Nationwide, FG students represent a large pool of potentialscientists, engineers, and mathematicians.

While the current study provides some promising strategiesthat instructors can use in their courses to promote the perfor-mance of FG students, it is important to acknowledge its limita-tions and generalizability.

That is, when the professor communicated growthmindset beliefs (vs. control) it significantly increased FGstudents’ performance, F (1, 410) = 6.414, p = 0.012, butdid not increase CG students’ performance, F (1, 410) = 0.024,p = 0.877.

TABLE 1. Model results for all dependent variablesExam #2 Exam #3 Course GPAF (1, 409) p η2p F (1, 407) p η2p F (1, 410) p η2pCondition 1.26 0.263 0.003 3.87 0.050 0.009 4.61 0.032 0.011FG status 0.03 0.866 0.000 4.12 0.043 0.010 3.23 0.073 0.008Condition X FG 1.36 0.244 0.003 3.97 0.047 0.010 3.86 0.050 0.009Personal fixed mindset 0.57 0.452 0.001 1.62 0.203 0.004 0.28 0.595 0.001College GPA 136.14 0 < 0.001 0.250 121.55 0 < 0.001 0.230 285.24 0 < 0.001 0.410URM status 8.94 0.003 0.021 14.55 0 < 0.001 0.035 5.22 0.023 0.013Course content access Gradebook accessF (1, 410) p η2p F (1, 410) p η2pCondition 5.12 0.024 0.012 1.27 0.261 0.003FG status 0.02 0.903 0.000 0.67 0.414 0.002Condition X FG 0.12 0.731 0.000 0.13 0.719 0.000Personal fixed mindset 3.04 0.082 0.007 0.91 0.341 0.002College GPA 38.06 0 < 0.001 0.085 17.15 0 < 0.001 0.040URM status 2.27 0.132 0.006 0.61 0.435 0.001Note. FG = first-generation; URM = underrepresented racial/ethnic minority; Condition was coded: 1 = growth mindset, –1 = control; FG status was coded: 1 = FG,0 = CG; URM status was coded: 1 = URM (Black, Hispanic, Native American, or Pacific Islander), 0 = non-URM (White or Asian).TABLE 2. Raw means and standard deviations by condition and generational statusGrowth mindset condition Control conditionVariable Generational status M SD M SDExam #2 CG 72.00 14.36 71.69 14.69FG 69.07 15.12 65.44 14.90Exam #3 CG 68.84 14.19 68.62 14.21FG 64.51 16.81 58.39 16.32Course GPA CG 2.67 1.09 2.63 1.06FG 2.34 1.22 1.89 1.29Course Content Access CG 369.80 198.82 334.43 164.02FG 347.41 171.86 294.46 199.72Gradebook Access CG 70.07 98.14 58.18 67.29FG 53.34 61.89 45.48 61.00Note. FG = first-generation; CG = continuing-generation.

All analyses were conducted using SPSS Version 27. We con-ducted a two-way ANCOVA on all dependent variables to com-pare the intervention and control conditions and their interac-tion with FG status. Three covariates were included in themodel: 1) students’ college GPA, 2) personal mindset beliefs,and 3) URM status. Students’ self-reported current college GPAwas included as a covariate in all analyses to assess the effect ofthe instructor growth mindset manipulation independent ofstudents’ prior academic performance. Previous research exam-ining intervention effects on students’ performance typicallycontrols for students’ academic performance before the inter-vention (Harackiewicz et al., 2016; Yeager et al., 2016; Canninget al., 2018). Students’ personal mindset beliefs were entered asa covariate in all analyses to assess the effect of the instructorgrowth mindset intervention independent of students’ personalmindset beliefs. Previous research examining the effects ofinstructor mindset on student outcomes has controlled for stu-dents’ personal mindset beliefs when available (LaCosse et al.,2020; Muenks et al., 2020; Canning et al. 2022). URM statuswas included as a covariate to assess the effect of the manipula-tion for FG students independent of URM status. In our sample,36% of the FG students were also URM. See Table 1 for modelresults for all dependent variables, see Table 2 for means anddescriptive statistics by condition and FG status. See Supple-mental Table S1 in Supplemental Materials for all model resultswithout covariates.

Recently, a university-wide study conducted with STEMinstructors revealed that students earned higher grades whentheir instructor endorsed more of a growth (vs. fixed) mindset–and this was especially true for stigmatized students (Canninget al., 2019). Further studies illuminate several potential mech-anisms. Instructors with growth mindsets engender greatertrust, sense of belonging, academic engagement, and fewerfeelings of being an imposter among their students (Cavanaghet al., 2018; Rattan et al., 2018; LaCosse et al., 2020; Muenkset al., 2020; Canning et al., 2022; Hecht et al., 2022). In onestudy, a college instructor built trust with their students in partby communicating a growth mindset, which resulted in stu-dents becoming more engaged in the course and earning highergrades (Cavanagh et al., 2018).

Indeed, many introduc-tory science courses are designed to “weed out” those students deemed capable andthose that are not.

This field experiment took place in a large enrollment Introduc-tory Biology course at a research intensive public university inthe Pacific Northwest. We chose an introductory biology courselargely because the instructor was willing to collaborate with usand because most introductory biology courses serve as import-ant gateways to persistence in STEM fields (Seymour andHunter, 2019). This Introductory Biology course is a criticalgateway course to further study in the biological sciences.Students typically take this course in their freshmen or sopho-more year, and their experiences in this foundational coursemay determine whether they pursue subsequent coursework ina variety of STEM disciplines. The instructor for this course had7.5 years of experience teaching the course. The experimenttook place during the Spring 2021 semester. This semester wasunique in a historical sense in that the course was taught com-pletely online due to COVID-19 precautions. All lectures weredelivered synchronously via Zoom and all exams were admini-stered online and proctored by the instructor and graduateteaching assistants.

First-generation (FG) college students (i.e., those for whom neither parent/guardian ob-tained a bachelor’s degree) experience more barriers in college, compared with continu-ing-generation students.