To whom do students believe a growth mindset applies?

Alysa Malespina1 , Christian Schunn2 and Chandralekha Singh1

1. Department of Physics and Astronomy, University of Pittsburgh, Pittsburgh, PA, USA 15260 and
2. Learning, Research, and Development Center, University of Pittsburgh, Pittsburgh, PA, USA 15260

Intelligence mindset has been studied extensively in education research, but domain-specific intelligence
mindset research is relatively new in the physics context. Additionally, recent mindset research has uncovered
separable factors within the intelligence mindset construct. In this study, we test a model involving four factors
(My Ability, My Growth, Others’ Ability, and Others’ Growth) to pre and post survey data from Physics 1
classes. In particular, we explore how these mindset factors change over time as well as their ability to predict
course grade. We find that students are less likely to endorse a growth mindset for themselves and others at the
end of their first calculus-based introductory physics course than at the beginning. We also find that decrease
in mindset measures are more drastic for female students than male students. Finally, we find that the best
predictor of course grades is the My Ability component of the mindset construct, which has implications both
for creating equitable and inclusive learning environment and determining how educators implement mindset
interventions.

2022 PERC Proceedings edited by Frank, Jones, and Ryan; Peer-reviewed, doi.org/10.1119/perc.2022.pr.Malespina
Published by the American Association of Physics Teachers under a Creative Commons Attribution 4.0 license.
Further distribution must maintain the cover page and attribution to the article's authors.

292
 I. INTRODUCTION significantly greater academic achievement. The varying ef-
fectiveness of these interventions may be due in part to proce-
Gender differences in introductory physics performance dural details (e.g., whether the intervention needs to be cus-
and persistence have been linked to many factors, such as tomized to the particular concerns that students have in a par-
societal stereotypes and biases about who can succeed in ticular context). Additionally, the focus of each interventions
physics [1–3], an unwelcoming climate in physics classrooms may account for their effectiveness. For example, did the in-
for women [1, 2], and gender differences in motivational fac- tervention seek only to address the growth mindset but ig-
tors [4–6]. Among the many motivational factors that have nore the ability mindset? Prior interventions may have tried
been investigated, considerable attention has been placed on to convince students that people in general can grow their in-
the role of intelligence mindsets [7], which describe a per- telligence, leaving relatively untouched the beliefs they have
son’s beliefs about the nature of intelligence. A growth mind- about themselves.
set is one in which intelligence is viewed as something that
A third issue might also exist in domain-specificity of in-
can be cultivated with effort, like a muscle, while a fixed
telligence mindsets. Students might believe that intelligence
mindset is one in which intelligence is thought to be in-
in general can change through hard work but still have fixed
nate and unchangeable [8]. The mindsets held by learners
mindsets about particular domains. Physics-specific mindset
are thought to shape how students engage in learning. With
research has just begun in recent years [5, 7]. Early analy-
a fixed mindset, a student is likely to disengage from or
sis has found that physics-specific mindset is both different
avoid difficult tasks [9]. On the other hand, the engagement,
than and a better predictor of physics class performance than
propensity to attempt challenging problems, and persistence
general intelligence mindsets [5]. There is also evidence that
that often come with growth mindsets have been linked to
physics intelligence mindset becomes more fixed after taking
positive learning outcomes [8, 10], even after controlling for
a physics course, especially for female students [5]. This may
prior academic achievement [11, 12].
be in part due to brilliance beliefs in physics - both the physics
Intelligence mindsets may also play a role in shaping
community [3] and general public [2] believe that success in
learner self-efficacy [13]. As a result, growth mindsets are not
physics requires innate talent. Brilliance beliefs are not the
only relevant to creating equitable classroom environments,
same as a fixed mindset, though they work in tandem. If a
but they also may be an important factor in improving learn-
student thinks raw talent is needed to succeed in a domain
ing outcomes for all students. Growth mindsets have also
(a brilliance belief), and they believe that intelligence is un-
been linked to greater participation in STEM fields for stu-
changeable (a fixed mindset), then they will see no path to
dents in underrepresented groups [14], and can be a useful
success unless they believe they have innate talent [24]. This
resource for underrepresented students to combat stereotype
combination of brilliance beliefs and a fixed mindset can re-
threat or anxiety [15]. Fixed mindsets can cause students to
sult in fewer students from underrepresented groups entering
withdraw from a subject of interest due to fear of representing
physics [3].
their identities (e.g., race or gender) poorly[15], while anxiety
Notably, some research supports division of mindset com-
can limit working memory [16].
ponents in the physics context [21, 22] However, since
While intelligence mindset was originally conceptualized
physics-specific mindset is still a very recently explored con-
as a continuum (with fixed and growth mindsets on either
cept, many fundamental questions about its nature and rela-
end [8, 11], a two-factor model has gained popularity in re-
tionship to gendered performance in physics are still open.
cent years [17, 18]. In this model, endorsement of fixed and
Specifically, we address the following questions:
growth mindsets are measured separately. The primary evi-
dence in favor of treating them separately as two dimensions RQ1. Do students have different mindset beliefs about them-
was psychometric evidence in which a two-factor model pro- selves and others, and do they change over time?
duced a better fit to the data [17, 18]. Another conceptual RQ2. Do any mindset components predict the course grade?
divide in mindset research involves beliefs about self ver-
sus others. One study [19] found that Australian high-school II. METHODOLOGY
students conceptualized intelligence mindsets differently for
themselves than for others. They also found that intelligence This study took place at a large, public university in the
"self-theory" was a stronger predictor of academic perfor- United States. Participants were enrolled in calculus-based
mance and motivation than general intelligence mindsets. Physics 1 over one semester and across four course sections,
Although intelligence mindsets are carried by students into each taught by a different instructor. The surveys were ad-
various learning contexts (i.e., have some stability over time ministered in the first and last week of the required teach-
and context), they can be malleable through strategic (and rel- ing assistant-led recitations. The mindset items were a subset
atively brief) interventions with positive results for students’ of a larger survey, which took approximately ten minutes to
learning outcomes [11, 12], especially if students are at a high complete. Students received either a participation grade or a
risk of failing a class [20]. However, a recent meta-analysis small amount of extra credit for completing the survey, de-
[23] has found that both the methodology and effectiveness pending on the instructor’s preference. Survey results were
of these interventions vary greatly. For example, only 12% collected, de-identified by an honest broker, and then com-
of the interventions included in the meta-analysis resulted in bined with similarly de-identified demographic information

293
and academic history. In the student sample (N = 683), TABLE I: Mindset survey questions.
63% were first-semester engineering students. Female stu-
dents made up 36% of the sample - only binary data were My Growth (α = 0.84)
available for this sample, but we acknowledge that gender is
1 I can become even better at solving physics problems through
more complex than our data suggest. hard work.
The mindset survey was adapted from previously validated 2 I am capable of really understanding physics if I work hard.
surveys [5]. The survey was designed to measure mindsets 3 I can change my intelligence in physics quite a lot by working
about self and others, as well as growth and ability mind- hard.
sets. Therefore, to be able to separately evaluate these differ-
My Ability (α = 0.84)
ent aspects of mindset, additional questions were created and
some questions were adapted to make the specific focus more 4 Even if I were to spend a lot of time working on difficult
physics problems, I cannot develop my intelligence in physics
salient. For example, "People can change their intelligence in
further.
physics quite a lot by working hard," becomes "I can change
5 I won’t get better at physics if I try harder.
my intelligence in physics quite a lot by working hard." After
the questions were drafted, we used twenty hours of semi- 6 I could never excel in physics because I do not have what it
takes to be a physics person.
structured think-aloud interviews to refine the questions and
ensure that students interpreted them as intended. 7 I could never become really good at physics even if I were to
work hard because I don’t have natural ability.
Confirmatory Factor Analysis (CFA) using the R package
Others’ Growth (α = 0.84)
"lavaan" was used to both provide quantitative validation of
the survey items and to test the proposed conceptual divi- 8 People can change their intelligence in physics quite a lot by
working hard.
sion into four components in terms of growth/ability and my-
self/others. For model fit, we chose the following standards: 9 If people were to spend a lot of time working on difficult
physics problems, they could develop their intelligence in
standardized factor loadings above 0.5 [25], a Comparative
physics quite a bit.
Fit Index (CFI) and Tucker Lewis Index (TLI) ≥ 0.95, a Root
Mean Square Error of Approximation (RMSEA) ≤ 0.08 , and 10 People can become good at solving physics problems through
hard work.
a Standardized Root Mean Square Residual (SRMR) ≤ 0.06
[26]. In the final model, which can be seen in Table I, the Others’ Ability (α = 0.68)
CFI was 0.95, the TLI was 0.95, the RMSEA was 0.073, and 11 Only a few specially qualified people are capable of really un-
the SRMR was 0.052. All fit indices met the chosen cutoffs. derstanding physics.
To create latent variables, we calculated the average score for 12 To really excel in physics, people need to have a natural ability
the questions in each validated category, and we named the in physics.
resulting variables "My Ability" (MA), "My Growth" (MG), 13 If a student were to often make mistakes on physics assign-
"Others’ Ability" (OA), and "Others’ Growth" (OG). All the ments and exams, I would think that maybe they are just not
mindset factors are scored from 1 to 4, and are coded such that smart enough to excel in physics.
a high score corresponds with a growth physics mindset, and
a low score corresponds with a fixed mindset. After averag-
ing scores we winsorized each mindset factor so that outliers tional midterm and final exams, with a smaller portion based
were set at a cutoff two standard deviations from the mean on homework, quizzes, and recitation attendance.
of each factor (in order to maintain the direction of outliers To characterize change in mean attitudes over time, and
while eliminating extreme values [27]). differences by gender in mean attitudes at pre and post as well
High school Grade Point Average (HS GPA) and Scholastic as grades, we used Cohen’s d to describe the size of the mean
Achievement Test math (SAT math) scores were used as mea- differences and t-tests to evaluate the statistical robustness of
sures of prior academic preparation. HS GPA was reported the differences. Cohen’s d can be considered small if d ∼ 0.2,
using the weighted 0-5 scale, which is based on the standard 0 medium if d ∼ 0.5, and large if d ∼ 0.8 [29]. Pre-post
(Failing) - 4 (A) scale with adjustments for advanced courses. differences were conducted using a paired t-test and gender
SAT math scores were winsorized using a two standard devi- differences were conducted with unpaired t-tests. We used
ation cutoff to maintain the direction of outliers without in- a significance level of 0.05 in the t-tests and later regression
troducing extreme values. If only American College Testing models as a balance between Type I and Type II errors [29].
(ACT) scores were provided, scores were converted to the Multiple linear regression analysis was used to find par-
SAT equivalent for analysis [28]. Course performance was tial correlations between mindset factors and grades control-
measured using final grades on a 0 - 4 scale. The suffixes ’+’ ling for gender and prior preparation. Multiple models were
and ’-’ respectively add or subtract 0.25 grade points (e.g. B- tested in order to find which was the best predictor of learning
= 2.75 and B+ = 3.25), except for the A+, which is reported outcomes and show robustness of relationships across model
as 4. Each instructor determined their own grading scheme specification. A baseline model predicted grade using only
and exam content. From examination of syllabi across all gender, HS GPA, and SAT math scores. Next, we added the
sections, course grades were predominantly based upon tradi- mindset variables with the strongest correlation to grade one-

294
by-one until all mindset variables were present. All models TABLE II: Means (M) and standard deviations (SD) of
with significant mindset variables were kept, along with the students’ academic and mindset factors. 164 women and 264
final model with all variables induced as a robustness test. men took the pre-survey; 70 women and 147 men took the
The regression analyses were conducted with the scores from post-survey. Cohen’s d is negative if scores decreased from
the pre-survey, then from the average of pre- and post-survey pre to post. ∗ = p < 0.05, ∗∗ = p < 0.01, and ∗∗∗ = p < 0.001.
scores. The average group included only students who took
the survey both times. Average scores were used as a proxy Women Men
for students’ mindset during the semester, when they were
Variable M SD d M SD d
taking the course, rather than after the class. Using average
rather than only pre is particularly important given the size- MG Pre 3.59 0.46 3.62 0.48
able changes from pre to post that were observed in several MG Post 3.12 0.58 -0.96∗∗∗ 3.48 0.54 -0.31∗∗
of the attitudinal variables. MA Pre 3.32 0.46 3.49 0.46
MA Post 2.98 0.59 -0.71∗∗∗ 3.28 0.62 -0.40∗∗∗
OG Pre 3.48 0.49 3.50 0.49
III. RESULTS AND DISCUSSION
OG Post 3.17 0.44 -0.66∗∗∗ 3.45 0.54 -0.15
Do students have different mindset beliefs about them- OA Pre 3.15 0.53 3.16 0.55
selves and others, and do these change over time? In OA Post 2.91 0.49 -0.49∗∗∗ 3.08 0.61 -0.14
validation, survey items separated along both myself/others
SAT Math 695 64 712 60
and ability/effort dimensions, resulting in four factors: My
Ability (MA), My Growth (MG), Others’ Ability (OA), and HS GPA 4.25 0.35 4.09 0.42
Others’ Growth (OG). This is consistent with previous work Grade 2.49 0.90 2.76 0.98
[19], which found the same division of mindset factors in
Australian high school students. Our results also agree with
other studies that find a divide along the ability/effort di- average motivational characteristics, such as self-efficacy and
mensions [17, 18]. Table II shows descriptive statistics for sense of belonging, then male students [4–6]. In physics, and
each measure by gender at pre and post. On the pre-surveys, other fields in which success is viewed as a result of brilliance
men and women have nearly identical scores for both OA [3], students become less engaged and more likely to avoid
and OG. Men and women also have similar scores for MG attempting difficult problems if they don’t believe they have
pre. This means that at the beginning of the semester, most the necessary talent to succeed [9]. Female students gener-
students have growth rather than fixed mindsets, particularly ally receive few messages that they are brilliant and can thus
when considering others. The only pre-survey category with succeed in physics. Women also make up less than a third of
a significant gender difference is My Ability. In this category students who take advanced high school physics (Physics 2 or
men had higher scores than women, with a medium effect AP Physics C), so they are also likely to be less experienced
size (d = 0.37, p < 0.001). This means women were more in physics than their male peers [30]. If the student believes
likely than men to believe that natural ability is important for that ability is static, then a student attributing early difficulty
themselves to succeed in physics. to lack of ability instead of experience can discourage the en-
By the end of the semester, all students saw declines in MA gagement needed to master content.
and MG, and women also saw declines in OA and OG. The Do any mindset components predict course grade? We
declines in all four categories were larger for women than conducted multiple regression analysis to find which mindset
for men. This suggests that classroom experiences that influ- factors best predicted course grade (see Tables III and IV).
enced student mindsets affected men and women differently. Models 1-3 used only pre-survey results, while Models 4-6
The larger declines in MA and MG than OA and OG suggest used the mean of pre- and post-survey mindset scores (be-
that intelligence mindset "self-theory" [19] may be more mal- cause of the large changes in mindset across the semester).
leable than general physics mindset. This also suggests that In Model 1, only gender, SAT math scores, and HS GPA are
the classroom environment led to students becoming more included as predictors and all three were statistically signif-
fixed mindset away from growth mindset in the physics con- icant. This model shows that women have lower Physics 1
text. Trends were similar across instructors, though some re- grades than men when controlling for prior academic prepa-
sults were non-significant when calculated for individual in- ration, formally establishing that other factors are needed to
structors’ classes, due to low sample size. account for gender differences in course performance. Model
Further, following their first university-level experience, 2 includes My Ability (MA) as a fourth predictor. MA was
women became more likely than men (for MA post d = 0.50, chosen because it is the single strongest correlate of grades.
p < 0.001) to believe that natural ability is important to suc- Here pre-survey MA is a significant predictor beyond aca-
ceed in physics for both themselves and others. This suggests demic preparation. Its addition weakens the relationship be-
that classroom experiences that influenced student mindsets tween gender and grade, though gender remains significant.
affected men and women differently. These results mirror Model 3 adds the remaining pre-survey mindset factors: MG,
other studies that have shown that female students have lower OA, and OG. None of the newly-added factors are statistically

295
TABLE III: Standardized β coefficients of models predicting ing on teaching students about the brain’s ability to change
course grade using pre-survey results. For the gender and grow [11, 12]. The latter approach appears to be well-
variable, Women = 1 and Men = 0. N = 497. suited to students who hold a general fixed mindset. How-
ever, it may not be useful to students who endorse a general
Model 1 Model 2 Model 3
growth mindset but a fixed self-mindset. In addition to show-
ing students that changing one’s intelligence is possible, we
Predictor: β p β p β p
must show them that they can change their own intelligence.
Gender -0.13 <0.001 -0.12 <0.001 -0.11 0.003 Moreover, the mindset of the instructors can predict the
SAT Math 0.39 <0.001 0.38 <0.001 0.38 <0.001 motivation and achievement of students in their class [34]. In-
HS GPA 0.32 <0.001 0.32 <0.001 0.32 <0.001 structors with fixed mindsets tend to have low expectations of
MG Pre 0.03 0.515 students they believe lack natural talent [35]. Instructors with
MA Pre 0.09 0.016 0.10 0.038 growth mindsets encourage students to embrace mistakes and
OG Pre -0.01 0.910 failures as a part of a normal learning process, congratulate
persistence, and praise effort rather than intelligence when
OA Pre -0.05 0.212
students succeed [32, 35]. Students in classes taught by pro-
R2 0.32 0.33 0.33
fessors with a growth mindset report that the instructors "mo-
tivated them to do their best work" and "emphasized learning
and development" [34]. Thee studies suggest that, even in
significant, and their addition leaves fully intact or slightly courses that may not have the time or resources to conduct a
strengthens the predictive power of the other predictors, sug- mindset intervention, some benefits of a growth mindset can
gesting robust relationship estimates. The predictive power be brought about via instructor-level changes.
of gender decreases slightly. Importantly, using pre-survey IV. FUTURE DIRECTIONS
results, MA is the mindset variable with the most predictive
power and the variable with the largest gender difference. Our results are most likely to translate directly to other
Models 4-6 (in Table IV) are focused on the sample that large universities. However, results in different contexts, such
completed both pre and post. Model 4 is identical to Model as liberal arts and community colleges, would be valuable to
1, but now provides the baseline model for the reduced sam- examine and compare to our findings. Further, due to low
ple set. The parameter values are similar to Model 1, although sample size, we were unable to study whether mindset beliefs
the SAT estimate is smaller and the gender estimate is larger. differ or predict grades differently for students of different
Model 5 adds average MA as a predictor. Average MA has racial/ethnic backgrounds. We plan to do this in future years
more than twice the predictive power of pre-MA, and the gen- once we have an adequate number of students for this type of
der estimate decreases in size by 38%. Model 6 introduces the analysis. We also note that the relationship between mindset
remaining average mindset factors, none of which are stati- and grades is correlational in our study. Other work has sug-
cally significant predictors (similar to the findings of Model gested a causal relationship [36] but this relationship needs to
3). There are no major changes in the predictive power of be further explored. We believe that the decrease in growth
MA, HS GPA, or SAT math from Model 5 to Model 6, again mindset and growing gender disparity over the semester is a
suggesting robust relationship estimates and that MA in par- solvable problem. Thus, future work must focus on develop-
ticular was the most likely mediator of gender differences in ment and implementation of teaching methods and interven-
grades among the mindset factors. When performing regres- tions to cultivate a growth mindset in all students.
sions, MG, OA and OG do not predict course grade, while
both pre- and average-MA do. TABLE IV: Standardized β coefficients of models predicting
Physics self-mindset is a predictor of Physics 1 grade, so course grade using averaged survey results. For the gender
increasing MA beliefs may increase all students’ performance variable, Women = 1 and Men = 0. N = 197.
as well as mitigate gendered grade differences. In this popula-
tion (primarily engineering students), women are more likely Model 4 Model 5 Model 6
to leave the major due to concerns about low grades than men Predictor: β p β p β p
are, even when they have an A or B average [1], so enhanc-
Gender -0.15 0.019 -0.09 0.152 -0.09 0.179
ing students’ MA beliefs may increase retention. Importantly,
average-MA is a stronger predictor of course grade than pre- SAT Math 0.34 <0.001 0.31 <0.001 0.31 <0.001
MA. Educators have an opportunity to intervene and poten- HS GPA 0.28 <0.001 0.28 <0.001 0.28 <0.001
tially improve grades and cultivate growth mindsets, since MG Avg 0.07 0.485
self-mindset appears to be simultaneously more malleable MA Avg 0.25 <0.001 0.24 <0.001
and have a stronger correlation to learning outcomes. Prior OG Avg -0.02 0.791
research suggests that mindset interventions in this context OA Avg -0.06 0.426
should focus on students’ individual experiences or the expe- R2 0.27 0.33 0.32
riences of people they can relate to [31–33], rather than focus-

296
 [1] I. Goodman, Final report of the women’s experiences in college Psychol. Test Assess. Model. 62, 3 (2020).
engineering project, (Goodman Research Group, 2002). [19] K. De Castella and D. Byrne, My intelligence may be more
[2] E. Reuben, P. Sapienza, and L. Zingales, How stereotypes im- malleable than yours: The revised implicit theories of intelli-
pair women’s careers in science, in Proceedings of the National gence (self-theory) scale is a better predictor of achievement,
Academy of Sciences 111, 12 (2014). motivation, and student disengagement. Eur. J. Psychol. Educ.
[3] S. Leslie, A. Chimpian, M. Meyer, and E. Freeland, Expec- 30, 245 (2015).
tations of brilliance underlie gender distributions across aca- [20] D. Yeager, P. Hanselman, and G. Walton, A national experi-
demic disciplines, Science 347, 6219 (2015). ment reveals where a growth mindset improves achievement,
[4] V. Sawtelle, E. Brewe, and L. Kramer, Exploring the rela- Nature 573, 364 (2019).
tionship between self-efficacy and retention in introductory [21] Z. Y. Kalender , E. Marshman, C. D. Schunn, T. J. Nokes-
physics, J. Res. Sci. Teach. 49, 1096 (2012). Malach, and C. Singh, Framework for unpacking studentsâ
[5] E. Marshman, Z. Kalender, C. Schunn, T. Nokes-Malach, and mindsets in physics by gender. Phys. Rev. Phys. Educ. 18,
C. Singh, A longitudinal analysis of students’ motivational 010116 (2022).
characteristics in introductory physics courses: Gender differ- [22] A. Malespina , C. D. Schunn, and C. Singh, Whose ability and
ences, Can. J. Phys. 96, 391 (2017). growth matter? Gender, mindset and performance in physics.
[6] J. Nissen and J. Shemwell, Gender, experience, and self- Int. J. STEM Educ. 9, 28 (2022).
efficacy in introductory physics, Phys. Rev. Phys. Educ. Res. [23] V. Sisk, A. Burgoyne, J. Sun, J. Butler and B. Macnamara, To
12, 020105 (2016). what extent and under which circumstances are growth mind-
[7] A. Little, B. Humphrey, A. Green, A. Nair, and V. Sawtelle, sets important to academic achievement? Two meta-analysis,
Exploring mindset’s applicability to students’ experiences with Psychol. Sci. 29, 549 (2018).
challenge in transformed college physics courses, Phys. Rev. [24] A. Deiglmayr, E. Stern, and R. Schubert, Beliefs in "brilliance"
Phys. Educ. Res. 15, 010127 (2019). and belonging uncertainty in male and female STEM students,
[8] C. Dweck, Mindset: The new psychology of success (Random Front. Psychol. 10, 1114 (2019).
House, 2006). [25] R. Kline, Principles and practice of structural equation mod-
[9] D. Yeager and C. Dweck, Mindsets that promote resilience: eling (Guilford Press, 2016).
When students believe that personal characteristics can be de- [26] L. Hu and P. Bentler, Fit indices in covariance structure model-
veloped, Educ. Psychol. 47, 302 (2012). ing: Sensitivity to underparameterized model misspecification,
[10] L. Limeri, N. Carter, H. Harper, H. Martin, A. Benton and E. Psychol. Methods 3, 4 (1998).
Dolan, Growing a growth mindset: Characterizing how and [27] B. Frey, editor, The SAGE Encyclopedia of Educational Re-
why undergraduate students’ mindset change, Int. J. STEM search, Measurement, and Evaluation, (SAGE, 2018).
Educ. 7, 35 (2020). [28] The College Board and ACT, Inc, Guide to the 2018 ACT/SAT
[11] L. Blackwell, K. Trzesniewski, and C. Dweck, Implicit the- Concordance, 2018.
ories of intelligence predict achievement across an adolescent [29] J. Cohen, P. Cohen, S. West, and L. Aiken, Applied Multiple
transition: A longitudinal study and an intervention, Child Dev. Regression/Correlation Analysis for the Behavioral Sciences,
78, 246 (2007). (Lawrence Erlbaum, 2003).
[12] C. Good, J. Aronson and M. Inzlicht, Improving adoles- [30] A. Porter and R. Ivie, Statistical Research Center of the Amer-
cents’ standardized test performance: An intervention to re- ican Institute of Physics, Women in Physics and Astronomy,
duce the effects of stereotype threat, Appl. Dev. Psychol. 24, 2019.
645 (2003). [31] K. Binning et al., Changing social contexts to foster equity in
[13] A. Bandura, Self-efficacy: The Exercise of Control (Macmillan, college science courses: An ecological-belonging intervention,
1997). Psychol. Sci. 31, 1059 (2020).
[14] K. Kricorian, M. Seu, D. Lopez, E. Ureta and O. Equils, Fac- [32] C. Mueller and C. Dweck, Praise for intelligence can under-
tors influencing participation of underrepresented students in mine children’s motivation and performance, J. Person. Soc.
STEM fields: Matched mentors and mindsets, Int. J. STEM Psychol. 75, 33 (1998).
Educ. 7, 16 (2020). [33] G. Walton, and G. Cohen, A brief social-belonging interven-
[15] C. Steele, and J. Aronson, Stereotype threat and the intellectual tion improves academic and health outcomes among minority
test performance of African Americans, J. Pers. Soc. Psychol. students, Science 331, 1447-1451 (2011).
69, 5 (1995). [34] E. Canning, K. Muenks, D. Green, and M. Murphy, STEM fac-
[16] S. Beilock, R. Rydell, and A. McConnell, Stereotype threat and ulty who believe ability is fixed have larger racial achievement
working memory: Mechanisms, alleviations, and spillover, J. gaps and inspire less student motivation in their classes, Sci.
Exp. Psych. General 136, 256 (2007). Adv. 5, 2 (2019).
[17] D. Cook, R. Castillo, B. Gas, and A. Artino, Measuring [35] A. Rattan, C. Good, and C. Dweck, "It’s ok-Not everyone can
achievement goal motivation, mindsets and cognitive load: be good at math": Instructors with an entity theory comfort
Validation of three instruments’ scores, Med. Educ. 51, 10 (and demotivate) students. J. Exp. Soc. Psychol. 48, 3 (2012)
(2017). [36] R. Felder, G. Felder, M. Mauney, C. Hamrin, and E. Dietz, A
[18] S. Troche and A. Kunz, The factorial structure and construct longitudinal study of student performance and retention. Gen-
validity of a German translation of Dweck’s Implicit Theories der differences in student performance and attitude, J. Engi-
of Intelligence Scale under consideration of the wording effect, neering Educ. 84, 151 (1995).

297