henhaobio
Sign in

bioRxivScientific Communication & EducationPreprint

Family matters! The influence of family, peers, mentors, and professors on STEM college students motivational beliefs and career decision making

Trevor T. TumaiD

PDFDOI
In-depth full-text digest available · Jump to digest ↓

Abstract

Support from family, peers, mentors, and professors has a crucial role in the development of students motivational beliefs about learning. However, the relative importance of each of these social supports in affecting students motivation, and the precise processes through which these social factors motivate students, are not well understood at the college level. The present study examined how family, peers, mentors, and professors shaped science, technology, engineering, and mathematics (STEM) college students motivational beliefs for career decision making, defining motivational beliefs in accordance with situated expectancy-value theory. We collected survey and open-ended responses from a large sample of college students (n = 2,229) in 42 unique STEM courses at a large, public southeastern university in the United States. We examined how, why, and for whom different social supports influenced students task-value and/or competence related motivational beliefs for career-decision making. Results indicated that students most often selected interactions with family members as influencing their career decision-making, both in terms of overall influences and as the strongest influence, relative to interactions with peers, mentors, or professors. For all social influences, most students reported being attracted in a positive way towards certain careers as a result of social interactions, as opposed to socializers making students disenchanted with careers. Most students described how social supports influenced motivational beliefs related to task values, with comparatively fewer referencing social influences on competence-related beliefs. Findings highlight the important role of multiple social supports, particularly family, in shaping students motivational beliefs and career decision-making throughout emerging adulthood. Educational Impact and Implications StatementPreparing a qualified and diverse STEM workforce requires a robust understanding of the factors that influence college students motivational beliefs and decisions about their career paths. We examined how, why, and for whom support from family, peers, mentors, and professors impacted STEM college students career motivation and decision making, using both qualitative and quantitative data. Results suggest that family plays a particularly crucial role in shaping college students career motivation, especially by helping them see the positive value of particular career paths. Findings help illustrate the importance of different kinds of social support in guiding career choices throughout emerging adulthood.

In-depth summary

Claude Opus 5.5 · xhigh effort · 8 Nov 2026 · from the full text of version 2 (22 May 2025) · CC BY 4.0

Self-reports from 2,229 STEM undergraduates rank family as the most common top career influence (38.4%), acting chiefly by highlighting a career's value rather than students' competence.

Most research on how parents, teachers and peers socialize motivation comes from K-12 settings, so it is unclear which socializers matter most in college career choices, how they act, and whether this varies by gender, race/ethnicity, or year. This mixed-methods survey ranks four socializers against non-social supports and codes open-ended answers through situated expectancy-value theory (SEVT) to map attraction versus disenchantment and value versus competence pathways.

Key results

  1. ShownSTEM undergraduates named family more often than any other social support, both as any influence on career decisions and as a top influence. Peers, mentors/advisors/role models, and professors/TAs followed in that order.

    Any influence: family n = 1495 (67.1%), peers 1195 (53.6%), mentors/advisors/role models 740 (33.2%), professors/TAs 577 (25.9%). Most influential: family 857 (38.4%), peers 362 (16.2%), mentors 207 (9.2%), professors/TAs 113 (5.0%); analytic sample n = 2,229; students reported M = 3.19 (SD = 1.46) supports. · What social supports do STEM college students…; The importance of family for promoting college…

  2. ShownCompared with non-social supports, coursework was chosen slightly more often than family as any influence. When students named their top influence, family ranked above coursework, professional experiences, and university-provided resources.

    Any influence: high school or college coursework n = 1557 (69.9%) vs family 67.1%; job/internship/research experiences 966 (43.3%); university resources 407 (18.3%). Most influential: family 38.4% vs coursework 734 (32.9%), professional experiences 554 (24.8%), university resources 84 (3.7%). · How impactful are these social supports on…; The importance of family for promoting college…

  3. ShownFor all four socializers, students explaining their top social influence most often described being drawn toward a career through its perceived value or interest. Competence-related or disenchanting influences were much less common, and mentors had the highest share of competence-related attraction.

    Attraction-coded: family 800 (95.1%), peers 323 (90.9%), mentors 195 (96.5%), professors 109 (97.3%). Value vs competence attraction: family 54.8% vs 9.6%, peers 51.0% vs 10.4%, mentors 48.0% vs 18.8%, professors 55.3% vs 4.4%. Disenchantment: family 9.4%, peers 14.9%, mentors 14.3%, professors 10.7%. · Family members; Peers and friends; Mentors, advisors, & role models; Professors & Teaching Assistants

  4. SuggestedEach socializer worked through different behaviors. Family inspired, advised, and set expectations. Peers served as models for social comparison. Mentors and professors introduced careers students did not know about and gave practical guidance, and mentors also gave emotional support.

    Second-cycle pattern and axial coding of free-text responses (n = 2,059 coded), with themes illustrated by student quotations; frequencies for these behavioral themes are not given in the text. · Coding how social supports influenced career…; Mentors, advisors, & role models; Summary of coding results; Different socializers provide both overlapping…

  5. SuggestedFamily was the top influence in every gender, racial/ethnic, and year group. Exploratory tests found peers mattered more for men and mentors for women. After the second year, students named family less often and mentors and professors more often.

    Peers as top factor, men > women: χ2(1) = 22.15, p < 0.0001. Mentors as top factor, women > men: χ2(1) = 6.23, p = 0.012. Family as top factor, 1st/2nd-year > 3-6th year: χ2(1) = 31.76, p < 0.0001. Professors overall, 3-6th year higher: χ2(1) = 21.06, p < 0.0001. Family as top factor, Asian and Black > White and Hispanic: χ2(3) = 13.51, p = 0.003. · Are there variations in how these social…; The importance of family for promoting college…; Individual differences in social and non-social…

Methods, data and reproducibility

Cross-sectional Spring 2022 survey of STEM-course undergraduates. Students checked which social and non-social supports influenced their career decisions and picked their top one or two. Free-text explanations were coded with SEVT categories (attraction/disenchantment; value/competence), and chi-square tests compared gender, race/ethnicity, and year groups.

  • Undergraduates enrolled in STEM courses at a single large public research university (Southeastern US), surveyed Spring 2022; analysis restricted to intended STEM majors · human · n = 2,229 analyzed (2,312 surveyed; 83 non-STEM majors excluded)
  • Free-text explanations of how each student's most influential support(s) shaped career decisions, coded for attraction/disenchantment and value/competence · human · n = 2,059 coded (232 for reliability; 1,827 by two coders)
Methods & tools
Researcher-designed career-support survey (checklist plus open-ended item), Situated expectancy-value theory (SEVT) coding framework, Qualitative content analysis (open, first-cycle, second-cycle coding), Pattern and axial coding, Cohen's kappa inter-rater reliability, Chi-square tests of independence, R (Version 4.2.2)
Reagents & environment
R, Version 4.2.2
Reproducibility gaps
Survey data and analysis code are shared only by email request to the corresponding author. Free-text student responses are withheld because they contain indirectly identifying content. The design and analysis were not preregistered.
Limitations (6)
  • Data come from one institution in the Southeastern US with modest demographic diversity (largely women, White, and first-year students), so the findings may not generalize to other institutions, regions, or cultural backgrounds. · stated by the authors
  • The sample was too small to test intersections of gender, race/ethnicity, and year in college. · stated by the authors
  • The researcher-generated checklist of supports was not exhaustive, and coding used only one motivation framework (SEVT), so other supports and motivation-related beliefs may be missed. · stated by the authors
  • Data are cross-sectional and retrospective. They cannot show how perceptions change over time, they do not support causal or directional claims, and they are open to recall bias. · stated by the authors
  • Many subgroup chi-square tests were run at α = 0.05 without correcting for multiple comparisons. At least one reported statistic looks internally inconsistent (χ2(1) = 12.01 reported with p = 0.005), so the demographic differences should be treated as tentative. · noted in this summary
  • First-cycle codes have reported reliability (κ 0.77 to 0.91). The second-cycle behavioral themes for each socializer are described without frequencies or inter-rater statistics in the text, so the socializer-specific mechanisms are less firmly supported. · noted in this summary

Read this if you design STEM retention or career-advising programs and want survey plus qualitative evidence on whether family, peers, mentors, or professors most shape undergraduates' career motivation.

Citations
5

Reproducibility ledger

Nobody has reported trying to reproduce this paper yet. Did you replicate an experiment or run the code? Your report saves the next lab a week.

Sign in to file a reproducibility report.

Related papers

Similarity Map
SelectedSame fieldCross-field
THISMcDonough '21Zajic '26Marriott '18Grinnell '25Lordan '24Coté '26
63% semantic matchScientific Communication & Education
E. McDonough, K. S. Sawyer, J. Wilks+1 more·bioRxiv·Nov 29, 2021