University of Southern California Dissertations and Theses

Difficulty-as-importance and difficulty-as-impossibility: unpacking the context-sensitivity and consequences of identity-based inferences from difficulty

(USC Thesis Other)

Difficulty-as-importance and difficulty-as-impossibility: unpacking the context-sensitivity and consequences of identity-based inferences from difficulty

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content DIFFICULTY-AS-IMPORTANCE AND DIFFICULTY-AS-IMPOSSIBILITY:
UNPACKING THE CONTEXT-SENSITIVITY AND CONSEQUENCES OF IDENTITYBASED INFERENCES FROM DIFFICULTY
By
Alysia Elizabeth Burbidge
A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
PSYCHOLOGY
May 2024
Copyright 2024 Alysia Elizabeth Burbidge

ii
Acknowledgments
First, I would like to express my deepest appreciation to my advisor, Daphna Oyserman,
for guiding and mentoring me through the past six years. Her invaluable advice, feedback, and
encouragement have helped me become a researcher and mentor in my own right. It has been a
privilege to learn from her and collaborate with her. I look forward to continuing to work
together for years to come.
Next, I thank my committee members: Mary Helen Immordino-Yang, John Monterosso,
Arthur Stone, and Sarah Townsend. Undoubtedly, their suggestions have improved the quality of
the work presented here. Their support and unique insights have inspired me to expand my
perspectives as a scholar.
I also give special thanks to prior members of the USC Mind and Society Center who
helped make the work presented here possible. I thank Oliver Fisher, David Newman, Casey
O’Donnell, and Gulnaz Kiper who contributed to this work with their theoretical ideas, data
collection efforts, and statistical knowledge. Additionally, I extend my thanks to past and current
lab mates who have graciously given their time and energy to support my intellectual endeavors.
Lastly, I would be remiss in not mentioning my partner, Jake Morales, for his endless
love, selflessness, and encouragement while I have pursued my dreams. No words can fully
express my gratitude for his unwavering belief in my ability to succeed and accomplish this
monumental achievement.

iii
Table of Contents
Acknowledgments........................................................................................................................... ii
List of Tables.................................................................................................................................. iv
List of Figures................................................................................................................................. v
Abstract.......................................................................................................................................... vi
Introduction..................................................................................................................................... 1
Chapter 1: Believing that Difficulty Signals Importance Improves School Outcomes by
Bolstering School-Focused Possible Identities............................................................................. 11
Abstract..................................................................................................................................... 11
Method ...................................................................................................................................... 21
Results....................................................................................................................................... 26
Discussion................................................................................................................................. 37
Chapter 2: As Inferences from Difficulty Shift, So Do Action and Self-Regard.......................... 42
Abstract..................................................................................................................................... 42
Study 1 ...................................................................................................................................... 48
Studies 2a and 2b ...................................................................................................................... 51
Study 3 ...................................................................................................................................... 57
Studies 4 to 6............................................................................................................................. 61
General Discussion ................................................................................................................... 74
Discussion..................................................................................................................................... 81
References..................................................................................................................................... 85
Appendices.................................................................................................................................. 101
Appendix A: Chapter 1 Additional Tables.............................................................................. 101
Appendix B: Chapter 2 Measures........................................................................................... 104
Appendix C: Chapter 2 Multilevel Equations......................................................................... 107

iv
List of Tables
Chapter 1
Table 1: Path Analysis Step 1: Academic Outcomes and School-Focused Possible Identity
Certainty.......................................................................................................................................33
Table 2: Path Analysis Step 2: Academic Outcomes, School-Focused Possible Identity
Certainty, Metacognitive Inferences from Difficulty, and Culture-Based Beliefs......................34
Chapter 2
Table 1: Study 3 Scale Means, Standard Deviation (SD), and α Reliability ...............................59
Table 2: Studies 4 to 6 Sample Size and Characteristics by Study and Pooled...........................61
Table 3: Studies 4 to 6 Pooled Scale Means, Standard Deviation (SD), and α Reliability..........65
Table 4: Studies 4 to 6 Pooled Trait Measure and Day Measure Correlations............................66
Table 5: Studies 4 to 6 Number of Daily Reports and Intraclass Correlations (ICCs)................67
Appendix A: Chapter 1 Additional Tables
Table A1: Age, Gender (% Female), and Proportion of Sample in Each School at Each Time
Point.............................................................................................................................................101
Table A2: Scale Means (M), Standard Deviations (SD), and α Reliability at Time 1, Time 2,
and Time 3 ...................................................................................................................................101
Table A3: Snapshot of Sample Descriptive Results: Means and Standard Deviations (SD) ......102
Table A4: Bivariate Correlations of the Self and Motivation Scales at Each Time Point...........102
Table A5: Intra-Measure Correlations Between T1 and T2 and Between T2 and T3.................103

v
List of Figures
Chapter 1
Figure 1: Theoretical Process Models..........................................................................................19
Figure 2: Data Collection Timeline .............................................................................................20
Figure 3: Path Analysis Step 2: Controlling for Academic Achievement, Difficulty-AsImportance and Possible Identity Certainty are Recursively Related. Other Associations are
Not Recursive...............................................................................................................................30
Chapter 2
Figure 1: Example Item Screen....................................................................................................52
Figure 2: Difficulty-as-Impossibility Endorsement Increases in Difficult Contexts and
Decreases in Easier Ones, While Difficulty-as-Importance Endorsement Does Not Shift .........56
Figure 3: Comparison of ICCs for Difficulty-as-Importance and Difficulty-as-Impossibility
Endorsement Among High School and Pre-High School Students.............................................60
Figure 4: Difficulty Mindset Endorsement Distributions Across 14 Days..................................65
Figure 5: Comparison of ICCs for Difficulty-as-Importance and Difficulty-as-Impossibility
Endorsement Across Studies and Pooled.....................................................................................67
Figure 6: Trait Difficulty-as-Importance and Not Trait Difficulty-as-Impossibility Predicts
Daily Meaningful Engagement with School................................................................................69
Figure 7: Self-Regulatory Means are Less Positive When Today’s
Difficulty-as-Impossibility is Higher and More Positive When Today’s
Difficulty-as-Importance is Higher..............................................................................................70
Figure 8: Predicting Daily Self-Regard and Meaningful Engagement From Trait-Level
Difficulty Beliefs, State-Level Difficulty Belief Instability, and Their Interaction Effects........73

vi
Abstract
When tasks or goals feel hard to think about or do, people can draw two inferences: “This
is important to me, worthwhile” (difficulty-as-importance) and “This is not for me, a waste of
time” (difficulty-as-impossibility). Identity-based motivation theory predicts that people’s beliefs
in each inference vary between persons and between situations, but studies to date have not
examined how or why beliefs fluctuate and the extent to which fluctuations matter for selfregulation during task or goal pursuit. In this dissertation, I present two chapters that address
these research gaps in unique ways. In Chapter 1, I present evidence of a recursive relationship
between difficulty-as-importance beliefs, academic outcomes, and possible identity certainty.
These results suggest difficulty-as-importance beliefs support goal progress by boosting possible
identity certainty which in turn affects goal-directed behavior. Doing well on a school placement
test also minimizes difficulty-as-impossibility beliefs, but this shift does not further benefit goal
progress. In Chapter 2, I document that difficulty beliefs are shaped by both features of contexts
and individuals. Difficulty-as-importance and difficulty-as-impossibility beliefs are not sensitive
to the same situational cues, and both trait and state beliefs are consequential. Trait difficulty-asimportance beliefs predict meaningful schoolwork engagement while trait difficulty-asimpossibility beliefs predict a preference to avoid difficulty. Daily difficulty-as-impossibility
beliefs relate negatively to daily self-regard (goal self-efficacy, self-esteem, and selfcompassion), while daily difficulty-as-importance beliefs relate positively but to a weaker extent.
Taken together, this dissertation addresses how difficulty-as-importance beliefs increase selfregulatory focus on ends (making goal progress) and difficulty-as-impossibility beliefs increase
focus on means (the process of goal progress).

1
Introduction
People can and often do encounter difficulties while thinking about or engaging with
tasks and goals. Identity-based motivation theory (Oyserman, 2007, 2009) predicts that when a
task or goal feels hard to think about or do, people can make metacognitive inferences about
what this feeling of difficulty implies. People can infer to varying degrees that the task or goal is
self-relevant and important to them (a difficulty-as-importance interpretation) and that it is selfirrelevant and a waste of their motivational energy (a difficulty-as-impossibility interpretation).
Research shows people’s interpretations matter for how they perceive themselves, how they
prefer to act, and what they focus on during task or goal pursuit. Experimental research has
largely focused on the effects when either a difficulty-as-importance or difficulty-asimpossibility interpretation is “on the mind,” accessible at the moment of judgment. These
studies find that which interpretation is momentarily more accessible affects perceptions and
actions (e.g., Aelenei et al., 2017; Choi & Oyserman, 2024; Smith & Oyserman, 2015). On the
other hand, correlational research has focused on individual differences in difficulty-asimportance and difficulty-as-impossibility scale endorsement scores. These correlational studies
find that individual differences in how much people endorse each interpretation are associated
with self-perceptions and self-regulatory focus (e.g., Kiper et al., 2024; O’Donnell et al., 2023;
Yan et al., 2023). Research to date has not examined daily fluctuations in how much people
endorse difficulty-as-importance and difficulty-as-impossibility, how these fluctuations in beliefs
matter for self-regulation, and the extent to which situated (state-like) beliefs relate to selfregulation in similar ways compared to previously documented effects and associations of
between-person (trait-like) beliefs and situated accessibility. In this dissertation, I address these

2
gaps using a diverse set of methodological approaches and extend our understanding of the
distinct self-regulatory profiles of difficulty-as-importance and difficulty-as-impossibility beliefs.
Identity-Based Motivation and Interpretations of Difficulty
Identity-based motivation (IBM) theory is a situated social cognition theory of
motivation, goal pursuit, and self-regulation (Oyserman, 2007, 2009). IBM theory conceptualizes
difficulty-as-importance and difficulty-as-impossibility as two distinct interpretations of
difficulty with unique self-regulatory consequences rather than as flip sides (Fisher & Oyserman,
2017). Though the nomenclature is unique to IBM theory, these interpretations are familiar
concepts that exist in everybody’s knowledge base. A person with fitness goals may try to psych
themselves up with some pep talk like “no pain, no gain” (difficulty-as-importance) while a
person struggling with a math exam may think “this is too hard–I’m just not a math person!”
(difficulty-as-impossibility). These interpretations are mentally available constructs that guide
people’s judgements and behavior when they are accessible. As outlined by accessibility theory
(Higgins, 1996; Higgins & Bargh, 1987), an accessible construct is a construct that is “on the
mind” and therefore more likely to be used in subsequent judgments and decisions. IBM theory
posits that while people may vary in the extent to which a difficulty-as-importance and
difficulty-as-impossibility interpretation is more chronically accessible (trait-like), each
interpretation is context-sensitive (state-like) such that features of situations facilitate or hinder
the accessibility of each construct (Oyserman, 2007). Critically, a difficulty-as-importance or
difficulty-as-impossibility interpretation can be accessible and applied even if it is not explicitly
endorsed. For instance, although a difficulty-as-impossibility interpretation is chronically
accessible among Americans, Americans also tend to explicitly reject it (O’Donnell et al., 2023).
In the following sections, I will briefly summarize how the chronic accessibility, momentary

3
accessibility, and explicit endorsement of each interpretation relates to task preferences,
behavior, and self-perceptions that bear on self-regulation.
Accessibility of Difficulty-as-Importance and Difficulty-as-Impossibility As Interpretations of
Difficulty
IBM theory predicts that how people interpret their experiences of difficulty is both
culture-based and situated. The chronic accessibility of each difficulty interpretation should
differ depending on the culture in which a person is immersed. More culturally normative
associations will be more fluent, come to mind more easily, and feel truer (Oyserman, 2017; Yan
& Oyserman, 2018). Indeed, research finds that the chronic accessibility of each difficulty
interpretation varies across cultures (O’Donnell et al., 2021). Among people in the U.S., a
difficulty-as-impossibility interpretation is more chronically accessible than a difficulty-asimportance interpretation; among people in India or China, both interpretations seem to be
equally accessible. These culture-based differences in accessibility reflect the cultural milieu.
While chronic accessibility captures the extent to which a particular interpretation is
likely to be the default interpretation on the mind (i.e., applied in a trait-like manner), situated
accessibility should not be overlooked. Situated, context-sensitive accessibility is a feature of
adaptive self-regulation. When people experience difficulty while thinking about or engaging
with a task or goal, they may either keep going or stop and switch to something else. If difficulty
implies importance, then they should persist. If difficulty implies impossibility, they should
disengage and switch to something else. While premature disengagement can have adverse
consequences such as opportunity and reputational costs (Doody, 2019), failing to disengage
from an unattainable goal can negatively affect self-regulation by undermining subjective well-

4
being (Wrosch & Scheier, 2020). Thus, people should be sensitive to situational cues about what
difficulty implies, as this guides which course of action they are likely to take.
Experimental research shows that priming methods such as autobiographical recall can
cue momentary shifts in the accessibility of a difficulty-as-importance or difficulty-asimpossibility interpretation. When a difficulty-as-impossibility interpretation is made accessible,
students are less likely to perceive academic success as central to their identities (Smith &
Oyserman, 2015) and more likely to perceive school tasks as effortful (Oyserman et al., 2018).
When a difficulty-as-importance interpretation is made accessible, students are more certain of
attaining their school-focused possible identities (Aelenei et al., 2017) and spend more time on
school tasks (Smith & Oyserman, 2015). Together, these studies suggest that the situated
accessibility of a difficulty-as-importance or difficulty-as-impossibility interpretation can affect
self-regulation by influencing how people perceive their own identities, how they perceive an
identity-relevant task, and how they behave on an identity-relevant task.
Endorsements of Difficulty-as-Importance and Difficulty-as-Impossibility as Interpretations of
Difficulty
While experimental research has focused on the situated (state-like) accessibility of
people’s difficulty interpretations, correlational research has focused on between-person (traitlike) differences in people’s explicit endorsements. Using traditional survey methods, this
correlational research documents associations between people’s once-measured difficulty-asimportance and difficulty-as-impossibility endorsement scores and various perceptions and
preferences that relate to self-regulation. Some prior research suggests that endorsement matters
separately from accessibility. Elmore and colleagues (2016) found that students with an
accessible difficulty-as-importance interpretation did not necessarily do better on difficult logic

5
problems compared to students with an accessible difficulty-as-impossibility interpretation. In
fact, students with an accessible but lowly endorsed difficulty-as-importance interpretation
performed just as poorly as students with an accessible but highly endorsed difficulty-asimpossibility interpretation. The reverse was also true; students with an accessible but highly
endorsed difficulty-as-importance interpretation performed just as well as students with an
accessible but lowly endorsed difficulty-as-impossibility interpretation.
Correlational research finds that students who believe more in a difficulty-as-importance
interpretation are more likely to perceive time as an expandable resource while students who
believe more in a difficulty-as-impossibility interpretation are more likely to perceive time as a
limited resource (Choi & Oyserman, 2024). When beliefs of each interpretation are
simultaneously assessed, such that they are entered as separate variables in the same statistical
models, beliefs separately contribute to the amount of variance explained in self-regard and task
engagement. That is, students who endorse a difficulty-as-impossibility interpretation perceive
less effortful means of goal attainment as effective, controlling for their difficulty-as-importance
beliefs which do not predict preferences for goal attainment means (Kiper et al., 2024). Across
samples in Australia, Canada, China, India, Iran, the U.K., the U.S., and Turkey, difficulty-asimportance beliefs are associated with perceiving oneself as more conscientious and difficultyas-impossibility beliefs are associated with perceiving oneself as less conscientious, controlling
for each other (Yan et al., 2023).
Interpretations of Difficulty and Self-Regulatory Focus
The goal literature has long theorized about and examined self-regulation “ends” and
“means.” “Ends” refers to the goals that people are pursuing and hope to achieve. For example,
the achievement orientation literature (Dweck, 1986; Hulleman et al., 2010) concerns itself with

6
the types of goals students pursue, namely goals that pertain to meeting a standard (performance
goals) and goals that pertain to attaining skills and proficiencies (mastery goals). Similarly,
expectancy-value theory (Wigfield & Eccles, 2000) posits that people pursue goals that they
deem to strike the right balance of potential payoff (value) and likelihood of success
(expectations). “Means'' refers to the process of goal pursuit. Researchers have identified various
psychological processes and behaviors that affect goal pursuit. Gollwitzer (1999) highlights
implementation intentions as if-then strategies that people may use during the self-regulation of
goal pursuit. Brandstätter and colleagues (2013) identify action crises as internal conflicts people
may encounter during goal pursuit that may lead to goal disengagement or recommitment. Selfregard can also bear on the process of goal pursuit. Among students who recently failed to attain
a good grade on a recent exam, greater self-compassion was associated with greater use of
adaptive emotion-focused coping strategies (Neff et al., 2005). Similarly, self-esteem may shape
goal pursuit by influencing one’s threshold for experiencing failure. When students were asked
to repeatedly attempt a rigged-to-fail task that purportedly measured cognitive skill but given the
option to switch to a different task after each failure, students with high self-esteem were more
likely to switch than those with lower self-esteem (Di Paula et al., 2002).
Only one paper (Kiper et al., 2024) has considered how the ways that people interpret
difficulty during goal pursuit may shape their preferred means of goal attainment. This
dissertation aims to extend this prior research by considering how people’s difficulty
interpretations may generally shape self-regulatory focus on ends or means. Taking the results
from prior experimental and correlational research together, regardless of whether studies
examined effects of accessibility or endorsement, we see that each difficulty interpretation is
associated with a distinct self-regulatory focus. A difficulty-as-importance interpretation is

7
primarily associated with a focus on ends–the task or goal that one wants to achieve. When
difficulty is interpreted as a signal of importance, school-focused possible identity certainty
increases, time investment in school tasks increases, time feels like something one can make
more of, and there’s no preference for how one goes about attaining their goals. A difficulty-asimpossibility interpretation is primarily associated with a focus on means–the process of
engaging with one’s tasks or goals. When difficulty is interpreted as a signal of impossibility,
school tasks feel more effortful to work on, time feels more limited, and there’s a preference to
attain goals with less effort.
A Methodological Gap
Students achieve better academic outcomes after participating in interventions designed
to encourage the application of a difficulty-as-importance interpretation (Destin et al., 2018;
Horowitz et al., 2018; Oyserman et al., 2002; Oyserman et al., 2006; Oyserman et al., 2021) and
discourage the application of a difficulty-as-impossibility interpretation (Horowitz et al., 2018;
Oyserman et al., 2002; Oyserman et al., 2006; Oyserman et al., 2021). For example, the six-week
Pathways-to-Success intervention (Horowitz et al., 2018; Oyserman et al., 2021) includes
multiple activities that build on each other to help students practice applying a difficulty-asimportance interpretation when faced with a problem. In theory, participation in these activities
should make a difficulty-as-importance interpretation more chronically accessible than a
difficulty-as-impossibility interpretation. Yet, there is a mismatch in the theorized effects and the
measurement of these effects. Despite aiming to increase chronic accessibility, researchers
evaluate “changes” in difficulty interpretations by assessing differences in belief scores from preintervention to post-intervention. This discrepancy is indicative of a larger methodological gap.

8
Altogether, research to date has primarily focused on two ways in which difficulty-asimportance and difficulty-as-impossibility interpretations matter for self-regulation.
Experimental research has focused on effects from situated accessibility–the extent to which the
accessibility of each difficulty interpretation is context-sensitive and hence can be made to be
“on the mind” to affect at-hand perceptions and behavior. Correlational research has focused on
associations with individual difference belief scores–the extent to which a person’s explicit
beliefs in each difficulty interpretation are associated with their perceptions and preferences.
Research to date has not yet examined situated differences in difficulty beliefs. That is, the extent
to which a person’s explicit beliefs in each difficulty interpretation are also situated and contextsensitive. Consequently, prior research has also not examined how the context-sensitive
fluctuations in these beliefs may matter for self-regulation.
Addressing these research gaps is a much-needed next step that can advance our
understanding of how difficulty interpretations factor into the complex process of self-regulation.
Specifically, the studies presented in this dissertation span various methodological approaches
that together can help close the existing methodological gap by answering three questions. First,
are difficulty-as-importance and difficulty-as-impossibility beliefs more trait-like (differing
between persons) or are they both state-like and trait-like (differing between persons and across
situations)? Second, how do situated difficulty-as-importance and difficulty-as-impossibility
beliefs each uniquely relate to self-regulation? Third, to what extent do the relationships between
daily difficulty beliefs and self-regulation align with or differ from prior research on selfregulation and situated accessibility and between-person beliefs?

9
Overview of Dissertation
In Chapter 1, I aim to address the highlighted research gaps by leveraging a three-timepoint, two-month study during a high-stakes school placement process. I examine the potential
roles of difficulty-as-importance and difficulty-as-impossibility beliefs in moderating recursive
effects between academic outcomes and possible identity certainty. IBM theory predicts that
goal-directed behavior is more likely when people feel certain of attaining their possible
identities, and people are more likely to feel certain of attaining their possible identities when
they interpret difficulty as a signal of importance and self-relevance and not as a signal of
impossibility and self-irrelevance (Oyserman & Horowitz, 2023). The reverse should also be
true, such that possible identity certainty fosters interpreting relevant difficulty as a signal of
importance. Research documents associations between difficulty-as-importance and identity
certainty (Aelenei et al., 2017; Smith & Oyserman, 2015) and between identity certainty and
academic outcomes (O’Donnell & Oyserman, 2023) but this evidence exists as disjointed pieces.
By evaluating repeated measurements of difficulty beliefs and possible identity certainty during a
high-stakes school transition, we can better understand how indicators of goal progress (i.e.,
placement test results) may shift beliefs which in turn affect future goal progress by shaping
possible identity certainty.
In Chapter 2, I use additional methodological approaches to provide a fuller explanation
of how features of situations may cause fluctuations in difficulty beliefs and how these
fluctuations matter for self-regulatory focus. I aim to address these gaps across six studies with
over 1,500 participants. First, I build on prior work by Kiper and colleagues (2024) by examining
whether difficulty-as-impossibility endorsers have insight into their preference for easier means
of task engagement. Second, I go beyond prior experimental research documenting situated

10
effects of construct accessibility (Aelenei et al., 2017; Smith & Oyserman, 2015) and
correlational research documenting individual differences in endorsement (Kiper et al., 2024;
O’Donnell et al., 2022; Yan et al., 2023) by employing experimental designs and multilevel
analytic approaches to capture and quantify situated differences in endorsement. Lastly, I present
novel results from daily diary studies comparing how individual differences and natural, daily
fluctuations in difficulty-as-importance and difficulty-as-impossibility beliefs are associated with
self-regard and action. Specifically, I evaluate how difficulty-as-importance endorsement relates
to a focus on making goal progress via meaningful engagement with schoolwork and how
difficulty-as-impossibility endorsement relates to a focus on self-regulatory means of goal-self
efficacy, self-esteem, and self-compassion.
By advancing our understanding of the situated nature of difficulty-as-importance and
difficulty-as-impossibility beliefs, this work can shed light on how difficulty beliefs vary
between persons and between situations and shape self-regulation. I conclude this dissertation
with a summary of how these results contribute to theory development, touching on implications
for practical applications, and suggestions for future research.

11
Chapter 1: Believing that Difficulty Signals Importance Improves School Outcomes by
Bolstering School-Focused Possible Identities
Burbidge, A., Zhu, S., Cheung, S. H., & Oyserman, D. (2024). Manuscript submitted for
publication.
Abstract
Identity-based motivation theory predicts that how well students do in school and how
certain they are of attaining their school-focused possible identities recursively interact with what
they infer when school tasks and goals feel hard to think about or do. Recursivity implies
bidirectionality, but studies to date focus on unidirectional paths. To address this gap, we devised
a three-time-point, two-month study during a secondary school transition (Chinese students
N=818, Mage=12, 44% female). We obtained prior grades (T0) and placement test scores (T1+)
and asked whether secondary school placement fit student preference (T2+). Students filled out
questionnaires at critical junctures: a month before the placement test (T1), the day test results
were announced (T2), and after learning their school placement (T3). Structural equation model
results show both recursive and unidirectional paths. The recursive paths are between schoolfocused possible identity certainty and belief in difficulty-as-importance (inferring value when
schoolwork feels hard to think about or do) and school achievement; the unidirectional path is
between identity certainty and belief in difficulty-as-impossibility (inferring self-irrelevance
when schoolwork feels hard to think about or do). T1 school-focused possible identity certainty
boosted T1+ placement test scores and T2 difficulty-as-importance (controlling for T0 grades,
T1 difficulty-as-importance respectively); T1 difficulty-as-importance boosted T2 schoolfocused possible identity certainty (controlling for T1 school-focused possible identity certainty).
T1+ test scores boosted T2 difficulty-as-importance; T2 difficulty-as-importance boosted T3

12
school-focused possible identity certainty. Lower T1 school-focused possible identity certainty
carries over to higher T2 difficulty-as-impossibility (controlling for T1 difficulty-asimpossibility) and this process recurs from T2 to T3.
Keywords: possible self; motivation; metacognition; culture; identity; primary school transition

13
According to identity-based motivation (IBM) theory, people’s certainty of attaining their
possible identities varies across situations (Oyserman, 2007; 2009). Moreover, IBM theory
predicts that when thinking about or working on tasks and goals feels hard, people can use that
feeling of difficulty to draw inferences about themselves. Hence, when people interpret difficulty
as a signal of importance and self-relevance, they may also feel more certain about attaining their
relevant possible identities (Oyserman & Horowitz, 2023). The reverse should also be true.
When people feel more certain of attaining a possible identity they may also interpret relevant
difficulty as a signal of importance. A recent review of the voluminous literature on possible
future selves and identities (Oyserman & Horowitz, 2023) documents a variety of conditions
under which an accessible future self matters for behavior, including academic outcomes, but it
also highlights a dearth of research on the recursive association between people’s future
identities and the inferences they make from difficulty. To address this gap, we applied IBM
theory to examine school-focused possible identity certainty during an ecologically valid time:
the primary-to-secondary school transition. We focus on certainty because research suggests that
just having a school-focused possible identity is not enough to impact academic outcomes
(Oyserman, 2013), but that certainty of attaining school-focused possible identities is associated
with later academic success (O’Donnell & Oyserman, 2023). We focus on the recursive
relationship between certainty and academic outcomes to contribute to the literature on
associations between self-beliefs and school motivation (Jansen et al., 2022) and school
attainment (Huang, 2011; Valentine et al., 2004; for reviews, see Brandstätter & Bernecker,
2022; Oyserman et al., 2012).
High-Stakes School Transitions in China and Around the World
Students in China attend six years of primary school and three years of junior secondary

14
school, after which about half continue to senior secondary school and the others enter the
workforce directly (Bush et al. 1998). In China and elsewhere, secondary school placement is
determined by test scores and other factors that are more within a student’s control and by
lotteries, quotas, and other factors that are not (Lai et al., 2011; Liu, 2015; OECD, 2012). Highstakes tests are used to allocate competitive high school placement even in the United States
(e.g., the High School Placement Test in San Francisco and the Specialized High Schools
Admissions Test in New York).
Students who are invested in schoolwork are likely to achieve better outcomes than
students who are not, and students who do well in school are more likely to see schoolwork as
self-relevant (for reviews, see Brandstätter & Bernecker, 2022; Oyserman et al., 2012; for metaanalyses, see Huang, 2011; Jansen et al., 2022; Valentine et al., 2004). Not only is the
relationship between investment and achievement likely to be bidirectional, but Chinese students
are likely to feel pressure to do well in school. For Chinese primary students, strong placement
test scores can help students get into “key” junior secondary schools to prepare for higher
education and increase their odds of attending elite colleges (Liu & Dunne, 2009; Ye, 2015). The
implication is that students in high-stakes testing contexts experience grades and test scores as
outcomes not to be taken lightly. In the next section, we describe two culture-based beliefs that
Chinese students may use to make sense of difficulties experienced during high-stakes school
transitions.
Culture-Based Beliefs: Optimism and Accepting Fate
Chinese culture emphasizes education as the path to opportunity (Chen et al., 1996; Zuo
et al., 2020) and fate as predetermining life outcomes (Liang et al., 2008; Zhang et al., 2000).
Experiencing difficulty is seen as a necessary part of the path to success, an idea described as

15
“eating bitter” (Loyolka, 2012). As such, people can accept their fate (Liang et al., 2008; Zhang
et al., 2000) and they can experience optimism about their future possibilities (Purol & Chopik,
2021) in the face of hardships. Students may apply culturally available inferences of optimism
for the future and accepting fate to make sense of the difficulties they experience during highstakes school transitions. Empirically, having optimism for the future seems to benefit students
while accepting fate seems to be detrimental. Accepting fate is associated with worse school
outcomes–lower academic aspirations among Chinese students (Leung et al., 2010), fewer
school-focused possible identities among Chinese students (Bi & Oyserman, 2015), less school
engagement among Australian students (Horstmanshof & Zimitat, 2007), and worse grades
among American students (Mello & Worrell, 2006). Optimism for the future is associated with
better academic outcomes (meta-analysis of mostly Western college students, Richardson et al.,
2012; in China, Chen et al., 2020) and having more positive possible identities (Carver et al.,
1994; Meevissen et al., 2011; Peters et al., 2010).
Identity-Based Motivation Theory
In the prior section, we considered accepting fate while maintaining optimism for the
future and having school-focused possible identities as supporting school performance. In this
section, we broaden the focus to consider identity-based motivation theory, which posits a
recursive relationship between certainty and inferences from metacognitive experience. Certainty
of attaining one’s possible identities is expected to vary across situations (Oyserman, 2007;
2009) and this certainty is expected to shape what people infer from their metacognitive
experiences of difficulty while thinking about or engaging with self-relevant tasks or goals
(Oyserman & Horowitz, 2023). If a task feels hard to think about or do, people can think ‘no
pain, no gain, this is valuable for me’–a difficulty-as-importance inference (Fisher & Oyserman,

16
2017). At the same time, people can also think ‘this task feels hard, maybe it is not for me’–a
difficulty-as-impossibility inference (Fisher & Oyserman, 2017). People vary in how much they
endorse each inference from difficulty (Fisher & Oyserman, 2017; O’Donnell et al., 2023).
Moreover, controlling for how much people endorse difficulty-as-importance, higher difficultyas-impossibility predicts preference for low effort means of goal attainment (Kiper et al., 2024).
In contrast, controlling for how much people endorse difficulty-as-impossibility, higher
difficulty-as-importance does not consistently predict preference for easier or more effortful
means of goal attainment (Kiper et al., 2024). The implication may be that difficulty-asimpossibility is associated with preference for easier means of attaining school-focused possible
identities while difficulty-as-importance is associated with certainty of attaining school-focused
possible identities. Whether this is the case is not yet clear.
Indeed, a comprehensive review of the literature on the link between future selves and
behavior reveals conflicting results (Oyserman & Horowitz, 2023). The review revealed three
seemingly disjointed literatures focused on different ways of imagining a future self and how this
affects behavior. The first literature focuses on self-continuity, the effects of including the future
self in the current self. The second literature focuses on self-contrast, the effects of feeling
efficacious about overcoming the gap between the current and the future self. The third literature
focuses on other aspects of future selves including their valence and link to strategies. Each
literature provides supporting evidence but cannot sufficiently explain why supporting evidence
also exists for the alternative models. If what matters is self-continuity, why would imagining a
gap between the current and future self sometimes trigger future-focused behavior? Similarly, if
either self-continuity or self-contrast matters, why would features of the imagined future identity
sometimes trigger future-focused behavior? To make sense of these heterogeneous findings, the

17
authors of the review apply identity-based motivation theory to suggest that two typically
unmeasured features of future selves matter for behavior. First, people’s certainty of attaining a
particular future identity in the moment. Second, how people make sense of experiences of
difficulty while thinking about or working on that identity (Oyserman & Horowitz, 2023).
Results from prior experimental studies focusing on unidirectional effects are in line with
this proposal. When researchers vary whether students have a difficulty-as-importance or
difficulty-as-impossibility belief are on the mind, they find the expected shifts in student
certainty of attaining their school-focused possible identities (Aelenei et al., 2017; Smith &
Oyserman, 2015). Students with difficulty-as-impossibility on the mind report that the same
difficult standardized test was harder and perform worse on it (Oyserman et al., 2018). Research
examining the association between reported boredom during demanding test situations and
poorer performance suggests that American students (and teachers) may lack the vocabulary to
describe the inferences they draw from their metacognitive experiences of difficulty (Goetz et
al., 2023). Given the right terminology, they may instead label their feelings of boredom as
difficulty-as-impossibility–the sense that it is not for them, and hence a waste of their time.
Beliefs in difficulty-as-importance and difficulty-as-impossibility may be differentially
accessible across countries, as evidenced in comparisons between people from the U.S., India,
and China (O’Donnell et al., 2023). Cross-cultural comparisons conducted by Yan and
colleagues (2023) suggest that people from Western countries are more likely to endorse
difficulty-as-impossibility (though no more likely to endorse difficulty-as-importance) than
people from China, India, Iran, and Turkey. A number of studies have examined the discriminant
validity and measurement invariance of the difficulty-as-importance and difficulty-asimpossibility scales, reporting evidence of measurement invariance in samples from the U.S.,

18
other Western countries, and less Westernized countries like China, Iran, India, and Turkey
(O’Donnell et al., 2023; Yan et al., 2023).
Several studies also provide evidence that difficulty-as-importance and difficulty-asimpossibility beliefs are distinct, not flip sides of the same construct (Fisher & Oyserman, 2017;
O’Donnell et al., 2023). For example, people who score higher in difficulty-as-importance are
more likely to believe that they can make time for important tasks and goals, controlling for their
difficulty-as-impossibility scores (Choi & Oyserman, 2024). In contrast, higher difficulty-asimpossibility scorers are less optimistic for the future (Yan et al., 2023), report lower self-esteem
(Burbidge & Oyserman, 2024), and are more likely to endorse a fixed mindset (Fisher &
Oyserman, 2017; O’Donnell et al., 2023). In these studies, analyses controlled for difficulty-asimportance scores and none of these measures were uniquely associated with difficulty-asimportance scores.
Summary and Proposed Process Model
Literature to date documents pieces of the posited recursive process between certainty,
academic outcomes, and metacognitive inferences from difficulty but has not tested the full
recursive process. Specifically, it documents paths from difficulty-as-importance and difficultyas-impossibility to identity certainty (Aelenei et al., 2017; Smith & Oyserman, 2015) and from
identity certainty to school outcomes (O’Donnell & Oyserman, 2023). It also documents
associations between possible identities and accepting fate (Bi & Oyserman, 2015) and between
optimism and difficulty-as-impossibility (Yan et al., 2023). To make progress, in the current
study, we applied identity-based motivation theory to a high-stakes school transition to examine
the recursive relationship among students’ certainty of attaining their school-focused possible
identities, the inferences they make about themselves when school tasks and goals feel hard to

19
think about or work on, and their school achievement.
Figure 1
Theoretical Process Models
Note. Green bolded text = association predicted to be positive. Red italicized text = association
predicted to be negative. Dashed line = association predicted to be weak.
Following identity-based motivation theory, we focus on difficulty-as-importance and
difficulty-as-impossibility as central inferences people make from their metacognitive
experiences of difficulty while thinking about or working on self-relevant tasks and goals. We
broaden our focus to include optimism for the future and accepting fate as culture-based beliefs
that may motivate or demotivate, as described in our review of the literature. We present our
synthesized process model in Figure 1. Panel 1 outlines the posited recursive relationship
between academic outcomes and school-focused possible identity certainty. Panel 2 outlines the

20
posited, mediating recursive relationship between school-focused possible identity certainty and
motivating or demotivating beliefs (difficulty-as-importance, difficulty-as-impossibility,
optimism for the future, and accepting fate).
The Current Study
Figure 2 presents the flow of our data collection. We collected data at three time points
during a two-month segment as Chinese primary school students transitioned to secondary
school. This allowed us to leverage multiple time points during a naturally occurring temporal
process. First, we obtained prior semester academic grades, labeled as T0. We label as T1 the
first time that students completed a survey, a month prior to their high-stakes secondary
placement test. We label the test itself as T1+. A week after the test, students learned their test
scores and completed the second survey; we label this as T2. Four weeks after the test, they
completed the third survey; we label this as T3. Students learned their school placement between
T2 and T3; we label this T2+.
Throughout our analyses we control for past achievement and self-beliefs (certainty of
attaining school-focused possible identities, difficulty-as-importance, and difficulty-asimpossibility, optimism for the future, accepting fate) allowing us to examine how changes in
achievement or self-beliefs can shift trajectories.
Figure 2
Data Collection Timeline

21
Method
Twelve-year-old students (N = 818) attending three elementary schools in China
completed three paper-and-pencil surveys at school (T1, T2, T3). Students participated if they
signed assent forms and their parents returned signed consent forms. We coordinated data
collection so that two research assistants visited each school on the same day and asked students
to fill out surveys. Teachers were not in the classroom during data collection. The alternate
activity was to sit and read either in the classroom or in a room outside the classroom. Students
filled in their T1 surveys a month prior to the high-stakes test (T1+) on a normal school day. A
week after the test, students came to school in the morning to receive their test results and could
return that afternoon to collect their transcripts from their classrooms, at which time research
assistants asked them to complete the T2 survey. Four weeks later, students learned their school
placement and could return to their classrooms during their summer break to collect their
graduation certificates, at which time research assistants asked them to complete T3 the survey.
We obtained age and gender data only at T1.
Students averaged 12 years and three months of age at T1. About 4 in 10 students were
female. The students remaining at T2 and T3 were similar in demographic profile (Appendix
Table A1). Regarding schools, the three schools provided about equally to the sample. We do not
have further information on features of the schools and do not use school as a variable in further
analyses (Appendix Table A1).
Sampling
Students lived in a middle-sized Chinese city with 15 primary schools. To obtain an
adequate sample size, we recruited three schools by sending recruitment letters one by one to
randomly selected schools, sending five letters in total. The letters invited schools to let their 6th-

22
graders participate in a study examining the transition from primary school to secondary school.
To the best of our knowledge, accepting and declining schools did not differ in neighborhood
characteristics. Declining schools could not fit our study into their school calendar and
management.
Human Subjects
Before starting data collection, we obtained university IRB approval as part of a seed
grant. Participation was voluntary. Once schools agreed to participate, we mailed parents a
consent form and study information sheet. Students could sign assent forms if their parents
returned signed consent forms and received a token (pencil case) for each time survey.
Data Transparency and Collection
Our surveys in Chinese and the English translation, data, code, and supplemental
materials are in OSF: https://osf.io/ew3kn/?view_only=52386fbc3f3a4b5db666a0fcaa07b348.
Measures
Measures were written in a way that was understandable to primary school students. We
used a translation-back translation method (Maneesriwongul & Dixon, 2004) for scales in
English. Primary measures were collected at T1, T2, and T3. Descriptive measures were
collected at T1 or T2.
Descriptive Measures
To provide a snapshot of the sample, we asked students at T1 to estimate the time they
typically spend on daily activities, indicate their favorite subject (choice of mathematics,
Chinese, English, or “Other”), and rate how much effort they were exerting in 6th grade (1=not
working hard, 10=working very hard). Also at T1, students rated their likelihood of getting
placed into their ideal secondary school (0=not likely at all, 9=very likely), how much they

23
hoped for this (0=not hopeful at all, 9=very hopeful), and how much they worried about not
getting in (0=not at all worried, 9=very worried). At T2, we asked students to what extent they
felt their placement test score was a measure of their true ability, effort, luck, and fate (0=not at
all true, 9=completely true).
Primary Measures
Academic Outcomes. Schools shared prior semester scores and placement test scores
with us. Both scores were a combination of Chinese, English, and Math subject tests (upper
bound of 300 points). At T1, students named their ideal secondary school. At T3, they indicated
whether their school placement was their ideal school (1=yes, 2=no).
Certainty of Attaining School-Focused Possible Identities. Students rated their
certainty (0=not possible at all to 9=very possible) of attaining their school-focused possible
identities in the coming year with Kemmelmeier and Oyserman’s (2001) 3-item scale (e.g.,
“Getting good grades”). They also rated their certainty of using effective strategies to attain these
possible identities with Kemmelmeier and Oyserman’s (2001) 4-item scale but, as detailed in the
preliminary analysis section, we did not use this scale in further analyses.
Accepting Fate. Students rated how much they agreed or disagreed (1=strongly disagree,
5=strongly agree) with Bi & Oyserman’s (2015) Chinese-language 10-item children’s fatalism
scale (e.g., “I have not realized my ideals, mainly because I do not have that fate.”).
Optimism for the Future. Students rated how much they agreed or disagreed
(1=strongly disagree, 5=strongly agree) with Scheier and Carver’s (1992) 7-item Life
Orientation Test-Revised scale (e.g., “I am always optimistic about my future.”).
Difficulty-as-Importance and Difficulty-as-Impossibility. Students rated how much
they agreed or disagreed (1=strongly disagree, 5=strongly agree) with conceptual translations of

24
Oyserman and colleagues’ (2017) 6-item difficulty-as-importance scale (e.g., “To achieve good
student results, there is no doubt that real hard work is required. Those things that require me to
really work hard mean these things are important to me.”) and difficulty-as-impossibility scale
(e.g., “Sometimes, it feels too difficult to learn—it is even impossible to learn well. This may be
a good thing because it makes me understand that I should do other things.”). The conceptual
translations concretized school tasks as learning, studying, reading, and test taking.
Analysis Plan
We conducted our preliminary and main analyses using the statistical software R (R Core
Team, 2023).
Preliminary Analysis
We described and verified measurement invariance and factor structures for our repeated
measures and described our sample.
Main Analysis
Given evidence of measurement invariance, we opted for a structural equation modeling
(SEM) path analysis approach to model temporal processes and control for prior scores for each
repeated measure delineated in Figure 1. We explored the temporal relationships between
academic outcomes, metacognitive inferences from difficulty (difficulty-as-importance,
difficulty-as-impossibility), and culture-based beliefs (optimism for the future, accepting fate) in
two steps. First, we modeled the temporal and bidirectional relationships between academic
outcomes and certainty of attaining school-focused possible identities. Then we added the
theory-based recursive paths for the metacognitive inferences from difficulty and culture-based
beliefs. We controlled for prior grades when predicting high-stakes test scores and for test scores
when predicting T1 self-beliefs. We also controlled for prior (T1 or T2) self-beliefs when

25
predicting subsequent (T2 or T3) self-beliefs. We did not use prior grades or test scores as
controls when predicting ideal school placement in our main analysis due to low correlations
(described below).
We chose to use first-order SEM models, as second-order models would require
estimating the value of the latent variables. This would decrease the degrees of freedom and
necessitate a substantially larger sample than ours to detect significant effects. A post hoc power
analysis using G*Power (Faul et al., 2007) determined our analytic sample size (N = 818) had
sufficient power of .82 to detect small effect sizes (f
2 = 0.02) with an alpha of .05 in our most
complex paths (e.g., a path with nine predictors) and power of .96 in our simplest paths (e.g., a
path with two predictors). We used maximum likelihood estimation with robust (Huber-White)
SE to estimate each SEM model.
We imputed missing data (see Supplemental Materials Table S1 for the proportion of
missing data for each analytic variable) using the R mice package (Multivariate Imputation by
Chained Equations; van Buuren & Groothuis-Oudshoorn, 2011). Multivariate imputation by
chained equations can appropriately impute missing values when both dependent variables and
independent variables have missing data points and data is missing at random (Biering et al.,
2015; Jakobsen et al., 2017). Multiple imputation restores the natural variability in the missing
data while accounting for the uncertainty caused by estimating missing data with consideration
to how variables in the analytic dataset correlate with the missing data (Wayman, 2003). As per
Enders (2017) and Mainzer and colleagues (2021), we imputed missing values at the item-level
and created unique multiple imputation models for each missing variable. Following Mainzer
and colleagues (2021), we used a mixed item-scale approach to determine predictors of each
multiple imputation model. We imputed only variables in our SEM analysis, not descriptive

26
variables.
Other modeling approaches such as cross-lagged panel models (CLPMs) were not
appropriate given the structure of our data and our specific research questions. First, our
ecologically valid data collection timeline (Figure 2) meant that our data structure did not meet
the assumption of synchronicity for cross-lagged panel model (CLPM) approaches (Kenny &
Harackiewicz, 1979). Second, our focus on the high-stakes testing period meant our third school
outcome data point is the match between school placement and student preference. A third
academic score data point would be needed for a random intercept CLPM (Hamaker et al.,
2015). Third, a CLPM would not allow us to simultaneously evaluate how school placement
match, a one-time outcome, may affect subsequent self-beliefs.
Results
Preliminary Analysis
Measure Construction
Confirmatory factor analysis results showed that school-focused possible identity
certainty and certainty of using effective strategies to attain these possible identities were
correlated but separate constructs. We attempted to include the latter in our SEM model, but this
resulted in a 33% increase in AIC and BIC statistics, indicating the model’s increased
complexity was not counterbalanced by increased model fit. Hence, we did not include certainty
of using school-focused strategies in further analyses.
We used confirmatory factor analysis to confirm our theory-based decision to model each
self-belief measure as a separate construct (for detailed results, see Supplemental Materials
Tables S4 and S5). The confirmatory factor analysis revealed a small number of poor-loading
items (< 0.6): one of six items in the difficulty-as-impossibility scale, two of ten items in the

27
accepting fate scale, and three of seven items in the optimism for the future scale. We dropped
these items to ensure an acceptable fit for our analytic model. Then we explored consistency in
each measure’s internal structure over time using longitudinal measurement invariance analyses
(Supplemental Materials Table S6). Following Chen (2007), we determined invariance via
changes in CFI supplemented by changes in SRMR and RMSEA. We found metric and scalar
invariance for each self-belief. These results suggest our measures are internally stable over time,
so repeated measurements can be meaningfully compared. Measures were reliable (Appendix
Table A2).
Repeated Measures Stability and Invariance
Prior grades and high-stakes test results correlated highly (r = .93 p < .001) but prior
grades (r = .08, p = .031) and test results (r = .12, p = .001) had only a small correlation with
placement into one’s ideal secondary school. The implication is that what constitutes an “ideal
school” is weakly determined by achievement. Certainty of attaining school-focused possible
identities correlated moderately with metacognitive inferences from difficulty and culture-based
beliefs (|.19| < rs < |.37|, ps < .001). Metacognitive inferences from difficulty and culture-based
beliefs correlated moderately with each other at each time point (|.26| < rs < |.56|, ps <.001).
Correlations were in the expected directions (positive associations between accepting fate and
difficulty-as-impossibility and between difficulty-as-importance and optimism for the future;
negative associations between difficulty-as-impossibility and difficulty-as-importance and
accepting fate and optimism for the future). Between T1 and T2 and between T2 and T3, each
repeated measure correlated strongly with itself (.62 < rs < .74, ps <.001). Appendix Table A4
provides inter-measure correlations at each time point. Appendix Table A5 provides intrameasure correlations over time.

28
Sample Description
Appendix Table A3 shares a detailed, descriptive snapshot of our sample. On average,
students’ favorite subject was math. They spent over an hour daily on homework, tutoring,
reading, and sports and less than an hour on physical education, the internet, television, video
games, and music or dance. Before the high-stakes test, students reported working moderately
hard in school, believed they had a modest chance of getting into their ideal school, had high
hopes that this would occur, and were modestly worried they would not.
On the day students learned their high-stakes test results, we asked them to what extent
their test scores were due to ability, effort, luck, and fate. Students' attributions about the reason
for their scores were distinct (see Supplemental Materials Table S2 for correlations). Across
linear regression models controlling for high-stakes test scores (see Supplemental Materials
Table S3 for detailed results), students' attributions about the reason for their scores were
associated with their T1 metacognitive inferences from difficulty and culture-based beliefs.
Thus, students were more likely to believe their test scores were due to their ability if at T1 they
endorsed difficulty-as-importance more. Students were more likely to believe their test scores
were due to their effort if at T1 they endorsed difficulty-as-importance and optimism for the
future more and difficulty-as-impossibility less. They were more likely to believe their test
scores were due to luck if at T1 they endorsed difficulty-as-impossibility and accepting fate more
and optimism for the future less. They were more likely to believe their test scores were due to
fate if at T1 they endorsed difficulty-as-importance and optimism for the future less and
difficulty-as-impossibility and accepting fate less. We take these distinct patterns of associations
as suggesting that difficulty-as-importance, difficulty-as-impossibility, optimism for the future,
and accepting fate each may play a unique, motivational role in how students make sense of their

29
academic outcomes.
Students averaged an 81.5% prior semester grade (0-300 scale, M = 244.50 SD = 41.25)
and an 84.33% score on the high-stakes test (0-300 scale, M = 253.00, SD = 39.87). 60.15% of
students got into their ideal school at T3. Students were certain they would attain their schoolfocused possible identities, endorsed difficulty-as-importance and optimism for the future and
rejected difficulty-as-impossibility and accepting fate (Appendix Table A2).
Main Analysis
Step 1: Academic Outcomes and Possible Identity Certainty
We tested the relationship between students’ academic outcomes and their certainty of
attaining their school-focused possible identities. We share complete path statistics in Table 1.
The model had adequate fit: CFI = .99, RMSEA = .07, SRMR = .02.
T1 possible identity certainty was higher among students with better T0 grades (B = 1.43,
z = 11.28, p < .001). Controlling for T0 grades, students who felt more certain about attaining
their school-focused possible identities at T1 scored higher at T1+ on the high-stakes test (B =
0.01, z = 2.49, p = .013). Controlling for T1 possible identity certainty, T2 possible identity
certainty was higher among students with better T1+ high-stakes test scores (B = 0.93, z = 5.89, p
< .001). Controlling for T1 and T2 possible identity certainty, T3 ideal school placement and T3
possible identity certainty were not significantly associated.
Step 2: Adding Metacognitive Inferences from Difficulty and Culture-Based Beliefs to the
Academic-Identity Model
The model had adequate fit: CFI=.96, RMSEA=.08, SRMR = .08. We detail the recursive
process next, piece by piece. We share complete path statistics in Table 2. We present a
visualization of the results in Figure 3.

30
Figure 3
Path Analysis Step 2: Controlling for Academic Achievement, Difficulty-As-Importance and
Possible Identity Certainty are Recursively Related. Other Associations are Not Recursive
Note. Path estimates and statistics are presented in Table 2. Red lines are significantly negative,
and green lines are significantly positive associations. Single-headed gray arrows are control
paths. The double-headed gray arrows represent the covariances between our self and motivation
constructs at each time point.
The first path of focus is from T0 grades to T1 metacognitive inferences from difficulty
and culture-based beliefs. T0 grades were associated with T1 metacognitive inferences from
difficulty and culture-based beliefs. Students who had better grades in the prior semester scored
higher in difficulty-as-importance (B = 0.44, z = 7.15, p < .001) and optimism for the future (B =
0.46, z = 6.49, p < .001) and lower in difficulty-as-impossibility (B = -0.62, z = -9.50, p < .001)
and accepting fate (B = -0.55, z = -8.22, p < .001).
The second path of focus is from T1 metacognitive inferences from difficulty and
culture-based beliefs to T1+ high-stakes test scores. Accounting for the strong effect of T0
grades on T1+ high-stakes test scores and of T1 school-focused possible identity certainty, T1
metacognitive inferences from difficulty and culture-based beliefs did not significantly predict
T1+ high-stakes test scores.
The third path of focus is from T1 metacognitive inferences from difficulty and culture-

31
based beliefs to T2 possible identity certainty. Accounting for the effects of T1 possible identity
certainty and T1+ high-stakes test scores, T2 possible identity certainty was higher among
students who at T1 scored higher in difficulty-as-importance (B = 0.18, z = 2.02, p = .043) and
optimism for the future (B = 0.22, z = 3.36, p = .001). T2 possible identity certainty was not
significantly associated with T1 difficulty-as-impossibility and accepting fate.
The fourth path of focus is from T1+ high-stakes test Scores and T1 possible identity
certainty to T2 metacognitive inferences from difficulty and culture-based beliefs. T2 was about
one month after T1, on the day students learned their high-stakes test results. Accounting for the
effect of T1 difficulty-as-importance, T2 difficulty-as-importance was higher among students
who felt more certain about attaining their school-focused possible identities at T1 (B = 0.04, z =
2.45, p = .014) and who did better on the T1+ high-stakes test (B = 0.12, z = 2.10, p = .036). The
reverse was true for difficulty-as-impossibility. Accounting for the effect of T1 difficulty-asimpossibility, T2 difficulty-as-impossibility was lower among students who felt more certain
about attaining their school-focused possible identities at T1 (B = -0.03, z = -2.21, p = .027) and
who did better on the T1+ high-stakes test (B = -0.13, z = -2.28, p = .023). Accounting for the
effect of T1 accepting fate, T2 accepting fate was lower among students who did better on the
T1+ high-stakes test (B = -0.21, z = -3.08, p = .002) but was not significantly associated with T1
possible identity certainty. Neither T1 possible identity certainty nor T1+ high-stakes test scores
were significantly associated with T2 optimism for the future.
The fifth path of focus is from T2 metacognitive inferences from difficulty and culturebased beliefs to T3 possible identity certainty. Accounting for the effects of T1 and T2 possible
identity certainty and ideal school placement, T3 possible identity certainty was higher among
students who scored higher in T2 difficulty-as-importance (B = 0.18, z = 2.82, p = .005). T2

32
difficulty-as-impossibility, accepting fate, and optimism for the future were not significantly
associated with T3 possible identity certainty.
The sixth and final path of focus is from T2+ ideal school placement and T2 possible
identity certainty to T3 metacognitive inferences from difficulty and culture-based beliefs. T3
was about one month after T2, when students knew the secondary school they would attend in
the fall. Accounting for the effects of T1 and T2 difficulty-as-importance, T3 difficulty-asimportance was higher among students with higher T2 possible identity certainty (B = 0.07, z =
4.29, p < .001) but was not significantly associated with ideal school placement. Accounting for
the effects of T1 and T2 difficulty-as-impossibility, T3 difficulty-as-impossibility was lower
among students who got into their ideal school (B = -0.21, z = -5.36, p < .001) and felt more
certain about attaining their school-focused possible identities at T2 (B = -0.04, z = -3.12, p =
.002). Accounting for T1 and T2 accepting fate, T3 accepting fate was lower among students
who felt more certain about attaining their school-focused possible identities at T2 (B = -0.04, z
= -2.59, p = .010) but was not significantly associated with ideal school placement. Neither T2
possible identity certainty nor ideal school placement were significantly associated with T3
optimism for the future.

33
Table 1
Path Analysis Step 1: Academic Outcomes and School-Focused Possible Identity Certainty
R2
0.14
0.87
0.47
0.57
Note. B = unstandardized coefficient. SE B = standard error of unstandardized coefficient. β = standardized
coefficient.
p
<.001
<.001
0.013
<.001
<.001
<.001
<.001
0.344
z
11.28
47.79
2.49
18.11
5.89
8.48
15.2
-0.95
β
0.37
0.92
0.04
0.58
0.22
0.28
0.54
-0.02
SE B
0.13
0.02
0.00
0.04
0.16
0.03
0.03
0.08
B
1.43
0.89
0.01
0.63
0.93
0.29
0.50
-0.07
Effect
T0 Prior Academic Scores
T0 Prior Academic Scores
T1 Possible Identity Certainty
T1 Possible Identity Certainty
T1+ High-Stakes Test Scores
T1 Possible Identity Certainty
T2 Possible Identity Certainty
T2+ Ideal School Placement
Outcome Variable
T1 Possible Identity Certainty
T1+ High-Stakes Test Scores
T2 Possible Identity Certainty
T3 Possible Identity Certainty

34
Table 2
Path Analysis Step 2: Academic Outcomes, School-Focused Possible Identity Certainty,
Metacognitive Inferences from Difficulty, and Culture-Based Beliefs
Outcome Variable Effect B SE B β z p R2
T1 Possible Identity
Certainty
T0 Prior
Academic Scores 1.43 0.13 0.37 11.28 <.001 .14
T1 Difficulty-asImportance
T0 Prior
Academic Scores 0.44 0.06 0.28 7.15 <.001 .08
T1 Difficulty-asImpossibility
T0 Prior
Academic Scores -0.62 0.07 -0.36 -9.50 <.001 .13
T1 Accepting Fate T0 Prior
Academic Scores -0.55 0.07 -0.31 -8.22 <.001 .10
T1 Optimism for the
Future
T0 Prior
Academic Scores 0.46 0.07 0.23 6.49 <.001 .05
T1+ High-Stakes
Test Scores
T0 Prior
Academic Scores 0.88 0.02 0.91 43.77 <.001 .87
T1 Possible
Identity Certainty 0.01 0.00 0.03 1.81 .071
T1 Difficulty-asImportance 0.00 0.01 0.01 0.43 .671
T1 Difficulty-asImpossibility -0.01 0.01 -0.03 -1.23 .218
T1 Accepting
Fate -0.01 0.01 -0.02 -0.88 .378
T1 Optimism for
the Future -0.01 0.01 -0.01 -0.53 .598
T2 Possible Identity
Certainty
T1 Possible
Identity Certainty 0.58 0.04 0.54 15.49 <.001 .49
T1+ High-Stakes
Test Scores 0.88 0.17 0.20 5.23 <.001
T1 Difficulty-asImportance 0.18 0.09 0.07 2.02 .043
T1 Difficulty-asImpossibility 0.04 0.09 0.02 0.51 .608
T1 Accepting
Fate 0.07 0.08 0.03 0.84 .403
T1 Optimism for
the Future 0.22 0.07 0.11 3.36 .001

35
T2 Difficulty-asImportance
T1 Difficulty-asImportance 0.53 0.05 0.52 11.86 <.001 .33
T1 Possible
Identity Certainty 0.04 0.01 0.08 2.45 .014
T1+ High-Stakes
Test Scores 0.12 0.06 0.07 2.10 .036
T2 Difficulty-asImpossibility
T1 Difficulty-asImpossibility 0.54 0.03 0.53 16.04 <.001 .35
T1 Possible
Identity Certainty -0.03 0.02 -0.07 -2.21 .027
T1+ High Stakes
Test Scores -0.13 0.06 -0.07 -2.28 .023
T2 Accepting Fate T1 Accepting
Fate 0.61 0.04 0.55 15.29 <.001 .35
T1 Possible
Identity Certainty -0.01 0.07 -0.02 -0.72 .471
T1+ High Stakes
Test Scores -0.21 0.07 -0.10 -3.08 .002
T2 Optimism for the
Future
T1 Optimism for
the Future 0.64 0.04 0.60 18.33 <.001 .39
T1 Possible
Identity Certainty 0.02 0.02 0.03 0.89 .375
T1+ High Stakes
Test Scores 0.08 0.06 0.03 1.21 .227
T3 Possible Identity
Certainty
T1 Possible
Identity Certainty 0.29 0.03 0.29 8.47 <.001 .58
T2 Possible
Identity Certainty 0.50 0.03 0.53 14.93 <.001
T2+ Ideal School
Placement 0.14 0.07 0.06 2.08 .038
T2 Difficulty-asImportance 0.18 0.06 0.08 2.82 .005
T2 Difficulty-asImpossibility -0.09 0.05 -0.05 -1.68 .093
T2 Accepting Fate 0.04 0.05 0.02 0.75 .452
T2 Optimism for
the Future -0.11 0.08 -0.03 -1.49 .136

36
T3 Difficulty-asImportance
T1 Difficulty-asImportance
0.18 0.05 0.16 3.66 <.001 .35
T2 Difficulty-asImportance 0.45 0.04 0.42 11.49 <.001
T2 Possible
Identity Certainty 0.07 0.02 0.15 4.29 <.001
T2+ Ideal School
Placement 0.01 0.04 0.01 0.26 .792
T3 Difficulty-asImpossibility
T1 Difficulty-asImpossibility 0.13 0.04 0.13 3.48 .001 .42
T2 Difficulty-asImpossibility 0.49 0.04 0.50 13.65 <.001
T2 Possible
Identity Certainty -0.04 0.01 -0.10 -3.12 .002
T2+ Ideal School
Placement -0.21 0.04 -0.14 -5.36 <.001
T3 Accepting Fate T1 Accepting Fate 0.14 0.04 0.13 3.27 .001 .47
T2 Accepting Fate 0.56 0.04 0.58 13.92 <.001
T2 Possible
Identity Certainty -0.04 0.01 -0.08 -2.59 .010
T2+ Ideal School
Placement -0.06 0.04 -0.04 -1.42 .156
T3 Optimism for the
Future
T1 Optimism for
the Future 0.28 0.04 0.28 7.84 <.001 .41
T2 Optimism for
the Future 0.41 0.03 0.42 12.48 <.001
T2 Possible
Identity Certainty 0.01 0.02 0.03 0.93 .354
T2+ Ideal School
Placement 0.06 0.05 0.03 1.25 .212
Note. B = unstandardized coefficient. SE B = standard error of unstandardized coefficient.
β=standardized coefficient.

37
Discussion
We collected data from a large sample of 12-year-old Chinese students at three points in
time during a two-month period that marked their transition from primary to secondary school.
This period began a month before the school year ended, was punctuated by taking a high-stakes
test, and ended with students knowing where they would attend secondary school. We predicted
a recursive relationship between students’ certainty of attaining their school-focused possible
identities and their academic achievement. We sought to understand the underlying process by
examining the mediating effect of culture-based beliefs (optimism for the future, accepting fate)
and inferences from metacognitive experiences of difficulty (difficulty-as-importance, difficultyas-impossibility). We found evidence of two recursive relationships with school-focused possible
identity certainty–one with school attainment and another with endorsing difficulty-asimportance (when schoolwork feels hard, it is important to you). We also found evidence of a
unidirectional relationship from school-focused possible identity certainty and endorsing
difficulty-as-impossibility (when schoolwork feels hard to think about or do, that implies it may
not be for you). Lower certainty of attaining school-focused possible identities was associated
with subsequently higher difficulty-as-impossibility endorsement, but difficulty-as-impossibility
endorsement did not carry over to subsequent possible identity certainty. We found associations
between certainty of attaining school-focused possible identities and changes in optimism for the
future and in accepting fate at only some time points.
Implications for Theory Development
Our results are relevant to the literature on identity and school outcomes in three ways.
First, our results converge with and build on research documenting a positive relationship
between some aspects of identity and academic outcomes (for reviews, see Brandstätter &

38
Bernecker, 2022; Oyserman et al., 2012; Carver & Scheier, 2012). Existing meta-analyses
include an array of operationalizations of the self including self-concept, self-efficacy, selfesteem, expectancy-value, among others and include an array of operationalizations of
achievement (Huang, 2011; Jansen et al., 2022; Valentine et al., 2004). In the current study, we
focused on certainty of attaining school-focused possible identities and real-time achievement on
a high-stakes test and found a recursive relationship even though prior semester grades captured
much of the variability in high-stakes test scores. Our results advance this literature by
highlighting both a specific process (via identity certainty) and capturing a recursive process
feeding into self-certainty (believing difficulty implies importance).
Second, our results advance research on identity-based motivation theory (Oyserman,
2007; Oyserman & Horowitz, 2023; Oyserman & Destin, 2010). Research to date has
documented that school-focused possible selves are associated with academic outcomes but has
not separately examined the recursive processes among certainty of attaining school-focused
possible selves, optimism for the future, accepting fate, endorsing difficulty-as-importance, and
difficulty-as-impossibility. Our results imply an indirect process in which the difficulty-asimportance inferences students draw when thinking about school feels hard matter for goal
pursuit by affecting possible identities which in turn affect future-focused behavior.
Third, we add to the cultural generalizability of literatures concerning both identity-based
motivation and links between identity and school outcomes. Most research focuses on the United
States. Our results highlight generalizability to a non-Western sample of children in China. We
document the positive effect of prior academic achievement on students’ certainty of attaining
their school-focused possible identities (for a prior review with mostly Western data, see
O’Donnell & Oyserman, 2023). Regarding difficulty-as-importance and difficulty-as-

39
impossibility, our results are in line with prior research conducted with Western samples (Fisher
& Oyserman, 2017), documenting that difficulty-as-importance and difficulty-as-impossibility
are related but distinct ways of making sense of difficulty. We document that difficulty-asimportance and difficulty-as-impossibility have consistent relationships with certainty of
attaining school-focused possible identities, implying that future research would benefit from
applying these constructs to understanding student motivation and outcomes.
Limitations and Future Directions
In this section, we consider limitations to generalizability given our sample, our
methodological approach, and the measures we used. We also consider potential future research
that could build on our findings. First, regarding our sample, a strength of our sample was that
we collected data from a large sample of Chinese students at multiple time points during a
critical transition. Data collection was timed to increase the ecological validity of our results.
Regarding limitations, we did not use a survey procedure to draw a representative sample of all
Chinese students so we cannot be sure that our effect sizes generalize. Our IRB-approved plan
did not entail intensive outreach, so we lost some students in our summertime data collection.
We used multiple imputation to impute the missing data, allowing us to move forward with a
statistically powered analysis. Future research with a representative sample and IRB-approved
follow-up outreach would increase the likely stability of effect sizes and confidence intervals.
Second, regarding the approach, we used a quasi-experimental approach with a real
school transition. A strength of our approach is that it allowed us to explore the relationships of
interest with high external validity. Though ecologically valid and statistically powered, our
approach was not a randomized controlled intervention. Moreover, advancements in statistical
modeling (i.e., next-generation cross-lagged panel modeling) suggest caution in interpreting

40
reciprocal effects (Burns et al., 2020; Núñez-Regueiro et al., 2022). A minimum of three data
points for both academic outcomes and identity is necessary for such next-generation crosslagged analyses. We did not have a third ecologically-valid academic outcome data point. Future
research could address these limitations by considering what might be a third academic outcome
that could be used with a better-funded longitudinal panel, such as academic grades in the first
semester of secondary school. Alternatively, future research could use a diary method to obtain
additional data points during a smaller set of academic milestones than the school transition we
studied.
Third, regarding our measures, a strength is that to our knowledge we are the first to
assess certainty of attaining school-focused possible identities, difficulty-as-importance,
difficulty-as-impossibility, optimism for the future, and accepting fate across time in a Chinese
sample. As noted in the measures section, we did assess students’ certainty of using effective
strategies to work on their school-focused possible identities, a concept akin to academic selfefficacy, but this measure reduced model fit and could not be used. Future research could try to
assess academic self-efficacy with differently worded measures (Bandura, 1997). We also did
not measure how much students valued schooling, as we did not expect much variability given
the high cultural normativity of valuing school (Chen et al., 1996). Future research could explore
how capturing students’ expectancy-value beliefs (Eccles & Wigfield, 2020) might provide
insight into students’ certainty of attaining their possible identities and their future-focused
behavior.
Conclusion
We find an energizing effect of feeling more certain of attaining one’s school-focused
possible identities, such that certainty boosts and is boosted by believing that when schoolwork

41
feels hard to think about or do, it is probably for you (difficulty-as-importance). Intervening to
increase the accessibility of a difficulty-as-importance mindset may help foster a virtuous cycle
in which students feel empowered to “give it their all” in high-stakes situations which can open
doors to future opportunities.

42
Chapter 2: As Inferences from Difficulty Shift, So Do Action and Self-Regard
Burbidge, A., & Oyserman, D. (in prep).
Abstract
When tasks or goals feel hard to think about or do, people can draw two inferences: “this
is important to me, worthwhile” (difficulty-as-importance) and “this is not for me, a waste of
time” (difficulty-as-impossibility). Difficulty-as-impossibility endorsers are aware that they
prefer to avoid engaging the hard way (difficulty-as-importance endorsers are agnostic, Study 1,
N=216, scenario study). Difficulty-as-importance beliefs are unchanged by a difficult task;
difficulty-as-impossibility beliefs shift depending on features of the context (Study 2a-2b,
N=464, experiments). High difficulty-as-impossibility endorsers perform worse in success-likely
contexts than low endorsers. Difficulty-as-importance and difficulty-as-impossibility beliefs
have state-like and trait-like features (Study 3, N=733, secondary analysis). Trait difficulty-asimportance (not difficulty-as-impossibility) is associated with subsequent daily task engagement;
compared with daily fluctuations in difficulty-as-importance, daily fluctuations in difficulty-asimpossibility are more associated with daily fluctuations in self-esteem, self-efficacy, and selfcompassion (Studies 4-6, N=260, n=2,789 entries, daily diaries).
Keywords: within-person and between-person variation; identity-based motivation;
metacognition; daily well-being; daily self-esteem

43
Identity-based motivation (IBM) theory is a situated social cognition theory of
motivation, goal pursuit, and self-regulation (Oyserman, 2007, 2009). It posits that people draw
on their experiences of difficulty when thinking about or engaging with tasks or goals to make
inferences about who they are, how to act, and what to focus on (Oyserman et al., 2017). When a
task or goal feels hard to think about or do, people can infer to varying degrees that the task or
goal is self-relevant and important to them (difficulty-as-importance) and that it is self-irrelevant
and a waste of their limited motivational energy (difficulty-as-impossibility). Evidence of the
self-regulatory consequences of difficulty-as-importance and difficulty-as-impossibility comes
from experimental and correlational studies. Experiments using autobiographical recall to shift
construct accessibility find that people with difficulty-as-importance on the mind experience
time as more of an expandable resource compared to people with difficulty-as-impossibility on
the mind (Choi & Oyserman, 2024). People also feel more certain of attaining their schoolfocused goals when difficulty-as-importance is on the mind (Aelenei et al., 2017). Studies
assessing the unique effects of endorsing each difficulty mindset find that people who endorse
difficulty-as-impossibility prefer less effortful means of goal attainment, controlling for their
difficulty-as-importance beliefs which are not associated with preferences for means (Kiper et
al., 2024), suggesting that difficulty-as-importance is associated with a focus on ends and
difficulty-as-impossibility is associated with a focus on means. What is less clear is whether
difficulty-as-impossibility endorsers have insight into their preference for easier means, how
contexts shift endorsement of each mindset, how much endorsements fluctuate, and how much
trait and state endorsements matter for focus during self-regulation. We address these open
questions in the current studies.

44
Two Inferences from Metacognitive Experiences of Difficulty
IBM theory conceptualizes difficulty-as-importance and difficulty-as-impossibility as
two separate, useful inferences rather than as two sides of the same inference (Fisher &
Oyserman, 2017). IBM theory posits that features of situations can facilitate or hinder the
construct accessibility of each difficulty mindset and that people vary in their chronic
endorsement of each belief (Oyserman, 2007). As detailed in the next sections, research to date
finds support for situated effects of construct accessibility and individual differences in
endorsement but has not yet examined situated differences in endorsement.
Self-Regulatory Consequences of Difficulty Mindset Accessibility
Prior research has used random assignment to assess the effects of difficulty mindset
accessibility on self-regulation. Studies document that features of situations can momentarily
increase the accessibility of difficulty-as-importance and difficulty-as-impossibility mindsets and
that accessibility bears on action and judgment (for a review, Oyserman & Horowitz, 2023).
When a difficulty-as-impossibility mindset is accessible, students are less likely to consider
academic success to be central to their identities (Smith & Oyserman, 2015), more likely to
experience time as limited (Choi & Oyserman, 2024), and perceive school tasks as more effortful
(Oyserman et al., 2018). When a difficulty-as-importance mindset is accessible, students spend
more time on school tasks (Smith & Oyserman, 2015), are more certain they will attain their
school-focused possible identities (Aelenei et al., 2017), and believe they can find the time to
work on important tasks (Choi & Oyserman, 2024). Having an accessible difficulty-asimportance mindset also leads to lower temptation to overeat among adults with weight loss
goals (Lewis & Earl, 2018).
In addition to momentary accessibility, difficulty-as-impossibility may be chronically

45
accessible among Americans (O’Donnell et al., 2023). This chronic accessibility may be why
people prefer not to use effective but effortful learning strategies such as spacing learning over
time (Carpenter et al., 2022). Indeed, school outcomes among American students improve after
participation in school interventions that aim to reduce reliance on a difficulty-as-impossibility
mindset (Horowitz et al., 2018; Oyserman et al., 2002; Oyserman et al., 2006; Oyserman et al.,
2021) and encourage applying a difficulty-as-importance mindset (Destin et al., 2018; Horowitz
et al., 2018; Oyserman et al., 2002; Oyserman et al., 2006; Oyserman et al., 2021). We infer from
these intervention results that trait-like, chronic accessibility of difficulty-as-impossibility and
difficulty-as-importance bears on self-regulation. However, we also underscore that people do
not necessarily endorse the mindset that is accessible. For instance, although difficulty-asimpossibility is chronically accessible among Americans, Americans generally reject a
difficulty-as-impossibility interpretation (O’Donnell et al., 2023). As such, we turn to the selfregulatory consequences of difficulty mindset endorsement in the next section.
Self-Regulatory Profiles and Difficulty Mindset Endorsement
In addition to documenting the effects of construct accessibility, prior research has
documented effects of construct endorsement. While accessibility studies manipulate
accessibility between persons, endorsement can be assessed within persons. This allows for an
examination of the extent to which the two constructs are functionally distinct. On average,
people endorse difficulty-as-importance and reject difficulty-as-impossibility but people’s
endorsements of each mindset are not strongly correlated and the direction of correlation varies
across studies (Fisher & Oyserman, 2017; O’Donnell et al., 2023; Yan et al., 2023).
Difficulty-as-importance and difficulty-as-impossibility endorsement scores have distinct
profiles of correlation with other measures of motivation. For example, difficulty-as-importance

46
scores are distinct from growth or fixed mindset scores while difficulty-as-impossibility scores
are moderately correlated with fixed mindset scores (Fisher & Oyserman, 2017) and both
difficulty-as-importance and difficulty-as-impossibility endorsement are distinct from learning or
performance goals (Kiper et al., 2024). Endorsements of difficulty mindset scores also have
distinct associations with preferences and perceptions that bear on self-regulation. Students who
endorse difficulty-as-importance prefer working with complicated numbers (i.e., fractions or
percentages) and see themselves as capable of solving problems involving these numbers
(Mielicki et al., 2022). Students who endorse difficulty-as-importance are also more likely to
experience time as an expandable resource while students who endorse difficulty-asimpossibility are more likely to experience time as a limited resource (Choi & Oyserman, 2024).
Adults who endorse difficulty-as-importance support aid programs that are more resourceintensive, separate from willingness to be personally involved (Viola et al., 2023).
When endorsement of each mindset is jointly assessed, difficulty-as-importance and
difficulty-as-impossibility endorsement scores separately contribute to variance explained in
self-beliefs and task engagement. Students who endorse difficulty-as-impossibility perceive less
effortful means of goal attainment as effective, controlling for their difficulty-as-importance
scores which do not predict preferences for goal attainment means (Kiper et al., 2024). Across
samples in Australia, Canada, China, India, Iran, the U.K., the U.S., and Turkey, higher
difficulty-as-importance endorsers see themselves as virtuous and conscientious, controlling for
how much they endorse difficulty-as-impossibility (Yan et al., 2023).
We looked for studies examining the features of situations that might affect how much
people endorse difficulty-as-impossibility and difficulty-as-importance. Oyserman and
colleagues (2017) manipulated college students’ perceptions of their possible selves and how

47
college was likely to unfold, making either matches or mismatches salient. Students led to
consider a mismatch between their possible selves and the way college was likely to unfold
endorsed difficulty-as-impossibility more compared to students led to consider a match between
their possible selves and the way college was likely to unfold (Oyserman et al., 2017). The
reverse was true for difficulty-as-importance, though the effects were not robust. Students led to
consider a match between their possible selves and the way college was likely to unfold endorsed
difficulty-as-importance more compared with students led to consider a mismatch between their
possible selves and the way college was likely to unfold. The implication we draw is that
difficulty-as-impossibility endorsement may be more sensitive to features of immediate
situations that highlight the likelihood of success than difficulty-as-importance endorsement.
Current Studies
Research to date suggests that people differ in their difficulty-as-importance and
difficulty-as-impossibility beliefs and situations trigger the accessibility of each. We address four
remaining gaps. First, do difficulty-as-importance and difficulty-as-impossibility endorsers have
insight into their task preferences? Given evidence to date, we predict H1 (Study 1, scenario
design): Difficulty-as-impossibility endorsers are sensitive to their preference for easier tasks.
Second, do difficulty mindset beliefs prior to entering a situation and the features of the situation
interact to predict subsequent beliefs? Given evidence to date, we predict H2 (Studies 2a and 2b,
experimental design): Difficulty-as-impossibility endorsement shifts depending on the context of
a task, increasing after experiencing a context of difficulty or impossibility and decreasing after
experiencing a context of ease. Third, do difficulty mindset beliefs meaningfully fluctuate within
persons? Given evidence to date, we predict H3 (Study 3, secondary data analyses; Studies 4 to 6
two-week diary studies): Difficulty-as-importance endorsement varies between persons more

48
than difficulty-as-impossibility endorsement. Fourth, are individual differences and daily
fluctuations in difficulty mindset beliefs associated with daily self-regard and action? Given
evidence to date, we predict H4 (Studies 4 to 6, two-week daily diaries): Trait difficulty-asimportance endorsement is associated with daily meaningful schoolwork engagement (focus on
goal progress); daily difficulty-as-impossibility endorsement is associated with daily self-esteem,
goal self-efficacy, and self-compassion (self-regulatory means).
Our anonymized data, code, pre-registrations, and supplemental materials are available in
OSF: https://osf.io/s6ugv/?view_only=967bc089faf548d985c007125024fa6d. For ease,
Appendix B details all scales. To maintain a focused discussion, we present analyses from preregistrations that are directly relevant to our research questions. Remaining analyses are in
Supplemental Materials. All analyses were conducted using the statistical software R (R Core
Team, 2023). All studies except for Study 3 involve college student subject pool samples. We
chose to use the subject pool for three reasons: it facilitated time-lagged studies, in Studies 1 to
2b we normed our materials with students, and students were motivated to complete a difficult
word-puzzle task as indicated by our pilot (see Supplemental Materials).
Study 1
Method
Sample
We pre-registered our study and obtained University of Southern California IRB approval
(#UP-15-0055). Undergraduate students participated for subject pool credit at two points in the
semester, receiving credit each time (N = 216, 65.74% female, Mage = 20.07, SD = 1.89; 33.80%
Asian American, 28.70% White American, 15.28% Hispanic American).

49
Procedure and Measures
Students filled out a prescreening questionnaire at T1 (mostly demographics other
researchers requested) with the Fisher and Oyserman (2017) 4-item difficulty-as-importance and
4-item difficulty-as-impossibility scales near the end in randomized order. Students rated each of
the eight statements from 1 = strongly disagree to 6 = strongly agree (Appendix B). At T2,
students read: “You will work on a set of 10 word-puzzles for this task. Working on wordpuzzles is a useful way to develop the kinds of sharp analytic skills that help people get good
grades in school. Working on word-puzzles is particularly useful for developing this gradebolstering skill when the word-puzzles are difficult.” They rated how much they preferred (1 =
not at all, 5 = very much) easy (“I would prefer the puzzles most people can solve.”) and difficult
(“I would prefer the puzzles most people cannot solve.”) puzzles, and a set of ten puzzles to
work on: “Okay, put together a training set of 10 puzzles. How many of those 10 would you
select from each kind of puzzle? How many would you include from the puzzles that most
people can solve? How many would you include from the puzzles that most people cannot
solve?”. Students then provided gender, age, and race-ethnicity.
Analysis Plan
We used linear regression to test H1, predicting each T2 preference from T1 difficulty-asimpossibility endorsement while controlling for T1 difficulty-as-importance endorsement. Our a
priori power analysis using G*Power (Faul, et al., 2007) recommended 200 participants to
achieve a power of .80 to test the effect of each independent variable using a two-tailed test, a
small effect size (f2 = 0.04), and an alpha of .05. As such, our N = 216 is sufficient. We excluded
5 participants who did not follow instructions to add exactly 10 puzzles to their set.

50
Results
Descriptive Analyses
Students endorsed difficulty-as-importance (M = 4.69, SD = 0.93, � = .69) and rejected
difficulty-as-impossibility (M = 2.67, SD = 1.24, � = .80). The two scales were distinct, r = -.03,
p = .640. Students preferred easy puzzles (M = 3.54, SD = 1.10) and felt neutral about difficult
ones (M = 2.95, SD = 1.18). On average, students added 3.80 (SD = 2.01) difficult puzzles to
their puzzle sets. Students who preferred easy puzzles did not prefer difficult ones (r = -.55, p <
.001). Preferences corresponded to action; students who preferred easy puzzles put fewer
difficult ones in their set (r = -.56, p < .001); those who preferred difficult puzzles put more
difficult ones in their set (r = .69, p < .001).
H1: Difficulty-as-Impossibility Endorsers Are Sensitive to Their Preference for Ease
Controlling for students’ T1 difficulty-as-importance endorsement, students’ T1
difficulty-as-impossibility endorsement predicted their T2 preferences. At T2, higher T1
difficulty-as-impossibility endorsers preferred easy puzzles (β = 0.20, t = 3.49, p < .001), not
difficult ones (β = -0.24, t = -3.89, p < .001) and were less likely to add difficult puzzles to their
puzzle set (β = -0.35, t = -3.21, p = .002). T1 difficulty-as-importance endorsement did not affect
T2 preferences (prefer easy puzzles β = 0.13, t = 1.72, p = .087; prefer difficult puzzles β = 0.05,
t = 0.58, p = .566; adding difficult puzzles β = -0.13, t = -0.89, p = .379).
Discussion
Study 1 results replicate and extend correlational evidence that people who endorse
difficulty-as-impossibility find easier means of goal attainment to be more effective and say they
will use them while people who endorse difficulty-as-importance are agnostic as to the means of
goal attainment (Kiper et al., 2024). We find that difficulty-as-impossibility endorsers have

51
insight into their preference not to engage with difficulty and that difficulty-as-importance
endorsers have insight into their indifference as to means. In Studies 2a and 2b, we turn to the
process by which difficulty-as-impossibility endorsement and features of situations interact.
Studies 2a and 2b
Method
Sample
We pre-registered both studies and obtained University of Southern California IRB
approval (#UP-15-0055). Undergraduate students participated for subject pool credit at two
points in the semester, receiving credit each time (N = 464, 65.30% female, age M = 19.98, SD =
2.07, 38.88% Asian American, 29.59% White American, 11.45% Hispanic American). Sample
characteristics by study are shared in the Supplemental Materials.
Procedure and Measures
Students filled out a prescreening questionnaire at T1 (mostly demographics other
researchers requested) with the Fisher and Oyserman (2017) 4-item difficulty-as-importance and
4-item difficulty-as-impossibility scales near the end (statements presented in randomized order,
Appendix B). At T2, they read: “In the next few screens, we will present word-puzzles for you to
solve. You will see a scrambled letter representation of a word which presents a cognitive
puzzle. To solve the puzzle, you have to rearrange the letters to form a correctly spelled word
(not an abbreviation) in English using every letter and each letter only once. For example, the
letters TERDN can be rearranged to form the correctly spelled word TREND. For each puzzle,
show your solution by typing it into the box.” Figure 1 presents an example of what participants
saw.

52
Figure 1
Example Item Screen
Each participant was presented with sixteen word-puzzles. Eight puzzles formed the
moderately difficult single-solution core set (Gilhooly & Johnson, 1978) and were constant
across conditions. The other eight puzzles formed the context, as detailed next. In Study 2a and
2b, contexts were a wide-open path (eight easier to solve multiple-solution puzzles,
EnchantedLearning.com, 2019) and a blocked path (eight unsolvable puzzles, Calef et al., 1992).
In Study 2b, we added a third, open-path context (eight puzzles as difficult as the core set,
Gilhooly & Johnson, 1978). All word-puzzles are shared in Supplemental Materials Table S1.
Students then completed the difficulty-as-importance and difficulty-as-impossibility scales again,
rated “How easy or difficult was the word-puzzle task?” from 1 = very easy, to 7 = very difficult,
and in Study 2b rated "To me, these puzzles felt important to solve." and "To me, these puzzles
felt impossible to solve." from 1 = strongly disagree to 6 = strongly agree. Finally, for
descriptive purposes, students shared their gender, age, race-ethnicity, and verbal SAT/ACT
percentiles. We do not report the latter as most said they could not recall.
Analysis Plan
We pre-registered a repeated-measures ANOVA to test H2. We tested the main and
interaction effects of time and context on difficulty-as-impossibility endorsement. T1 and T2
difficulty-as-impossibility endorsement scores formed the within-subjects factor and context was
the between-subjects factor. Though not pre-registered, we conducted an ANCOVA analysis
pooling across Study 2a and Study 2b to increase the robustness of our estimates with Study and

53
difficulty-as-importance endorsement scores as covariates. We conducted two additional
exploratory analyses. First, we looked at the context sensitivity of difficulty-as-importance
endorsement by substituting difficulty-as-importance for difficulty-as-impossibility in our H2
analyses. Second, we looked at whether T1 difficulty-as-impossibility influenced T2
performance (number of core word-puzzles solved correctly) and persistence (seconds from
opening the screen to responding) using simple regression models with T1 difficulty-asimpossibility as the predictor and performance or persistence as the outcome.
Power. Our a priori power analysis using G*Power (Faul, et al., 2007) recommended
200 participants to achieve a power of .80 to test the interaction effect of time and context on
difficulty-as-impossibility using a two-tailed test, a small effect size (f
2 = 0.02), and an alpha of
.05. We aimed to collect 100 participants per condition but fell a bit short after exclusions
(pooled data, n = 186 wide-open path condition, n = 94 open path condition, Study 2b only, and
n = 184 blocked path condition).
Exclusions. As per our pre-registration, we excluded outlier puzzle responses from mean
performance and persistence scores. We defined outlier responses as responses that were too fast
(M-2.5 SD) or too slow (M+2.5 SD) compared to the sample (between-person) mean. We
excluded participants who did not attempt any of the eight core puzzles (n = 26) or spent less
than five seconds on average on each puzzle (n = 15). Finally, we checked for suspicious data
and excluded participants who mistook the puzzle task for a mnemonics task (n = 2).
Results
Descriptive Analyses
Students endorsed difficulty-as-importance (T1 M = 4.73, SD = 0.95, � = .70; T2 M =
4.72, SD = 1.08, � = .80) and rejected difficulty-as-impossibility (T1 M = 2.75, SD = 1.30, � =

54
.81; T2 M = 2.77, SD = 1.30, � = .84). Difficulty-as-importance and difficulty-as-impossibility
endorsement scores were distinct (T1 r = .02, p = .713; T2 r = -.02, p = .674).
Regarding the context manipulation, our manipulation check suggests that our
experimental design was successful in manipulating students’ experience of difficulty. Students
in the wide-open path context (M = 4.23, SD = 1.35) rated the task as less difficult than students
in the other contexts (open path M = 4.68, SD = 1.31, t= 2.68, p = .008, blocked path M = 5.53,
SD = 0.87, t = 11.02, p < .001). Students in the open path context rated the task as less difficult
than students in the blocked path context, t = 5.68, p < .001. In Study 2b, we added a second set
of manipulation checks assessing subjective importance and impossibility. Students found the
puzzle task important (M = 4.15, SD = 1.78) no matter the context they were randomly assigned
(for details, see Supplemental Materials). Random assignment to context did shift how
impossible the task felt, with significant differences between the blocked (M = 3.75, SD = 1.72)
and the other contexts (open M = 2.82, SD = 1.65, t = 3.81, p < .001, wide-open M = 2.39, SD =
1.45, t = 5.88, p < .001), which did not significantly differ from each other, t = 1.87, p = .062.
Students on average were able to solve 66.25% of the core puzzles (M = 5.30, SD = 2.24)
no matter the context they were randomly assigned (for details, see Supplemental Materials).
Students randomly assigned to the wide-open path context spent more time on the core puzzles
(M = 28.82, SD = 14.63) than those assigned to the open path context (M = 24.43, SD = 12.54, t
= -2.61, p = .010) and the blocked path context (M = 24.94, SD = 13.49, t = -2.65, p = .008),
which did not differ from each other, t = 0.31, p = .754.

55
H2: Difficulty-As-Impossibility Endorsement Shifts Depending on the Context of a Task,
Increasing After Experiencing a Context of Difficulty or Impossibility and Decreasing After
Experiencing a Context of Ease
We found the predicted context x time interaction, F(2, 461) = 4.27, p = .015, f= 0.12,
suggesting that difficulty-as-impossibility endorsement shifted as a function of both time and
context. No main effects were significant (context, F(2, 461) = 0.35, p = .707, f= 0.04; time F(1,
461) = 0.59, p = .444, f= 0.03; study number F(1, 460) = 1.06, p = .305, f= 0.04; difficulty-asimportance F(1, 858) = 0.01, p = .910, f= 0.01).
Next, we conducted a pairwise comparison analysis to unpack the significant condition x
time interaction effect, examining the effect of time separately in each context. As depicted in
Figure 2 (red bars), students in the wide-open path context endorsed difficulty-as-impossibility
less at T2 than they did at T1 (t = 2.02, p = .044). The reverse pattern occurred among students in
the other contexts (open path t = -1.30, p = .194; blocked path t = -1.71, p = .088) though these
effects did not attain significance at the p < .05 level. Exploratory analyses reveal that this
reverse pattern of decline was stable at the p < .05 level when both difficult path conditions were
combined into a single condition (t = -2.15, p = .032). Moreover, effects were specific to
difficulty-as-impossibility. As depicted in Figure 2 (green bars), difficulty-as-importance
endorsement did not shift (context x time interaction F(2, 463) = 0.56, p = .574, f= 0.04; context
F(2, 460) = 1.78, p = .170, f= 0.08; time F(1, 461) = 0.01, p = .946, f= 0.00; study number F(1,
460) = 0.02, p = .896, f = 0.01; difficulty-as-impossibility F(1, 858) = 0.01, p = .939, f = 0.00).

56
Figure 2
Difficulty-as-Impossibility Endorsement Increases in Difficult Contexts and Decreases in Easier
Ones, While Difficulty-as-Importance Endorsement Does Not Shift
Exploratory Analysis: Does T1 Difficulty-as-Impossibility Affect T2 Performance and
Persistence?
We explored whether higher T1 difficulty-as-impossibility endorsers performed worse
(solved fewer puzzles) and persisted less (fewer average seconds per puzzle) at T2 when faced
with a difficult task. We found significant context-based differences for only performance (for
persistence: wide-open path β = -1.33, t = -1.64, p = .103; open path β = -1.12, t = -1.09, p =
.281; blocked path β = 0.49, t = 0.63, p = .527). Controlling for T1 difficulty-as-importance
endorsement (for details, see Supplemental Materials), T1 difficulty-as-impossibility
endorsement predicted worse performance for students randomly assigned to the wide-open path
context (β = -0.30, t = -2.45, p = .015) but not for students in the other contexts (open path β = -
0.28, t = -1.41, p = .163; blocked path β = 0.15, t = 1.26, p = .210).
Discussion
Students endorsed difficulty-as-impossibility less after experiencing a difficult puzzle
task (the core puzzle set) in an easier context of puzzles that most people can solve. In contrast,

57
people who experienced the same difficult task in a context of difficult or even unsolvable
puzzles endorsed difficulty-as-impossibility more. As a general rule of thumb, traits are less
likely to matter in strong situations that suggest a particular inference and more likely to matter
in weaker situations that are more open to interpretation. We had predicted that our wide-open
path situation would be a strong situation, one which would draw attention to the possibility of
success and hence engage efforts. Instead, our wide-open path situation might have been weak
enough to allow for an effect of trait difficulty-as-impossibility on engagement. People who
endorsed difficulty-as-impossibility more going into the success-likely (wide-open path)
situation performed worse than people who went into the same situation with lower difficulty-asimpossibility endorsement. We did not see such situation-sensitivity for difficulty-as-importance.
In Study 3, we use secondary data analysis to explore situational sensitivity across longer time
lags and earlier developmental phases.
Study 3
Method
Sample
The analysis is a secondary data analysis of the control groups in two data sets, as
described below. Data were from SBIR #2R44 MD008582-02 National Institute of Minority
Health and Health Disparities, IRB #2016-002, Sterling IRB; #U411C150011 Department of
Education, IRB #18020, University of Illinois IRB, McREL IRB #00005432, and the University
of Southern California IRB #HS-18-00646.
Our analytic sample (N = 733, 53.02% female, age M = 14.05, SD = 1.57; 58.23% White,
29.66% Hispanic) consists of two samples of students assigned to the control groups in their
respective studies (Data Set 1: n = 88, 9 to 15 year old elementary to middle school after-school

58
program students in Alabama, Missouri, and Florida; Data Set 2: n = 645, 11 to 19 year old
middle to high school students in Colorado public schools). Sample characteristics by study are
shared in Supplemental Materials Table S2.
Procedure and Measures
All students completed the Oyserman, Destin, and Novin (2017) difficulty-as-importance
and difficulty-as-impossibility scales (Appendix B) and self-reported demographics. Data were
collected at three points (T1 n = 733, T2 n = 660, T3 n = 348). From T1 to T2, 1 to 37 weeks
elapsed (after-school students M = 4.73 weeks, SD = 2.29, in-school students M = 33.83 weeks,
SD = 0.93). From T2 to T3, 1 to 56 weeks elapsed (after-school M = 9.37 weeks, SD = 8.62, inschool M = 35.84 weeks, SD = 6.99). From T1 to T3, 41 to 99 weeks elapsed (after-school
students M = 47.48 weeks, SD = 5.09, in-school students M = 82.58 weeks, SD = 10.89). As
such, the T1 to T3 time gap spanned at least one semester for after-school students and at least
one academic school year for the in-school students.
Analysis Plan
To test H3, we calculated ICCs and their 95% confidence intervals (CIs) using the ICC R
package (v2.4.0; Wolak et al., 2012) with measurement occasions nested within persons. The
ICC is a descriptive statistic calculated as the proportion of within-person variance to betweenperson variance. An ICC under .50 indicates more situated variability (larger differences in
repeated responses within persons). We also explored endorsement scores over the three time
points. To do so, we created two two-level models in which time (months from T1 to T3),
developmental phase (T1 before or during high school), and their interaction were the predictor
variables and difficulty-as-importance endorsement or difficulty-as-impossibility endorsement
was the predicted variable (see Appendix C for multilevel equations)

59
Results
Descriptive Analyses
Table 1
Study 3 Scale Means, Standard Deviation (SD), and α Reliability
Time Point Difficulty-as-Importance Difficulty-as-Impossibility
Mean SD α Mean SD α
Time 1 3.37 0.82 .88 2.49 0.84 .85
Time 2 3.17 0.85 .91 2.47 0.81 .86
Time 3 3.23 0.74 .89 2.60 0.78 .88
Note: Scales from 1 = strongly disagree to 5 = strongly agree.
Difficulty-as-importance and difficulty-as-impossibility scales were reliable (�s > .85)
and distinct (T1 r = -.10, p = .006; T2 r = .07, p = .054; T3 r = .10, p = .068). Though the scales
were significantly correlated at T1, Ratner’s (2009) rules of thumb for interpreting correlations
indicate the correlation is still very small (r ≤ 0.19 is very small, 0.2 ≤ r ≤ 0.39 is small, 0.40 ≤ r
≤ 0.59 is moderate, 0.6 ≤ r ≤ 0.79 is large and 0.8 ≤ r ≤ 1 is very large). As Table 1 details,
students endorsed difficulty-as-importance and rejected difficulty-as-impossibility.
H3: Difficulty-as-Importance Endorsement Varies Between Persons More Than Difficulty-asImpossibility Endorsement
The difficulty-as-importance endorsement ICC was .45 (95% CI [.40, .50]) and the
difficulty-as-impossibility endorsement ICC was .42 (95% CI [.36, .47]), suggesting that
difficulty-as-importance may vary more between-persons than difficulty-as-impossibility and
that the latter is particularly more state-like than trait-like. We then explored whether this pattern
was stable by developmental phase. Given our data distribution, we created two relatively equalsized groups: students who at T1 were not yet in high school (n = 280) and students who at T1
were already in high school (n = 453). As shown in the top panel of Figure 3, among students in
high school at T1, the difficulty-as-impossibility ICC 95% CI is to the left of the .5 mark,
suggesting more situation-based than person-based variability (red line). The difficulty-as-

60
importance ICC 95% CI crosses the .5 mark, suggesting about equal state-like and trait-like
variability (green line). As the lower panel of Figure 3 depicts, the reverse pattern occurs among
students not yet in high school at T1. The difficulty-as-impossibility ICC 95% CI crosses the .5
mark, suggesting about equal state-like and trait-like variability (red line) and the difficulty-asimportance ICC 95% CI is to the left of .5, suggesting more situation-based than person-based
variability (green line).
Figure 3
Comparison of ICCs for Difficulty-as-Importance and Difficulty-as-Impossibility Endorsement
Among High School and Pre-High School Students
We explored whether difficulty-as-importance or difficulty-as-impossibility endorsement
changed over time (months from T1 to T3) and developmental phase (T1 before or during high
school). Difficulty-as-importance endorsement decreased (β = -0.01, t = -3.54, p < .001) and
difficulty-as-impossibility endorsement increased (β = 0.01, t = 2.41, p = .016) over time.
Developmental phase mattered for difficulty-as-importance endorsement (higher among students
not yet in high school at T1, β = 0.26, t = 3.87, p < .001) but not difficulty-as-impossibility
endorsement (β = -.02, t = -0.28, p = .781). Time did not interact with developmental phase

61
(difficulty-as-importance β = -.01, t = -1.53, p = .127; difficulty-as-impossibility β = -.01, t = -
1.15, p = .251).
Discussion
Difficulty-as-impossibility and difficulty-as-importance endorsement varied
meaningfully between and within students over time. Our results suggest that as students
advance from middle school to high school, they come to endorse difficulty-as-impossibility
more and difficulty-as-importance less, though on average they reject the former and endorse the
latter. We infer from Study 3 results that our Study 1 and Study 2 results are not due to
differences in the extent that difficulty-as-importance and difficulty-as-impossibility are more
trait-like or state-like but rather due to differences in sensitivity to specific features of the
situations we created. To begin addressing this gap, in Studies 4 to 6 we used a daily diary
approach to obtain a more ecologically valid sampling of situations and we assessed both focus
on goal progress (meaningful engagement with schoolwork) and self-regulatory means of
making progress (goal self-efficacy, self-esteem, and self-compassion).
Studies 4 to 6
Method
We pre-registered Study 6. We obtained University of Southern California IRB approval
(IRB #UP-19-00570). Undergraduates (N = 260) signed up for a 14-day daily diary study for
subject pool course credit (see Table 2 for sample characteristics).
Table 2
Studies 4 to 6 Sample Size and Characteristics by Study and Pooled
Study 4 Study 5 Study 6 Pooled
Sample Size 123 76 61 260
Number of Daily Reports 1209 840 740 2789
Mean Reports per Person (SD) 11.11 (2.24) 11.30 (2.12) 12.33 (1.66) 11.45 (2.13)
Mean Age (SD) 20.01 (2.26) 19.73 (1.48) 19.98 (1.92) 19.95 (2.05)
% Female 65.85 70.00 77.05 69.64

62
% Asian American 26.13 23.19 45.00 30.00
% White American 21.62 31.88 20.00 24.17
% Hispanic American 13.51 11.59 6.67 11.25
Note. SD = standard deviation. Mean Reports per Person (of a maximum of 14) varies by
measure (see Table 5).
Procedure and Measures
Appendix B provides our trait-level and day-level scales. Students completed a baseline
trait questionnaire with the Fisher and Oyserman (2017) 4-item difficulty-as-importance and 4-
item difficulty-as-impossibility scales followed by the self-regard variables (self-esteem, goal
self-efficacy, self-compassion), and demographics (gender, age, and race-ethnicity). Students
then watched an introductory video explaining that they would receive a link each evening for 14
days and should complete each daily diary as close to bedtime as possible. In Study 4, students
had from 9 p.m. to 11 a.m. the next day to complete the daily link, following prior diary studies
(e.g., Newman et al., 2020; Oishi et al., 2007). Given student feedback indicating that 9 p.m. was
too late, we opened the link at 5 p.m. in Studies 5 and 6 and retained the 11 a.m. next day cut off.
The daily diary included day-level versions of our trait scales as well as items assessing
meaningful engagement with schoolwork [“Today I did schoolwork the hard way (e.g., I took
notes, I read the assignments before class)”] and less meaningful engagement with schoolwork
[“Today I did schoolwork the easy way (e.g., I skimmed the readings, came to class but didn’t
pay much attention)”]. Correcting an omission in Study 4, in Studies 5 and 6, we presented the
daily difficulty-as-importance and difficulty-as-impossibility measures only to students who
reported that they had experienced difficulty that day (“Did you experience difficulty working on
a task today?”) and only presented the daily goal self-efficacy measure to students who reported
that they engaged with a goal that day (“Did you work on any of your goals today?”). Across

63
Studies 4 to 6, we presented the daily schoolwork engagement items, self-esteem measure, and
self-compassion measure to all students each day.
Analysis Plan
To test H3, we calculated ICCs and their 95% confidence intervals (CIs) using the ICC R
package (v2.4.0; Wolak et al., 2012) with measurement occasions nested within persons. To test
H4, we created four two-level models. In the first two-level model, trait difficulty-as-importance
endorsement was the level-2 predictor and daily meaningful school engagement was the
predicted variable. In the other three two-level models, daily difficulty-as-impossibility
endorsement was a group-mean centered, level-1 predictor and the predicted variables were daily
goal self-efficacy, daily self-esteem, or daily self-compassion. By group-mean centering daily
predictor variables, we center scores on each person’s 14-day average (Hoffman, 2015, Chapter
9). All multilevel equations are shared in Appendix C. For effect sizes, we follow Rights and
Sterba’s (2019) framework and report R2 effect sizes calculated using model-implied variances.
Specifically, we report the proportion of total outcome variance explained by all predictors via
fixed slopes for each model. As recommended by Nezlek (2012) and Kreft and de Leeuw (1998),
we do not report R2 separately for level-1 and level-2. We conducted exploratory analyses to
address alternative explanations for our H4 results (see Supplemental Materials).
Power. Our a priori power analysis using G*Power (Faul, et al., 2007) recommended
196 participants to achieve a power of .80 to test the effect of each independent variable using a
two-tailed test, a small-to-medium effect size (f2 = 0.05), and an alpha of .05. We collected as
large a sample as possible each semester. For trait analyses, each sample was underpowered, so
we combined samples to create a pooled dataset N = 260. We are sufficiently powered to test

64
daily-level relationships separately for each individual study as we have above the recommended
minimum of 100 level-2 observations (Maas & Hox, 2005).
Exclusions. We retained 83.3% of daily reports (analytic daily report n = 2,789; M = 11.5
per person daily reports, see Supplemental Materials for exclusion details). As per Nezlek (2012)
and Meade and Craig (2012), we excluded people completing less than five daily entries, kept
only the first report on days a multiple was provided, and excluded daily reports completed
outside the time window or with failed attention check.
Results
Descriptive Analyses
Across Studies 4 to 6, meaningful (58.11% of daily reports) and less meaningful (42.57%
daily reports) engagement with schoolwork occurred at about equal frequency. Whether one or
the other occurred on a particular day was weakly, negatively correlated (r = -.07, p < .001),
suggesting the two were not flip sides of one another. In Studies 5 and 6, students reported each
day if they had experienced difficulty while working on a task and if they had worked on a goal.
On average, at least once during the two weeks 98.23% of students reported experiencing
difficulty while working on a task and 98.04% reported working on their goals. They reported
experiencing difficulty (Study 5 48.33%, Study 6 49.32%) and working on a goal (Study 5
57.35%, Study 6 65.68%) in about half of their daily reports, but one did not necessarily imply
the other, r = 0.22, p < .001. Difficulty-as-importance and difficulty-as-impossibility
endorsement scores were distinct (trait-level r = -.21, p < .001; day-level r = -0.13, p < .001). As
shown in Table 3, scales were reliable and students endorsed difficulty-as-importance, goal selfefficacy, self-esteem, and self-compassion and rejected difficulty-as-impossibility (see
Supplemental Materials Table S3 for statistics by study).

65
Table 3
Studies 4 to 6 Pooled Scale Means, Standard Deviation (SD), and α Reliability
Variable Trait Measure Day Measure
Mean SD α Mean SD α
Difficulty-as-Importance 4.28 0.86 .80 4.13 1.10 .78
Difficulty-as-Impossibility 2.33 0.88 .75 2.58 1.05 .77
Goal Self-Efficacy 4.16 0.73 .86 3.98 0.92 .76
Self-Esteem 4.12 0.93 .90 4.27 1.11 .80
Self-Compassion 3.08 0.94 .83 4.02 1.10 .69
Note. Response scales 1 = strongly disagree, 6 = strongly agree. We computed day-measure
reliabilities using a three-level model approach with scale items nested within days and days
within persons (Nezlek, 2017).
Figure 4 depicts difficulty-as-importance (green) and difficulty-as-importance (red)
endorsement distributions across each day of the daily diary using violin plots. As can be seen,
endorsement scores were distributed across the full scale on most days and did not converge
across the 14 days. Our following analyses seek to uncover patterns in these distributions.
Figure 4
Difficulty Mindset Endorsement Distributions Across 14 Days
As shown in Table 4, difficulty-as-impossibility was moderately and negatively
correlated with the other measures and difficulty-as-importance was weakly and positively
correlated with them, except for trait self-esteem and trait self-compassion. Rules of thumb for
interpreting correlations suggests the trait-level correlation between difficulty-as-importance and

66
difficulty-as-impossibility is small while the day-level correlation is very small (Ratner, 2009).
Table 4
Studies 4 to 6 Pooled Trait Measure and Day Measure Correlations
Scale Trait Measures Day Measures
1 2 3 4 5 1 2 3 4 5
(1) Difficulty-as-Importance -- -.21 .20 .05 -.03 -- -.13 .17 .19 .17
(2) Difficulty-as-Impossibility -- -.54 -.46 -.26 -- -.36 -.41 -.30
(3) Goal Self-Efficacy -- .51 .33 -- .51 .40
(4) Self-Esteem -- .65 -- .73
(5) Self-Compassion -- --
Note. ps < .001 except trait difficulty-as-importance and trait self-esteem (p = .780) and trait
difficulty-as-importance and trait self-compassion (p = .780). We calculated day measure
correlations by group-mean centering variables to account for the clustered nature of the data.
H3: Difficulty-as-Importance Endorsement Varies Between Persons More Than Difficulty-asImpossibility Endorsement
As detailed in Table 5, difficulty-as-importance endorsement ICCs ranged from .51 to .62
while difficulty-as-impossibility endorsement ICCs ranged from .44 to .50. In comparison, ICCs
for goal self-efficacy, self-esteem, and self-compassion never exceeded .48 and reached as low
as .34. Difficulty mindset endorsement ICCs with 95% CIs are depicted in Figure 5. In all studies
and in the pooled dataset, the 95% CIs of the difficulty-as-impossibility ICCs cross the .5 mark,
suggesting about equal state-like and trait-like variability (red lines). In Study 1 and the pooled
dataset, the 95% CIs of the difficulty-as-importance ICCs are to the right of the .5 mark,
suggesting more person-based than situation-based variability (green lines).

67
Table 5
Studies 4 to 6 Number of Daily Reports and Intraclass Correlations (ICCs)
Variable Study 4 Study 5 Study 6 Pooled
n ICC
[95% CI]
n ICC
[95% CI]
n ICC
[95% CI]
n ICC
[95% CI]
Difficulty-asImportance
1209 .62
[.55, .68]
406 .54
[.44, .65]
365 .51
[.40, .63]
1980 .59
[.55, .64]
Difficulty-asImpossibility
1209 .44
[.37, .52]
406 .45
[.35, .56]
365 .50
[.39, .62]
1980 .46
[.41, .52]
Goal SelfEfficacy
1209 .44
[.37, .52]
483 .46
[.36, .57]
486 .37
[.27, .49]
2178 .45
[.39, .50]
Self-Esteem 1209 .45
[.38, .53]
838 .34
[.26, .43]
740 .48
[.38, .58]
2787 .43
[.38, .48]
Self-Compassion 1209 .48
[.41, .55]
839 .41
[.32, .51]
740 .44
[.35, .55]
2788 .45
[.40, .50]
Less Meaningful
School
Engagement
1206 .16
[.11, .23]
840 .26
[.19, .35]
740 .17
[.11, .26]
2786 .20
[.16, .24]
Meaningful
School
Engagement
1208 .19
[.14, .26]
840 .22
[.15, .30]
740 .27
[.19, .37]
2788 .22
[.18, .27]
Note. n = number of daily reports. CI = confidence interval. An ICC < .50 suggests greater
contextual sensitivity. Daily reports vary in number because in Studies 5 and 6, students only
rated difficulty-as-importance and difficulty-as-impossibility on days they experienced difficulty
and goal self-efficacy on days they engaged with a goal.
Figure 5
Comparison of ICCs for Difficulty-as-Importance and Difficulty-as-Impossibility Endorsement
Across Studies and Pooled

68
H4: Trait Difficulty-As-Importance Endorsement is Associated with Meaningful Schoolwork
Engagement; Daily Difficulty-As-Impossibility Endorsement is Associated with Daily SelfRegulatory Means
Trait-level analyses were underpowered at the study level, so we focus on the pooled
dataset. As shown in Figure 6, controlling for study number and trait difficulty-as-impossibility
endorsement, people higher in trait difficulty-as-importance endorsement were more likely to
report engaging with school in a meaningful way on any given day (pooled data β = 0.28, z =
2.78, p = .005, R2 = .04). Trait difficulty-as-impossibility endorsement was not associated with
meaningful school engagement (pooled data β = 0.03, z = 0.34, p = .738).
As detailed in the Supplemental Materials, exploratory analyses found that daily
difficulty-as-importance endorsement was also positively associated with meaningful
schoolwork engagement (in 2 of 3 studies) while daily difficulty-as-impossibility endorsement
was not. We also explored whether trait and daily difficulty mindset endorsements were
associated with less meaningful engagement with schoolwork. Trait difficulty mindset
endorsements were not associated with less meaningful schoolwork engagement. Daily
difficulty-as-importance endorsement was negatively associated with less meaningful
schoolwork engagement (in 1 of 3 studies) while daily difficulty-as-impossibility endorsement
was not.

69
Figure 6
Trait Difficulty-as-Importance and Not Trait Difficulty-as-Impossibility Predicts Daily
Meaningful Engagement with School
Next, we turn to effects on self-regulatory means, as depicted in Figure 7. Controlling for
study number, time (day of diary report), and daily difficulty-as-importance endorsement, on
days students endorsed difficulty-as-impossibility more, they felt less efficacious (pooled data β
= -0.32, t = -15.57, p < .001, R2= .08), liked themselves less (pooled data β = -0.45, t = -19.02, p
< .001, R2 = .11), and were less self-compassionate (pooled data β = -0.32, t = -13.42, p < .001, R2
= .07). Daily difficulty-as-importance endorsement was also positively associated with daily goal
self-efficacy (pooled data β = 0.15, t = 6.55, p < .001), self-esteem (pooled data β = 0.18, t =
6.88, p < .001), and self-compassion (pooled data β = 0.17, t = 6.49, p < .001). We followed up
by using unstandardized regression coefficients and Wald tests to determine if the magnitudes of
the daily difficulty-as-impossibility endorsement effects were significantly larger than the
magnitudes of the daily difficulty-as-importance endorsement effects. Indeed, the effects of daily
difficulty-as-impossibility endorsement on daily goal self-efficacy (X2
= 263.90, p < .001), self-

70
esteem (X2 = 368.29, p < .001), and self-compassion (X2 = 220.02, p < .001) were significantly
larger than the effects of daily difficulty-as-impossibility endorsement.
Figure 7
Self-Regulatory Means are Less Positive When Today’s Difficulty-as-Impossibility Endorsement
is Higher and More Positive When Today’s Difficulty-as-Importance Endorsement is Higher
As detailed in the Supplemental Materials, daily difficulty-as-impossibility and difficultyas-importance endorsement effects were significant for each study except for one statistically
insignificant effect of daily difficulty-as-importance endorsement on daily goal self-efficacy in
Study 5. Additionally, we found that trait difficulty-as-importance and daily self-regulatory
means were not significantly associated, but trait difficulty-as-impossibility and daily selfregulatory means were.
EQ1: How Do Trait and State Difficulty Beliefs Differ?
First, we explored how trait and state difficulty beliefs differed descriptively. Following
Nezlek and colleagues (2001), we evaluated the correlations between trait and state beliefs by
examining the reduction in residual variance of state beliefs once trait beliefs were included as a
person-level predictor. Starting with an unconditional (null) model, the residual variance of daily

71
difficulty-as-importance endorsement decreased by 27% (.64 to .47) once trait difficulty-asimportance beliefs were included. This yielded a .52 correlation (the square root of .27) between
daily and trait difficulty-as-importance beliefs. For difficulty-as-impossibility beliefs, we found
that residual variance decreased by 26% (.52 to .38) once trait difficulty-as-impossibility beliefs
were included, yielding to a .51 correlation between daily and trait difficulty-as-impossibility
endorsements. Following Ratner’s (2009) rules of thumb, these correlations between trait and
state difficulty beliefs are moderately correlated; trait and state difficulty beliefs are related to
each other, but they are not redundant by any means. This is also displayed in the reduction in
unexplained variance of each daily difficulty belief by about a quarter when the corresponding
trait beliefs are accounted for.
Next, we explored how trait and state difficulty beliefs differed in their associations with
daily self-regulatory means and daily goal engagement. Following Kernis (2005), we examined
the extent to which the stability of daily difficulty-as-importance and difficulty-as-impossibility
beliefs may matter separately from or interact with trait difficulty endorsement levels. To
calculate instability, we computed the within-person standard deviation of each participant’s
daily difficulty-as-importance and difficulty-as-impossibility endorsement scores. Higher
standard deviation values indicate greater instability. We then predicted each measure of daily
self-regulatory means and daily goal engagement from daily difficulty-as-importance beliefs
instability and trait-level difficulty-as-importance beliefs (and their interaction) and daily
difficulty-as-impossibility beliefs instability and trait difficulty-as-impossibility beliefs (and their
interaction).
As depicted in panel (a1) of Figure 8, greater trait difficulty-as-impossibility beliefs and
instability of daily difficulty-as-impossibility beliefs are negatively associated with daily goal

72
self-efficacy, and there is a significant interaction such that the negative relationship between
trait beliefs and daily goal self-efficacy is less pronounced when instability is high. At the same
time, as shown in panel (a2), greater trait difficulty-as-importance beliefs and lesser instability of
daily difficulty-as-importance beliefs are positively associated with daily goal self-efficacy, and
there is a significant interaction such that the relationship between trait beliefs and daily goal
self-efficacy is attenuated when instability is high. We find the same pattern of effects for daily
self-esteem in panels (b1) and (b2) and self-compassion in panels (c1) and (c2), though the
significance of these effects at a p < .05 level varies (for details, see Supplemental Materials
Table S4). As shown in panels (d1) and (d2), daily meaningful schoolwork engagement, main
effects of trait beliefs, main effects of daily belief instability, and interaction effects were not
significant.

73
Figure 8
Predicting Daily Self-Regard and Meaningful Schoolwork Engagement From Trait-Level
Difficulty Beliefs, State-Level Difficulty Belief Instability, and Their Interaction Effects

74
Discussion
Descriptively, difficulty-as-importance is slightly more trait-like than difficulty-asimportance, but how much people endorse each mindset varies meaningfully both across
situations and between people. At the same time, the outcome variables each mindset is
associated with differs. Trait difficulty-as-importance is associated with daily goal engagement,
trait difficulty-as-impossibility is not. At the day-level, both difficulty mindsets are associated
with daily self-regulatory means. On days people endorse difficulty-as-impossibility more, they
also experience lower goal self-efficacy, self-esteem, and self-compassion. On days people
endorse difficulty-as-importance more, they experience greater goal self-efficacy, self-esteem,
and self-compassion. However, the depleting effects of daily difficulty-as-impossibility beliefs
on these means of self-regulation are significantly stronger than the facilitative effects of daily
difficulty-as-importance beliefs.
General Discussion
We examined how difficulty mindsets and contexts intersect across six studies with over
1,500 participants using four methodologies. People who endorse difficulty-as-impossibility
have insight into their preference for less effortful means, opting to limit their engagement with
difficult tasks (Study 1). They also perform worse on difficult tasks embedded in a success-likely
context (Studies 2a, 2b). At the same time, we found effects of context on how much people
endorse difficulty-as-impossibility after engaging with a difficult task (Studies 2a, 2b). People
endorse difficulty-as-impossibility less after engaging with a difficult task in a context which
suggests that success is likely and endorse difficulty-as-impossibility more after engaging in a
difficult task in contexts which suggest that success is unlikely. In these studies, we did not see
shifts in difficulty-as-importance endorsement. How much people endorse difficulty-as-

75
importance and difficulty-as-impossibility varies by situation and between persons (Studies 3 to
6) and difficulty-as-importance endorsement declines in the transition from middle to high
school (Study 3). On any given day, students who endorse difficulty-as-importance are more
likely to engage meaningfully with their schoolwork, suggesting a focus on goal progress
(synthesized across Studies 4 to 6). On any given day, higher difficulty-as-impossibility is
consistently associated with decreases in self-regulatory means (less efficacy, self-esteem, and
self-compassion, Studies 4 to 6). On any given day, endorsing difficulty-as-importance is
associated with more self-regulatory means, though this boost is more modest than the decrease
associated with endorsing difficulty-as-impossibility (Studies 4 to 6).
Theoretical Contributions
Identity-based motivation theory predicts that difficulty-as-importance and difficulty-asimpossibility beliefs are situated and shape action (for a review, Oyserman & Horowitz, 2023).
Our results advance identity-based motivation theory and contribute to the broader literature on
self-regulation and goal pursuit by addressing four research gaps. First, prior research by Kiper
and colleagues (2024) found that difficulty-as-importance is associated with a focus on ends
(making goal progress) and difficulty-as-impossibility is associated with a preference for easier
means (the process of making goal progress). We document that people who endorse difficultyas-impossibility are aware of their preference for easier means while people who endorse
difficulty-as-importance are aware of their indifference to means. Regarding the broader
literature on self-regulation, our results are relevant to expectancy-value theories which predict
people will engage with goals they find important and worthwhile and disengage when the
probability of success is low enough (e.g., motivational intensity theory, Brehm & Self, 1989;
Richter et al., 2016). Motivational intensity theory specifies that people will adjust the amount of

76
effort they are willing to invest based on perceptions of worthwhileness (i.e., a good use of
motivational resources) when task difficulty is unfixed. In Study 1, we created a scenario in
which students dictated the difficulty of their puzzle set and found that difficulty-as-impossibility
endorsers added less difficult puzzles to their puzzle set. Our results suggest difficulty-asimpossibility endorsement may influence the amount of effort people are willing to invest by
shaping perceptions of how self-relevant and hence worthwhile a difficult task seems.
Second, prior research has shown how context can make difficulty mindsets momentarily
accessible (Aelenei et al., 2017; Oyserman et al., 2018; Smith & Oyserman, 2015) but not how
context can shift people’s difficulty mindset endorsements. We show that difficulty-asimpossibility endorsement is context-sensitive, shifting up when there is a challenge and
decreasing when success seems likely. Difficulty-as-importance endorsement was not sensitive
to contextual cues signaling odds of success and remained high across contexts. Regarding the
broader literature on goal pursuit, our results are relevant to the mindset theory of action phases
(Gollwitzer & Keller, 2016), which posits that people use a deliberative mindset to assess the
feasibility and desirability of a potential goal before initiating goal-focused behavior (Gollwitzer,
2012). However, people may return to a deliberative mindset after goal-focused behavior has
begun, and ultimately even abandon the goal, if during goal pursuit they become frustrated with
how difficult it is to achieve the goal (Brandstätter et al., 2013). Our results suggest that people
may begin to redeliberate because engaging with tasks or goals in a challenging context increases
their difficulty-as-impossibility beliefs, hence increasing their preference for easier means of
goal attainment and frustration with effortful means.
Third, our studies are the first to examine whether difficulty mindset endorsements
include meaningful between-person and within-person variability. Endorsement ICCs suggest

77
difficulty-as-importance varies more between persons than difficulty-as-impossibility, though
endorsements of both mindsets are typically about equal in state and trait variability. Regarding
the broader literature on self-regulation, our results add to a growing body of literature interested
in using multilevel modeling and ICCs to further our understanding of constructs' trait and state
components (Wolak et al., 2012). By employing these methods, our results can contribute novel
evidence of the situated nature of reasoning theorized by general models of socially situated
cognition (Smith & Semin, 2004).
Fourth, our studies are the first to examine daily fluctuations in difficulty mindset beliefs
and the subsequent self-regulatory consequences. Past research has examined the self-regulatory
profile of each mindset by documenting associations between mindset endorsement, selfperceptions, and self-reported behavior in the same moment (Fisher & Oyserman, 2017; Kiper et
al., 2024). We extend this work by presenting evidence of trait and state effects. We find that
people’s trait difficulty-as-impossibility beliefs carry over to self-regulatory means (goal selfefficacy, self-esteem, self-compassion) but not goal progress (meaningful engagement with
schoolwork). In contrast, trait difficulty-as-importance beliefs carry over to goal progress but not
self-regulatory means. These results align with prior findings that trait-difficulty as impossibility
is associated with a preference for less effortful means of goal attainment (Kiper et al., 2024)
while trait difficulty-as-importance is not associated with a preference for means (Kiper et al.,
2024) but is associated with better academic outcomes (Burbidge et al., 2024). Regarding the
broader literature on self-regulation and goal pursuit, our results are novel in considering how
difficulty mindset beliefs relate to aspects of self-regard that bear on self-regulation. Selfcompassion and self-esteem both uniquely and positively predict life outlook and better mental
health (Neff, 2011; Neff & Vonk, 2007). Congruent with prior research (Hernandez et al., 2021),

78
we found that trait difficulty-as-impossibility beliefs were associated with lower self-esteem. Our
results also suggest that while both daily difficulty-as-importance and difficulty-as-impossibility
beliefs are associated with daily self-regulatory means, difficulty-as-impossibility beliefs are
particularly detrimental to people’s self-regard.
Limitations and Future Directions
As is the case for any set of studies, our studies have limitations that might be addressed
in future research. We consider limitations regarding our sample, measures, and methods. First,
regarding our sample, our student samples were diverse in educational level, gender, and raceethnicity. The generalizability of our findings is limited given that all participants were students
and all participants lived in the United States. We present exploratory, cross-sectional results
suggesting difficulty-as-impossibility endorsement becomes more state-like and difficulty-asimportance endorsement becomes less trait-like as American students transition from middle
school to high school. We also do not know if difficulty mindset endorsements remain relatively
state-like in adulthood. One possibility is that since opportunities generally decrease and
constraints on motivational resources generally increase with age (motivational theory of lifespan development, Heckhausen et al., 2010), difficulty-as-impossibility endorsement may
become more trait-like in older age. Future research could evaluate the stability of our results by
utilizing longitudinal survey designs, non-student samples, and samples outside of the U.S.
Second, regarding our measures, our current studies focused on operationalizations of
tasks and goals in the domain of academics. We looked at which features of situations may shift
people’s difficulty-as-impossibility endorsements when they are faced with a challenging wordpuzzle task. In our daily diary studies, we focused on meaningful progress toward academic

79
goals. Future research should consider other operationalizations and ways of testing the stability
of our results across different domains.
Third, regarding our methods, we used various methodological approaches–a scenario
study, experiments, secondary analysis of survey data, and daily diaries–to address gaps in
research to date. Of course, our results are only a first step in filling these gaps. Prior reviews of
the literature on goal pursuit tell us that the antecedents of goal disengagement are not well
understood (Brandstätter & Bernecker, 2022; Wrosch & Scheier, 2020). With our experimental
results, we offer evidence of only one of likely many contextual features that may trigger
difficulty-as-impossibility endorsement and subsequent goal disengagement. Future research is
needed to identify other aspects of situations that may shift difficulty-as-impossibility or
difficulty-as-importance endorsement. Researchers should consider leveraging a daily diary
approach to capture descriptive features of days that are associated with difficulty mindset
endorsement fluctuations and use experimental designs to test the causality of these
relationships.
Conclusion
Our results concretize a key inference of identity-based motivation theory, that
endorsements of difficulty mindsets are shaped by both features of the context and features of
individuals. We document that difficulty-as-important and difficulty-as-impossibility
endorsements vary both between people and between situations but are not sensitive to the same
situational cues. Each mindset is also associated with a unique self-regulatory profile. Endorsing
a difficulty-as-importance mindset is associated with a focus on goal progress marked by an
indifference toward the means to do so, while endorsing a difficulty-as-impossibility mindset is
associated with a preference for easier means of goal attainment. While prior research has largely

80
focused on the effects of trait-like endorsement on self-regulation, we present evidence that daily
fluctuations in endorsement not only matter for self-regard but also momentary increases in
difficulty-as-impossibility endorsement may be particularly demotivating and may offset the
benefits of momentary increases in difficult-as-importance endorsement.

81
Discussion
In this dissertation, I showed that difficulty-as-importance and difficulty-as-impossibility
beliefs are shaped by both features of context and features of individuals. I replicated prior
research, finding that trait difficulty mindset beliefs uniquely shape self-regulatory focus, and
then went beyond this prior work by examining state fluctuations in difficulty mindset beliefs.
Taken together, the results presented here document that difficulty-as-importance beliefs support
goal progress by focusing attention on ends while difficulty-as-impossibility beliefs may
undermine goal progress by focusing attention on the required motivational resources. Next, I
consider the theoretical contributions of these findings and how they may inform practical
applications.
Theoretical Contributions
The results presented in this dissertation highlight the distinct self-regulatory profiles of
difficulty-as-importance and difficulty-as-impossibility beliefs. The implication is that
interventions may be most effective when they not only foster difficulty-as-importance beliefs
but also inoculate against believing difficulty is a signal of impossibility. Endorsing a difficultyas-importance interpretation may support goal-directed behavior and functional self-regard
(possible identity certainty, goal self-efficacy, self-esteem, and self-compassion), but it will not
protect against temptations to disengage if challenges along the way trigger upward shifts in
difficulty-as-impossibility beliefs.
These results may also be of interest to educators who practice process-oriented teaching
(Bolhuis, 2003). Process-oriented teaching intends to help students shift their goal focus from inclass performance to lifelong learning by building up their self-regulation competencies.
Educators with such intentions could also consider implementing existing identity-based

82
motivation interventions such as the Pathways-to-Success intervention (Oyserman et al., 2021).
Pathways-to-Success is designed to inoculate students against a difficulty-as-impossibility
interpretation, helping them see that difficulties in the classroom are normal and overcomeable.
Researchers interested in designing behavioral interventions should consider how these
unique self-regulatory profiles might inform the types of behaviors they aim to promote. For
instance, researchers interested in closing the gap between pro-environmental attitudes and proenvironmental behaviors (for a review, see Kollmuss & Agyeman, 2022) can consider how
difficulty-as-importance and difficulty-as-impossibility beliefs may contribute to this gap. For
example, people who believe pro-environmental action is important and worthwhile may still fail
to take even low-effort actions such as recycling if they believe those actions do not
meaningfully contribute to the end goal of climate mitigation. Moreover, barriers to proenvironmental action may trigger increases in difficulty-as-impossibility endorsement and
subsequently lower feelings of efficacy even among people who generally consider proenvironmental action to be important, potentially undermining the positive benefits of believing
difficulty is a signal of importance.
Limitations and Future Directions
Like any research, the current studies presented here have limitations that future research
can address. I consider limitations regarding the sample, measures, and methodological
approaches. First, regarding the sample, this dissertation examined data collected from Chinese
and American students. Students’ educational levels ranged from primary school to college.
These samples were appropriate given our focus on the domain of academics. Some correlational
research conducted among adults living in Australia, Canada, China, India, Iran, the U.K., the
U.S., and Turkey (Yan et al., 2023) suggests the results presented here are generalizable.

83
Additional research using non-student samples employing experimental and lagged research
designs is needed to know if the context-sensitivity and self-regulatory profiles of difficulty-asimportance and difficulty-as-impossibility beliefs documented here generalize to other
populations.
Second, regarding the measures, this dissertation evaluated possible identity certainty and
various task-oriented actions and attitudes, but did not measure people’s idiographic possible
identities and corresponding behaviors. By utilizing self-generated, open-ended measures of
students’ possible identities and their linked strategies, researchers can assess the extent to which
possible identities and strategies are apt and actionable (Oyserman et al., 2004; O’Donnell &
Oyserman, 2023). Although using open-ended data is typically inefficient and unmanageable at a
large scale, researchers should employ such measures to improve the ecological validity of their
work. Hence, my future work intends to minimize barriers to using open-ended measures by
comparing three methods of coding possible identity and strategy data: traditional, human-coded
scores of plausibility (Oyserman et al., 2004), machine learning algorithms similar to those
implemented by past researchers (Horowitz et al., 2020; O’Donnell & Oyserman, 2023), and
student self-coding in which students generate open-ended possible identities and strategies and
then self-label features of their responses (e.g., label possible identity as school-focused or not,
label strategy as recently used or not).
Third, regarding methodological approaches, the studies presented in this dissertation
each had different methodological strengths but shared the same limitation. The current studies
did not employ methods that could accurately capture how difficulty beliefs shape real-world
task and goal engagement. Though students in Study 1 of Chapter 2 were led to believe that they
would have to complete the puzzle set they were asked to create, this type of scenario is unlikely

84
to occur in the real world. Similarly, the daily diary studies captured ecologically-valid
fluctuations in beliefs but could not capture ecologically-valid behavior. Given that self-reported
behaviors may be biased overestimates of actual behavior (for examples, see Brenner &
DeLamater, 2016; Koller et al., 2023), future research would benefit from evaluating behaviors
outside of the lab. Thanks to the ubiquity of modern technology like smartphones (Harari et al.,
2016), objective behavioral data collection is much less time-consuming and costly. Sensorbased behavioral estimates automatically collected by participant’s own smartphones can be used
to assess a variety of information, including but not limited to location, noise level, locomotion,
and phone usage. Researchers have used such data to accurately track students’ studying
behaviors (Wang et al., 2015). Future research could draw on sensor data to evaluate how
difficulty mindset beliefs predict real-world, goal-directed behavior.
Conclusion
In this dissertation, I documented how difficulty-as-importance and difficulty-asimpossibility beliefs vary meaningfully between persons and between situations. Difficulty
beliefs differ in their context-sensitivity and motivational consequences. This is true of beliefs
when they are measured as traits and as momentary fluctuations, with each contributing in
separate ways to self-regulation. In summary, beliefs in difficulty-as-importance and difficultyas-impossibility function as distinct, situated inferences that matter in different ways for action
and self-regard during self-regulation.

85
References
Aelenei, C., Lewis Jr, N., & Oyserman, D. (2017). No pain no gain? Social demographic
correlates and identity consequences of interpreting experienced difficulty as
importance. Contemporary Educational Psychology, 48, 43-
55. https://doi.org/10.1016/j.cedpsych.2016.08.004
Bandura, A. (1997). Self-efficacy: The exercise of control. W. H. Freeman.
Bandura, A. (2006). Guide for constructing self-efficacy scales. In T. Urdan & F. Pajares
(Eds.) Self-efficacy beliefs of adolescents (pp. 307-337). Information Age Publishing.
Bi, C., & Oyserman, D. (2015). Left behind or moving forward? Effects of possible selves and
strategies to attain them among rural Chinese children. Journal of Adolescence, 44, 245-
258. https://doi.org/10.1016/j.adolescence.2015.08.004
Biering, K., Hjollund, N., & Frydenberg, M. (2015). Using multiple imputation to deal with
missing data and attrition in longitudinal studies with repeated measures of patientreported outcomes. Clinical Epidemiology, 7. https://doi.org/10.2147/CLEP.S72247
Bolhuis, S. (2003). Towards process-oriented teaching for self-directed lifelong learning: A
multidimensional perspective. Learning and Instruction, 13, 327-347.
https://doi.org/10.1016/S0959-4752(02)00008-7
Brandstätter, V., Herrmann, M., & Schüler, J. (2013). The struggle of giving up personal goals:
Affective, physiological, and cognitive consequences of an action crisis. Personality and
Social Psychology Bulletin, 39, 1668–1682. https://doi.org/10.1177/0146167213500151
Brandstätter, V., & Bernecker, K. (2022). Persistence and disengagement in personal goal
pursuit. Annual Review of Psychology, 73. https://doi.org/10.1146/annurev-psych020821-110710

86
Brehm, J. W., & Self, E. A. (1989). The intensity of motivation. Annual Review of Psychology,
40(1), 109-131. https://doi.org/10.1146/annurev.ps.40.020189.000545
Brenner, P. S., & DeLamater, J. (2016). Lies, damned lies, and survey self-reports? Identity as a
cause of measurement bias. Social Psychology Quarterly, 79(4), 333-354.
https://doi.org/10.1177/0190272516628298
Burbidge, A. & Oyserman, D. (2024, February 9). What does task-based difficulty imply today?
Using a daily diary approach to understand correlates and consequences of difficulty-asimpossibility and difficulty-as-importance mindsets [Poster presentation]. Society for
Personality and Social Psychologists Annual Convention, San Diego, California.
Burns, R., Crisp, D., & Burns, R. (2020). Re-examining the reciprocal effects model of selfconcept, self-efficacy, and academic achievement in a comparison of the cross-lagged
panel and random-intercept cross-lagged panel frameworks. British Journal of
Educational Psychology, 90, 77-91. https:// doi.org/10.1111/bjep.12265
Bush, T., Coleman, S., & Xiaohong, S. (1998). Managing secondary schools in China. Compare,
28, 183-195. https://doi/org/10.1080/0305792980280206
Calef, R.S., Choban, M.C., Calef, R.A., Brand, R.L., Rogers, M.J. & Geller, E.S. (1992). Effects
of unsolvable anagrams on retention. Bulletin of the Psychonomic Society, 30, 164-166.
https://doi.org/10.3758/BF03330428
Carpenter, S., Pan, S., & Butler, A. (2022). The science of effective learning with spacing and
retrieval practice. Nature Reviews Psychology, 1, 496-511.
https://doi.org/10.1038/s44159-022-00089-1

87
Carver, C. S., Reynolds, S. L., & Scheier, M. F. (1994). The possible selves of optimists and
pessimists. Journal of Research in Personality, 28, 133-141.
https://doi.org/10.1006/jrpe.1994.1011
Carver, C. S., & Scheier, M. (2012). Attention and self-regulation: A control-theory approach to
human behavior. Springer Science & Business Media.
Carver, C. S., Scheier, M. F., & Segerstrom, S. C. (2010). Optimism. Clinical Psychology
Review, 30, 879–889. https://doi.org/10.1016/j.cpr.2010.01.006
Chen, C., Lee, S., & Stevenson, H. (1996). Academic achievement and motivation of Chinese
students: A cross-national perspective. In S. Lau, (Ed.), Growing up the Chinese way:
Chinese child and adolescent development (pp. 69-92). The Chinese University Press.
Chen, J., Huebner, E. S., & Tian, L. (2020). Longitudinal relations between hope and academic
achievement in elementary school students: Behavioral engagement as a mediator.
Learning and Individual Differences, 78. https://doi.org/10.1016/j.lindif.2020.101824
Choi, K., & Oyserman, D. (2024). If it feels hard, can I find the time to do anything about it?
Interpretations of difficulty shape beliefs about time [Poster presentation]. Society for
Personality and Social Psychologists Annual Convention, San Diego, California.
Destin, M., Castillo, C., & Meissner, L. (2018). A field experiment demonstrates near peer
mentorship as an effective support for student persistence. Basic and Applied Social
Psychology, 40, 269-278. https://doi.org/10.1080/01973533.2018.1485101
Di Paula, A., & Campbell, J. D. (2002). Self-esteem and persistence in the face of failure.
Journal of Personality and Social Psychology, 83, 711–724.
https://doi.org/10.1037/0022-3514.83.3.711

88
Doody, R. (2019). The sunk cost "fallacy" is not a fallacy. Ergo: An Open Access Journal of
Philosophy, 6, 1153-1190. https://doi.org/10.3998/ergo.12405314.0006.040
Dweck, C. S. (1986). Motivational processes affecting learning. American Psychologist, 41,
1040–1048. https://doi.org/10.1037/0003-066X.41.10.1040
Eccles, J. & Wigfield, A. (2020). From expectancy-value theory to situated expectancy-value
theory: A developmental, social cognitive, and sociocultural perspective on motivation.
Contemporary Educational Psychology, 61.
https://doi.org/10.1016/j.cedpsych.2020.101859
Elmore, K., Oyserman, D., Smith, G., & Novin, S. (2016). When the going gets tough:
Implications of reactance for interpretations of experienced difficulty in the classroom.
AERA Open, 2, 2332858416664714. https://doi.org/10.1177/23328584166647
EnchantedLearning.com (2019, August). Five-letter Anagrams.
https://www.enchantedlearning.com/english/anagram/numberofletters/5letters.shtml
Enders, C. (2017). Multiple imputation as a flexible tool for missing data handling in clinical
research. Behaviour Research and Therapy, 98.
https://doi.org/10.1016/j.brat.2016.11.008
Faul, F., Erdfelder, E., Lang, A., & Buchner, A. (2007). G*Power 3: A flexible statistical power
analyses program for the social, behavioral, and biomedical sciences. Behavior Research
Methods, 39, 175-191. https://doi.org/10.3758/BF03193146
Fisher, O., & Oyserman, D. (2017). Assessing interpretations of experienced ease and difficulty
as motivational constructs. Motivational Science, 3, 133-163.
http://dx.doi.org/10.1037/mot0000055

89
Gilhooly, K. J., & Johnson, C. E. (1978). Effects of solution word attribute on anagram
difficulty: A regression analysis. Quarterly Journal of Experimental Psychology, 30, 57-
70. https://doi.org/10.1080/14640747808400654
Goetz, T., Bieleke, M., Yanagida, T., Krannich, M., Roos, A.-L., Frenzel, A., Lipnevich, A., &
Pekrun, R. (2023). Test boredom: Exploring a neglected emotion. Journal of Educational
Psychology, 115(7), 911-931. https://doi.org/10.1037/edu0000807
Gollwitzer, P. M. (1999). Implementation intentions: strong effects of simple plans. American
Psychologist, 54, 493–503. https://doi.org/10.1037/0003-066X.54.7.493
Gollwitzer, P. M. (2012). Mindset theory of action phases. In P. A. van Lange (Ed.), Handbook
of theories of social psychology (pp. 526-545). Sage.
Gollwitzer, P. M., & Keller, L. (2016). Mindset theory. In V. Zeigler-Hill & T. K. Shackelford
(Eds.), Encyclopedia of personality and individual differences (pp. 1-8). Springer
International Publishing AG.
Hamaker, E., Kuiper, R., & Grasman, R. (2015). A critique of the cross-lagged panel model.
Psychological Methods, 20, 102. https://doi.org/10.1037/a0038889
Harari, G. M., Lane, N. D., Wang, R., Crosier, B. S., Campbell, A. T., & Gosling, S. D. (2016).
Using smartphones to collect behavioral data in psychological science: Opportunities,
practical considerations, and challenges. Perspectives on Psychological Science, 11(6),
838-854. https://doi.org/10.1177/1745691616650285
Heckhausen, J., Wrosch, C., & Schulz, R. (2010). A motivational theory of life-span
development. Psychological Review, 117, 32–60. https://doi.org/10.1037/a0017668
Hernandez, I., Silverman, D., & Destin, M. (2021). From deficit to benefit: Highlighting lowerSES students’ background-specific strengths reinforces their academic persistence.

90
Journal of Experimental Social Psychology, 92.
https://doi.org/10.1016/j.jesp.2020.104080
Higgins, E. T. (1996). Activation: Accessibility, and salience. In T. Higgins & A. Kruglanski
(Eds.), Social psychology: Handbook of basic principles (pp. 133-168). Guilford Press.
Higgins, E. T., & Bargh, J. A. (1987). Social cognition and social perception. Annual Review of
Psychology, 38, 369-425. https://doi.org/10.1146/annurev.ps.38.020187.002101
Hoffman, L. (2015). Longitudinal analysis. New York: Routledge.
Horowitz, E., Oyserman, D., Dehghani, M., & Sorensen, N. (2020). Do you need a roadmap or
can someone give you directions: When school-focused possible identities change so do
academic trajectories. Journal of Adolescence, 79, 26-38.
https://doi.org/10.1016/j.adolescence.2019.12.013
Horowitz, E., Sorensen, N., Yoder, N., & Oyserman, D. (2018). Teachers can do it: Scalable
identity-based motivation intervention in the classroom. Contemporary Educational
Psychology, 54, 12-28. https://doi.org/10.1016/j.cedpsych.2018.04.004
Horstmanshof, L., & Zimitat, C. (2007). Future time orientation predicts academic engagement
among first-year university students. British Journal of Educational Psychology, 77, 703-
718. https://doi.org/10.1348/000709906X160778
Huang, C. (2011). Self-concept and academic achievement: A meta-analysis of longitudinal
relations. Journal of School Psychology, 49. https://doi.org/10.1016/j.jsp.2011.07.001
Hulleman, C. S., Schrager, S. M., Bodmann, S. M., & Harackiewicz, J. M. (2010). A metaanalytic review of achievement goal measures: Different labels for the same constructs or
different constructs with similar labels?. Psychological Bulletin, 136, 422–449.
https://doi.org/10.1037/a0018947

91
Jakobsen, J. C., Gluud, C., Wetterslev, J., & Winkel, P. (2017). When and how should multiple
imputation be used for handling missing data in randomised clinical trials–a practical
guide with flowcharts. BMC Medical Research Methodology, 17.
https://doi.org/10.1186/s12874-017-0442-1
Jansen, T., Meyer, J., Wigfield, A., & Möller, J. (2022). Which student and instructional
variables are most strongly related to academic motivation in K-12 education? A
systematic review of meta-analyses. Psychological Bulletin, 148, 1-26.
https://doi.org/10.1037/bul0000354
Kemmelmeier, M. & Oyserman, D. (2001). The ups and downs of thinking about a successful
other: self-construals and the consequences of social comparisons. European Journal of
Social Psychology, 31, 311-310. https://doi.org/10.1002/ejsp.47
Kenny, D. A., & Harackiewicz, J. M. (1979). Cross-lagged panel correlation: Practice and
promise. Journal of Applied Psychology, 64, 372–379. https://doi.org/10.1037/0021-
9010.64.4.372
Kernis, M. H. (2005). Measuring self‐esteem in context: The importance of stability of self‐
esteem in psychological functioning. Journal of Personality, 73, 1569-1605.
https://doi.org/10.1111/j.1467-6494.2005.00359.x
Kiper, G., Atari, M., Yan, V., & Oyserman, D. (2022). The upside: How people make sense of
difficulty matters during a crisis. Self and Identity, 22, 19-41.
https://doi.org/10.1080/15298868.2022.2033309
Kiper, G., Oyserman, D., & Yan, V. (2024). I’ll take the high road: Paths to goal pursuit and
identity-based interpretations of difficulty. Self & Identity.
https://doi.org/10.1080/15298868.2024.2314920

92
Koller, K., Pankowska, P. K., & Brick, C. (2023). Identifying bias in self-reported proenvironmental behavior. Current Research in Ecological and Social Psychology, 4,
100087. https://doi.org/10.1016/j.cresp.2022.100087
Kollmuss, A., & Agyeman, J. (2002). Mind the gap: why do people act environmentally and
what are the barriers to pro-environmental behavior?. Environmental Education
Research, 8(3), 239-260. https://doi.org/10.1080/13504620220145401
Kreft, I. G., & De Leeuw, J. (1998). Introducing multilevel modeling. London: Sage.
Lai, F., Sadoulet, E., & de Janvry, A. (2011). The contributions of school quality and teacher
qualifications to student performance: Evidence from a natural experiment in Beijing
middle schools. The Journal of Human Resources, 46, 123-153.
https://doi.org/10.3368/jhr.46.1.123
Leung, C. H. Y., Chen, S. X., & Lam, B. C. P. (2010). Where there’s a will, there’s a way: The
mediating effect of academic aspiration between beliefs and academic outcomes. Journal
of Psychology in Chinese Students, 11, 53-72.
Lewis Jr, N., & Earl, A. (2018). Seeing more and eating less: Effects of portion size granularity
on the perception and regulation of food consumption. Journal of Personality and Social
Psychology, 114, 786-803. https://doi.org/10.1037/pspp0000183
Liang, W., Wang, J., Chen, M., Shibao Feng, Lee, M., Schwartz, M., Pasick, R., & Mandelblatt,
J. (2008). Developing and validating a measure of Chinese cultural views of health and
cancer. Health education & behavior: The official publication of the Society for Public
Health Education, 35, 361-375. https://doi.org/10.1177/1090198106294893
Liu, J. (2015). Understanding inequality in public school admission in urban China: Analysis of
public discourses on Ze Xiao. Asian Education and Development Studies, 4, 434-447.

93
Liu, Y., & Dunne, M. (2009). Educational reform in China: tensions in national policy and local
practice. Comparative Education, 45, 461–476. http://www.jstor.org/stable/40593190
Loyalka, M. (2012). Eating bitterness: stories from the front lines of China’s great urban
migration. Berkeley: University of California Press.
Maas, C., & Hox, J. (2005). Sufficient sample sizes for multilevel modeling. Methodology, 1, 86-
92. https://doi.org/10.1027/1614-2241.1.3.86
Mainzer, R., Apajee, J., Nguyen, C., Carlin, J., & Lee, K. (2021). A comparison of multiple
imputation strategies for handling missing data in multi‐item scales: Guidance for
longitudinal studies. Statistics in Medicine, 40. https://doi.org/10.1002/sim.9088
Maneesriwongul, W., & Dixon, J. K. (2004). Instrument translation process: A methods review.
Journal of Advanced Nursing, 48, 175-186. https://doi.org/10.1111/j.1365-
2648.2004.03185.x
Masinga N. C., & Dumont, K. B. (2018) The motivational implications of adolescents' schooloriented possible identities in a social change context. Journal of Applied Social
Psychology, 48. https://doi.org/10.1111/jasp.12512
Meade, A., & Craig, S. (2012). Identifying careless responses in survey data. Psychological
Methods, 17, 437–455. http://dx.doi.org/10.1037/a0028085
Meevissen, Y., Peters, M., & Alberts, H. (2011). Become more optimistic by imagining a best
possible self: effects of a two week intervention. Journal of Behavior Therapy and
Experimental Psychiatry, 42, 371-378. https://doi.org/10.1016/j.jbtep.2011.02.012
Mello, Z., & Worrell, F. (2006). The relationship of time perspective to age, gender, and
academic achievement among academically talented adolescents. Journal for the
Education of the Gifted, 29, 271-289.

94
Mielicki, M., Schiller, L., Fitzsimmons, C., Scheibe, D. & Thompson, C. (2022). Perceptions of
ease and difficulty, but not growth mindset, relate to specific math attitudes. British
Journal of Educational Psychology, 92, 707-729. https://doi.org/10.1111/bjep.12472
Neff, K. (2003). Development and validation of a scale to measure self-compassion. Self and
Identity, 2, 223-250. https://doi.org/10.1080/15298860309027
Neff, K. (2011). Self-compassion, self-esteem, and well-being. Social and Personality
Psychology Compass, 5, 1-12. https://doi.org/10.1111/j.1751-9004.2010.00330.x
Neff, K. D., Hsieh, Y. P., & Dejitterat, K. (2005). Self-compassion, achievement goals, and
coping with academic failure. Self and Identity, 4, 263-287.
https://doi.org/10.1080/13576500444000317
Neff, K., & Vonk, R. (2009). Self‐compassion versus global self‐esteem: Two different ways of
relating to oneself. Journal of Personality, 77, 23-50. https://doi.org/10.1111/j.1467-
6494.2008.00537.x
Newman, D., Sachs, M., Stone, A., Schwarz, N. (2020). Nostalgia and well-being in daily life:
An ecological validity perspective. Journal of Personality and Social Psychology, 118,
325-347. https://dx.doi.org/10.1037/pspp0000236
Nezlek, J. (2005). Distinguishing affective and non-affective reactions to daily events. Journal of
Personality, 73, 1539–1568. https://doi.org/10.1111/j.1467-6494.2005.00358.x
Nezlek, J. (2012). Diary methods for social and personality psychology. Sage.
http://doi.org/10.4135/9781446287903
Nezlek, J. (2017). A practical guide to understanding reliability in studies of within-person
variability. Journal of Research in Personality, 69, 149–155.
http://dx.doi.org/10.1016/j.jrp.2016.06.020

95
Nezlek, J. B., Feist, G. J., Wilson, F. C., & Plesko, R. M. (2001). Day-to-day variability in
empathy as a function of daily events and mood. Journal of Research in Personality, 35,
401–423. https://doi.org/10.1006/jrpe.2001.2332
Núñez-Regueiro, F., Juhel, J., Bressoux, P., & Nurra, C. (2022). Identifying reciprocities in
school motivation research: A review of issues and solutions associated with cross-lagged
effects models. Journal of Educational Psychology, 114, 945-965.
https://doi.org/10.1037/edu0000700
O’Donnell, S. & Oyserman, D. (2023). Apt and actionable possible identities matter: The case of
academic identities. Journal of Adolescence, 95, 354-371.
https://doi.org/10.1002/jad.12118
O’Donnell, S., Yan, V., Bi, C., & Oyserman, D. (2023). Is difficulty mostly about impossibility?:
What difficulty implies may be culturally variant. Personality and Social Psychology
Bulletin, 49, 309-328. https://doi.org/10.1177/0146167221106559
OECD. (2012). Education at a Glance 2012: OECD Indicators. OECD Publishing, Paris.
https://doi.org/10.1787/eag-2012-en
Oishi, S., Diener, E., Choi, D., Chu, K., & Choi, I. (2007). The dynamics of daily events and
well-being across cultures: When less is more. Journal of Personality and Social
Psychology, 93, 685-698. http://doi.org/10.1037/0022-3514.93.4.685
Oyserman, D. (2007). Social identity and self-regulation. In A. Kruglanski & T. Higgins
(Eds.), Social psychology: Handbook of basic principles (pp. 432-453). Guilford Press.
Oyserman, D. (2009). Identity-based motivation: Implications for action-readiness, proceduralreadiness, and consumer behavior. Journal of Consumer Psychology, 19, 250-260.
https://doi.org/10.1016/j.jcps.2009.05.008

96
Oyserman, D. (2013). Not just any path: Implications of identity-based motivation for disparities
in school outcomes. Economics of Education Review, 33, 179-190.
http://dx.doi.org/10.1016/j.econedurev.2012.09.002
Oyserman, D. (2015). Identity‐based motivation. Emerging trends in the social and behavioral
sciences: An interdisciplinary, searchable, and linkable resource, 1-11.
https://doi.org/10.1002/9781118900772.etrds0171
Oyserman, D. (2017). Culture three ways: Culture and subcultures within countries. Annual
Review of Psychology, 68, 435-463. https://doi.org/10.1146/annurev-psych-122414-
033617
Oyserman, D., Bybee, D., & Terry, K. (2006). Possible selves and academic outcomes: How and
when possible selves impel action. Journal of Personality and Social Psychology, 91,
188–204. https://doi.org/10.1037/0022-3514.91.1.188
Oyserman, D., Bybee, D., Terry, K., & Hart-Johnson, T. (2004). Possible selves as roadmaps.
Journal of Research in Personality, 38(2), 130-149. https://doi.org/10.1016/S0092-
6566(03)00057-6
Oyserman, D., & Destin, M. (2010). Identity-based motivation: Implications for intervention.
The Counseling Psychologist, 38(7), 1001-1043.
https://doi.org/10.1177/0011000010374775
Oyserman, D., Destin, M., & Novin, S. (2017). The context-sensitive future self: Possible selves
motivate in context, not otherwise. Self and Identity, 14(2), 173–188.
https://doi.org/10.1080/15298868.2014.965733
Oyserman, D., Elmore, K., Novin, S., Fisher, O., & Smith, G. (2018). Guiding people to interpret
their experienced difficulty as importance highlights their academic possibilities and

97
improves their academic performance. Frontiers in Psychology, 9.
https://doi.org/10.3389/fpsyg.2018.00781
Oyserman, D., Elmore, K., & Smith, G. (2012). Self, self-concept, and identity. In M. R. Leary
& J. P. Tangney (Eds.), Handbook of self and identity (pp. 69–104). The Guilford Press.
Oyserman, D., & Horowitz, E. (2023). From possible selves and future selves to current action:
An integrated review and identity-based motivation synthesis. Advances in Motivation
Science, 10, 73-147. https://doi.org/10.1016/bs.adms.2022.11.003
Oyserman, D., Lewis, N. Jr., Yan, V. X., Fisher, O., O’Donnell, S., C., & Horowitz, E. (2017).
An identity-based motivation framework for self-regulation. Psychological Inquiry, 28,
139-147. https://doi.org/10.1080/1047840X.2017.1337406
Oyserman, D., O'Donnell, S., Sorensen, N., & Wingert, K. (2021). Process matters: Teachers
benefit their classrooms and students when they deliver an identity-based motivation
intervention with fidelity. Contemporary Educational Psychology, 66.
https://doi.org/10.1016/j.cedpsych.2021.101993
Oyserman, D., Terry, K., & Bybee, D. (2002). A possible selves intervention to enhance school
involvement. Journal of Adolescence, 25, 313-316.
https://doi.org/10.1006/jado.2002.0474
Peters, M., Flink, I., Boersma, K., & Linton, S. J. (2010). Manipulating optimism: Can imagining
a best possible self be used to increase positive future expectancies?. The Journal of
Positive Psychology, 5, 204–211. https://doi.org/10.1080/17439761003790963
Purol, M., & Chopik, W. (2021). Optimism: Enduring resource or miscalibrated perception?.
Social and Personality Psychology Compass, 15, e12593.
https://doi.org/10.1111/spc3.12593

98
R Core Team (2023). R: A language and environment for statistical computing. R Foundation for
Statistical Computing, Vienna, Austria. https://www.R-project.org.
Ratner, B. (2009). The correlation coefficient: Its values range between +1/− 1, or do they?.
Journal of Targeting, Measurement and Analysis for Marketing, 17(2), 139-142.
https://doi.org/10.1057/jt.2009.5
Richardson, M., Abraham, C., & Bond, R. (2012). Psychological correlates of university
students' academic performance: A systematic review and meta-analysis. Psychological
Bulletin, 138, 353–387. https://doi.org/10.1037/a0026838
Richter, M., Gendolla, G. H. E., & Wright, R. A. (2016). Three decades of research on
motivational intensity theory: What we have learned about effort and what we still don’t
know. In A. J. Elliot (Ed.), Advances in Motivation Science, 149-186.
https://doi.org/10.1016/bs.adms.2016.02.001
Rights, J. D., & Sterba, S. K. (2019). Quantifying explained variance in multilevel models: An
integrative framework for defining R-squared measures. Psychological Methods, 24(3),
309-338. https://doi.org/10.1037/met0000184
Rosenberg, M. (1965). Rosenberg self-esteem scale. Journal of Religion and Health.
https://doi.org/10.1037/t01038-000
Scheier, M. F., & Carver, C. S. (1992). Effects of optimism on psychological and physical wellbeing: Theoretical overview and empirical update. Cognitive Therapy and Research, 16,
201-228. https://doi.org/10.1007/BF01173489
Smith, E. R., & Semin, G. R. (2004). Socially situated cognition: Cognition in its social context.
Advances in Experimental Social Psychology, 36, 57-121.

99
Smith, G., & Oyserman, D. (2015). Just not worth my time? Experienced difficulty and time
investment. Social Cognition, 33, 85-103. https://doi.org/10.1521/soco.2015.33.2.1
Valentine, J., DuBois, D., & Cooper, H. (2004). The relation between self-beliefs and academic
achievement: A meta-analytic review. Educational Psychologist, 39, 111-113.
https://doi.org/10.1207/s15326985ep3902_3
van Buuren, S., & Groothuis-Oudshoorn, K. (2011). “mice: Multivariate Imputation by Chained
Equations in R.” Journal of Statistical Software, 45. https://doi.org/10.18637/jss.v045.i03
Viola, M., Kruglanski, A. W., Ellenberg, M., & Pierro, A. (2023). Ambitious people are more
prone to support resource-intensive aid programs. Motivation and Emotion, 47, 1027-
1039. https://doi.org/10.1007/s11031-023-10044-z
Wang, J. L., Wang, H. Z., Gaskin, J., & Wang, L. H. (2015). The role of stress and motivation in
problematic smartphone use among college students. Computers in Human Behavior, 53,
181-188. https://doi.org/10.1016/j.chb.2015.07.005
Wayman, J. C. (2003). Multiple imputation for missing data: What is it and how can I use it. In
Annual Meeting of the American Educational Research Association, 2,16. Chicago, IL
Wigfield, A., & Eccles, J. S. (2000). Expectancy–value theory of achievement motivation.
Contemporary Educational Psychology, 25, 68-81.
https://doi.org/10.1006/ceps.1999.1015
Wolak, M. E., Fairbairn, D. J., & Paulsen, Y. R. (2012). Guidelines for estimating repeatability.
Methods in Ecology and Evolution, 3, 129-137. https://doi.org/10.1111/j.2041-
210X.2011.00125.x

100
Wrosch, C., & Scheier, M. F. (2020). Adaptive self-regulation, subjective well-being, and
physical health: The importance of goal adjustment capacities. Advances in Motivation
Science, 7, 199-238. https://doi.org/10.1016/bs.adms.2019.07.001
Yan, V. X., & Oyserman, D. (2018). The world as we see it: The culture—identity—
metacognition interface. In J. Proust & M. Fortier (Eds.), Metacognitive diversity: An
interdisciplinary approach (pp. 225-244). Oxford University Press.
Yan, V. X., Oyserman, D., Kiper, G., & Atari, M. (2023). Difficulty-as-improvement: the
courage to keep going in the face of life’s difficulties. Personality and Social Psychology
Bulletin. https://doi.org/10.1177/01461672231153680
Ye, H. (2015) Key-point schools and entry into tertiary education in China, Chinese Sociological
Review, 47, 128-153. https://doi.org/10.1080/21620555.2014.990321
Zhang, J., Zhao, Y., Yang, D., & Yu, H. (2000). Study of Chinese attribution of success and
failure. Journal of Southwest China Normal University (Natural Science), 25 (in
Chinese).
Zuo, B., Zhang, X., Wen, F., & Zhao, Y. (2020). The influence of stressful life events on
depression among Chinese university students: Multiple mediating roles of fatalism and
core self-evaluations. Journal of Affective Disorders, 260, 84-90.
https://doi.org/10.1016/j.jad.2019.08.083

101
Appendices
Appendix A: Chapter 1 Additional Tables
Table A1
Age, Gender (% Female), and Proportion of Sample in Each School at Each Time Point
Time 1 Time 2 Time 3
Mean Age (SD) at T1 12.28 (0.61) 12.26 (0.61) 12.27 (0.60)
% Female at T1 44.10 44.47 44.44
% of Total Sample by School
School A 42.75 43.43 38.54
School B 23.45 21.51 22.92
School C 33.81 35.06 38.54
Note. SD=standard deviation. We asked for age and gender at Time 1 only. The mean and SD of
age and % female vary across time points as a function of missing data across time points.
Table A2
Scale Means (M), Standard Deviations (SD), and α Reliability at Time 1, Time 2, and Time 3
Scale Time 1 Time 2 Time 3
M SD α M SD α M SD α
Difficulty-as-Importance 4.27 0.65 .86 4.23 0.69 .88 4.11 0.74 .91
Difficulty-as-Impossibility 1.75 0.72 .85 1.78 0.76 .88 1.86 0.73 .90
Accepting Fate 1.82 0.73 .86 1.93 0.85 .92 1.95 0.83 .93
Optimism for the Future 3.56 0.84 .79 3.50 0.92 .82 3.59 0.90 .87
School-Focused Possible Identity
Certainty
6.72 1.59 .86 6.80 1.72 .89 6.98 1.61 .91
Note. All scales are 1=strongly disagree, 5=strongly agree, except for school-focused possible
identity certainty which is 0=not possible at all, 9=very possible.

102
Table A3
Snapshot of Sample Descriptive Results: Means and Standard Deviations (SD)
Variable Mean (%, Hours, or Score) SD
% Chose Math as Favorite Subject 35.69% 47.95
Daily Activities (in hours)
Doing Homework 1.84 0.74
Receiving Tutoring 1.31 1.20
Reading 1.17 0.83
Playing Sports 1.08 0.74
In PE 0.94 0.54
Using the Internet 0.77 0.78
Watching Television 0.76 0.76
Playing Video Games 0.63 0.78
Playing Music or Dancing 0.58 0.71
Belief and Attribution Items (responses from 0 to 9 except as noted)
6th-Grade Effort (1 to 10) 6.51 1.71
Likely to Get into Ideal School 5.16 2.29
Hope to Get into Ideal School 7.76 1.98
Worry if Get into Ideal School 5.75 2.78
Ability Impacts Exam Score 5.60 2.34
Effort Impacts Exam Score 6.59 2.15
Luck Impacts Exam Score 3.76 2.52
Fate Impacts Exam Score 3.16 2.69
Table A4
Bivariate Correlations of the Self and Motivation Scales at Each Time Point
Scale Time 1 Time 2 Time 3
2 3 4 5 2 3 4 5 2 3 4 5
1 -.45 -.28 .26 .37 -.50 -.30 .26 .30 -.49 -.30 .29 .30
2 -- .47 -.43 -.34 -- .56 -.40 -.30 -- .55 -.45 -.26
3 -- -.48 -.27 -- -.49 -.19 -- -.42 -.21
4 -- .31 -- .27 -- .19
5 -- -- --
Note. 1=Difficulty-as-Importance, 2=Difficulty-as-Impossibility, 3=Accepting Fate, 4=Optimism
for the Future, 5=School-Focused Possible Identity Certainty. All correlations p < .001.

103
Table A5
Intra-Measure Correlations Between T1 and T2 and Between T2 and T3
Scale Correlations Between
T1 and T2 T2 and T3
Difficulty-as-Importance .59 .58
Difficulty-as-Impossibility .64 .65
Accepting Fate .63 .71
Optimism for the Future .65 .64
Certainty of Attaining School-Focused Possible Identities .66 .73
Note. All correlations p < .001.

104
Appendix B: Chapter 2 Measures
All response scales range from 1=strongly disagree to 6=strongly agree except for Study 3. Study
3 response scales range from 1=strongly disagree to 5=strongly agree.
Difficulty-as-Importance
Studies 1 to 2b, 4 to 6 Trait Measure, Fisher and Oyserman (2017)
1. If a task feels difficult, my gut says that it really matters for me.
2. I know a goal is a key one for me when it feels difficult to work on.
3. When a goal feels difficult to attain, then it is probably worth my effort.
4. When a task feels difficult, the experience of difficulty informs me that succeeding in the
task is important for me.
Study 3 Trait Measure, Oyserman, Destin, & Novin (2017)
1. If I’m working on a task that feels difficult, it means that the task is important.
2. A sign that a task is important to me is how difficult it feels while working on it. If it
feels difficult, it's important.
3. If a task is difficult, it is probably important for me to do well at it.
4. Struggling to complete a task reminds me that the task is important.
5. Tasks that feel difficult are important tasks.
6. If a task is difficult, it means that it's important for me.
Studies 4 to 6 Daily Measure, Study-Created
1. Today when I experienced difficulty on a task, my gut said that it really mattered for me.
2. Today when a goal was difficult to attain, I knew that the goal was a key one to work on.
3. Today when a goal was difficult to attain, I felt that it was worth my effort.
4. Today when I experienced difficulty on a task, I felt that it was important for me.
Difficulty-as-Impossibility
Studies 1 to 2b, 4 to 6 Trait Measure, Fisher and Oyserman (2017)
1. If a task feels difficult, my gut says that it may be impossible for me.
2. I know a goal is impossible for me when it feels difficult to work on.
3. When a goal feels difficult to attain, then it is probably out of my reach.
4. When a task feels difficult, the experience of difficulty informs me that succeeding in the
task is just not possible for me.
Study 3 Trait Measure, Oyserman, Destin, & Novin (2017)
1. If I feel stuck on a task, it's a sign that my effort is better spent elsewhere.
2. If working on a task feels very difficult, that type of task may not be possible for me.
3. If a task feels too difficult, I should move on to something else.
4. When working on a task feels hard, that feeling means it's not for me.
5. Finding a task really difficult tells me that I can't complete that task.
6. If a task feels really difficult, it may not be possible for me.

105
Studies 4 to 6 Daily Measure, Study-Created
1. Today when I experienced difficulty on a task, my gut said that it might be impossible for
me.
2. Today when a goal was difficult to attain, I knew that the goal was impossible for me.
3. Today when a goal was difficult to attain, I felt that it was out of my reach.
4. Today when I experienced difficulty on a task, I felt that succeeding at it was just not
possible for me.
Goal Self-Efficacy
Studies 4 to 6 Trait Measure, Study-Created
Per Bandura’s (2006) recommendation to create scales to target study issues
1. If I try to work on my goals, I can.
2. When I need to take actions to work on my goals, I can do that.
3. I am capable of working on my goals when I need to.
4. I am capable of coming up with strategies to make progress on my goals.
5. When I try to implement the strategies I come up with, I often cannot. (R)
6. I am not sure how to turn my goals into concrete actions. (R)
7. I don’t know how to make progress on the goals I have for myself. (R)
8. When I need to start working on a new goal, I have trouble getting started. (R)
9. When I encounter an obstacle while working on a goal, I have trouble keeping going. (R)
10. When I encounter a setback while working on a goal, I become disheartened. (R)
Studies 4 to 6 Daily Measure, Study-Created
1. Today when I needed to take actions to work on my goals, I could do that.
2. Today, I was capable of coming up with strategies to make progress on my goals.
3. Today when I tried to implement the strategies I came up with, I could not. (R)
4. Today, I was not sure how to turn my goals into concrete actions. (R)
5. Today, I didn’t know how to make progress on the goals I have for myself. (R)
6. Today when I needed to start working on a new goal, I had trouble getting started. (R)
7. Today when I encountered an obstacle while working on a goal, I had trouble keeping
going. (R)
Self-Esteem
Studies 4 to 6 Trait Measure, Rosenberg (1965)
1. I feel that I am a person of worth, at least on an equal plane with others.
2. I feel that I have a number of good qualities.
3. All in all, I am inclined to feel that I am a failure. (R)
4. I am able to do things as well as most other people.
5. I feel I do not have much to be proud of. (R)
6. I take a positive attitude toward myself.
7. On the whole, I am satisfied with myself.
8. I wish I could have more respect for myself. (R)
9. I certainly feel useless at times. (R)
10. At times I think I am no good at all. (R)

106
Studies 4 to 6 Daily Measures, Nezlek (2005)
1. Today, I felt like a failure. (R)
2. Today, I felt I had many good qualities.
3. Today, I thought I was no good at all. (R)
4. Today, I was satisfied with myself on the whole.
Self-Compassion
Studies 4 to 6 Trait Measure, Neff (2003)
1. When I fail at something important to me I become consumed by feelings of inadequacy.
(R)
2. I try to be understanding and patient towards those aspects of my personality I don’t like.
3. When I’m going through a very hard time, I give myself the caring and tenderness I need.
4. When I’m feeling down, I tend to obsess and fixate on everything that’s wrong. (R)
5. I’m disapproving and judgmental about my own flaws and inadequacies. (R)
6. I’m intolerant and impatient towards those aspects of my personality I don’t like. (R)
Studies 4 to 6 Daily Measure, Study-Created
1. Today, I felt compassion toward myself.
2. Today I showed caring, understanding, and kindness toward myself.
3. Today I was disapproving and judgmental about my own flaws and inadequacies. (R)

107
Appendix C: Chapter 2 Multilevel Equations
General Notation
The subscript j is for the persons (j = 1…J) and the subscript i is for the repeated measurements
of each person (i = 1…nj). eij is the residual error term at the repeated measurements level (Level
1) and u0j is the residual error term at the person level (Level 2). y00 is the grand mean for the
outcome variable score.
Equation 1: Time and Developmental Phase Predicting Difficulty Mindset Endorsements
Level-1 predictor variable: time (months from T1 to T3)
Level-2 predictor variable: developmental phase (T1 before or during high school)
Cross-level predictor variable: time x developmental phase interaction
Level-1 outcome variable: difficulty-as-importance endorsement or difficulty-as-impossibility
endorsement
(outcome variable)ij = y00 + y10(time)ij + y01(developmental phase)j +
y11[(time)ij x (developmental phase)j] + u0j + eij
Equation 2: Trait Difficulty Mindset Endorsements Predicting Daily School Engagement
Level-2 predictor variables: trait difficulty-as-importance endorsement, trait difficulty-asimpossibility endorsement, study 5 dummy variable, and study 6 dummy variable
Level-1 outcome variable: daily meaningful schoolwork engagement or daily less meaningful
schoolwork engagement
(outcome variable)ij = πij ; π ~ Binomial (μ)
πij = logistic(ηij)
ηij = y00 + y01(trait difficulty-as-importance endorsement)j +
y02(trait difficulty-as-impossibility endorsement)j +
y03(study5dummy)j + y04(study6dummy)j + u0j

108
Equation 3: Daily Difficult Mindset Endorsements Predicting Daily School Engagement
Level-1 predictor variables: group-mean centered daily difficulty-as-importance endorsement,
group-mean centered daily difficulty-as-impossibility endorsement, and time (day of diary
report)
Level-1 outcome variable: daily meaningful schoolwork engagement or daily less meaningful
schoolwork engagement
(outcome variable)ij = πij ; π ~ Binomial (μ)
πij = logistic(ηij)
ηij = y00 + y10(daily difficulty-as-importance endorsement)ij +
y20(daily difficulty-as-importance endorsement)ij + y30(time)ij + u0j
Equation 4: Daily Difficulty Mindset Endorsements Predicting Daily Self-Regard
Level-1 predictor variables: group-mean centered daily difficulty-as-impossibility
endorsement, group-mean centered daily difficulty-as-importance endorsement, and time (day of
diary report)
Level-1 outcome variable: daily goal self-efficacy, daily self-esteem, or daily self-compassion
(outcome variable)ij = y00 + y10(daily difficulty-as-importance endorsement)ij +
y20(daily difficulty-as-importance endorsement)ij + y30(time)ij + u0j + eij
Equation 5: Trait Difficulty Mindset Endorsements Predicting Daily Self-Regard
Level-2 predictor variables: trait difficulty-as-impossibility endorsement, trait difficulty-asimportance endorsement, study 5 dummy variable, and study 6 dummy variable
Level-1 outcome variable: daily goal self-efficacy, daily self-esteem, or daily self-compassion
(outcome variable)ij = y00 + y01(trait difficulty-as-importance endorsement)j +
y02(trait difficulty-as-impossibility endorsement)j + y03(study5dummy)j +
y04(study6dummy)j + u0j + eij

Abstract (if available)

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Can I make the time or is time running out? That depends in part on how I think about difficulty

PDF

The antecedents and consequences of believing that difficulties are character-building

PDF

Difficulty-as-sanctifying: when difficulties build character, purify the self, and elevate the soul

PDF

People can change when you want them to: changes in identity-based motivation affect student and teacher Pathways experience

PDF

Psychological correlates of certainty strength: measuring the relative influence of evidence, opinion of personal contacts & importance to self-identity

PDF

Selflessness takes time: altruistic (but not cooperative) prosocial behavior increases with decision time

PDF

Classrooms are game settings: learning through and with play

PDF

A roadmap for changing student roadmaps: designing interventions that use future “me” to change academic outcomes

PDF

Bridging possible identities and intervention journeys: two process-oriented accounts that advance theory and application of psychological science

PDF

Intuitions of beauty and truth: what is easy on the mind is beautiful and true

PDF

Assessing the psychological correlates of belief strength: contributing factors and role in behavior

PDF

Culture's consequences: a situated account

PDF

Help, I need somebody: an examination of the role of model minority myth and goal orientations in Asian American college students' academic help-seeking practices

PDF

The relationship between small ensemble experiences, empathy, and emotional self-regulation skills in music students

PDF

Inclusive vogue: exploring the demand for body diversity in the fashion and modeling world

PDF

A neuropsychological exploration of low-SES adolescents’ life goals and their motives

PDF

Self-reported and physiological responses to hate speech and criticisms of systemic social inequality: an investigation of response patterns and their mediation…

PDF

Metacognitive experiences in judgments of truth and risk

PDF

Underemployment among U.S. military veterans in the private sector after transitioning from the military

PDF

Identity is the lens through which moral values predict action

Asset Metadata

Creator Burbidge, Alysia (author)

Core Title Difficulty-as-importance and difficulty-as-impossibility: unpacking the context-sensitivity and consequences of identity-based inferences from difficulty

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Psychology

Degree Conferral Date 2024-05

Publication Date 05/21/2024

Defense Date 03/14/2024

Publisher Los Angeles, California (original), University of Southern California (original), University of Southern California. Libraries (digital)

Tag culture,daily self-esteem,daily well-being,identity,identity-based motivation,metacognition,motivation,OAI-PMH Harvest,possible self,primary school transition,within-person and between-person variation

Format theses (aat)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor Oyserman, Daphna (committee chair), Immordino-Yang, Mary Helen (committee member), Monterosso, John (committee member), Stone, Arthur (committee member), Townsend, Sarah (committee member)

Creator Email alysiaburbidge@gmail.com,burbidge@usc.edu

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC113950123

Unique identifier UC113950123

Identifier etd-BurbidgeAl-12984.pdf (filename)

Legacy Identifier etd-BurbidgeAl-12984

Document Type Dissertation

Format theses (aat)

Rights Burbidge, Alysia

Internet Media Type application/pdf

Type texts

Source 20240521-usctheses-batch-1157 (batch), University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu