View in EDS HTML Full Text PDF Full Text

Seeing Women Who Fit: Girls' Forecasted Fit in STEM Fosters Career Interest

Saved in:
Bibliographic Details
Title: Seeing Women Who Fit: Girls' Forecasted Fit in STEM Fosters Career Interest
Language: English
Authors: Emily N. Cyr (ORCID 0000-0002-1640-1840), Steven J. Spencer, Stephen C. Wright, Jennifer R. Steele, Kathryn M. Kroeper, Patricia Colaco, Tara C. Dennehy, Priscilla Lok-Chee Shum, Taylor Ballinger, Haemi Nam, Stephanie L. Reeves, Mary Wells, Toni Schmader, Hilary B. Bergsieker
Source: Social Psychology of Education: An International Journal. 2025 28(1).
Availability: Springer. Available from: Springer Nature. One New York Plaza, Suite 4600, New York, NY 10004. Tel: 800-777-4643; Tel: 212-460-1500; Fax: 212-460-1700; e-mail: customerservice@springernature.com; Web site: https://link.springer.com/
Peer Reviewed: Y
Page Count: 18
Publication Date: 2025
Document Type: Journal Articles
Reports - Research
Education Level: Elementary Education
Descriptors: Females, STEM Education, STEM Careers, Vocational Interests, Role Models, Science Interests, Predictor Variables, Elementary School Students
DOI: 10.1007/s11218-025-10056-2
ISSN: 1381-2890
1573-1928
Abstract: Girls often express less interest in STEM (science, technology, engineering, mathematics) education and careers than boys, despite having comparable aptitude. We randomly assigned 242 girls (Mdn[subscript age] = 12 years; 38% East Asian; 37% White) at Canadian STEM camps to control conversations about generic camp experiences, or intervention conversations where STEM role models discussed how STEM education and careers align with each girl's most important value and emphasized the social community inside and outside of STEM. Key measures were collected at baseline, and several days after the intervention (or control). Girls' "current" STEM fit did not differ by condition, but as expected, the intervention (vs. control) significantly improved girls' forecasts regarding "future" STEM fit (ds = 0.27-0.35) and girls' interest in STEM careers (d = 0.42), with a marginally significant boost in girls' interest in STEM high school classes (d = 0.23). Pre-post increases in forecasted STEM fit mediated increases in STEM interest. Forecasted fit (beyond current fit) appears pivotal for promoting girls' sustained interest in STEM.
Abstractor: As Provided
Entry Date: 2025
Accession Number: EJ1472361
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwGTGdkfscUX5aiJxwDCGGegAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDOwGUbGRsh9GNuXeywIBEICBm53j0Zf1FVQ6LHX03dtPYcazstv9MNjrUw1u_ZGORJibJnY9GyPj5DsKJyxbrh8-tYq1SqeJOhVDNRdUOq538u31-HrwSjawpbFfWLTi5npftXZWYGtR5FM_03HD8URA-ScI89WYsVBBn-wCCZJgWIT9O4xMPR8ePaBIeOnV3usLBxMfvxF0QBajWsNtb4qRTzHnnAQIcEK3-kn5
Text:
  Availability: 1
  Value: <anid>AN0185469581;luo28may.25;2025May30.03:31;v2.2.500</anid> <title id="AN0185469581-1">Seeing women who fit: Girls' forecasted fit in STEM fosters career interest </title> <p>Girls often express less interest in STEM (science, technology, engineering, mathematics) education and careers than boys, despite having comparable aptitude. We randomly assigned 242 girls (Mdn<sub>age</sub> = 12 years; 38% East Asian; 37% White) at Canadian STEM camps to control conversations about generic camp experiences, or intervention conversations where STEM role models discussed how STEM education and careers align with each girl's most important value and emphasized the social community inside and outside of STEM. Key measures were collected at baseline, and several days after the intervention (or control). Girls' current STEM fit did not differ by condition, but as expected, the intervention (vs. control) significantly improved girls' forecasts regarding future STEM fit (ds = 0.27–0.35) and girls' interest in STEM careers (d = 0.42), with a marginally significant boost in girls' interest in STEM high school classes (d = 0.23). Pre-post increases in forecasted STEM fit mediated increases in STEM interest. Forecasted fit (beyond current fit) appears pivotal for promoting girls' sustained interest in STEM.</p> <p>Keywords: Field intervention; Gender; Identity fit; STEM; Role models</p> <p>Note. Affiliation pertains to where that author's work was conducted, except in the case of the first author (who was affiliated with University of Waterloo during the majority of this research program, and is now affiliated with York University).</p> <hd id="AN0185469581-2">Introduction</hd> <p>"As a high schooler, I don't feel that I've had many barriers to success yet. I can see them coming, though."- Sloane Sell, high school girl interested in STEM (social media post).</p> <p>Girls are good at STEM (science, technology, engineering, and mathematics): In childhood, girls and boys achieve similar scores on early standardized tests of mathematical, scientific, and statistical ability (Hyde & Linn, [<reflink idref="bib19" id="ref1">19</reflink>]; Kersey et al., [<reflink idref="bib20" id="ref2">20</reflink>]). Yet girls report lower interest in advanced STEM education and careers than their male peers (Master et al., [<reflink idref="bib26" id="ref3">26</reflink>]; Sadler et al., [<reflink idref="bib32" id="ref4">32</reflink>]). Why are girls' current abilities and potential decoupled from their long-term interest in STEM? Building on recent theorizing (Schmader & Sedikides, [<reflink idref="bib34" id="ref5">34</reflink>]), we propose that anticipating a lack of fit in future adult STEM environments can undermine girls' interest in studying and building careers in STEM. In the current research, we developed an intervention to help girls envision themselves pursuing value-congruent, socially supported STEM activities into adulthood, with the aim of increasing their forecasted future fit and long-term interest in STEM education and careers.</p> <hd id="AN0185469581-3">Theoretical framework: Changing perceptions of future STEM fit</hd> <p>Schmader and Sedikides ([<reflink idref="bib34" id="ref6">34</reflink>]) propose that when people do not feel they can embody their authentic selves within a specific environment, they leave. This sense of "fit" is a three-fold construct encompassing a range of potential "matches" (or "mismatches") between the self and the environment. <emph>Self-concept fit</emph> is the degree to which the environment activates and validates aspects of one's default (or "true") self (e.g., chronically accessible traits, preferences, or memories). <emph>Goal fit</emph> is how much the environment affords enacting one's goals or values. <emph>Social fit</emph> emerges when the environment includes people who respect and reaffirm one's contributions. Although related to other social psychological constructs (e.g., social fit shares similarities with belonging), fit more comprehensively addresses multiple potential points of (dis)connection. Importantly, fit is implicated in key education and career decisions, especially for members of marginalized groups, including girls and women in STEM (Sax et al., [<reflink idref="bib33" id="ref7">33</reflink>]; Walton & Cohen, [<reflink idref="bib37" id="ref8">37</reflink>]). Girls' awareness of gender-STEM stereotypes predicts girls' lower personal belonging in computer science and engineering (Master et al., [<reflink idref="bib26" id="ref9">26</reflink>]), and girls are already wary of gendered barriers that emerge in adult STEM spaces (Steele, [<reflink idref="bib35" id="ref10">35</reflink>]). By adulthood, women report lower fit in STEM fields relative to men, and women with low STEM fit are less likely to enter or remain in STEM careers (Block et al., [<reflink idref="bib1" id="ref11">1</reflink>], [<reflink idref="bib2" id="ref12">2</reflink>]; Cheryan et al., [<reflink idref="bib5" id="ref13">5</reflink>]).</p> <p>Less is known, however, about how appraisals of <emph>current</emph> fit may align with (or diverge from) <emph>forecasts</emph> of future fit. For example, past research on belonging draws conceptual distinctions between experienced and anticipated belonging, but ultimately treats them as essentially interchangeable (Murphy & Zirkel, [<reflink idref="bib29" id="ref14">29</reflink>]). We additionally posit that, even in childhood, current fit and forecasted fit tend to correlate highly but hold distinct implications for academic and occupational decision making. Like Murphy and Zirkel ([<reflink idref="bib29" id="ref15">29</reflink>]), we theorize that students "implicitly or explicitly consult their anticipated sense of belonging [i.e., future fit] when making future-oriented academic decisions" (p. 6), but we further theorize that forecasted fit exerts a more powerful influence than current fit on decision making about long-term options (e.g., future STEM careers). Specifically, we propose that a child's likelihood of pursuing STEM education and career pathways (e.g., by taking pre-requisite courses) may hinge on <emph>forecasting</emph> a sense of future fit in STEM environments as an adult. Here, we probe children's interest in high school STEM classes and careers as a proxy for their likelihood of pursuing later STEM education and career pathways.</p> <hd id="AN0185469581-4">The present study: A multi-part intervention for girls in STEM</hd> <p>We focus our attention on girls entering grades 5 to 9, a critical educational period before children select key prerequisite classes necessary for STEM higher education and careers (Tyson, [<reflink idref="bib36" id="ref16">36</reflink>]; Zheng & Weeden, [<reflink idref="bib39" id="ref17">39</reflink>]). We designed a multi-part intervention to help girls imagine themselves fitting in <emph>future</emph> STEM environments. Each intervention component was tailored toward one of the three subtypes of fit—self-concept fit, goal fit, and social fit—theorized by Schmader and Sedikides ([<reflink idref="bib34" id="ref18">34</reflink>]). The intervention was contrasted against an "empty" control that was similar in structure yet devoid of key content.</p> <p>The intervention (and control) conditions were led by a female role model in STEM, who conducted a one-on-one semi-structured in-person conversation. A growing number of studies have provided evidence that female role models can have positive effects on the attitudes and outcomes of young adult female STEM students (Dasgupta et al., [<reflink idref="bib9" id="ref19">9</reflink>]; Dennehy & Dasgupta, [<reflink idref="bib10" id="ref20">10</reflink>]; Wu et al., [<reflink idref="bib38" id="ref21">38</reflink>]). As with prior studies, we leverage an undergraduate female role model—but here focus on girls' forecasted future fit in STEM. Additionally, as part of the intervention, girls watched a video of 14 undergraduates (i.e., students further along on STEM trajectories) reflecting on their educational journeys and anticipated careers in STEM.</p> <p>Self-concept fit was encouraged by both the in-person role model and the video, via providing girls with multiple role models representing a diversity of gender, racial, and cultural groups—increasing the likelihood that each girl would be able to focus on a self-relevant role model (O'Brien et al., [<reflink idref="bib30" id="ref22">30</reflink>]) and see a point of connection (Gladstone & Cimpian, [<reflink idref="bib14" id="ref23">14</reflink>]; Marx & Ko, [<reflink idref="bib25" id="ref24">25</reflink>]). Further, the video highlighted a variety of role models (both stereotypic and non-stereotypic; Cheryan et al., [<reflink idref="bib6" id="ref25">6</reflink>]), potentially broadening conceptualizations of which identities can fit in STEM. Critically, all role models (in-person and videotaped) were young adult university students, as we anticipated this "interim" career stage would make the role models' STEM successes seem attainable (Lockwood & Kunda, [<reflink idref="bib23" id="ref26">23</reflink>]) while still maintaining a future focus. We also extend typical paradigms (often featuring novel role models in artificial settings; see Wu et al., [<reflink idref="bib38" id="ref27">38</reflink>]) by employing an in-person role model with whom participants have already built an authentic relationship (i.e., a camp counselor they interacted with each day).</p> <p>Goal fit was reinforced using a value-affordances induction (Fuesting et al., [<reflink idref="bib12" id="ref28">12</reflink>]). Prior to the intervention, participants ranked several personal values, then later had a conversation with the in-person role model that highlighted how STEM pursuits would support that girls' top-ranked value. We tailored the conversation to each participant's top-ranked value because girls often express communal and prosocial values (Block et al., [<reflink idref="bib1" id="ref29">1</reflink>], [<reflink idref="bib2" id="ref30">2</reflink>]) and may see STEM fields as incongruous with these communal goals (Boucher et al., [<reflink idref="bib3" id="ref31">3</reflink>]; Diekman et al., [<reflink idref="bib11" id="ref32">11</reflink>]). Further, past research suggests that girls may find STEM classes more meaningful when educational content connects to personal or communally oriented goals (e.g., Gentry & Owen, [<reflink idref="bib13" id="ref33">13</reflink>]; Harackiewicz et al., [<reflink idref="bib16" id="ref34">16</reflink>]). The videotaped role models further highlighted how they were pursuing their core values and goals, many of which were also communally oriented, through STEM (e.g., biomedical engineering research into a relative's ailment).</p> <p>The intervention was also designed to strengthen social fit through descriptions of a strong social community. Specifically, the in-person role model highlighted friendships between young women seen in the video (all of whom were in STEM), then discussed her personal sense of community with peers inside and outside of STEM, despite initial misgivings. This was designed to address any emerging concerns about "double isolation" (Cheryan et al., [<reflink idref="bib4" id="ref35">4</reflink>]): Being socially rejected by both STEM and non-STEM peers. Female friendships were highlighted because female friends in STEM have been found to predict girls' STEM-related decisions (Riegle-Crumb et al., [<reflink idref="bib31" id="ref36">31</reflink>]).</p> <p>We tested the effectiveness of this novel intervention among girls in STEM summer camps geared toward youth entering grades 5 to 9. We hypothesized that the intervention would be associated with increased forecasted fit and interest in STEM, and that increased interest in STEM education and careers would be mediated by increases in forecasted STEM fit. We contrast the effects of forecasted STEM fit against those for current STEM fit, expecting that our intervention would be associated with stronger effects for forecasts. We further test for specificity to STEM (vs. non-STEM) outcomes, hypothesizing that intervention effects will be strongest for STEM forecasted fit and interest.</p> <hd id="AN0185469581-5">Method</hd> <p></p> <hd id="AN0185469581-6">Participants and procedure</hd> <p>Given the practical constraints of a large-scale field intervention, we recruited as many participants as possible, by partnering with three STEM-focused camps in 2019. These 5-day camps were designed for youth entering grades 5 to 9, and took place during typical school hours from Monday-Friday over the summer break. Camps were broadly advertised to the surrounding communities and included financial needs-based fee structures. The three camp sites (hosted by three major Canadian universities) offered various educational topics (15 total; e.g., engineering, chemistry), with topics typically split into successive classes. In total, we recruited girls from 52 distinct classes (across all available topics).</p> <p>This study focused on girls was run simultaneously with a separate study focused on boys (see Cyr, [<reflink idref="bib8" id="ref37">8</reflink>]). Study assignment was based on participants' self-reported gender (not sex assigned at birth) as our hypotheses relate to gendered socialization. All associated terms (e.g., "female" as an adjective, "girl" as a noun) refer to gender. Non-binary youth were assigned to the control condition but not analyzed given their small sample size.</p> <hd id="AN0185469581-7">Participants</hd> <p>In the current study, a total of 258 girls received parental consent, agreed to participate, and completed either the intervention or control conversation. Analyses excluded girls who did not complete the Exit Survey (<emph>n</emph> = 11), or had serious protocol errors (<emph>n</emph> = 5).</p> <p>The final sample of 242 girls had a median age of 12 (72% from 11–13; full range: 8–15). Our relatively diverse sample broadly reflected the demographic composition of the surrounding communities: 38% identified as East Asian, 37% White, 14% multi-racial, and 6% South Asian (< 3% in each remaining category).</p> <p>See the Supplement (section "Moderation by Demographics") for lack of moderation by participant race, and a singular instance (out of five potential instances) of significant moderation by participant age. Additionally, we did not see significant variation by site (see Table S3), topic, or class (see Supplement section "Multi-Level Modelling").</p> <hd id="AN0185469581-8">Procedure</hd> <p>Our pre-post design took place over 5 days (Monday to Friday).</p> <p> <bold>Monday.</bold> On Monday, participants completed the Intake Survey and were randomly assigned to either the intervention (<emph>n</emph> = 124) or control (<emph>n</emph> = 118) condition. As part of the Intake Survey, girls completed a value-affordances pre-test, on which they ranked a list of eight values (e.g., family, friends, education) from least to most personally important. Most girls (82%) selected "family" as their top-ranked value; 5% chose "friends" and 3% chose "education" (each other value was top-ranked by < 3%). Top value ("family" vs. all others) did not moderate any intervention condition effects (<emph>p</emph>s >.14).</p> <p> <bold>Midweek</bold> <bold> <emph>.</emph> </bold> The semi-structured intervention and control conversations were conducted by one of ten role models across the three camp sites (i.e., role models were fully nested within sites), all of whom were trained camp staff members and university students in STEM. When possible, conversations were conducted by one of the six female role models (99% for intervention; 97% for control), with the remaining handful of conversations conducted by male role models. All role models conducted both intervention and control conversations (randomly assigned). About 90% of intervention conversations lasted 12–20 min, and about 90% of control conversations lasted 5–15 min.</p> <p>Conversations took place midweek, typically on Tuesday (68%) with the rest on Wednesday (21%) or on Monday sometime after the Intake Survey (11%). Girls completed the Midweek Survey immediately after their intervention or control conversation. We additionally collected several proxies of conversation effectiveness, as rated by both role models and participants (see Supplement section "Moderation by Conversation Characteristics").</p> <p> <bold>Friday</bold> <bold> <emph>.</emph> </bold> The Exit Survey was completed on Friday. Each participant received a $5 gift card and raffle entry to win a science museum pass.</p> <hd id="AN0185469581-9">Experimental design</hd> <p>Regardless of condition, conversations between the participant and the in-person role model always began with introductions (e.g., their major and interest in STEM) then icebreakers (e.g., "If you could plan a perfect day, what would it be?").</p> <hd id="AN0185469581-10">Control condition</hd> <p>Following introductions and icebreakers, the control condition conversation ended with the participant telling the in-person role model about their favorite camp experiences, as the control conversation was designed to mimic a typical "empty" interaction.</p> <hd id="AN0185469581-11">Intervention condition</hd> <p>Following icebreakers in the intervention condition, the in-person role model used the participant's top-ranked value (from the Intake Survey) to tell a genuine anecdote demonstrating how they pursued that specific value by going into STEM (e.g., "When my sister was in grade 3 and I was in grade 2, she was always excited to teach me something she'd learned in school that day. When she learned about currency, she and I sat in our basement and she taught me what each coin was worth and how to use them to pay for things. I remember that her excitement about math was so infectious that I became excited too..."). Immediately after, the role model asked the participant to reflect and comment on how this idea might apply to them (e.g., asking them if they have done any STEM projects with a family member). The in-person role model then showed participants a 5-minute video of university STEM students (of diverse genders, ethnicities, and interests), each describing their STEM projects, motivations to pursue them, and their future STEM career (see video at osf.io/h7v5d/). The video ended with a call to action, asking participants to contemplate what they could personally achieve through STEM. After the video, the in-person role model recounted another genuine anecdote about feeling simultaneously included in social communities inside and outside of STEM, despite initial concerns.</p> <hd id="AN0185469581-12">Measures</hd> <p>The value ranking pre-test and measures of age, gender, and racial/ethnic background were collected only in the Intake Survey. Measures of current fit, forecasted fit, interest in high school classes and interest in careers were collected in both the Intake and Exit Surveys. The Midweek Survey included only forecasted fit. See Table S1 for full measure details.</p> <hd id="AN0185469581-13">Value ranking pre-test</hd> <p>Participants ranked a list of eight values from "the least important to you to the most important to you."</p> <hd id="AN0185469581-14">Fit</hd> <p>All fit measures were rated from 1 (<emph>not at all</emph>) to 5 (<emph>extremely</emph>). STEM and non-STEM items were randomly interspersed within blocks.</p> <p> <bold>Current STEM Fit</bold>. Feelings of fit in current STEM environments were assessed using four items (e.g., "How much do you feel you [belong (fit in) / can "be yourself"]... in [Math / Science] classes?"; αs > 0.79). The Exit Survey added the preamble, "Think about the classes you will take this coming year" and slightly adjusted grammar.</p> <p> <bold>Current Non-STEM Fit</bold>. Two items (<emph>r</emph>s > 0.52, <emph>p</emph>s <.001) probed current fit in English classes (e.g., "How much do you feel you [belong (fit in) / can "be yourself" in]... in English classes?").</p> <p> <bold>Forecasted STEM Fit</bold>. Forecasted feelings of fit in future STEM environments was operationalized as fit in university STEM classes ("Imagine that you are finishing high school and choosing your university classes."), assessed using eight items (e.g., [How much do you feel like you <emph>will</emph> [belong (fit in) / be able to "be yourself"]... in [Computer science / Engineering / Math / Science] classes?"; αs > 0.85).</p> <p> <bold>Forecasted Non-STEM Fit</bold>. Two items (<emph>r</emph>s > 0.66, <emph>p</emph>s <.001) also probed forecasted fit in university English classes (e.g., "How much do you feel like you <emph>will</emph> [belong (fit in) / be able to "be yourself"]... in English classes?").</p> <hd id="AN0185469581-15">Interest</hd> <p>As with fit measures, interest measures were rated from 1 (<emph>not at all</emph>) to 5 (<emph>extremely</emph>), and STEM and non-STEM items were randomly interspersed within blocks.</p> <p> <bold>Interest in High School STEM Classes</bold>. To assess interest in taking high school STEM classes, participants were prompted, "How interested are you in each of these high school classes?" and rated three STEM classes (<emph>Advanced Math</emph>, <emph>Advanced Chemistry</emph>, <emph>Advanced Physics</emph>; αs > 0.72).</p> <p> <bold>Interest in High School Non-STEM Class</bold>. One additional item probed interest in a non-STEM class (<emph>Advanced English</emph>).</p> <p> <bold>Interest in STEM Careers</bold>. Participants were asked, "How interested are you in going into each of these careers?" and rated their interest in 7 STEM-related careers (e.g., <emph>Engineer</emph>, <emph>Scientist</emph>; αs > 0.82).</p> <p> <bold>Interest in Non-STEM Careers</bold>. Finally, participants rated their interest in 12 non-STEM careers (e.g., <emph>Journalist</emph>, <emph>Lawyer</emph>; αs > 0.80).</p> <hd id="AN0185469581-16">Results</hd> <p></p> <hd id="AN0185469581-17">Analytic approach</hd> <p>Minimal evidence of clustered responses within classes emerged (all within-class |ICC|s < 0.1, <emph>p</emph>s >.09; see Supplement section "Multi-Level Modelling" for more details), so analyses were conducted using OLS regression and PROCESS mediational models (Hayes, [<reflink idref="bib17" id="ref38">17</reflink>]) in SPSS 29. Condition was effects coded (control = − 1; intervention = 1). Rather than excluding datapoint outliers, we used the more conservative approach of winsorization: Continuous variables were winsorized to ± 3 <emph>SD</emph>s (affecting < 1% of cases per fit variable; no winsorizing was needed for interest variables), and then grand mean-centered. Effects sizes are Cohen's <emph>d</emph>s based on covariate-adjusted standard deviations from the omnibus models (Howell, [<reflink idref="bib18" id="ref39">18</reflink>]). Using the calculated effect sizes from our critical tests of condition differences in forecasted fit in a post hoc power analysis (using G*Power 3.1.9.7), we had 77.36% power to detect our effect at the Midweek Survey (<emph>d</emph> = 0.35) and 55.22% power to detect our effect at the Exit Survey (<emph>d</emph> = 0.27). Per a sensitivity analysis, we had 80% power to detect effects of <emph>d</emph> = 0.36.</p> <hd id="AN0185469581-18">Descriptive statistics</hd> <p>Means and standard deviations are in Table 1 (see bivariate correlations in Table S2).</p> <p>Table 1 Descriptive Means and Standard Deviations for STEM and Non-STEM Outcomes by Condition</p> <p> <ephtml> <table frame="hsides" rules="groups"><thead><tr><th align="left" rowspan="3"><p>Measure</p></th><th align="left" rowspan="3"><p>Survey</p></th><th align="left" colspan="2"><p>STEM</p></th><th align="left" colspan="2"><p>Non-STEM</p></th></tr><tr><th align="left"><p>Control</p></th><th align="left"><p>Intervention</p></th><th align="left"><p>Control</p></th><th align="left"><p>Intervention</p></th></tr><tr><th align="left"><p>M (SD)</p></th><th align="left"><p>M (SD)</p></th><th align="left"><p>M (SD)</p></th><th align="left"><p>M (SD)</p></th></tr></thead><tbody><tr><td align="left"><p>Current fit</p></td><td align="left"><p>Intake</p></td><td align="left"><p>3.79 (0.72)</p></td><td align="left"><p>3.84 (0.73)</p></td><td align="left"><p>3.78 (0.76)</p></td><td align="left"><p>3.76 (0.83)</p></td></tr><tr><td align="left"><p>Forecasted fit</p></td><td align="left"><p>Intake</p></td><td align="left"><p>3.61 (0.70)</p></td><td align="left"><p>3.59 (0.69)</p></td><td align="left"><p>3.65 (0.86)</p></td><td align="left"><p>3.61 (0.84)</p></td></tr><tr><td align="left"><p>Interest in high school classes</p></td><td align="left"><p>Intake</p></td><td align="left"><p>3.29 (0.97)</p></td><td align="left"><p>3.35 (1.04)</p></td><td align="left"><p>3.28 (1.22)</p></td><td align="left"><p>3.13 (1.16)</p></td></tr><tr><td align="left"><p>Interest in careers</p></td><td align="left"><p>Intake</p></td><td align="left"><p>2.76 (0.88)</p></td><td align="left"><p>2.71 (0.82)</p></td><td align="left"><p>2.53 (0.73)</p></td><td align="left"><p>2.48 (0.68)</p></td></tr><tr><td align="left"><p>Forecasted fit</p></td><td align="left"><p>Midweek</p></td><td align="left"><p>3.56 (0.71)</p></td><td align="left"><p>3.66 (0.64)</p></td><td align="left"><p>3.61 (0.84)</p></td><td align="left"><p>3.58 (0.85)</p></td></tr><tr><td align="left"><p>Current fit</p></td><td align="left"><p>Exit</p></td><td align="left"><p>3.76 (0.72)</p></td><td align="left"><p>3.88 (0.72)</p></td><td align="left"><p>3.72 (0.80)</p></td><td align="left"><p>3.74 (0.89)</p></td></tr><tr><td align="left"><p>Forecasted fit</p></td><td align="left"><p>Exit</p></td><td align="left"><p>3.61 (0.73)</p></td><td align="left"><p>3.70 (0.70)</p></td><td align="left"><p>3.66 (0.85)</p></td><td align="left"><p>3.62 (0.86)</p></td></tr><tr><td align="left"><p>Interest in high school classes</p></td><td align="left"><p>Exit</p></td><td align="left"><p>3.36 (0.97)</p></td><td align="left"><p>3.54 (1.02)</p></td><td align="left"><p>3.35 (1.16)</p></td><td align="left"><p>3.26 (1.13)</p></td></tr><tr><td align="left"><p>Interest in careers</p></td><td align="left"><p>Exit</p></td><td align="left"><p>2.74 (0.95)</p></td><td align="left"><p>2.89 (0.83)</p></td><td align="left"><p>2.55 (0.74)</p></td><td align="left"><p>2.61 (0.73)</p></td></tr></tbody></table> </ephtml> </p> <p>Control <emph>n</emph>s range across outcomes from 110–118, intervention <emph>n</emph>s range from 123–124</p> <hd id="AN0185469581-19">Effect of the intervention</hd> <p>We examined intervention condition effects on current STEM fit, forecasted STEM fit, interest in STEM high school classes, and interest in STEM careers (see Fig. 1), with parallel models of non-STEM measures to test specificity to STEM outcomes (see Table 2). To account for pre-existing individual differences across conditions, analyses covaried for the matched Intake Survey version of each respective Midweek and Exit Survey outcome measure. (Baseline values never significantly moderated condition effects; see Table S4.) An additional three-level covariate accounted for potential variation due to camp site, which did not change the pattern of results (see Table S3), and supplemental multi-level models (with participants nested within classes) showed nearly identical effects (see Supplement section "Multi-Level Modelling").</p> <p>Graph: Fig. 1 Girls' Average Fit and Interest in STEM by Condition. Note. Boxplots display medians and interquartile ranges of raw data; dashed black lines indicate baseline-adjusted means. "H.S." indicates "high school". "Midweek" and "Exit" indicate measures collected in the Midweek versus Exit Surveys, respectively</p> <p>Table 2 Condition effects: Regression coefficients</p> <p> <ephtml> <table frame="hsides" rules="groups"><thead><tr><th align="left" rowspan="3"><p>Outcome</p></th><th align="left" rowspan="3"><p>Survey</p></th><th align="left" colspan="4"><p>Intervention effects</p></th></tr><tr><th align="left" colspan="2"><p>STEM Items</p></th><th align="left" colspan="2"><p>Non-STEM Items</p></th></tr><tr><th align="left"><p><italic>b</italic></p></th><th align="left"><p><italic>t</italic></p></th><th align="left"><p><italic>b</italic></p></th><th align="left"><p><italic>t</italic></p></th></tr></thead><tbody><tr><td align="left"><p>Current fit</p></td><td align="left"><p>Exit</p></td><td char="." align="char"><p><bold>0.04</bold></p></td><td char="." align="char"><p><bold>1.37</bold></p></td><td char="." align="char"><p>0.03</p></td><td char="." align="char"><p>0.65</p></td></tr><tr><td align="left"><p>Forecasted fit</p></td><td align="left"><p>Midweek</p></td><td char="." align="char"><p><bold>0.07</bold></p></td><td char="." align="char"><p><bold>2.64</bold><sup><bold>**</bold></sup></p></td><td char="." align="char"><p>0.01</p></td><td char="." align="char"><p>0.25</p></td></tr><tr><td align="left"><p>Forecasted fit</p></td><td align="left"><p>Exit</p></td><td char="." align="char"><p><bold>0.06</bold></p></td><td char="." align="char"><p><bold>2.03</bold><sup><bold>*</bold></sup></p></td><td char="." align="char"><p>0.00</p></td><td char="." align="char"><p>0.08</p></td></tr><tr><td align="left"><p>Interest in high school classes</p></td><td align="left"><p>Exit</p></td><td char="." align="char"><p><bold>0.07</bold></p></td><td char="." align="char"><p><bold>1.73</bold><sup><bold>†</bold></sup></p></td><td char="." align="char"><p>0.01</p></td><td char="." align="char"><p>0.22</p></td></tr><tr><td align="left"><p>Interest in careers</p></td><td align="left"><p>Exit</p></td><td char="." align="char"><p><bold>0.10</bold></p></td><td char="." align="char"><p><bold>3.24</bold><sup><bold>**</bold></sup></p></td><td char="." align="char"><p>0.05</p></td><td char="." align="char"><p>2.12<sup>*</sup></p></td></tr></tbody></table> </ephtml> </p> <p>Condition was effects coded (− 1 = control; 1 = intervention). Boldface indicates key condition effects. All models included the corresponding Intake Survey baseline measure as a covariate (always <emph>p</emph> <.001) and a three-level factor representing site (which was never significant, <emph>ps</emph> ≥.108; see Table S3 for these non-focal parameters). Degrees of freedom range (226–236) by varying item completion rates **<emph>p</emph> <.01, *<emph>p</emph> <.05, <bold>†</bold><emph>p</emph> <.1</p> <hd id="AN0185469581-20">Fit</hd> <p> <bold>Current Fit</bold> <bold> <emph>.</emph> </bold> Girls in the intervention (vs. control) condition did not significantly differ on current fit in STEM classes (<emph>p</emph> =.171, <emph>d</emph> = 0.18) or English classes (<emph>p</emph> =.514, <emph>d</emph> = 0.08).</p> <p> <bold>Forecasted Fit.</bold> Our hypothesized condition effect for forecasted STEM fit emerged. Girls in the intervention condition reported stronger forecasted fit in STEM university classes (e.g., Engineering) than girls in control, on both the Midweek Survey (<emph>p</emph> =.009, <emph>d</emph> = 0.35) and the Exit Survey (<emph>p</emph> =.043, <emph>d</emph> = 0.27). The effect of the intervention was specific to girls' forecasted <emph>STEM</emph> fit: No condition differences emerged for forecasted fit in English classes on the Midweek or Exit Surveys (<emph>ps</emph> ≥.80, |<emph>d</emph>|s ≤ 0.03).</p> <hd id="AN0185469581-21">Interest in classes and careers</hd> <p> <bold>Interest in High School Classes.</bold> Reported levels of interest in later taking high school STEM classes were marginally higher for girls in the intervention compared to the control condition (<emph>p</emph> =.085, <emph>d</emph> = 0.23). As expected, the intervention did not affect girls' interest in high school English classes (<emph>p</emph> =.830, <emph>d</emph> = 0.03).</p> <p> <bold>Interest in Careers.</bold> Mirroring the forecasted fit measures, girls in the intervention (vs. control) condition reported stronger interest in pursuing STEM careers (<emph>p</emph> =.001, <emph>d</emph> = 0.42). Unexpectedly, interest in non-STEM careers was also higher among girls in the intervention (vs. control) condition (<emph>p</emph> =.035; <emph>d</emph> = 0.27), although this effect was notably (over 35%) smaller than that for STEM careers.</p> <hd id="AN0185469581-22">Mediational models</hd> <p>We hypothesized that increases in forecasted STEM fit would mediate condition differences in increased interest in high school STEM classes and interest in STEM careers. We tested this hypothesized mediational link from condition (− 1 = control; 1 = intervention) to forecasted STEM fit (measured on the Midweek Survey, which was typically on a Tuesday) to interest in STEM (measured Friday on the Exit Survey) using PROCESS (with 5,000 bootstrap samples; Hayes, [<reflink idref="bib17" id="ref40">17</reflink>]). Following best practices for repeated-measures mediation (Montoya & Hayes, [<reflink idref="bib27" id="ref41">27</reflink>]), the mediator and outcome variables were each constructed as change scores by subtracting Intake Survey values (from Midweek Survey values for forecasted STEM fit, and from Exit Survey values for interest in STEM measures). To control for mean-level variation in the mediator, we customized PROCESS Model 4 to covary for the average of forecasted STEM fit on the Intake and Midweek Surveys. See Table S5 for full model details, plus alternate models (using change in current STEM fit and change in forecasted STEM fit as mediators, both again constructed as change scores from Intake to Exit Survey).</p> <p>As expected, change in forecasted STEM fit from Intake to Midweek fit varied by condition: Girls in the intervention condition had significantly larger gains than girls in the control condition, <emph>b</emph> = 0.07, <emph>p</emph> =.012 (see Fig. 2). Further, supplementary analysis within the intervention condition showed significant Intake-to-Midweek increases in forecasted STEM fit, <emph>t</emph>(<reflink idref="bib121" id="ref42">121</reflink>) = 2.33,<emph> p</emph> =.011.</p> <p>Graph: Fig. 2 Increased Forecasted Fit in STEM Mediates Intervention Effect on Increased STEM Interest. Note. Delta (Δ) indicates change over time (since the Intake Survey). Path estimates are unstandardized regression coefficients. Bracketed values indicate 95% bootstrap confidence intervals. Per Montoya and Hayes ([<reflink idref="bib27" id="ref43">27</reflink>]), models covaried for average forecasted STEM fit (Intake and Midweek) when predicting each outcome. "H.S." indicates "high school"</p> <p>These mediation models provide correlational evidence that this boost in forecasted STEM fit is associated with increased interest in STEM high school classes and STEM careers, <emph>p</emph>s <.001 (see Fig. 2). Finally, bootstrapped 95% CIs indicated an indirect effect for both STEM interest outcomes, such that increased forecasted STEM fit mediated the relation from condition to increased interest in STEM education and careers. Further, speaking to the robustness of these findings, the reported indirect effects remained significant in comparable mediational models using raw forecasted fit and interest measures (instead of change scores) and covarying for respective baseline measurements from the Intake Survey (per recommendations by Loh & Ren, [<reflink idref="bib24" id="ref44">24</reflink>]).</p> <hd id="AN0185469581-23">Discussion</hd> <p>Our novel field intervention improved girls' forecasted fit in future STEM environments, as well as girls' interest in future STEM education and careers. This intervention built upon Schmader and Sedikides' ([<reflink idref="bib34" id="ref45">34</reflink>]) theorizing about fit in social environments to highlight to young girls how they could experience self-concept, goal, and social fit in STEM into adulthood. In addition, the findings presented here extend that theoretical account by emphasizing how <emph>forecasts</emph> regarding adult selves (in future STEM environments) may sway long-term educational and occupational decision-making. Consistent with this framing, we found that the intervention improved girls' forecasted fit in future STEM environments (i.e., when on the cusp of entering an undergraduate program), but did not significantly impact their current sense of fit in STEM. Moreover, increases in forecasted (but not current) STEM fit predicted improved interest in future STEM high school classes and careers (see Supplement for full details).</p> <p>This research makes several important theoretical contributions. First, these findings highlight the potential for <emph>forecasts</emph> (not just current experiences) to shape long-term STEM trajectories. Further, extending the "leaky pipeline" metaphor (Clark Blickenstaff, [<reflink idref="bib7" id="ref46">7</reflink>]), we see girls and women as active agents who anticipate the cracks they and their female peers may fall (or be pushed) through. Such wariness of future barriers in STEM may prove adaptive, helping some girls and women avoid sexist interpersonal interactions or workplace cultures (e.g., Hall et al., [<reflink idref="bib15" id="ref47">15</reflink>]; Moss-Racusin et al., [<reflink idref="bib28" id="ref48">28</reflink>]). However, expecting the outright hostility faced by prior generations may deter girls and women from discovering or developing inclusive downstream STEM settings, or lead them to discount their own feelings of fit and interest, creating a lag in girls and women choosing to enter a now "less leaky" pipeline. As such, the intervention presented here encouraged girls to see potential points of connection between themselves and relatively welcoming future environments in STEM.</p> <p>Our intervention also featured a diverse mix of STEM role models—varying on demographics, values and goals (e.g., communal, agentic) and interests (e.g., biomedical, mathematics). Although we did not explicitly provide a choice of in-person role models (e.g., O'Brien et al., [<reflink idref="bib30" id="ref49">30</reflink>]), we anticipated that the wide array of options would raise the odds that each girl could feel a sense of similarity and connection with at least one role model (Gladstone & Cimpian, [<reflink idref="bib14" id="ref50">14</reflink>]). As suggested by our own previous pilot testing (see Supplement section "Additional Related Studies"), providing a single in-person undergraduate role model (with no video; Pilot 1), or combining an in-person undergraduate role model with a video of a highly successful late-career STEM role model (a dean; Pilot 2) were not effective.</p> <p>This research also has clear practical implications for STEM outreach initiatives and other efforts to increase girls' interest in STEM. These findings underscore the value of featuring an array of role models who represent an aspirational (but still realistic) depiction of potential future selves (Lockwood & Kunda, [<reflink idref="bib23" id="ref51">23</reflink>]). Further, this research suggests broad benefits of directly interacting with role models through genuine conversations.</p> <hd id="AN0185469581-24">Specificity and robustness</hd> <p>Our intervention was designed to benefit girls' fit and interest in STEM, distinct from any potential effects regarding fit and interest in non-STEM domains. Demonstrating specificity, the intervention primarily affected fit and interest in STEM domains, with only one instance of spillover onto non-STEM outcomes (higher interest in non-STEM careers). Further, the effect of this intervention on girls' STEM outcomes transcended any effects on non-STEM outcomes: Covarying for non-STEM outcomes did not decrease intervention effects on STEM outcomes (on the contrary, effects on high school STEM class interest grew stronger; see Supplement section "Non-STEM Outcomes as Covariates").</p> <hd id="AN0185469581-25">Constraints on generalizability and future directions</hd> <p>Ideally, we would have tracked participants' actual future choices (e.g., taking high school math and science classes), in light of evidence that educational preferences can shift from idealistic to pragmatic factors when choices are imminent versus in the more distant future (Kivetz & Tyler, [<reflink idref="bib21" id="ref52">21</reflink>]; Studies 1 & 3). However, the intervention-based increases in girls' forecasted fit and interest in STEM persisted across several days (i.e., from Tuesday to Friday), unlike many that fade within 24 hours (Lai et al., [<reflink idref="bib22" id="ref53">22</reflink>]). In addition, although we were unable to fully disentangle which specific elements of the multi-faceted intervention are necessary, our pilot work (see Supplement section "Additional Related Studies") suggested that relatable role models presented both via a personalized in-person conversation plus a comprehensive video may be necessary components of the intervention.</p> <p>Further, the effects of our intervention could plausibly be specific to the age range of participants in this study (mostly 11- to 13-year-olds). However, we found only a singular instance (out of five models) of significant moderation by participant age; such that younger (vs. older) girls' interest in STEM careers shifted more in response to the intervention (see Supplement section "Moderation by Demographics"). The overall lack of moderation by participant age suggests that our intervention may be broadly applicable, at least across the range of recruited ages (full range: 8–15 years). In addition, most participants were Asian (44%) and White (37%). Although intervention condition effects were not moderated by participant race (see Supplement section "Moderation by Demographics"), further work should examine the generalizability of these findings to more diverse samples, including girls from racial/ethnic groups who face negative gender and race-based STEM-ability stereotypes, as well as non-binary youth. Further, we tested this intervention with girls enrolled in STEM-focused summer camps. Plausibly, girls enrolled in STEM camps may have stronger feelings of fit and interest in STEM than girls in the general population, and thus may have less "room to grow" (potentially reducing intervention effects) or are perhaps more open to considering STEM pursuits (potentially increasing intervention effects). Future interventions with representative samples would provide a deeper understanding of the generalizability of these effects.</p> <hd id="AN0185469581-26">Conclusions</hd> <p>Girls begin making important educational and career decisions in early high school. This study provides evidence that an early intervention among girls in STEM camps increased their forecasted future fit in undergraduate STEM majors which in turn was associated with increased interest in STEM high school classes and careers. This research points to the value of intervening early to align girls' potential with their anticipated STEM persistence. These findings can inspire researchers and practitioners to consider how members of marginalized groups forecast their future and to prioritize interventions that can increase (and sustain) their sense of future fit in fields that urgently need their talent and expertise.</p> <hd id="AN0185469581-27">Acknowledgements</hd> <p>We thank research assistants Odilia Dys-Steenbergen, Brittany Dennett, Brian Mulvany, Ehsaan Elhagehassan, Jonathan Mendel, Jacob Pavicic, Jessie Shen, Kavana Ramesh, Kirsten Suhner, Seth Mahon, Tyler Hartwig, and Yingfan (Viola) Zhang; and organizational liaisons Alicia Ward, Caity Dyck, Emily Ritz, Jakob Manning, Jess Crawford-Brown, Leanne Predote, Soundous Ettayebi, and Virginia Hall.</p> <hd id="AN0185469581-28">Funding</hd> <p>Funding was provided by a Social Sciences and Humanities Research Council (SSHRC) of Canada Partnership Grant (#895-2017-1025); the Faculty of Arts and Social Sciences at Simon Fraser University; the Provost and Faculty of Arts at the University of Waterloo; and the International Work Learn Award and the Faculties of Science and Applied Science at the University of British Columbia.</p> <hd id="AN0185469581-29">Declarations</hd> <p></p> <hd id="AN0185469581-30">Conflict of interests</hd> <p>The authors have no conflicts of interest. Artificial intelligence was not used in any part of this research. All materials, including semi-structured intervention (and control) conversation details, are in the Supplement—as approved by research ethics boards at the University of Waterloo, University of British Columbia, and Simon Fraser University. We report all exclusions (see Method) and experiments (see Supplement). Although the main study was not preregistered, the research design and core hypotheses were reviewed by SSHRC prior to funding (in 2016). An online follow-up study was pre-registered (see Supplement). The de-identified datafile plus reproducible analytic code (written in SPSS syntax) are available at osf.io/r382a/.</p> <hd id="AN0185469581-31">Supplementary Information</hd> <p>Below is the link to the electronic supplementary material.</p> <p>Graph: Supplementary file1 (DOCX 68 KB)</p> <hd id="AN0185469581-32">Publisher's Note</hd> <p>Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.</p> <ref id="AN0185469581-33"> <title> References </title> <blist> <bibl id="bib1" idref="ref11" type="bt">1</bibl> <bibtext> Block K, Gonzalez AM, Schmader T, Baron AS. Early gender differences in core values predict anticipated family versus career orientation. Psychological Science. 2018; 29; 9: 1540-1547. 10.1177/0956797618776942</bibtext> </blist> <blist> <bibl id="bib2" idref="ref12" type="bt">2</bibl> <bibtext> Block K, Hall WM, Schmader T, Inness M, Croft E. Should I stay or should I go?. Social Psychology. 2018; 49; 4: 243-251. 10.1027/1864-9335/a000343</bibtext> </blist> <blist> <bibl id="bib3" idref="ref31" type="bt">3</bibl> <bibtext> Boucher, K. L, Fuesting, M. A, Diekman, A. B, & Murphy, M. C. (2017). Can I work with and help others in this field? How communal goals influence interest and participation in STEM fields. Frontiers in Psychology, https://doi.org/10.3389/fpsyg.2017.00901</bibtext> </blist> <blist> <bibl id="bib4" idref="ref35" type="bt">4</bibl> <bibtext> Cheryan S, Lombard EJ, Hudson L, Louis K, Plaut VC, Murphy MC. Double isolation: Identity expression threat predicts greater gender disparities in computer science. Self and Identity. 2020; 19; 4: 412-434. 10.1080/15298868.2019.1609576</bibtext> </blist> <blist> <bibl id="bib5" idref="ref13" type="bt">5</bibl> <bibtext> Cheryan S, Plaut VC, Davies PG, Steele CM. Ambient belonging: How stereotypical cues impact gender participation in computer science. Journal of Personality and Social Psychology. 2009; 97; 6: 1045-1060. 10.1037/a0016239</bibtext> </blist> <blist> <bibl id="bib6" idref="ref25" type="bt">6</bibl> <bibtext> Cheryan, S, Siy, J. O, Vichayapai, M, Drury, B. J, & Kim, S. (2011). Do female and male role models who embody STEM stereotypes hinder women's anticipated success in STEM? Social Psychological and Personality Science, 2(6), 656–664. https://doi.org/10.1177/1948550611405218</bibtext> </blist> <blist> <bibl id="bib7" idref="ref46" type="bt">7</bibl> <bibtext> Clark Blickenstaff J. Women and science careers: Leaky pipeline or gender filter?. Gender and Education. 2005; 17; 4: 369-386. 10.1080/09540250500145072</bibtext> </blist> <blist> <bibl id="bib8" idref="ref37" type="bt">8</bibl> <bibtext> Cyr, E. N, Kroeper, K. M, Bergsieker, H. B, Dennehy, T. C, Logel, C, Steele, J. R, & Spencer, S. J. (2024). Girls are good at STEM: Opening minds and providing evidence reduces boys' stereotyping of girls' STEM ability. Child Development, 95(2), 636–647.</bibtext> </blist> <blist> <bibl id="bib9" idref="ref19" type="bt">9</bibl> <bibtext> Dasgupta, N, Scircle, M. M, & Hunsinger, M. (2015). Female peers in small work groups enhance women's motivation, verbal participation, and career aspirations in engineering. Proceedings of the National Academy of Sciences, 112(16), 4988–4993.</bibtext> </blist> <blist> <bibtext> Dennehy, T. C, & Dasgupta, N. (2017). Female peer mentors early in college increase women's positive academic experiences and retention in engineering. Proceedings of the National Academy of Sciences, 114(23), 5964–5969.</bibtext> </blist> <blist> <bibtext> Diekman AB, Brown ER, Johnston AM, Clark EK. Seeking congruity between goals and roles: A new look at why women opt out of science, technology, engineering, and mathematics careers. Psychological Science. 2010; 21; 8: 1051-1057. 10.1177/0956797610377342</bibtext> </blist> <blist> <bibtext> Fuesting MA, Diekman AB, Boucher KL, Murphy MC, Manson DL, Safer BL. Growing STEM: Perceived faculty mindset as an indicator of communal affordances in STEM. Journal of Personality and Social Psychology. 2019; 117; 2: 260-281. 10.1037/pspa0000154</bibtext> </blist> <blist> <bibtext> Gentry M, Owen SV. Secondary student perceptions of classroom quality: Instrumentation and differences between advanced/honors and nonhonors classes. Journal of Secondary Gifted Education. 2004; 16; 1: 20-29. 10.4219/jsge-2004-464</bibtext> </blist> <blist> <bibtext> Gladstone JR, Cimpian A. Which role models are effective for which students? A systematic review and four recommendations for maximizing the effectiveness of role models in STEM. International Journal of STEM Education. 2021; 8; 1: 1-20. 10.1186/s40594-021-00315-x</bibtext> </blist> <blist> <bibtext> Hall W, Schmader T, Aday A, Croft E. Decoding the dynamics of social identity threat in the workplace: A within-person analysis of women's and men's interactions in STEM. Social Psychological and Personality Science. 2019; 10; 4: 542-552. 10.1177/1948550618772582</bibtext> </blist> <blist> <bibtext> Harackiewicz JM, Hecht CA, Asher MW, Beymer PN, Lamont LB, Wheeler NS, Else-Quest NM, Priniski SJ, Smith JL, Hyde JS, Thoman DB. A prosocial value intervention in gateway STEM courses. Journal of Personality and Social Psychology. 2023; 125; 6: 1265-1307. 10.1037/pspa0000356</bibtext> </blist> <blist> <bibtext> Hayes AF. PROCESS: A versatile computational tool for observed variable mediation, moderation, and conditional process modeling. 2012; University of Kansas</bibtext> </blist> <blist> <bibtext> Howell, D. C. (2012). Statistical methods for psychology. PWS-Kent Publishing Co. https://psycnet.apa.org/record/1992-97157-000</bibtext> </blist> <blist> <bibtext> Hyde JS, Linn MC. Gender similarities in mathematics and science. Science. 2006; 314; 5799: 599-600. 10.1126/science.1132154</bibtext> </blist> <blist> <bibtext> Kersey AJ, Braham EJ, Csumitta KD, Libertus ME, Cantlon JF. No intrinsic gender differences in children's earliest numerical abilities. Science of Learning. 2018. 10.1038/s41539-018-0028-7</bibtext> </blist> <blist> <bibtext> Kivetz, Y, & Tyler, T. R. (2007). Tomorrow I'll be me: The effect of time perspective on the activation of idealistic versus pragmatic selves. Organizational Behavior and Human Decision Processes, 102(2), 193–211.</bibtext> </blist> <blist> <bibtext> Lai, C. K, Skinner, A. L, Cooley, E, Murrar, S, Brauer, M, Devos, T, Calanchini, J, Xiao, Y. J, Pedram, C, & Marshburn, C. K. (2016). Reducing implicit racial preferences: II. Intervention effectiveness across time. Journal of Experimental Psychology: General, 145(8), 1001–1016. https://doi.org/10.1037/xge0000179</bibtext> </blist> <blist> <bibtext> Lockwood P, Kunda ZTesser A, Felson RB, Suls JM. Outstanding role models: Do they inspire or demoralize us?. Psychological perspectives on self and identity. 2000: Washington; American Psychological Association: 147-171. 10.1037/10357-006</bibtext> </blist> <blist> <bibtext> Loh, W. W, & Ren, D. (2023). Adjusting for baseline measurements of the mediators and outcome as a first step toward eliminating confounding biases in mediation analysis. Perspectives on Psychological Science: A Journal of the Association for Psychological Science, 18(5), 1254–1266. https://doi.org/10.1177/17456916221134573</bibtext> </blist> <blist> <bibtext> Marx DM, Ko SJ. Superstars "like" me: The effect of role model similarity on performance under threat. European Journal of Social Psychology. 2012; 42; 7: 807-812. 10.1002/ejsp.1907</bibtext> </blist> <blist> <bibtext> Master A, Meltzoff AN, Cheryan S. Gender stereotypes about interests start early and cause gender disparities in computer science and engineering. Proceedings of the National Academy of Sciences. 2021; 118; 48: 1-7. 10.1073/pnas.2100030118</bibtext> </blist> <blist> <bibtext> Montoya AK, Hayes AF. Two-condition within-participant statistical mediation analysis: A path-analytic framework. Psychological Methods. 2017; 22; 1: 6-27. 10.1037/met0000086</bibtext> </blist> <blist> <bibtext> Moss-Racusin CA, Sanzari C, Caluori N, Rabasco H. Gender bias produces gender gaps in STEM engagement. Sex Roles. 2018; 79; 11: 651-670. 10.1007/s11199-018-0902-z</bibtext> </blist> <blist> <bibtext> Murphy, M. C, & Zirkel, S. (2015). Race and belonging in school: How anticipated and experienced belonging affect choice, persistence, and performance. Teachers College Record, 117(12), 1–40. https://doi.org/10.1177/016146811511701204</bibtext> </blist> <blist> <bibtext> O'Brien LT, Hitti A, Shaffer E, Camp ARV, Henry D, Gilbert PN. Improving Girls' Sense of Fit in Science: Increasing the Impact of Role Models. Social Psychological and Personality Science. 2017; 8; 3: 301-309. 10.1177/1948550616671997</bibtext> </blist> <blist> <bibtext> Riegle-Crumb C, Farkas G, Muller C. The role of gender and friendship in advanced course taking. Sociology of Education. 2006; 79; 3: 206-228. 10.1177/00380407060790030</bibtext> </blist> <blist> <bibtext> Sadler PM, Sonnert G, Hazari Z, Tai R. Stability and volatility of STEM career interest in high school: A gender study. Science Education. 2012; 96; 3: 411-427. 10.1002/sce.21007</bibtext> </blist> <blist> <bibtext> Sax LJ, Blaney JM, Lehman KJ, Rodriguez SL, George KL, Zavala C. Sense of belonging in computing: The role of introductory courses for women and underrepresented minority students. Social Sciences. 2018; 7; 8: 122. 10.3390/socsci7080122</bibtext> </blist> <blist> <bibtext> Schmader T, Sedikides C. State authenticity as fit to environment: The implications of social identity for fit, authenticity, and self-segregation. Personality and Social Psychology Review. 2018; 22; 3: 228-259. 10.1177/1088868317734080</bibtext> </blist> <blist> <bibtext> Steele J. Children's gender stereotypes about math: The role of stereotype stratification. Journal of Applied Social Psychology. 2003; 33; 12: 2587-2606. 10.1111/j.1559-1816.2003.tb02782.x</bibtext> </blist> <blist> <bibtext> Tyson W. Modeling engineering degree attainment using high school and college physics and calculus coursetaking and achievement. Journal of Engineering Education. 2011; 100; 4: 760-777. 10.1002/j.2168-9830.2011.tb00035.x</bibtext> </blist> <blist> <bibtext> Walton GM, Cohen GL. A question of belonging: Race, social fit, and achievement. Journal of Personality and Social Psychology. 2007; 92; 1: 82. 10.1037/0022-3514.92.1.82</bibtext> </blist> <blist> <bibtext> Wu, D. J, Thiem, K. C, & Dasgupta, N. (2022). Female peer mentors early in college have lasting positive impacts on female engineering students that persist beyond graduation. Nature Communications, 13(1), 6837.</bibtext> </blist> <blist> <bibtext> Zheng H, Weeden KA. How gender segregation in higher education contributes to gender segregation in the USA labor market. Demography. 2023; 60; 3: 761-784. 10.1215/00703370-10653728</bibtext> </blist> </ref> <aug> <p>By Emily N. Cyr; Steven J. Spencer; Stephen C. Wright; Jennifer R. Steele; Kathryn M. Kroeper; Patricia Colaco; Tara C. Dennehy; Priscilla Lok-Chee Shum; Taylor Ballinger; Haemi Nam; Stephanie L. Reeves; Mary Wells; Toni Schmader and Hilary B. Bergsieker</p> <p>Reported by Author; Author; Author; Author; Author; Author; Author; Author; Author; Author; Author; Author; Author; Author</p> <p></p> <p>Emily N. Cyr Emily N. Cyr, is a Postdoctoral Fellow in the Department of Psychology at York University. Her research broadly focuses on using social network analysis of real-world groups to assess and improve intergroup relations.</p> <p>Steven J. Spencer Steven J. Spencer, is a Professor in the Department of Psychology at the Ohio State University. His research primarily focuses on how processes related to self, identity and motivation impact prejudices and stereotyping.</p> <p>Stephen C. Wright Stephen C. Wright, is a Professor in the Department of Psychology at Simon Fraser University. His research broadly focuses on intergroups relations, such as evaluating the responses of socially disadvantaged groups to prejudice and discrimination and the maintenance of minority languages and cultures.</p> <p>Jennifer R. Steele Jennifer R. Steele, is a Professor in the Department of Psychology at York University. Her research assesses biases and stereotyping based on a variety of social groups (i.e., racial, gender and wealth) with a focus on the development of implicit attitudes that can shape these group-based associations.</p> <p>Kathryn M. Kroeper Kathryn M. Kroeper, is an Assistant Professor in the Department of Psychology at Sacred Heart University. Her research examines the effects of situational cues that threaten social identity on motivation and well-being, while developing interventions that encourage inclusive and equitable environments.</p> <p>Patricia Colaco Patricia Colaco, is a Graduate Student in the Department of Psychology at York University. Her research primarily focuses on evaluating the development of implicit biases towards socially disadvantaged groups.</p> <p>Tara C. Dennehy Tara C. Dennehy, was a Postdoctoral Fellow in the Department of Psychology at the University of British Columbia. Her research examined how fundamental cognitive processes affect stereotyping, prejudice and social inequality.</p> <p>Priscilla Lok-chee Shum Priscilla Lok-Chee Shum, is a Postdoctoral Fellow at the Hong Kong Polytechnic University. She primarily conducts research focused on intergroup relations and communication, while aiming to tackle issues of gender equality and prejudice reduction.</p> <p>Taylor Ballinger Taylor Ballinger, was a Doctoral Graduate in the Department of Psychology at the Ohio State University, where he primarily conducted research around attitudes toward diversity, issues of race and socioeconomic inequalities, and the relations between student motivation and achievement.</p> <p>Haemi Nam Haemi Nam, was a Doctoral Graduate in the Department of Psychology at the Ohio State University. Her research evaluated the bi-directional relationship between the functions of the immune system and social experiences.</p> <p>Stephanie L. Reeves Stephanie L. Reeves, was a Doctoral Graduate in the Department of Psychology at the Ohio State University. Her research focused on how the experiences of marginalized groups in educational and work environments are shaped by factors that influence their sense of belonging.</p> <p>Mary Wells Mary Wells, is the Dean of Engineering at the University of Waterloo. Her research evaluates the processing, structure and properties of advanced metallic alloys. Additionally, she is a passionate advocate for diversity within STEM and continuously advances gender inclusion and equity within these male-dominated fields.</p> <p>Toni Schmader Toni Schmader, is a Professor and the Head of the Department of Psychology at the University of British Columbia. Her research aims to examine the dynamic between self and social identity, especially for members of socially disadvantaged groups, while drawing on topics like implicit gender bias, bias reduction, social stigma, and motivation and performance.</p> <p>Hilary B. Bergsieker Hilary B. Bergsieker, is an Associate Professor in the Department of Psychology at the University of Waterloo. Her research evaluates the interpersonal dynamics of intergroup interactions and relationships, focusing on antiracism and bias reduction, gender inequality, allyship and advocacy for marginalized groups, as well as trust and interdependence in diverse groups.</p> </aug> <nolink nlid="nl1" bibid="bib19" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib20" firstref="ref2"></nolink> <nolink nlid="nl3" bibid="bib26" firstref="ref3"></nolink> <nolink nlid="nl4" bibid="bib32" firstref="ref4"></nolink> <nolink nlid="nl5" bibid="bib34" firstref="ref5"></nolink> <nolink nlid="nl6" bibid="bib33" firstref="ref7"></nolink> <nolink nlid="nl7" bibid="bib37" firstref="ref8"></nolink> <nolink nlid="nl8" bibid="bib35" firstref="ref10"></nolink> <nolink nlid="nl9" bibid="bib29" firstref="ref14"></nolink> <nolink nlid="nl10" bibid="bib36" firstref="ref16"></nolink> <nolink nlid="nl11" bibid="bib39" firstref="ref17"></nolink> <nolink nlid="nl12" bibid="bib10" firstref="ref20"></nolink> <nolink nlid="nl13" bibid="bib38" firstref="ref21"></nolink> <nolink nlid="nl14" bibid="bib30" firstref="ref22"></nolink> <nolink nlid="nl15" bibid="bib14" firstref="ref23"></nolink> <nolink nlid="nl16" bibid="bib25" firstref="ref24"></nolink> <nolink nlid="nl17" bibid="bib23" firstref="ref26"></nolink> <nolink nlid="nl18" bibid="bib12" firstref="ref28"></nolink> <nolink nlid="nl19" bibid="bib11" firstref="ref32"></nolink> <nolink nlid="nl20" bibid="bib13" firstref="ref33"></nolink> <nolink nlid="nl21" bibid="bib16" firstref="ref34"></nolink> <nolink nlid="nl22" bibid="bib31" firstref="ref36"></nolink> <nolink nlid="nl23" bibid="bib17" firstref="ref38"></nolink> <nolink nlid="nl24" bibid="bib18" firstref="ref39"></nolink> <nolink nlid="nl25" bibid="bib27" firstref="ref41"></nolink> <nolink nlid="nl26" bibid="bib121" firstref="ref42"></nolink> <nolink nlid="nl27" bibid="bib24" firstref="ref44"></nolink> <nolink nlid="nl28" bibid="bib15" firstref="ref47"></nolink> <nolink nlid="nl29" bibid="bib28" firstref="ref48"></nolink> <nolink nlid="nl30" bibid="bib21" firstref="ref52"></nolink> <nolink nlid="nl31" bibid="bib22" firstref="ref53"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1472361
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Seeing Women Who Fit: Girls' Forecasted Fit in STEM Fosters Career Interest
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Emily+N%2E+Cyr%22">Emily N. Cyr</searchLink> (ORCID <externalLink term="http://orcid.org/0000-0002-1640-1840">0000-0002-1640-1840</externalLink>)<br /><searchLink fieldCode="AR" term="%22Steven+J%2E+Spencer%22">Steven J. Spencer</searchLink><br /><searchLink fieldCode="AR" term="%22Stephen+C%2E+Wright%22">Stephen C. Wright</searchLink><br /><searchLink fieldCode="AR" term="%22Jennifer+R%2E+Steele%22">Jennifer R. Steele</searchLink><br /><searchLink fieldCode="AR" term="%22Kathryn+M%2E+Kroeper%22">Kathryn M. Kroeper</searchLink><br /><searchLink fieldCode="AR" term="%22Patricia+Colaco%22">Patricia Colaco</searchLink><br /><searchLink fieldCode="AR" term="%22Tara+C%2E+Dennehy%22">Tara C. Dennehy</searchLink><br /><searchLink fieldCode="AR" term="%22Priscilla+Lok-Chee+Shum%22">Priscilla Lok-Chee Shum</searchLink><br /><searchLink fieldCode="AR" term="%22Taylor+Ballinger%22">Taylor Ballinger</searchLink><br /><searchLink fieldCode="AR" term="%22Haemi+Nam%22">Haemi Nam</searchLink><br /><searchLink fieldCode="AR" term="%22Stephanie+L%2E+Reeves%22">Stephanie L. Reeves</searchLink><br /><searchLink fieldCode="AR" term="%22Mary+Wells%22">Mary Wells</searchLink><br /><searchLink fieldCode="AR" term="%22Toni+Schmader%22">Toni Schmader</searchLink><br /><searchLink fieldCode="AR" term="%22Hilary+B%2E+Bergsieker%22">Hilary B. Bergsieker</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Social+Psychology+of+Education%3A+An+International+Journal%22"><i>Social Psychology of Education: An International Journal</i></searchLink>. 2025 28(1).
– Name: Avail
  Label: Availability
  Group: Avail
  Data: Springer. Available from: Springer Nature. One New York Plaza, Suite 4600, New York, NY 10004. Tel: 800-777-4643; Tel: 212-460-1500; Fax: 212-460-1700; e-mail: customerservice@springernature.com; Web site: https://link.springer.com/
– Name: PeerReviewed
  Label: Peer Reviewed
  Group: SrcInfo
  Data: Y
– Name: Pages
  Label: Page Count
  Group: Src
  Data: 18
– Name: DatePubCY
  Label: Publication Date
  Group: Date
  Data: 2025
– Name: TypeDocument
  Label: Document Type
  Group: TypDoc
  Data: Journal Articles<br />Reports - Research
– Name: Audience
  Label: Education Level
  Group: Audnce
  Data: <searchLink fieldCode="EL" term="%22Elementary+Education%22">Elementary Education</searchLink>
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22Females%22">Females</searchLink><br /><searchLink fieldCode="DE" term="%22STEM+Education%22">STEM Education</searchLink><br /><searchLink fieldCode="DE" term="%22STEM+Careers%22">STEM Careers</searchLink><br /><searchLink fieldCode="DE" term="%22Vocational+Interests%22">Vocational Interests</searchLink><br /><searchLink fieldCode="DE" term="%22Role+Models%22">Role Models</searchLink><br /><searchLink fieldCode="DE" term="%22Science+Interests%22">Science Interests</searchLink><br /><searchLink fieldCode="DE" term="%22Predictor+Variables%22">Predictor Variables</searchLink><br /><searchLink fieldCode="DE" term="%22Elementary+School+Students%22">Elementary School Students</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1007/s11218-025-10056-2
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 1381-2890<br />1573-1928
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Girls often express less interest in STEM (science, technology, engineering, mathematics) education and careers than boys, despite having comparable aptitude. We randomly assigned 242 girls (Mdn[subscript age] = 12 years; 38% East Asian; 37% White) at Canadian STEM camps to control conversations about generic camp experiences, or intervention conversations where STEM role models discussed how STEM education and careers align with each girl's most important value and emphasized the social community inside and outside of STEM. Key measures were collected at baseline, and several days after the intervention (or control). Girls' "current" STEM fit did not differ by condition, but as expected, the intervention (vs. control) significantly improved girls' forecasts regarding "future" STEM fit (ds = 0.27-0.35) and girls' interest in STEM careers (d = 0.42), with a marginally significant boost in girls' interest in STEM high school classes (d = 0.23). Pre-post increases in forecasted STEM fit mediated increases in STEM interest. Forecasted fit (beyond current fit) appears pivotal for promoting girls' sustained interest in STEM.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2025
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1472361
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1472361
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1007/s11218-025-10056-2
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 18
    Subjects:
      – SubjectFull: Females
        Type: general
      – SubjectFull: STEM Education
        Type: general
      – SubjectFull: STEM Careers
        Type: general
      – SubjectFull: Vocational Interests
        Type: general
      – SubjectFull: Role Models
        Type: general
      – SubjectFull: Science Interests
        Type: general
      – SubjectFull: Predictor Variables
        Type: general
      – SubjectFull: Elementary School Students
        Type: general
    Titles:
      – TitleFull: Seeing Women Who Fit: Girls' Forecasted Fit in STEM Fosters Career Interest
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Emily N. Cyr
      – PersonEntity:
          Name:
            NameFull: Steven J. Spencer
      – PersonEntity:
          Name:
            NameFull: Stephen C. Wright
      – PersonEntity:
          Name:
            NameFull: Jennifer R. Steele
      – PersonEntity:
          Name:
            NameFull: Kathryn M. Kroeper
      – PersonEntity:
          Name:
            NameFull: Patricia Colaco
      – PersonEntity:
          Name:
            NameFull: Tara C. Dennehy
      – PersonEntity:
          Name:
            NameFull: Priscilla Lok-Chee Shum
      – PersonEntity:
          Name:
            NameFull: Taylor Ballinger
      – PersonEntity:
          Name:
            NameFull: Haemi Nam
      – PersonEntity:
          Name:
            NameFull: Stephanie L. Reeves
      – PersonEntity:
          Name:
            NameFull: Mary Wells
      – PersonEntity:
          Name:
            NameFull: Toni Schmader
      – PersonEntity:
          Name:
            NameFull: Hilary B. Bergsieker
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 12
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 1381-2890
            – Type: issn-electronic
              Value: 1573-1928
          Numbering:
            – Type: volume
              Value: 28
            – Type: issue
              Value: 1
          Titles:
            – TitleFull: Social Psychology of Education: An International Journal
              Type: main
ResultId 1