difficulty Coping .12 .08 .13 .18* .40** .01 .25** .26** .33** Regret to Seek .30** .23** .16* .18* .31** .19* .08 .25** .42** .45** Regret to Avoid .13 .13 .01 .19* .24** .02 .35** .23** .09 .13 .06 Perceived Threat .31** .28** .10 .10 .38** .06 .06 .31** .58** .53** .48** .07 Perceived Utility .06 .05 .06 .33** .34** .01 .44** .18* .01 .37** .11 .52** .11 Note. N = 168. Correlations reported are Pearson moment correlations. Correlations for Dichot omized ToI are based on N = 153.** p < .01. p < .05.

PAGE 62

62 Table 4 2. Means for outcome m easures Dichotomized Incremental ToI Dichotomized Entity ToI Success n = 39 Failure n = 45 Success n = 32 Failure n = 37 M SD M SD M SD M SD Info Avoidance Sca le 2.0 (.92) 2.1 (1.05) 2.4 (1.36) 2.2 (.91) Perceived Control of Performance 5.6 (1.50) 5.3 (1.77) 5.0 (1.69) 4.6 (1.98) Current Negative Affect 1.6 (.73) 1.9 (.73) 1.7 (.67) 1.8 (.70) Anticipated Negative Affect 1.8 (1.22) 2.4 (1.49) 2.2 (1 .51) 3.5 (1.79) Anticipated Difficulty Coping 1.6 (1.22) 2.1 (1.60) 1.9 (1.18) 2.2 (1.38) Anticipated Regret to Seek feedback 1.2 (.51) 1.5 (.84) 1.6 (.91) 2.2 (1.94) Anticipated Regret to Avoid Feedback 3.8 (2.90) 3.2 (2.24) 4.1 (2.37) 4.1 ( 2.49) Perceived Threat of Feedback 1.5 (.87) 1.7 (.92) 2.0 (1.01) 2.3 (1.26) Perceived Utility of Feedback 5.6 (1.76) 5.6 (1.49) 5.3 (1.58) 5.5 (1.79) Note. N = 153. Avoidance of Performance Feedback As in Study 2a, I first tested Hypotheses 1 th rough 3 for the primary dependent variable separately. As evident in Table 4 1 and inconsistent with H1, the ToI composite was not significantly correlated with avoidance of feedback In addition participants were no more likely ents in the failure condition (31.4%) than in the success condition 2 (1) = .02, p that task outcome would influence avoidance of detailed feedback differently among different ToI I conducted a logistic regressi on with ToI, Task Outcome, and their interaction. Analyses revealed a nonsignificant ToI x Task Outcome interaction ( b = .03, SE = .25, p = ns, OR = .97), such that the effect of Task

PAGE 63

63 Outcome on feedback avoidance (a null effect) was consistent across the different ToI Together, these findings fail to support Hypotheses 1 through 3. Hypothesis 1 (Effects of ToI) I tested Hypothesis 1 by examining the correlations between ToI and perceived control, current negative affect, anticipated negative affect, ant icipated difficulty coping, regret to seek the feedback, regret to avoid the feedback, perceived threat posed by the feedback, and perceived utility of the feedback. Th e correlations appear in Table 4 1 As predicted and consistent with Study 2a the more s trongly participants endorsed an entity ToI, the lower their perceived control over their writing ability, the greater their anticipated negative affect, the greater their eat they believed the feedback posed. However, inconsistent with predictions, the association between implicit ToI was not significantly related to current negative affect or anticipated regret to avoid (consistent with Study 2a), or to anticipated difficu lty coping with the feedback or perceived utility of the feedback (inconsistent with Study 2a). These findings show mixed support for Hypothesis 1. Hypothesis 2 (Effects of Task Outcome) I tested Hypothesis 2 by conducting a series of t tests to examine m ean differences between the success and failure conditions. The results appear in Table 4 3 Consistent with Hypothesis 1 and with Study 2a compared with participants who succeeded on the task, participants who failed reported greater current negative affe ct, greater anticipated negative affect, greater difficulty coping (marginally significant), and greater anticipated regret to seek feedback. However, contrary to predictions participants experiencing success vs. failure on the task did not differ in antic ipated regret to avoid the feedback (consistent with Study 2a) or in perceived threat (inconsistent with Study 2a). Exploratory analyses further revealed that task outcome did not

PAGE 64

64 influence perceived control (consistent with Study 2a) or perceived utility (inconsistent with Study 2a). Overall, the findings show moderate support for Hypothesis 2. Table 4 3. Hypothesis 2 results: effects of task o utcome Success Failure M SD M SD t p d Perceived Control (of performance) 5.3 (1.63) 5.0 (1.81) 1.08 ns .17 Current Negative Affect 1.6 (0.68) 1.9 (0.70) 2.79 < .01 .44 Anticipated Negative Affect 2.0 (1.41) 3.0 (1.71) 4.38 < .001 .64 Anticipated Difficulty Coping 1.7 (1.18) 2.1 (1.44) 1.74 .08 .30 Anticipated Regret to Seek Feedback 1.4 (.80) 1.8 (1.42) 2.02 < .05 .34 Anticipated Regret to Avoid Feedback 3.8 (2.62) 3.7 (2.39) < 1 ns .04 Perceived Threat (of the feedback) 1.8 (.99) 2.0 (1.11) 1.24 ns .19 Perceived Utility (of the feedback) 5.4 (1.69) 5.6 (1.57) < 1 ns .12 Note N = 16 8. Table 4 4. Hypothesis 3 results: ToI x feedback i nteractions b SE p R 2 2 Perceived Control (of performance) .02 .17 ns .000 .000 Current Negative Affect .04 .07 ns .002 .002 Anticipated Negative Affect .20 .16 ns .01 .01 Anticipated Difficulty Co ping .12 .13 ns .01 .000 Anticipated Regret to Seek Feedback .11 .11 ns .01 .004 Anticipated Regret to Avoid Feedback .25 .25 ns .01 .003 Perceived Threat (of the feedback) .01 .10 ns .000 .001 Perceived Utility (of the feedback) .09 .17 ns .002 .001

PAGE 65

65 Hypothesis 3 (ToI x Task Outcome Interactions) To test Hypothesis 3, I conducted a series of multiple regression analyses on participants responses. The results appear in Table 4 4. Profiles of Seekers vs. Avoiders Because neither theories of intellige nce, task outcome, nor the interaction of the two predicted avoidance of performance feedback, I examined which variables correlated with the decision to seek vs. avoid feedback. As evident in Table 4 1 four variables significantly one only marginally so correlated with information avoidance. Specifically, and consistent with Study 2a, avoidance was associated with greater difficulty coping and lower perceived utility of the feedback. In addition, avoidance was associated with greater anticipated regret to seek the feedback and lower regret to avoid the feedback. Hypothesis 5 (Associations between ToI, Goal Preferences, and the Information Avoidance Scale) Theories of intelligence and inclination to avoid performance feedback As shown in Table 4 1 and i nconsistent with Hypothesis 5 and Study 1, ToI were unrelated to inclination to avoid performance feedback on the Information Avoidance Scale. Theories of intelligence and goal preferences I examined the dichotomous goal preference item (GP1) in which g oal preferences are directly pitted against one another. Inconsistent with expectations or with Studies 1 and 2a, ToI were unrelated to goal preferences (see Table 4 1 ) Also, for illustrative purposes alone, when examining the percentages participants who were classified as incremental theorists were equally likely to adopt a learning goal (56.0%) or a performance goal (44.0%), and participants who were classified as entity theorists were also equally likely to adopt a learning goal (52.2%) or a performanc e goal (47.8%), 2 Of course, the sample size of this study was

PAGE 66

66 much smaller than that of Study 1 which also examined the association between measured ToI and goal preferences and found significant differences and a smaller sample is m ore likely than is a larger sample size to deviate from true differences. Goal preferences and inclination to avoid performance feedback As shown in Table 4 1 and consistent with expectations and Studies 1 and 2a, goal preference significantly correlated with the Information Avoidance Scale such that preference for a good grade (performance goal) was associated with greater inclination to avoid performance feedback. As in Study 2a, however, goal preference did not significantly correlate with the actual a voidance decision. In summary, the findings in Study 2b failed to support the hypotheses for avoidance of performance feedback. However, I found mixed support for Hypotheses 1 and 2. Specifically, endorsement of an entity ToI was related to lower perceive d control, greater anticipated negative affect, greater regret to seek the feedback, and greater perceived threat of the feedback. In addition, a failure outcome was associated with greater current and anticipated negative affect, and greater regret to see k the feedback. Finally, internal analyses revealed that avoiders reported greater regret to seek the feedback, lower regret to avoid the feedback, greater difficulty coping with the feedback, and lower perceived utility of the feedback than did seekers. T he findings of Study 2b further highlight the roles played by difficulty coping and perceived utility of the feedback in avoidance of performance feedback. Finally, Study 2b replicated the finding in Studies 1 and 2a that goal preferences are related to in clination to avoid performance feedback. However, Study 2b failed to replicate the relationship between ToI and inclination to avoid performance feedback or between ToI and goal preferences. Study 2b also replicated the finding in Study 2a that the Informa tion Avoidance Scale significantly correlates with actual avoidance decisions.

PAGE 67

67 CHAPTER 5 DISCUSSION General Discussion In three studies I examined the question of whether theories of intelligence are linked to avoidance of performance feedback. The studie s generally supported the hypothesis that endorsing an entity theory of intelligence prompts a greater avoidance of intelligence feedback, and Study 2a revealed that this relationship occurs primarily when entity theorists experience failure. In addition, analyses revealed that participants tended to avoid feedback when they felt less able to cope with the feedback and when they perceived the feedback to be less useful. Furthermore, inclinations to avoid feedback were higher among participants who preferred a performance goal over a learning goal, and among participants who actually avoided the performance feedback. Not all of the findings were consistent across the studies. First, Study 2b found no significant relationship between implicit ToI and goal pre ferences. Second, Studies 2a and 2b failed to replicate the relationship between ToI and inclination to avoid academic feedback. Third and most notable, Study 2b failed to replicate the hypothesized relationship between ToI, task outcome, and performance f eedback avoidance. Specifically, in Study 2b neither ToI nor feedback outcome nor their interaction influenced feedback avoidance. In addition, whereas Study 2a produced significant ToI x Feedback interactions for anticipated difficulty coping and perceive d utility of the feedback which drove the effects on avoidance neither of these interactions were significant in Study 2b. Despite these null interactions, internal analyses revealed that avoiders reported greater difficulty coping and lower perceived util ity than did seekers. In other words, to the extent that participants felt that they were unable to cope with the feedback and/or perceived the feedback as less useful, they avoided the feedback.

PAGE 68

68 How can I explain these inconsistencies? Insufficient power could explain both the failure of Study 2b to replicate the relationship between ToI and goal preferences and the failure of Study 2a and 2b to find a link between ToI and inclination to avoid academic feedback. Study 1 had over 1200 participants whereas S tudies 2a and 2b had fewer than 400 combined participants. Alternatively, the null relationship in Studies 2a and 2b between ToI and inclination to avoid feedback could also be due to the proximity of the items within the survey. Whereas in Study 1 the ToI items, the dichotomous goal preference item, and the inclination to avoid academic feedback scale appeared in succession, in Studies 2a and 2b the ToI items appeared early on in the survey and dichotomous goal preference item and inclination to avoid acad emic feedback scale appeared very late in the survey (about 20 minutes after completing the ToI scale). If the associations between these items rely to some extent on priming, the distance between the items could conceivably have weakened the associations. Study 2b stood out because its findings were inconsistent with the other two studies. Several reasons for the inconsistency are plausible. First, the null effects of Study 2b could occur if participants did not feel threatened by the failure outcome becau se they did not interpret the writing task as a test of their intelligence. Specifically, Study 2a clearly stated that writing ability ToI scale and failure on the writing task as an indictment of their intelligence. In other words, participants may have perceived the task as neutral and this perception may have nullified the natural effects produced by theories of intelligence. Perhaps people more often conceptualize intelligence as verbal and/or mathematical ability rather than writing ability. If so, failure on a math test would likely generate similar feedback avoidance. This possibility gains credence in Study 1, where the scale meas uring inclination to avoid

PAGE 69

69 academic feedback was about avoiding academic test feedback (not writing). A second explanation could be that the effects of implicit ToI are more subtle than those generated by the ToI manipulation, and strong implicit entity th eorists may not exist in sufficient levels in the population to produce differences in performance feedback avoidance. Although these possibilities do not account for the failure to replicate the effect of task outcome on feedback avoidance, insufficient p ower could again be part of the explanation. A third explanation for the failure to replicate could be that the samples in Studies 1 and 2b are non comparable and that the relationship between ToI and inclination to avoid feedback is sample specific. Impl ications The current research has several implications. First, if ToI influences feedback avoidance only when manipulated and not when measured, are the findings from Study 2a important? 1 I would contend yes. Social psychology examines the influence of sit uational variables on houghts, feelings, and actions. Sometimes situation factors oriented people towards one theory vs. another, and these situationally caused orientations can produce differential feedback avoidance. In fact, given that two thi rds of the Study 1 sample scored as implicit incremental theorists, it is likely that the majority of participants in the entity condition in Study 2a did not naturally possess an entity ToI. Importantly then, inducing these participants to believe that in telligence is unchangeable was enough to make many of them display behavior in line with the Helpless response pattern that entity theorists tend to exhibit following failure. Indeed, it is 1 Although Study 2b failed to replicate the relationship between measured ToI and academic feedback avoidance inclinations, the failed replication may reveal and important caveat in ToI research. To my knowledge, no study has found effects of measured ToI in a non school aged sample. It is poss ible that measured ToI would not predict in an adult sample for two reasons. First, adults who are no longer in school, no longer confront the consequences of ve an implicit ToI, but it does not function in the same way that it does for a child. Second, adults may conceptualize intelligence more complexly or abstractly than do youth. Whereas youth may define intelligence concretely as whether their current readi definition of intelligence, or even question is the meaning of intelligence altogether.

PAGE 70

70 verbal remark, or teasing by performance goal, or a fear mindset when approaching available feedback. Second, the data suggest that avoidance of performance fee dback was a complex rather ToI x Task Outcome interaction was not significant, nor did current negative affect mediate the relationship between task outcome and performance feedback avoidance. Thus, the decision to seek or avoid feedback likely involved a more complex decision as supported by the mediation analyses. Third, the data in Studies 2a and 2b that revealed a positive relationship between the individ ual difference scale that measures inclination to avoid academic feedback and actual avoidance decisions suggests that this scale may serve as a proxy for actual avoidance when measuring actual avoidance is not possible. The correlation also suggests that the relationship between ToI and the information avoidance scale in Study 1 would translate to actual avoidance. However, the correlation between the individual difference scale and actual avoidance should be interpreted cautiously because these two measur es did not always correspond in their relationships with other constructs (i.e. ToI and goal preferences). Fourth, the data replicate past findings that experiencing failure means different things to entity and incremental theorists (Dweck, 2000). Compare d with participants in the incremental condition who experienced failure, participants in the entity condition who experienced failure interpreted the prospect of seeing their performance feedback as more threatening to their self beliefs, as something tha t they could not cope with, as something that would make them feel worse, and as something that was less useful. This pattern is consistent with the Helpless

PAGE 71

71 Response that entity theorists tend to exhibit in response to failure (Baird et al., 2009; Hong et al., unpublished, as cited in Dweck, 2000 ). I propose that feedback avoidance may also be part of the Helpless Response. Fifth, the current findings may shed light on an important problem Dweck calls the achievement gap; that is, incremental theorists ten d to earn higher grades than do entity theorists by middle school and continuing through the co llege years (Robins & Pals, unpublished, as cited in Dweck, 2000 ; Sorich & Dweck, 2000). One explanation for this gap is that entity theorists fail to increase t heir effort to meet increasing difficulty of academic tasks (Dweck, 2000). The current research ostensibly provided participants with an opportunity for feedback, a tool that can help people learn from past mistakes (failure) and minimize future ones, but can initially suggest deficiency. If entity theorists avoid feedback when they fail whereas incremental theorists do not the tool will benefit only some. Thus, the current research suggests that, to the extent that feedback can help people learn from past mistakes and prevent future ones, the achievement gap However, the implications do not apply exclusively to entity theorists. Main effects of task outcome revea led that participants reported the feedback as less useful and were more likely to avoid feedback in the failure condition than in the success condition. This finding is sobering given that people who fail have the most to gain from feedback. Alternative Explanation and Limitations Some might argue that an alternative explanation for avoiding performance feedback is pretation. First, participants who avoided the information reported greater anticipated difficulty coping and greater regret to seek the feedback. Such responses suggest a motivation to avoid rather than merely a disinterest in seeking. Second,

PAGE 72

72 although av the midpoint on the Likert scale, suggesting that they viewed the information to be useful. Furthermore, as just noted participants perceived the feedback as less useful i n the failure reflected something more complex than appraisal of the feedback itself. Rather, perceived utility may reflect an appraisal of the overall cir cumstances of the feedback opportunity, and low perceived utility may equate to perceived defeat or low self efficacy. Indeed, past research shows that failure decreases self efficacy (Bandura, 1977). In addition, perceived utility was lowest among partici pants in the entity condition who experienced failure. Participants in this condition had both an entity ToI (lower perceived changeability) and failure acting against self efficacy. And because researchers find that high self efficacy corresponds with gre ater persistence and effort (Schunk, 2010), it is quite possible that participants who reported lower perceived utility felt too defeated to use the feedback and thus avoided it. I would be remiss if I did not acknowledge the possibility that the relations hip between perceived utility and avoidance of performance feedback is bi directional and may function as a feedback loop. I tested only one side of the relationship. Future research could examine whether any downstream consequences of avoidance of perform ance feedback may include lower perceived utility of feedback. A second alternative explanation regards the effects of the ToI manipulation in Study 2a. It is possible that the manipulation manipulated something other than theories of intelligence, and tha t other thing in turn affected the dependent variables. For instance, the manipulation could have influenced the effort participants invested in the task (conceivably participants in the incremental condition could have invested more effort than did partic ipants in the entity condition in light of their respective essay statements), and this effort made them want to see

PAGE 73

73 their scores. However, I believe such an alternative explanation unlikely given that a) the ToI manipulation produced a large difference on the ToI scale in the entity and incremental temporarily), b) the item measuring effort was unaffected by the ToI manipulation but was sensitive to task outcome, and c) the ToI manipulation produced differences on measures negative affect, difficulty coping, etc. Alternative explanations aside, the studies described here have limitations. First, Study 1 assessed inclination to avoid feedback rather than actual avoidance. It remains to be seen how well the Information Avoidance Scale maps on to actual avoidance. However, the significant correlation between the Information Avoid ance Scale and actual avoidance in Studies 2a and 2b suggests that the two constructs overlap and inclinations to avoid feedback would map onto actual performance feedback avoidance in school. Second, Studies 2a and 2b examined actual avoidance in a non st udent population and it remains unknown whether the findings generalize to students. However, I performed additional analyses on Study 2a to examine whether age moderated the effects on avoidance, and it did not: that is, participants for whom the GRE writ ing task would likely be more relevant (younger participants) showed the same patterns as did older participants for whom the GRE task would be less relevant (older participants). Thus, the responses of this sample are likely similar to those of a student population. Last, Studies 2a and 2b examined actual avoidance of feedback in an artificial rather than an actual test/learning feedback in a real academic p erformance. However, the responses of participants in Study 2a to the questionnaire items suggest emotional and cognitive involvement in the task, the kind of

PAGE 74

74 involvement one might expect from students in an academic setting. Thus this study had high exper imental realism. Likewise, it is possible that avoidance of feedback online would not generalize to avoidance of physically present feedback (e.g. a hard copy). However, schools are increasingly using an online format for delivery of scores and/or feedback so the external validity of the online format is likely fairly high. Future Directions The current research lays the foundation for several other future studies. First, in Studies 2a and 2b, difficulty coping and perceived utility of the feedback predict ed feedback avoidance. Future studies might explore whether increasing coping resources or perceived utility of the feedback might decrease avoidance. Second, I suggested earlier that feedback avoidance may hinder academic performance and/or improvement. F uture research may examine the strength of the relationship between the utilizing feedback and academic achievement and/or improvement. If these studies are successful, they could have broader implications on school policy and/or teacher training to ensure that students understand the utility of effort, and to provide coping resources in the event of failure. A third future direction would be to examine the possibility that merely anticipating failure (rather than first experiencing it) could prompt feedbac k avoidance for entity theorists. If people oriented to an entity ToI can anticipate difficulty coping and defeat from merely anticipating failure, then they should exhibit the same decision patterns as did the participants in Study 2a. Conclusion The c urrent research extends past research on theories of intelligence and their response patterns to failure. Specifically, compared with a success experience, a failure experience increased avoidance of performance feedback only among participants with an ent ity orientation and not an incremental orientation. The research also replicated and extended past research

PAGE 75

75 showing that entity theorists respond particularly poorly to failure in terms of anticipated affect, coping, perceived threat, and perceived utility of feedback. Finally, the findings suggest that information avoidance may be a previously unexamined facet of the Helpless Response pattern.

PAGE 76

76 APPENDIX A THEORIES OF INTELLIGENCE ADULT SCALE Strongly Disagree Strongly Agree You h really do much to change it. 1 2 3 4 5 6 change very much. 1 2 3 4 5 6 your basic intellig ence. 1 2 3 4 5 6

PAGE 77

77 APPENDIX B INFORMATION AVOIDANCE SCALE Please answer the following items. Strongly Disagree Strongly Agree Even if it is going to upset me, I want to find out what grade I got on a school exam 1 2 3 4 5 It is foolish for me to avoid information about how I did on a school exam 1 2 3 4 5 Sometimes it is better that I not know how I did on a school exam 1 2 3 4 5 At times I avoid information about how I did on a school exam 1 2 3 4 5 When it comes to how I am doing in my classes, sometimes ignorance is bliss 1 2 3 4 5 I would never avoid information about how I did on a school exam 1 2 3 4 5 I would rather not know if I did poorly on a school exam 1 2 3 4 5 e I got I might see 1 2 3 4 5

PAGE 78

78 APPENDIX C GOAL PREFERENCES SCALE Strongly Disagree Strongly Agree Although I hate to admit it, I sometimes would rather do well in a class than learn a lot. 1 2 3 4 5 6 1 2 3 4 5 6 than it is to get the best grades. 1 2 3 4 5 6 I f I had to choose between getting a good grade and being Getting a good grade Being Challenged

PAGE 79

79 APPENDIX D CONSENT FORM Informed Consent Please read this consent form carefully before you decide to participate in this study. Purpose of the research study: We are examining how people respond to tasks in order to help future students. What you will be asked to do in the study: You will complete a writing task and a few person ality scales. As in many studies, we may withhold some information until the end of the study. Time required: About 40 minutes. Compensation: You will receive $1.00 for completing this study. Risks and Benefits: There are no risks fr om participating in this study. Confidentiality: Your responses will be confidential to the fullest extent provided by the law. You will not place any identifying information on any questionnaires. You will be assigned a code number, and your resp onses will be stored in a computer according to that code number and not by your name. As such, your name will never be associated with your responses and will not be used in any report. Moreover, all data will be analyzed by group averages and not by in dividual responses. Voluntary participation & right to withdraw: Your participation in this study is completely voluntary. You have the right to withdraw from this study at any time. However, to receive your monetary compensation, you must complet e the survey. Whom to contact if I have questions about the study: Principle Investigator: Corinne Novell , Dept of Psychology, University of Florida. Supervisor: Dr. James A. Shepperd , Dept of Psychology, Univer sity of Florida Whom to contact about my rights as a research participant in the study: UFIRB office, Box 112250, University of Florida, Gainesville, FL 32611 2250; ph 392 0433. By checking the box below I acknowledge that I have read the infor mation and agree to participate in this study. If you do not wish to participate, please close your browser at this time. o I agree to participate

PAGE 80

80 APPENDIX E ADDITIONAL QUESTIONNAIRE ITEMS Not at all difficult Very Difficult How dif 1 2 3 4 5 6 7 I did much worse than others I performed the same as others I did much better than others How do you think you did on the participants who h ave completed the same task? 1 2 3 4 5 6 7 No effort A lot of effort How much effort did you put into the 1 2 3 4 5 6 7 Not at all Somewhat Moderately so Very much so Calm 1 2 3 4 Nervous 1 2 3 4 Anxious 1 2 3 4 Relaxed 1 2 3 4 Tense 1 2 3 4

PAGE 81

81 Strongly Disagree Strongly Agree I believe that seeing the analysts' comments about my essay might make me feel bad (e.g., depressed, sad, angry, etc.). 1 2 3 4 5 6 7 I believe that seeing the analys ts' comments about my essay might require me to take action that I don't want to take (e.g., change my career or future plans, etc.). 1 2 3 4 5 6 7 I believe that seeing the analysts' comments about my essay might threaten an important belief about myself (such as that I'm intelligent or competent). 1 2 3 4 5 6 7 I believe that seeing the analysts' comments about my essay will challenge my belief that I will be successful in life. 1 2 3 4 5 6 7 There are things I can do to improve my writing score should I learn that I did not perform well. 1 2 3 4 5 6 7 It would be threatening to learn that I did not perform well on the "Analyze an Issue" task. 1 2 3 4 5 6 7 It would be useful to see the analyst's comments about my essay. 1 2 3 4 5 6 7 If I saw the an alyst's comments about my essay, I think I would cope pretty well. 1 2 3 4 5 6 7 I personally have what it takes to deal with the analyst's comments about my essay. 1 2 3 4 5 6 7

PAGE 82

82 Please answer the following items. Not at all Very much Imagine that you chose to see the analysts' comments about your essay. How much do you anticipate regretting that decision later? This is a hypothetical question. To answer this question, imagine choosing to see the analysts' comments about your essay in the previous part of the questionnaire, irrespectively of what option you actually chose. 1 2 3 4 5 Imagine that you chose NOT to see the analysts' comments about your essay. How much do you anticipate regretting that decision later? This is a hypothetica l question. To answer this question, imagine choosing not to see the analysts' comments about your essay in the previous part of the questionnaire, irrespectively of what option you actually chose. 1 2 3 4 5 It would be helpful for us to know your SAT s cores (if you remember them). Your answers will NOT be linked to your name ANYWHERE. SAT math score: ________ SAT verbal (critical reading) score: _______ SAT writing score: ________ SAT total: ________

PAGE 83

83 Strongly Disagree Stron gly Agree If I did poorly on a test I would rather not know 1 2 3 4 5 6 7 I would avoid learning if I did poorly on a test 1 2 3 4 5 6 7 Even if it will upset me, I want to know if I did poorly on a test 1 2 3 4 5 6 7 When it comes to knowing if I did poorly on a test, sometimes ignorance is bliss 1 2 3 4 5 6 7 I want to know if I did poorly on a test 1 2 3 4 5 6 7 I can think of situations in which I would rather not know if I did poorly on a test 1 2 3 4 5 6 7 It is important to know if I did poorl y on a test 1 2 3 4 5 6 7 I would want to know if I did poorly on a test immediately 1 2 3 4 5 6 7

PAGE 84

84 APPENDIX F DEBRIEFING FORM Debriefing The study is now over. Thank you for participating in this study. Your responses will be helpful purpose of the experiment for a couple of reason s. First, doing so might affect our results. For example, if we tell people the purpose of the experiment or how we predict people will act in the experiment, they may deliberately do whatever it is they think we want them to do, just to help us out and give us the results that they think we want. If this happened, we would not have a very good indication of how they would act in situations in everyday life. Second, the opposite might occur. That is, if we tell people our predictions, they might delibera them out. Again, our results would be invalid because we would not have a good sample of how people act in everyday life. In this study, we are interested in factors influence whether peopl e want to receive feedback. suspect that people are less w illing to see feedback if they believe that they cannot change or improve on the feedback and when they believe the likelihood of poor performance is high. To explore this suspicion it was necessary that we have some participants believe that intellige nce was changeable and other participants believe that intelligence was not changeable. We did this by having some people write an essay supporting that intelligence is stable and trait like and others write an essay that suggested that intelligence is cha We also tried to have some participants expect positive comments and some participants expect negative comments. We did this by randomly telling some people that they had scored highly on their essay and telling others that they had scored poorly on their essay. The essays you were asked to write were actually made up for the purpose of this study. In fact, the ETS and GRE are not involved in any of this research. This research is being conducted only by resear chers of the UF Psychology department. Also, research shows that effort can increase intelligence and that effort will typically improve performance. For us to draw any conclusions from this study, we must combine the data from you with the data fr om other people. What this means is that it is going to be necessary for us to ask you not to communicate anything about this study to anyone else. The only sure way to make sure that the

PAGE 85

85 hypothesis of the study never reaches a future participant before he /she is in the study is if you, the participant, do not tell others. Please feel free to ask any questions or offer comments or suggestions. How can we improve this study? We are really interested in your feedback. You may contact Dr. James Shepper d, Box 112250, phone number 392 0601 x 248, email, shepperd@ufl.edu, or Corinne Novell, email, cnovell@ufl.edu. If you have any questions about your rights as a research participant, you can contact the IRB office at UFIRB Office, Box 112250, University of Florida, Gainesville, F 32611 2250, ph 392 0433, IRB2@ufl.edu We would like to reiterate how helpful your participation has been in our research, so thanks again!

PAGE 86

86 LIST OF REFERENCES Afifi, W. A., Morgan, S. E ., Stephenson, M. T., Morse, C., Harrison, T., Reichert, T., & Long, S. D. (2006). Examining the decision to talk to family about organ donation: Applying the theory of motivated information management. Communication Monographs, 73, 188 215. Baird, G. L., Scott, W. D., Dearing, E., & Hamill, S. K. (2009). Cognitive self regulation in youth with and without learning disabilities: Academic self efficacy, theories of intelligence, learning vs. performance goal preferences, and effort attributions. Journal of Social and Clinical Psychology, 28, 881 908. Bandura, A. (1983). Self efficacy determinants of anticipated fears and calamities. Journal of Personality and Social Psychology, 45, 464 469. Baron, R. M., & Kenny, D. A. (1986). The moderator mediator variab le distinction in social psychological research: Conceptual, strategic, and statistical considerations. Journal of Personality and Social Psychology, 51, 1173 1182. Baumeister, R. F., Bratslavsky, E., Finkenauer, C., & Vohs, K. D. (2001). Bad is stronger than good. Review of General Psychology, 5, 323 370. Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561 571. Brewer, N. T, Chapman, G. B, Gibbons, F. X, G errard, M., McCaul, K. D., & Weinstein, N. D. (2007). Meta analysis of the relationship between risk perception and health behavior: the example of vaccination. Health Psychology, 26 136 45. echanical Turk: A new source of inexpensive, yet high quality, data? Perspectives on Psychological Science, 6, 3 5. Cai, H., Wu, Q., & Brown, J. D. (2009). Is self esteem a universal need? Evidence from The Asian Journal of Soc ial Psychology, 12, 104 120. Cappelli, M., Surh, L., Humphreys, L., Verma, S., Logan, D., Hunter, A., & Allanson, J. (1999). counseling and testing for breast cancer. Clinical Genetics, 55, 419 430. De la Ronde, C., & Swann, W. B., Jr. (1993). Caught in the crossfire: Positivity and self verification strivings among people with low self esteem. In R. F. Baumeister (Ed.), Self esteem: The puzzle of low self regard. (pp. 147 165) New York, NY, US: Plenum Press. Diener, C. I., & Dweck, C. S. (1978). An analysis of learned helplessness: Continuous changes in performance, strategy and achievement cognitions following failure. Journal of Personality and Social Psychology, 36, 451 462

PAGE 87

87 Diener, C. I., & Dweck, C. S. (1980). An analysis of learned helplessness: (II) The processing of success. Journal of Personality and Social Psychology, 39 940 952. Dunning, D. (1995). Trait importance and modifiability as factors influencing self ass essment and self enhancement motives. Personality and Social Psychology Bulletin, 21 1297 1306. Dweck, C. S. (1975). The role of expectations and attributions in the alleviation of learned helplessness. Journal of Personality and Social Psychology, 31, 674 685. Dweck, C. S. (2000). Self theories: Their role in motivation, personality, and development New York: Psychology Press. Dweck, C. S. (2006). Mindset: The new psychology of success New York: Random House. Dweck, C. S., & Bempechat, J. (1983). & H. Stevenson (Eds.), Learning and motivation in the classroom (pp.239 256). Hillsdale, NJ: Erlbaum. Dweck, C. S., Chiu, C. Y., & Hong, Y. Y. (1995). Implicit theories and their role in judgmen ts and reactions: A world from two perspectives. Psychological Inquiry, 6 267 285. Dweck, C. S., & Elliott, E. S. (1983). Achievement motivation. In P. Mussen and E. M. Hetherington (Eds.), Handbook of child psychology (pp. 643 692). New York: Wiley. Dw eck C.S., & Elliott Moskwa E. (2010). Self Theories: The roots of defensiveness. In J.E. Maddux and J.P. Tangney (Eds.) The Social Psychological Foundations of Clinical Psychology New York: Guilford Press. Dweck, C. S., & Leggett, E. L. (1988). A soci al cognitive approach to motivation and personality. Psychological Review, 95, 256 273. Dykman, B. M. (1998). Integrating cognitive and motivational factors in depression: Initial tests of a goal orientation approach. Journal of Personality and Social Psy chology, 74, 139 158. Elliott, E. S., & Dweck, C. S. (1988). Goals: An approach to motivation and achievement. Journal of Personality and Social Psychology, 54 5 12. Hart, W., Albarracin, D., Eagly, A. H., Brechan, I., Lindberg, M., & Merrill, M. (2009) Feeling validated versus being correct? A meta analysis of selective exposure to information. Psychological Bulletin, 135, 555 588. Henderson, V., & Dweck, C. S. (1990). Achievement and motivation in adolescence: A new model and data. In S. Feldman and G. Elliott (Eds.), At the threshold: The developing adolescence Cambridge, MA: Harvard University Press.

PAGE 88

88 Hong, Y. Y., Dweck, C. S., Lin, D. M. S., & Wan, W. (1999). Implicit theories, attributions, and coping: A Meaning System Approach. Journal of Person ality and Social Psychology, 77, 588 599. Hoorens, V. (1996). Self favoring biases for positive and negative characteristics: Independent phenomena? Journal of Social and Clinical Psychology, 15, 53 67. Katapodi, M. C., Lee, K. A., Facione, N. C., & Dodd M. J. (2004). Predictors of perceived breast cancer risk and the relations between perceived breast cancer risk and breast cancer screening: A meta analytic review. Preventive Medicine, 38, 388 402. Kirsch, I. (1985). Response expectancy as a determinan t of experience and behavior. American Psychologist, 40, 1189 1202. Kunda, Z. (1990). The case for motivated reasoning. Psychological Bulletin 108 480 498. Lee, W. (1971). Preference strength, expected value difference and expected regret ratio. Psycho logical Bulletin 75 186 191. Levy, S., Stroessner, S., & Dweck, C. S. (1998). Stereotype formation and endorsement: The role of implicit theories. Journal of Personality and Social Psychology, 74 1421 1436. Loomes, G., & Sugden, R. (1982). Regret the ory: An alternative theory of rational choice under uncertainty. The Economic Journal, 92, 805 822. Luce, M. F. (1998). Choosing to avoid: Coping with negatively emotion laden consumer decisions. Journal of Consumer Research, 24, 409 433. Lyter, D. W., V aldiserri, R. O., Kingsley, L. A., Amoroso, W. P., & Rinaldo Jr., C.R. (1987). The HIV antibody test: Why gay and bisexual men want or do not want to know their results. Public Health Reports, 102, 468 474. Melnyk, D., & Shepperd, J. A. (2012 ). Avoiding r isk information about breast cancer. Annals of Behavioral Medicine ( in press). Muller, D., Judd, C. M., & Yzerbyt, V. Y. (2005). When moderation is mediated and mediation is moderated. Journal of Personality and Social Psychology, 89, 852 863. Preacher, K. J., Rucker, D. D., & Hayes, A. F. (2007). Addressing moderated mediation hypotheses: Theory, methods, and prescriptions. Multivariate Behavioral Research, 42, 185 227. Roese, N., & Olson, J. M. (1995). Counterfactual thinking: An overview. In N. Roese and J. M. Olson (Eds.), What might have been: The social psychology of counterfactual thinking. Hillsdale, NJ: Erlbaum.

PAGE 89

89 Roth, S., & Cohen, L. J. (1986). Approach, avoidance, and coping with stress. American Psychologist, 38, 813 819. Rotter, J. B. (1954 ). Social learning and clinical psychology NY: Prentice Hall. Schunk, D. H. (1990). Goal setting and self efficacy during self regulated learning. Educational Psychologist, 25, 71 86. Smith, S. M., Fabrigar, L. R., & Norris, M. E. (2008). Reflecting on six decades of selective exposure research: Progress, challenges, and opportunities. Social and Personality Psychology Compass, 2 464 493. Sorum, P. C., Mullet, E., Shim, J., Bonnin Scaon, S., Chasseigne, G., & Cogneau, J. (2004). Avoidance of anticipate d regret: The ordering of prostate specific antigen tests. Medical Decision Making, 24, 149 159. Swann, W. B., Jr., De La Ronde, C. & Hixon, J. G. (1994). Authenticity and positivity strivings in marriage and courtship. Journal of Personality and Social P sychology, 66 857 869. Sweeny, K., Melnyk, D., Miller, W., & Shepperd, J. A. (2010). Information avoidance: Who, what, when, and why Review of General Psychology, 14, 340 353. Trope, Y., Gervey, B., & Bolger, N. (2003). The role of perceived control in overcoming defensive self evaluation. Journal of Experimental Social Psychology, 39 407 419. Yaniv, I., Benador, D., & Sagi, M. (2004). On not wanting to know and not wanting to inform others: Choices regarding predictive genetic testing. Risk Decision and Policy, 9, 317 336.

PAGE 90

90 BIOGRAPHICAL SKETCH Corinne grew up in Somers, New York. At Somers High School, s he graduated in the top 10% of her class in 2003. She then attended The University of Georgia as a Foundation Fellow from 2003 2007, and obt ained a Bachelor of Science in p sycho logy and a Bachelor of Arts in a nthropology. In addition, she received summa cum laude distinction and Highest Honors. Her senior Honors thesis in psychology examined how people react to a subtle ego threat (misspelling partic narcissism levels. at the University of Florida which hers. Now that she has completed her doctorate in psychology she plans to continue a career in research and contribute positively wherever she goes