CoChat
Sign Up
Cited by 138Higher Education Research StudiesCareer Development and DiversityEducation, Achievement, and Giftedness

The Importance of STEM Sense of Belonging and Academic Hope in Enhancing Persistence for Low-Income, Underrepresented STEM Students

Open access
Hansen, Michele J., Palakal, Mathew J., White, Le’Joy·2024·Journal for STEM Education Research
DOI 10.1007/s41979-023-00096-8
Open this paper in CoChat — free accountOpen PDF

Abstract

The purpose of this longitudinal investigation was to examine the effectiveness of a comprehensive, integrated curricular and co-curricular program designed to build community, provide academic and social support, and promote engagement in academically purposeful activities resulting in more equitable environments for historically underrepresented, low-income science, technology, engineering, and mathematics (STEM) information technology (IT) students. The study also focused on the role that the sense of belonging and academic hope play in enhancing persistence to degree completion. Program participants had significantly higher persistence rates compared to a matched comparison group. Additionally, STEM-specific belonging and academic hope significantly predicted students’ intentions to persist to degree completion in IT. A major finding was that STEM domain–specific belonging was a stronger predictor of persistence than general belonging. Our investigation has implications for the role that cohort-based programs, industry engagement, peer mentoring, proactive advising, undergraduate research opportunities, career preparation, and leveraging need-based financial aid play in ensuring equity in STEM.

AI snapshot

Written from the paper’s full text.

TL;DR

This longitudinal quasi-experimental study found that low-income, historically underrepresented IT students in the comprehensive LiFT program had significantly higher 1-, 2-, and 3-year persistence rates than a matched comparison group. Among LiFT scholars, STEM/IT-specific sense of belonging and academic hope significantly predicted intentions to persist to IT degree completion, with IT-specific belonging outperforming general belonging.

Key findings

  • LiFT participants had higher persistence than the matched comparison group: 1-year 81.5% vs. 54.2%, 2-year 75.3% vs. 41.3%, and 3-year 61.1% vs. 38.9%.
  • In logistic regressions controlling for HS GPA, LiFT participants were 8.76 times more likely to persist to year 1, 14.63 times more likely to persist to year 2, and 5.31 times more likely to persist to year 3.
  • In the LiFT questionnaire sample (N = 72), sense of belonging and academic hope significantly predicted intentions to persist to IT degree completion, R² = .32, F(5, 63) = 5.34, p < .001, even with gender and historically underrepresented status in the model.
  • IT domain-specific belonging significantly predicted intentions to persist (β = .40, p < .01) and academic hope significantly predicted intentions to persist (β = .42, p < .001), while general sense of belonging did not significantly predict intentions to persist.

Methods

  • Longitudinal quasi-experimental design with 170 LiFT scholars and 170 matched comparison students across a large public urban Midwestern 4-year university and a large urban public 2-year community college pipeline, for academic years 2017–2018, 2018–2019, and 2019–2020.
  • Institutional records provided demographics, HS GPA, SAT, financial aid, and 1-, 2-, and 3-year persistence in IT; an anonymous self-report questionnaire administered only to LiFT scholars (N = 72, 44% response rate) measured general belonging, IT-specific belonging, academic hope, and intention to persist.
  • Analyses included chi-square tests, hierarchical logistic regression adjusting for HS GPA, and hierarchical multiple regression predicting intentions to persist from gender, HUR status, general belonging, IT belonging, and academic hope.

Limitations

  • The paper states the sample was relatively small and primarily male, low-income, underrepresented minority IT students at two institutions, limiting generalizability and making it harder to understand women IT students’ experiences.
  • The paper states random assignment was not possible due to ethical and logistical constraints, so causal attributions about LiFT should be cautious; the study could not isolate which LiFT components caused differences, and time constraints limited assessment of long-term workforce outcomes.
  • Evident from the text: the questionnaire was administered only to LiFT scholars and was anonymous, so questionnaire results could not be linked to individual institutional records; financial aid data restrictions prevented 1:1 matching on Pell/FAFSA, and the LiFT group had significantly higher HS GPA than the comparison group, which was adjusted for in regression models.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1007/s41979-023-00096-8

Cites

85 of 86 references verified0 citation errors · 1 not checkable

Statistics

8 tests recomputed · 1 with the reported and recomputed p on opposite sides of .05

Claims the paper makes

15 of 41 cited sources checked · 5 supported · 0 flagged · accuracy not yet measured

  • Premisecitation

    Several institutional conditions, including undergraduate research, are effective in enhancing URM persistence in STEM.

    “Researchers have empirically identified numerous institutional conditions that are effective in enhancing URM persistence in STEM including providing opportunities for experiential learning experiences such as undergraduate research (Carpi et al., 2017;Hurtado et al., 2010)”
    Quote in the paperNot a substantial claim1 of 2 sources checked
    • cites Carpi et al., 2017 source

      The Literature on the Impacts of Undergraduate Research section reports studies linking undergraduate research programs to retention and persistence in math and science courses. But this paper’s own study focuses on self-efficacy and career ambitions at a minority-serving institution; the cited passages do not establish that its program enhanced URM persistence or identify numerous effective institutional conditions. (LLM, conf=0.83)

      Experimental — accuracy not yet measured
    • cites Hurtado et al., 2010 source
  • Premisecitation

    Cohort-based interventions have been effective in helping low-income and underrepresented students buffer against feelings of low sense of belonging, isolation, and alienation in competitive STEM majors.

    “These types of cohort-based interventions have been effective in helping low-income and underrepresented students buffer against feelings of low sense of belonging, isolation, and alienation in competitive STEM majors (Sax et al., 2018).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Sax et al., 2018 source

      Section 3.3.2 (Results; page unavailable) finds that peer and departmental support positively predict belonging among computing students. Section 4.2 (Implications for Practice; page unavailable) suggests living-learning communities, but does not show that cohort-based interventions were effective. The copy also does not establish an effect for low-income students or across competitive STEM majors. (LLM, conf=0.85)

      Experimental — accuracy not yet measured
  • Premisecitation

    When underrepresented STEM students feel supported and receive social support from institutional interventions, they persist despite obstacles and thrive to attain their learning goals.

    “However, research has shown that when underrepresented STEM students feel supported and receive social support from institutional interventions, they persist despite obstacles and thrive to attain their learning goals (Williams et al., 2017).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Williams et al., 2017 source

      The Results (Appraisal Support; page not provided) describe students feeling more confident and less stressed after receiving support through the coaching intervention. The Discussion says whether that support fosters persistence toward academic careers or has lasting effects remains for future research. The paper therefore does not establish that supported students persisted and attained their learning goals, and its study focused on underrepresented biomedical-science PhD students rather than underrepresented STEM students generally. (LLM, conf=0.87)

      Experimental — accuracy not yet measured
  • Premisecitation

    Institutional practices that promote STEM students' interpersonal relationships, perceived competence, personal interest, and science identity contribute to feelings of belonging.

    “More recent research focused specifically on STEM students suggests that institutional practices that promote STEM students' interpersonal relationships, perceived competence, personal interest, and science identity contribute to feelings of belonging (Rainey et al., 2018).”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Rainey et al., 2018 source

      The Discussion (page not provided) associates STEM students’ sense of belonging with interpersonal relationships, perceived competence, personal interest, and science identity. The Implications section recommends departmental structures that encourage relationships, but the paper does not establish that institutional practices promoting all four factors contribute to belonging. (LLM, conf=0.83)

      Experimental — accuracy not yet measured
  • Premisecitation

    STEM students' perceptions of sense of belonging affected their decisions to persist or withdraw from their STEM major.

    “Additionally, in their qualitative investigation, Rainey et al. found that STEM students' perceptions of sense of belonging affected their decisions to persist or withdraw from their STEM major.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Rainey et al., 2018 source
  • Classified as probably not a substantial claim. 21
  • May combine two claims. 18
  • The quote only partly states this claim. 1

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Cognitive motivational constructs such as academic self-efficacy, self-regulation, effort control, and goal setting affect STEM students' academic performance and intentions to pursue and persist in STEM fields.

    “such as academic self-efficacy (Chen & Zimmerman, 2007;Han et al., 2021;Luo et al., 2021;van Aalderen-Smeets et al., 2018), selfregulation (Shell & Soh, 2013;Zollman, 2012), effort control (Cromley et al., 2016), and goal setting (Shaw & Barbuti, 2010) affect STEM students' academic performance as well as their intentions to pursue and persist in STEM fields.”
    Quote in the paperNot a substantial claim2 of 3 sources checked
    • cites Chen & Zimmerman, 2007 source
    • cites Han et al., 2021 source

      The Self-efficacy and outcome expectancy section links student self-efficacy to academic success and career choices, and the Results section links STEM attitudes, which include self-efficacy, to STEM knowledge achievement. The citation is attached to self-efficacy, not the other listed constructs; the paper supports the performance and career-choice parts but does not establish an effect on intentions to persist in STEM. (LLM, conf=0.84)

      Experimental — accuracy not yet measured
    • cites Luo et al., 2021 source

      The section on relationships among self-efficacy, outcome expectations, and career interest reports that self-efficacy in STEM activities significantly predicted students’ STEM career interest. That supports the intention-to-pursue portion of the claim. The study concerns upper-elementary students and does not establish effects on academic performance or persistence in STEM fields; its discussion of persistence refers to learning, not remaining in a STEM field. The other constructs in the sentence are cited to other sources, not Luo et al. (LLM, conf=0.85)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Premisecitation

    A STEM Enrichment summer bridge early intervention program was effective in enhancing retention rates and academic performance among underrepresented Pell-eligible students.

    “Ghazzawi et al. (2021) used a rigorous propensity score matching methodology and found that a STEM Enrichment summer bridge early intervention program was effective in enhancing the retention rates and academic performance among underrepresented Pell-eligible students.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Ghazzawi et al. (2021) source

      The Data Analysis section reports propensity-score matching, and the Competing Risks Analysis finds lower dropout and greater STEM persistence among SEP summer bridge participants. However, the paper does not establish that the program improved academic performance: GPA was used as an analysis variable, not reported as an improved outcome. The passages also do not identify the studied students as Pell-eligible. (LLM, conf=0.86)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

    Premisecitation

    Creating supportive environments at the departmental and peer group level may help facilitate students' sense that they belong in computing and improve underrepresented student success in IT and computing professions.

    “Sax et al. contend that creating supportive environments, at the departmental and peer group level, may help in facilitating students' sense that they belong in the field of computing, improving underrepresented student success in IT and computing professions.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Sax et al., 2018 source

      The Conclusions (section 5) say that supportive departmental and peer-group environments foster students’ sense of belonging in computing. They also say that whether belonging affects decisions to pursue computing majors or careers remains a question for future research, so the claimed improvement in underrepresented students’ professional success is not established by this study. (LLM, conf=0.88)

      Experimental — accuracy not yet measured

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    High-impact educational practices such as learning communities, first-year seminars, and undergraduate research have been advocated to help students feel a sense of belonging and persist, particularly traditionally underserved populations.

    “Providing students with opportunities to engage in effective”
    Quote in the paperNot a substantial claim1 of 3 sources checked
    • cites Finley & McNair, 2013

      The Introduction (page unavailable) names first-year seminars, learning communities, and undergraduate research as high-impact practices and reports evidence suggesting particular benefits for underserved students, including retention gains. The Results (page unavailable) describe peer interaction and support networks as valuable, but the paper does not establish that these practices help students feel a sense of belonging. (LLM, conf=0.82)

      Experimental — accuracy not yet measured
    • cites Kuh, 2008

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

    • cites Kuh et al., 2013

      Couldn’t be matched to a paper in Crossref, so it can’t be checked.

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    The Meyerhoff Scholars Program focused on an integrated cohort approach for STEM to enhance student success.

    “Our results align with previous research on the effectiveness of programs such as the Meyerhoff Scholars Program which focused on an integrated cohort approach for STEM to enhance student success (Maton et al., 2012(Maton et al., , 2016)).”
    Quote in the paperNot a substantial claim2 of 2 sources checked
    • cites Maton et al., 2012 source
    • cites Maton et al., 2016 source

  • Classified as probably not a substantial claim.

    Premisecitation

    Sax et al. (2018) contend that dialog and action around improving student success in IT and computing professions have to move beyond the numbers and focus on improving the overall climate of inclusion.

    “Sax et al. (2018) contend that dialog and action around improving student success in IT and computing professions have to move beyond the numbers (e.g., of students or employees) and focus on improving the overall climate of inclusion.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Sax et al. (2018) source

  • May combine two claims.

    Premisecitation

    Supportive relationships can enhance student success through building both connections with the campus community and fostering hope can also help further increase student achievement and retention.

    “In their investigation of the role of social support in developing hope in college students, Guthrie and Fruiht (2020) found that supportive relationships can enhance student success through building both connections with the campus community and that fostering hope can also help further increase student achievement and retention.”
    Quote in the paperSubstantial claim1 of 1 sources checked
    • cites Guthrie and Fruiht (2020) source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Banchefsky et al.'s (2019) results suggested that there was a stronger relationship between social belonging and intentions to persist in STEM for women than men.

    “Banchefsky et al.'s (2019) results suggested that there was a stronger relationship between social belonging and intentions to persist in STEM for women than men.”
    Quote in the paperNot a substantial claim1 of 1 sources checked
    • cites Banchefsky et al. (2019) source

      The Results (Hypothesis 3) report that the relationship between end-of-semester social belonging and intentions to persist was significantly stronger for women than men. The study examined undergraduate majors in physical science, technology, engineering, and math (pSTEM), so the citing sentence’s broader reference to STEM is not fully supported. (LLM, conf=0.84)

      Experimental — accuracy not yet measured

  • May combine two claims.

    Findingown data

    The 3-year logistic regression model explained 24.2% (Nagelkerke R2) of the variance in persistence and correctly classified 71.0% of cases once HS GPA was taken into account in step 2.

    “The model explained 24.2% (Nagelkerke R 2 ) of the variance in persistence and correctly classified 71.0% of cases once HS GPA was taken into account in step 2.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    Compared to nonparticipants, LiFT participants were more likely to be underrepresented, first-generation, and low-income.

    “Compared to nonparticipants, participants were more likely to be underrepresented, first-generation (neither parent attended college), and low-income (received a Federal Pell Grant).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    The 2-year logistic regression model explained 40.9% (Nagelkerke R2) of the variance in persistence and correctly classified 79.7% of cases once HS GPA was taken into account in step 2.

    “The model explained 40.9% (Nagelkerke R 2 ) of the variance in persistence and correctly classified 79.7% of cases once HS GPA was taken into account in step 2.”
    Quote in the paperNot a substantial claim

  • The quote only partly states this claim.

    Findingown data

    In the questionnaire study, general belonging did not significantly predict intentions to persist to IT degree completion (β=-.20, ns) when IT belonging and academic hope were included.

    “General belonging | -.16 | .13 | -.20”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Findingown data

    Gender was negatively related to IT belonging and general belonging (not significantly).

    “Gender was negatively related to IT belonging and general belonging (not significantly).”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    Although general sense of belonging was not statistically significant, it was negatively related to intentions to persist in IT in the model once gender and HUR were taken into account.

    “Although general sense of belonging was not statistically significant, it was negatively related to intentions to persist in IT in the model once gender and HUR were taken into account.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Findingown data

    A domain-specific sense of belonging to IT significantly predicted intentions to persist to IT degree completion, while general sense of belonging did not significantly predict intentions to persist in IT.

    “Results also suggested that a domain-specific sense of belonging to IT significantly predicted intentions to persist to IT degree completion, while general sense of belonging did not significantly predict intentions to persist in IT.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Findingown data

    The 1-year logistic regression model explained 26.8% (Nagelkerke R2) of the variance in the persistence rate and correctly classified 69.4% of cases once HS GPA was taken into account in step 2.

    “The model explained 26.8% (Nagelkerke R 2 ) of the variance in the persistence rate and correctly classified 69.4% of cases once HS GPA was taken into account in step 2.”
    Quote in the paperNot a substantial claim

  • May combine two claims.

    Premisecitation

    Sense of belonging is positively associated with academic performance, retention, and persistence, and it is particularly important for college students from marginalized groups.

    “In his extensive review of sense of belonging research, Strayhorn (2011) found that sense of belonging is positively associated with academic performance, retention, and persistence, and it is particularly important for college students from marginalized groups.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Strayhorn (2011) source

  • Classified as probably not a substantial claim.

    Study design and conductnone given

    The questionnaire was only administered to LiFT scholars and not the comparison group students.

    “The questionnaire was only administered to the LiFT scholars and not the comparison group students.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S2 (checkable).

    Interpretationown data

    Academic hope and STEM major-specific sense of belonging are important theoretical frameworks to guide future research and practice in STEM student persistence.

    “what makes this study relevant is that it underscores that academic hope and STEM major-specific sense of belonging are important theoretical frameworks to guide future research and practice in the area of STEM student persistence.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Intentions to persist to IT degree completion are a valid proxy for actual persistence to degree completion.

  • Classified as probably not a substantial claim. May combine two claims.

    Study design and conductnone given

    The self-administered questionnaire was only administered to LiFT scholars and not the comparison group, resulting in a sample of 72 and a response rate of 44%.

    “The self-administered questionnaire was only administered to LiFT scholars and not the comparison group, resulting in a sample of 72 and a response rate of 44%.”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Premisecitation

    Domain- or major-specific STEM belonging may be more predictive of positive academic outcomes compared to general feelings of belonging.

    “They also concluded that domain-or major-specific STEM belonging may be more predictive of positive academic outcomes compared to general feelings of belonging.”
    Quote in the paperSubstantial claim0 of 1 sources checked
    • cites Lewis et al. (2016) source

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Study design and conductnone given

    Time constraints limited the authors' ability to assess the long-term impacts of the interventions.

    “Additionally, time constraints limited our ability to assess the long-term impacts of the interventions (e.g., IT employment and workforce outcomes).”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Xu and Lastrapes (2022) found that Latinx students' STEM sense of belonging had impacts on their sustained career interest in STEM.

    “For example, Xu and Lastrapes (2022) found that Latinx students' STEM sense of belonging had impacts on their sustained career interest in STEM.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Xu and Lastrapes (2022) source

  • May combine two claims.

    Study design and conductreasoning

    LiFT students participated in all the interventions and were not randomly assigned to different programs or interventions.

    “In other words, LiFT students participated in all the interventions and were not randomly assigned to different programs or interventions (summer bridge, residential learning communities, leadership and career development, industry engagement, peer mentoring, proactive advising, undergraduate research opportunities, and need-based financial aid support).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    The study used a process of stratifying observations into groups based on fixed values of potential confounding variables to attain a reasonable comparison group.

    “As such, we used a process of stratifying the observations into groups based on the fixed values of the potential confounding variables to attain a reasonable comparison group.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductreasoning

    The authors were not able to employ an experimental design and need to be cautious about making causal attributions about the LiFT program and differences in persistence rates.

    “As such, one of the most serious limitations of this study is that we were not able to employ an experimental design and need to be cautious about making causal attributions about the LiFT program and differences in persistence rates.”
    Quote in the paperSubstantial claim

  • Probably not a substantial claim: may not pass S3 (load-bearing).

    Premisecitation

    Alexander et al. found that hopeful future expectations positively predicted STEM career interest among high-and middle-school girls.

    “Although academic hope has been understudied in STEM college students, Alexander et al. found that hopeful future expectations positively predicted STEM career interest among high-and middle-school girls.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Alexander et al., 2022 source

  • Probably not a substantial claim: may not pass S3 (load-bearing). May combine two claims.

    Study design and conductnone given

    The sample of LiFT survey respondents was representative of the overall LiFT population in terms of gender distribution and ethnicity distribution but was slightly underrepresented in terms of African American students.

    “The sample of LiFT survey respondents was representative of the overall LiFT population in terms of gender distribution and ethnicity distribution but was slightly underrepresented in terms of African American students.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable), S3 (load-bearing).

    Premisecitation

    Cessna et al. (2018) contend that higher education leaders and scholars should conduct more research on what types of interventions work to prepare marginalized STEM students to do well in rigorous college coursework, instill confidence and agency to succeed, and facilitate feelings of belonging with the institution and specific major.

    “Cessna et al. (2018) contend that higher education leaders and scholars should conduct more research on what types of interventions work to prepare marginalized STEM students to do well in rigorous college coursework, instill confidence and agency to succeed, and facilitate feelings of belonging with the institution and specific major.”
    Quote in the paperNot a substantial claim0 of 1 sources checked
    • cites Cessna et al. (2018) source

  • May combine two claims.

    Study design and conductown data

    Students who began at the 4-year university had higher HS GPAs and 1-year retention rates than students who began at the community college (73% vs. 65% and 3.44 vs. 2.33, respectively).

    “The students who began at the 4-year university had higher HS GPAs and 1-year retention rates than students who began at the community college (73% vs. 65% and 3.44 vs. 2.33, respectively).”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    The authors employed logistic regression procedures to adjust for HS GPA as a potential confounding variable in their regression models.

    “As such, we employed logistic regression procedures to adjust for HS GPA as a potential confounding variable in our regression models.”
    Quote in the paperSubstantial claim

  • May combine two claims.

    Study design and conductown data

    The tolerance measures were 0.97, 0.99, and 0.95, suggesting that the beta coefficients were not affected by the presence of other predictor variables and that multicollinearity was not problematic.

    “The tolerance measures were as follows: 0.97, 0.99, and 0.95.This suggests that the beta coefficients were not affected by the presence of other predictor variables in a model and that multicollinearity was not problematic.”
    Quote in the paperNot a substantial claim

  • Classified as probably not a substantial claim.

    Study design and conductreasoning

    The authors were not able to link the students' questionnaire results to individual students' institutional data or official educational records.

    “As such, we were not able to link the students' questionnaire results to individual students' institutional data or official educational records (e.g., educational records of HS GPA, financial aid information, retention rates, persistence rates).”
    Quote in the paperSubstantial claim

  • Classified as probably not a substantial claim.

    Interpretationreasoning

    If HUR low-income STEM students are not provided with need-based financial aid, they may not have the basic necessities (food, clothing, housing) necessary to exert the intellectual effort necessary to meet rigorous course demands.

    “If HUR low-income STEM students are not provided with need-based financial aid, they may not have the basic necessities (food, clothing, housing) necessary to exert the intellectual effort necessary to meet rigorous course demands.”
    Quote in the paperSubstantial claim
    • Rests on an assumption: Need-based financial aid is a key component of the LiFT program's effectiveness.

  • May combine two claims.

    Study design and conductnone given

    The sample was relatively small and primarily consisted of male, low-income, underrepresented minority IT students at two institutions.

    “The sample was relatively small and primarily consisted of male, low-income, underrepresented minority IT students at two institutions.”
    Quote in the paperNot a substantial claim

  • Probably not a substantial claim: may not pass S2 (checkable). May combine two claims.

    Interpretationreasoning

    The results point to the necessity of providing financial support coupled with curricular and co-curricular programming.

    “As such, our results point to the necessity of providing financial support coupled with curricular and co-curricular programming.”
    Quote in the paperNot a substantial claim
    • Rests on an assumption: Need-based financial aid is a key component of the LiFT program's effectiveness.
36 of 36

  • The paper relies on this without support.

    The matched comparison group is equivalent to LiFT participants on unmeasured confounding variables.

    “The demographic, school, major, and academic preparation variables were used to find a matched comparison group of 170 students entering during the same cohort years in the same IT department and major who were closely matched on demographics, academic preparation, and financial aid information.”

  • The authors state this as an assumption.

    The institutional context and financial support create safe spaces and provide the resources students need to persist to degree completion.

    “The institutional context coupled with financial support was designed to serve as an effective mechanism to create safe spaces for IT students to develop an IT sense of belonging and hope.”

  • The paper relies on this without support.

    All LiFT students experienced all components of the program.

    “The program was designed so that all LiFT students experienced all components of the program (e.g., undergraduate research and career development planning).”

  • The paper relies on this without support.

    The LiFT program's comprehensive support caused the observed differences in persistence rates, rather than unmeasured confounding or selection effects.

    “As such, we employed logistic regression procedures to adjust for HS GPA as a potential confounding variable in our regression models.”

  • The paper relies on this without support.

    Need-based financial aid is a key component of the LiFT program's effectiveness.

    “The results may have implications in terms of the importance of considering need-based financial aid since this was a key component of the overall comprehensive LiFT program even though we were not able to isolate this component in our analyses.”

  • The paper relies on this without support.

    The comparison group is comparable to LiFT participants except for program participation and HS GPA.

    “We were able to employ a quasi-experimental design and develop a matched control group of IT, underrepresented, low-income students not participating in the LiFT interventions to investigate the effects on 1-, 2-, and 3-year retention to STEM IT rates.”

  • The paper relies on this without support.

    The matched comparison group is equivalent to LiFT participants on unmeasured confounders, so differences in persistence can be attributed to LiFT participation.

    “Persistence rates for the LiFT participants and the matched comparison group”

  • The paper relies on this without support.

    Intentions to persist to IT degree completion are a valid proxy for actual persistence to degree completion.

    “intentions to persist to IT degree completion”
8 of 8

Statistics that don’t add up

Each reported p-value recomputed from its test statistic and degrees of freedom.

  • Inconsistent p-value across alpha = .05: X 2 (1, N = 73) = 1.65, p < .05

    Major

    The paper reports p < .05 here; recomputing from the test statistic and its degrees of freedom gives p = .199. The two fall on opposite sides of alpha = .05. The check cannot tell whether the statistic or the p-value is the misprint.

    Reported p < .05 · recomputed p = .199

  • Inconsistent p-value: χ 2 (2) = 5.89, p < .05

    Minor

    The paper reports p < .05 here; recomputing from the test statistic and its degrees of freedom gives p = .053. The check cannot tell whether the statistic or the p-value is the misprint.

    Reported p < .05 · recomputed p = .053

  • p-value fits an unstated one-tailed test: t(340) = 1.81, p < .05

    Minor

    The paper reports p < .05 here; recomputing from the test statistic and its degrees of freedom gives p = .071 (two-tailed). The reported value is what a one-tailed test gives, but the paper does not say near this result that a one-tailed test was used.

    Reported p < .05 · recomputed p = .071

3 of 3

Statistics recomputed

Every reported test result, recomputed from its test statistic and degrees of freedom. A result shows as not adding up only once an independent check has confirmed it.

  • Doesn’t add upχ²(1) = 1.65reported p < .05 · recomputed p = .199The two fall on opposite sides of .05

    “X 2 (1, N = 73) = 1.65, p < .05”

  • Doesn’t add upt(340) = 1.81reported p < .05 · recomputed p = .071

    “t(340) = 1.81, p < .05”

  • Doesn’t add upχ²(2) = 5.89reported p < .05 · recomputed p = .053

    “χ 2 (2) = 5.89, p < .05”

  • Consistentχ²(1) = 22.7reported p < .001 · recomputed p < .001

    “X 2 (1, N = 340) = 22.7, p < .001”

  • Consistentχ²(1) = 13.14reported p < .001 · recomputed p < .001

    “X 2 (1, N = 161) = 13.14, p < .001”

  • Consistentχ²(2) = 22.2reported p < .001 · recomputed p < .001

    “χ 2 (2) = 22.20, p < .001”

  • Consistentχ²(2) = 21.73reported p < .001 · recomputed p < .001

    “χ 2 (2) = 21.73,p < .001”

  • ConsistentF(5, 63) = 5.34reported p < .001 · recomputed p < .001

    “F(5, 63) = 5.34, p < .001”

8 of 8

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

Each contradiction between two passages listed here was confirmed by an independent check.

  • Claims about identifying causal program components contradict each other

    Minor

    The limitations section says the design determined which aspects of LiFT caused differences, but the paper elsewhere explicitly says it could not isolate the contributions of individual components.

  • Reported tolerances conflict with the correlation between belonging measures

    Minor

    A correlation of .75 between general and IT belonging alone limits either variable's regression tolerance to at most 1 − .752 = .4375. Their reported tolerances of 0.97, 0.99, and 0.95 therefore cannot all describe the stated model.

2 of 2

Cites

Flagged first, then by citations

Research works printed but not linked: 3

  1. Kuh, G. D. (2008). High-impact practices: What they are, who has access to them, and why they matter. Association of American Colleges and Universities.
  2. Kuh, G. D., & O’Donnell, K. (2013). Ensuring quality and taking high-impact practices to scale. Association of American Colleges and Universities.
  3. Maples, S. C. (2003). Academic achievement and retention rate of students who did and did not participate in a university summer bridge program. Dissertation Abstracts International Section A: Humanities & Social Sciences, 63(7-A), 2468.

Cited by

Sorted by citations

Related

Authors
Hansen, M. J., Palakal, M. J., & White, L.
Affiliations
  • The Ohio State University (US)
  • Indiana University – Purdue University Indianapolis (US)
Corresponding author
Hansen, Michele J.
ORCID
Venue
Journal for STEM Education Research
Publisher
Springer Science and Business Media LLC
Published
2023-05-23
Volume
7
Issue
2
Pages
155-180
ISSN
  • 2520-8705 (ISSN-L)
  • 2520-8713
DOI
10.1007/s41979-023-00096-8
OpenAlex
W4377692532
Semantic Scholar
ad6dcdbecf6ff7e0dfd455dc39fd2ca558006bb1
Cited by
138
References
81
Citations by year
2026: 57 · 2025: 43 · 2024: 35 · 2023: 3
FWCI
102.27
Type
journal article
Language
en
Open access
Licence
Locations
Journal for STEM Education Research (publishedVersion, cc-by)
Funders
National Science Foundation (1643586.)
Topics
  • Social Sciences › Education › Higher Education Research Studies
  • Social Sciences › Safety Research › Career Development and Diversity
  • Social Sciences › Psychology › Experimental and Cognitive Psychology › Education, Achievement, and Giftedness
Keywords
sense of belonging, academic hope, student persistence, low-income students, underrepresented students, STEM education, information technology students, STEM equity, peer mentoring, undergraduate research, industry engagement, need-based aid, career preparation

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

Claims read by deepseek-v4.1-flash under definition version 3 (extract-v3) on 2026-11-03.

Classified by jev-1.13-20260917 (gold-guideline-v0.2) on 2026-11-03.

Cited sources are checked by gpt-6-sol.

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review