View in EDS HTML Full Text PDF Full Text

Examining STEM Preferences in Autistic Students: The Role of Contextual Support, Self-Efficacy, and Outcome Expectations

Saved in:
Bibliographic Details
Title: Examining STEM Preferences in Autistic Students: The Role of Contextual Support, Self-Efficacy, and Outcome Expectations
Language: English
Authors: Hyejung Kim (ORCID 0000-0002-4635-9292), Mack Ottens, Matthew Jacob, Xingye Qiao
Source: Exceptional Children. 2025 91(3):303-320.
Availability: SAGE Publications. 2455 Teller Road, Thousand Oaks, CA 91320. Tel: 800-818-7243; Tel: 805-499-9774; Fax: 800-583-2665; e-mail: journals@sagepub.com; Web site: https://sagepub.com
Peer Reviewed: Y
Page Count: 18
Publication Date: 2025
Document Type: Journal Articles
Reports - Research
Education Level: High Schools
Secondary Education
Higher Education
Postsecondary Education
Descriptors: High School Students, Undergraduate Students, Autism Spectrum Disorders, STEM Education, STEM Careers, Majors (Students), Course Selection (Students), Longitudinal Studies, Academic Achievement, Students with Disabilities, Self Efficacy, Student Educational Objectives, Context Effect, Occupational Aspiration
DOI: 10.1177/00144029241312777
ISSN: 0014-4029
2163-5560
Abstract: Over recent decades, there has been a significant increase in postsecondary STEM education among autistic individuals. Using data from the High School Longitudinal Study of 2009, this study examined the STEM pathways of autistic students, emphasizing key determinants like proximal context, self-efficacy, and outcome expectations within the framework of social cognitive theory. The results revealed that despite a lower college attendance rate, autistic students displayed a pronounced inclination for STEM majors, particularly in the fields of science, engineering, and mathematics. Notably, autistic students who pursue higher education tend to exhibit increased levels of self-efficacy and anticipate more positive outcomes within STEM disciplines. However, the levels of both constructs in mathematics had decreased by the 11th grade. Nonetheless, STEM self-efficacy played a significant role in influencing outcome expectations and major choices, with this relationship being more pronounced among autistic students. For autistic students, their choice of a STEM major was influenced by their self-efficacy, as well as factors like race and gender. On the other hand, for non-autistic students, their proximal context was an additional determinant in their decision. Insights gained from this research can inform educational strategies aimed at facilitating the participation of autistic individuals in postsecondary STEM education and related career paths.
Abstractor: As Provided
Notes: https://osf.io/u9sn4/?view_only=835a13c917dc4972bd877-d2e7fabeb28
Entry Date: 2025
Accession Number: EJ1466407
Database: ERIC
Full text is not displayed to guests.
FullText Links:
  – Type: pdflink
    Url: https://content.ebscohost.com/cds/retrieve?content=AQICAHj0k_4E0hTGH8RJwT4gCJyBsGNe_WN95AvKlDbXJGqwxwF8vG7AO2N4i9aby11nYjRtAAAA4zCB4AYJKoZIhvcNAQcGoIHSMIHPAgEAMIHJBgkqhkiG9w0BBwEwHgYJYIZIAWUDBAEuMBEEDHcj9jSQEDW-7ELNigIBEICBm74QebIXxaeLANsIWKH52T3JdR9dB1X3OT9bPmbuzwh1HUkJHqZFqWO0kXXUAUoOV9u0WIflGvrTmTZYNFYPhxUNIHbdPo8xiZm8kOHBjUIONGGrLTiUeAYMo_9knlmxT7PeFJ-IleyIQa248fAUnlQr9CO_TXUT-wXYng5lmvgoHiXlDUMqpVZU7iQZBd-KHeXwGkqKiDlvI6uQ
Text:
  Availability: 1
  Value: <anid>AN0183813417;exc01apr.25;2025Mar20.05:29;v2.2.500</anid> <title id="AN0183813417-1">Examining STEM Preferences in Autistic Students: The Role of Contextual Support, Self-Efficacy, and Outcome Expectations </title> <p>Over recent decades, there has been a significant increase in postsecondary STEM education among autistic individuals. Using data from the High School Longitudinal Study of 2009, this study examined the STEM pathways of autistic students, emphasizing key determinants like proximal context, self-efficacy, and outcome expectations within the framework of social cognitive theory. The results revealed that despite a lower college attendance rate, autistic students displayed a pronounced inclination for STEM majors, particularly in the fields of science, engineering, and mathematics. Notably, autistic students who pursue higher education tend to exhibit increased levels of self-efficacy and anticipate more positive outcomes within STEM disciplines. However, the levels of both constructs in mathematics had decreased by the 11<sup>th</sup> grade. Nonetheless, STEM self-efficacy played a significant role in influencing outcome expectations and major choices, with this relationship being more pronounced among autistic students. For autistic students, their choice of a STEM major was influenced by their self-efficacy, as well as factors like race and gender. On the other hand, for non-autistic students, their proximal context was an additional determinant in their decision. Insights gained from this research can inform educational strategies aimed at facilitating the participation of autistic individuals in postsecondary STEM education and related career paths.</p> <p>Keywords: autism; college; STEM; self-efficacy; outcome expectation; HSLS:09</p> <p>According to the Autism and Developmental Disabilities Monitoring Network, which is funded by the U.S. Centers for Disease Control and Prevention (CDC), there has been a marked increase in the prevalence of autism from.67% to 2.76% over the last 2 decades, equating to one in every 36 children in the United States in 2020 ([<reflink idref="bib36" id="ref1">36</reflink>]). This increase is also reflected in a corresponding rise in the number of neurodiverse[<reflink idref="bib6" id="ref2">6</reflink>] individuals pursuing post-secondary education. These students exhibit a notable preference for the science, technology, engineering, and mathematics (STEM) fields ([<reflink idref="bib65" id="ref3">65</reflink>]). Furthermore, it has been observed that once they embark on a STEM pathway, they are more likely to actively engage in their studies and successfully graduate ([<reflink idref="bib63" id="ref4">63</reflink>]). What mechanisms underlie this inclination?</p> <hd id="AN0183813417-2">Autism and STEM</hd> <p>Recently, research has recognized the distinctive attributes of autistic students[<reflink idref="bib7" id="ref5">7</reflink>] that may heighten their proficiency in STEM fields. For instance, autistic individuals are often characterized for their abilities of systemic thinking and a pronounced intrinsic motivation, which fosters an intense focus on their areas of interest ([<reflink idref="bib6" id="ref6">6</reflink>]; [<reflink idref="bib22" id="ref7">22</reflink>]). Furthermore, motivational factors inherent to many autistic individuals can be a driving force behind a firm commitment and comprehensive exploration in specific areas ([<reflink idref="bib15" id="ref8">15</reflink>]). This is bolstered by the resilience shown by some autistic people, allowing them to navigate through intricate tasks, which becomes invaluable in problem-solving within STEM disciplines ([<reflink idref="bib6" id="ref9">6</reflink>]). In addition, their notable meticulous attention to detail and advanced pattern recognition abilities are traits that could be advantageous in scientific and technical domains ([<reflink idref="bib43" id="ref10">43</reflink>]; [<reflink idref="bib47" id="ref11">47</reflink>]). Scholars also found that their preference for routine, structure, and consistency aligns well with the systematic and repetitive methodologies inherent in scientific investigations ([<reflink idref="bib7" id="ref12">7</reflink>]).</p> <p>Yet, autism is a spectrum condition: although many autistic individuals may possess these traits, they are not universally shared, nor do they singularly guarantee success in STEM. Nonetheless, researchers recently have paid attention to the strengths and unique perspectives offered by autistic individuals, especially considering less favorable postschool outcomes for the population (e.g., [<reflink idref="bib7" id="ref13">7</reflink>]; [<reflink idref="bib22" id="ref14">22</reflink>]). According to the National Longitudinal Transition Study 2, only about 44% of autistic students pursue postsecondary education eight years after high school, a figure considerably lower than the 67% enrollment rate among the general population; similarly, their employment rate of 37% is markedly below the 66% of their general population peers ([<reflink idref="bib44" id="ref15">44</reflink>]). This discrepancy underscores the need to illustrate the mechanisms behind the career path choices of autistic individuals, particularly their inclination towards STEM fields.</p> <hd id="AN0183813417-3">Understanding Major Choices: Social Cognitive Models</hd> <p>[<reflink idref="bib31" id="ref16">31</reflink>] provide a potential framework to understand what drives individuals to choose certain careers over others. They examined the career decisions of engineering students by integrating Bandura's Social Cognitive Theory with Social Cognitive Career Theory (SCCT). Their study shows the impact of <emph>proximal contexts</emph> on choice behaviors, such as the influence of mentors or economic factors, which can shape students' commitment to their majors. These proximal contexts are found to be indirectly guiding career-related actions through a pathway mediated by self-efficacy and outcome expectations.</p> <p>As [<reflink idref="bib31" id="ref17">31</reflink>] highlight, <emph>self-efficacy</emph> pertains to "people's beliefs about their ability to perform particular behaviors or courses of action" (p. 458). Originating from Bandura's Social Cognitive Theory, self-efficacy is not just about possessing requisite skills; it is also about believing in one's ability to utilize these skills effectively under varying circumstances ([<reflink idref="bib3" id="ref18">3</reflink>]; [<reflink idref="bib46" id="ref19">46</reflink>]). [<reflink idref="bib5" id="ref20">5</reflink>] demonstrate that self-efficacy can profoundly influence academic engagement, achievement, and choices—such as opting for STEM majors—particularly when navigating through both supportive and challenging environments ([<reflink idref="bib52" id="ref21">52</reflink>]; [<reflink idref="bib73" id="ref22">73</reflink>]).</p> <p> <emph>Outcome expectations</emph>, on the other hand, refer to an individual's anticipated consequences of performing particular behaviors. [<reflink idref="bib4" id="ref23">4</reflink>] defined outcome expectation as "a judgment of the likely consequence such behavior will produce" (p. 391). These anticipated outcomes range from tangible benefits like financial rewards to intrinsic ones, like personal satisfaction or societal respect ([<reflink idref="bib68" id="ref24">68</reflink>]). For instance, students' perceptions of the outcomes of their academic endeavors upon pursuing a STEM major can significantly sway their decisions ([<reflink idref="bib69" id="ref25">69</reflink>]). As students navigate their academic paths, the confluence of their belief in their own capabilities and their foresight of potential outcomes significantly influences their educational decisions.</p> <hd id="AN0183813417-4">Study Objectives</hd> <p>Despite our understanding of the roles of proximal contexts, self-efficacy, and outcome expectations in college major selection, and the recognized strengths of autistic individuals in STEM, a research gap persists in understanding educational pathways among autistic individuals in STEM fields. In response to this gap, our study seeks to address the following key research questions: (<reflink idref="bib1" id="ref26">1</reflink>) What are the STEM majors chosen by autistic college students relative to their non-autistic counterparts? (<reflink idref="bib2" id="ref27">2</reflink>) What distinguishes autistic college students from their non-autistic peers in terms of proximal context, self-efficacy, and outcome expectations in the context of STEM? (<reflink idref="bib3" id="ref28">3</reflink>) What factors, both directly and indirectly, influence the shaping of their STEM major pathways? By answering these questions, we aim to uncover the reasons behind their preference for STEM disciplines and further elucidate the paths to STEM majors for neurodiverse learners.</p> <hd id="AN0183813417-5">Methods</hd> <p>In order to gain a comprehensive understanding of the pathways autistic students take towards pursuing STEM majors, this study utilized one of the national surveys: the High School Longitudinal Study of 2009 (HSLS:09). The HSLS:09 survey placed a particular emphasis on STEM education, including courses, majors, and careers, offering valuable insights into the STEM paths of autistic students. Given its focus on students' educational decision-making processes, we adapted the social cognitive model to investigate the educational trajectories of autistic students within the area of STEM.</p> <hd id="AN0183813417-6">HSLS:09 Data</hd> <p>This nationally representative longitudinal study, which began in 2009, monitored the academic progression of over 23,000 students, starting from ninth grade, in 944 schools across the United States. The HSLS:09 tracked these students from secondary school through their postsecondary education, eventually leading into the workforce and adulthood. It focused on how students transitioned from high school to postsecondary institutions, and how their decisions during this transition impacted their subsequent academic, occupational, and social outcomes. Data were obtained through direct math assessment, and surveys from students, parents, math and science teachers, school administrators, and school counselors. The cohort of ninth grade students in 2009 was followed by subsequent data collections in 2012 (11<sups>th</sups> grade), 2013, and 2016, with another planned for 2025. Students' high school and postsecondary education transcripts were also collected in 2017 and 2018. Prior to beginning the data analysis process, we secured a license to use the restricted dataset from the National Center for Education Statistics, and obtained permission from the Institutional Review Board. Further, key quality indicators outlined by [<reflink idref="bib33" id="ref29">33</reflink>], p. 399) were incorporated to enhance the rigor of this analysis, although some indicators were beyond the scope of the current study.[<reflink idref="bib8" id="ref30">8</reflink>]</p> <hd id="AN0183813417-7">Sample</hd> <p>Within the HSLS:09 dataset, there were approximately 180 autistic students,[<reflink idref="bib9" id="ref31">9</reflink>] which represented about 1.4% (<emph>SE</emph> =.157) of ninth graders in 2009 in the United States, mirroring the prevalence of autism around that year ([<reflink idref="bib71" id="ref32">71</reflink>]). Table 1 displays the demographic characteristics of the sample used in this study, revealing slight differences in the profiles of autistic and non-autistic groups. For example, autistic students were less represented among Hispanic and female populations, but no significant disparity was observed in family income within the sample.</p> <p>Table 1. Weighted % of Sample Demographics</p> <p>Graph</p> <p></p> <p> <ephtml> <table><colgroup><col align="left" /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /></colgroup><thead><tr><th align="left">Variable</th><th align="left" colspan="2">Total</th><th align="left" colspan="2">ASD</th><th align="left" colspan="2">Non-ASD</th></tr><tr><th align="left" /><th align="left">%</th><th align="left"><italic>SE</italic></th><th align="left">%</th><th align="left"><italic>SE</italic></th><th align="left">%</th><th align="left"><italic>SE</italic></th></tr></thead><tbody><tr><td>Race/Ethnicity</td><td /><td /><td /><td /><td /><td /></tr><tr><td> White</td><td>54.2</td><td>1.1</td><td>60.9</td><td>6.1</td><td>54.1</td><td>1.1</td></tr><tr><td> Hispanic</td><td>21.8</td><td>.9</td><td>15.1</td><td>4.6</td><td>21.9***</td><td>.9</td></tr><tr><td> Black</td><td>12.0</td><td>.7</td><td>15.8</td><td>5.3</td><td>11.9</td><td>.7</td></tr><tr><td> Asian</td><td>3.3</td><td>.3</td><td>3.00</td><td>1.1</td><td>3.3</td><td>.3</td></tr><tr><td>Birth Sex</td><td /><td /><td /><td /><td /><td /></tr><tr><td> Male</td><td>49.9</td><td>.7</td><td>72.6</td><td>4.1</td><td>49.6***</td><td>.7</td></tr><tr><td> Female</td><td>50.1</td><td>.7</td><td>27.4</td><td>4.1</td><td>50.4***</td><td>.7</td></tr><tr><td>Family Income</td><td /><td /><td /><td /><td /><td /></tr><tr><td> ≤ 15,000</td><td>10.5</td><td>.6</td><td>10.8</td><td>3.6</td><td>10.5</td><td>.6</td></tr><tr><td> > 15,000 and ≤ 55,000</td><td>38.9</td><td>.9</td><td>40.3</td><td>6.5</td><td>38.9</td><td>.9</td></tr><tr><td> > 55,000 and ≤ 95,000</td><td>25.0</td><td>.6</td><td>30.9</td><td>5.5</td><td>24.9</td><td>.6</td></tr><tr><td> ≥ 95,000</td><td>25.6</td><td>.9</td><td>18.1</td><td>4.0</td><td>25.6</td><td>.9</td></tr><tr><td>Total <italic>n</italic></td><td>15,340</td><td align="center">−</td><td>180</td><td align="center">−</td><td>15,160</td><td align="center">−</td></tr></tbody></table> </ephtml> </p> <p>1 <emph>Note</emph>. In line with data reporting guidelines, any cell that contained less than three samples was excluded, and certain percentages have been rounded to the nearest whole number. Consequently, the sum of each category may not be equal to 100%. Data Source: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics, High School Longitudinal Study of 2009 (HSLS:09).</p> <p>2 *.05, **.01, ***.001.</p> <hd id="AN0183813417-8">Variables</hd> <p>We extracted variables that provide information about the sample (i.e., autism, race/ethnicity, gender, income), as well as those related to proximal context, self-efficacy, outcome expectations, and STEM major selection.[<reflink idref="bib10" id="ref33">10</reflink>]</p> <p>The variables that were used to define and describe samples were derived from the parent survey. In particular, the autism variable utilized came from the question "Doctor/school has told parent ninth grader has some form of autism." This variable, the only autism-specific one present in the HSLS:09 datasets, allowed us to compare differences between autistic and non-autistic student groups. Among respondents who indicated "yes" to the autism variable, 82% (<emph>SE</emph> = 3.7) reported receiving special education services or having an Individualized Education Program (IEP), 5% (<emph>SE</emph> = 1.9) reported not having an IEP, and 13% (<emph>SE</emph> = 3.3) did not disclose their IEP status (percentages rounded). Additional variables for sample descriptions included race/ethnicity, birth sex, and family income.</p> <p> <emph>Proximal contexts</emph>, as articulated by [<reflink idref="bib31" id="ref34">31</reflink>], are defined as the immediate influences or circumstances surrounding the decision-making process. These encompass both actual and perceived environmental facilitators and obstacles, all of which can potentially have a direct impact on career-related choices. This concept was represented in the HSLS datasets, particularly in the context of STEM-related interactions like parents' STEM occupation, and if the student had discussed STEM courses with people who served to influence them.</p> <p> <emph>Self-efficacy</emph>, as represented in the dataset, reflects students' confidence in their mathematical and scientific abilities. These were measured based on students' certainty in excelling in course tests and assignments, understanding complex textbook material, and mastering course-taught skills. Subsequently, composite variables were developed to represent students' self-efficacy in mathematics and science during their ninth and 11<sups>th</sups> grade years. The composite variables included in the HSLS:09 dataset were computed using a principal components factor analysis (PCA) with weights, which were standardized, having a mean of 0 and a standard deviation of 1, in accordance with the specifications outlined in NCES 2011-328 ([<reflink idref="bib27" id="ref35">27</reflink>]).</p> <p> <emph>Outcome expectations</emph>, in this context, refer to students' perceptions of the applicability and importance of mathematics and science to their future lives. To gauge these expectations, HSLS:09 included three key questions in 2009 and 2011, centered on whether the students believed the math and science course would be useful in their everyday life, if it would benefit them in college, and if it would prove valuable for their future career. Similar to the self-efficacy variables, the variables were standardized using PCA.</p> <p>Finally, the variable containing information about the students' majors is derived from their second student follow-up survey. To address the first research question, the variable related to the first major, X4RFMJSTNSF, was utilized. The variable is based on the National Science Foundation's (NSF) classification, derived from the U.S. Department of Education's Classification of Instructional Programs, 2010 edition (CIP 2010). For both the first and the third research questions, another binary variable was used to determine whether the first major was in an NSF-recognized STEM field, denoted as X4RFDGMJSTEM. Both variables, derived from the same survey question, are relate to the primary major students initially reported for their undergraduate studies.</p> <hd id="AN0183813417-9">Analytic Models</hd> <p>Research Questions 1 and 2 focus on describing the selection of STEM majors and the unique contexts within STEM fields for autistic college students compared to their non-autistic peers. Accordingly, we computed both weighted percentages and means. For these descriptive analyses, we used weights and stratified sampling information to account for the sampling process. For the third research question, we began by imputing the datasets before conducting our analysis using Structural Equation Modeling (SEM). Further details will follow.</p> <hd id="AN0183813417-10">Research Questions 1 and 2</hd> <p>To examine the distinct major pathways of autistic college students majoring in STEM fields (Research Question 1), we employed chi-square tests within SPSS Complex Samples, which facilitated the comparison in the variable <emph>majors</emph> between autistic and non-autistic students. To incorporate stratified sampling process into our descriptive analysis, we included strata, the Primary Sampling Unit (PSU), and respective weights for our analysis.[<reflink idref="bib11" id="ref36">11</reflink>] Standard errors were calculated to account for imbalances in the sample size.</p> <p>To address the second question regarding the differences in the characteristics of proximal context, self-efficacy, and outcome expectations, we utilized either chi-square tests or <emph>t</emph> tests for comparison between the two groups. For this specific analysis and later inference models, we limited our sample to individuals who had attended college, comprising 9,680 non-autistic and 80 autistic students which represent about.9% (<emph>SE</emph> =.159) of college students surveyed in 2016, 3 years after high school.</p> <hd id="AN0183813417-11">Research Question 3</hd> <p>The third research question, which examines the pathways to choosing STEM majors, was explored using SEM in AMOS. The overall percentage of missing data was 17.93%, thus we employed multiple imputation to produce 18 complete datasets, as suggested by [<reflink idref="bib67" id="ref37">67</reflink>]. Auxiliary variables, including the sampling stratum, PSU, and cross-wave weights (W4W1W2W3STU), were incorporated into the model as per [<reflink idref="bib48" id="ref38">48</reflink>]. Each imputed dataset was compared with original datasets in the SPSS Complex Samples module, both numerically and graphically, looking at histograms. To derive pooled estimates and standard errors for further analysis, the formula outlined in the AMOS 28 User's Guide ([<reflink idref="bib2" id="ref39">2</reflink>], pp. 487–488) was used.</p> <p>In our SEM models, we referenced Lent's social cognitive model (2003, p. 459). We considered three potential confounding variables: Non-Hispanic White (NHW), Sex, and Income, which are theoretically influential for STEM pathways (e.g., [<reflink idref="bib11" id="ref40">11</reflink>]; [<reflink idref="bib12" id="ref41">12</reflink>]; [<reflink idref="bib38" id="ref42">38</reflink>]). The latent variables were defined as Proximal Contexts (PC), Self-Efficacy (SE), and Outcome Expectations (OE). The original variables for PC include the STEM occupations of both the father and the mother, as well as whether the ninth grader talked to their father, mother, and friends about STEM courses.[<reflink idref="bib12" id="ref43">12</reflink>] For SE, four variables measuring mathematics and science self-efficacy, collected during the ninth and 11<sups>th</sups> grades, were included. For OE, mathematics and science utility scale variables during the ninth and 11<sups>th</sups> grades were included. Additionally, the final STEM pathway outcome, termed "Major," was incorporated.</p> <p>An a-priori sample size calculator for SEM ([<reflink idref="bib13" id="ref44">13</reflink>]; [<reflink idref="bib57" id="ref45">57</reflink>]; [<reflink idref="bib66" id="ref46">66</reflink>]) recommended a minimum sample size of 1,258 to adequately detect an effect size of at least 0.1. Our study's total sample size exceeded this threshold, ensuring a desired statistical power of 0.8 and a significance level of 0.05, while accounting for three latent variables and 13 observed variables in the SEM. Although the sample size for the autism group was smaller, it remained above the minimum of 89 required for estimating the SEM model structure. Given the significance of the research topic, we decided to proceed with the analysis. To ensure the measurement of each construct was unbiased across different autism status groups, we conducted Differential Item Functioning (DIF) analyses before constructing SEM models ([<reflink idref="bib62" id="ref47">62</reflink>]). DIF detection utilized model-based likelihood ratio tests, as recommended by [<reflink idref="bib59" id="ref48">59</reflink>]. Our approach to detecting nonuniform DIF—characterized by inconsistent variations in the values of each original variable across groups—involves comparing two models. The comprehensive model regresses the original variable on both the autism status and a composite score of other original variables for the latent variable, and then it includes an interaction term between the autism status and the composite score. This model is then compared to a corresponding model that excludes this interaction term. We conducted a log-likelihood ratio test against the chi-square distribution (1 degree of freedom) under the null hypothesis at the 5% level. The absence of significant differences led us to further compare the non-interaction model with a reduced model that contained only the composite score, aiming to identify uniform DIF—situations where the original variable differs consistently based on autism status. In this study, no significant differences were observed in these comparisons, suggesting that the original variables functioned equivalently across groups.</p> <p>Then, latent variables were constructed using confirmatory factor analysis within the omnibus model (CFA; [<reflink idref="bib8" id="ref49">8</reflink>]). Using an unconstrained SEM framework in AMOS, we first assessed the entire college student body, making no distinctions based on autism diagnoses. After checking the model's fit (e.g., CFI, RMSEA) and ensuring its validity (convergent and discriminant), we advanced to a multigroup analysis. Initially, to verify that the factor structure was consistent across groups, we conducted a CFA for each group ([<reflink idref="bib10" id="ref50">10</reflink>]). Once validated, we drew paths to represent the theoretical relationships and potential disparities between the groups. There were 18 paths included in this model: NHW to PC, SE, OE, and Major; Sex to PC, SE, OE, and Major; Income to PC, SE, OE, and Major; PC to SE, OE, and Major; SE to OE and Major; and finally, OE to Major. Pairwise comparisons further facilitated the identification of differences in effects between the groups ([<reflink idref="bib28" id="ref51">28</reflink>]).</p> <hd id="AN0183813417-12">Results</hd> <p>Within the HSLS:09 respondents, autistic students exhibited a heightened preference for majors within the STEM fields, particularly in science, engineering, and mathematics, despite their lower overall college enrollment rates. They engaged more in STEM discussions with their parents and less with peers compared to their non-autistic peers. In ninth grade, autistic students exhibited stronger STEM self-efficacy and outcome expectations than their non-autistic counterparts. By the 11<sups>th</sups> grade, however, while their confidence and interest in science persisted, their self-efficacy and expectations in mathematics had diminished. Further SEM models highlighted the differences in the influence of proximal context on major choices depending on the students' autism status. Furthermore, self-efficacy exerted a more substantial influence on the selection of STEM majors among autistic students, both directly and indirectly through outcome expectations.</p> <hd id="AN0183813417-13">Postsecondary Education and STEM Major</hd> <p>Out of all the students surveyed, 75.4% (<emph>SE</emph> =.8) went to college, and 22% (<emph>SE</emph> =.6) of those college students chose a STEM major. A closer examination of autistic students, as depicted in Figure 1, reveals a significantly lower propensity towards college attendance. Additionally, autistic students were less likely to attend 4-year institutions and more likely to enroll in 2-year colleges, or engage in shorter educational programs.</p> <p>Graph: Figure 1. College Attendance and Type of Postsecondary Institution by Autism Status (%)</p> <p>A larger proportion of autistic students among those who were surveyed chose STEM majors compared to their non-autistic counterparts, while the Chi-Square test indicated no significant difference (Figure 2). The graph shows the percentage distribution of majors, categorized as per the National Science Foundation (NSF) classifications. Notably, autistic students demonstrated a stronger inclination towards science, engineering, and mathematics majors while being less inclined to pursue majors in psychology and health-related fields. However, the substantial standard errors for the autistic group signify considerable variation in these estimates.</p> <p>Graph: Figure 2. Distribution of NSF STEM Majors Among College Students by Autism Status (%)</p> <hd id="AN0183813417-14">Distinct Characteristics</hd> <p>Table 2 presents the differences in college students' proximal contexts, STEM self-efficacy, and outcome expectations based on their autism status. Since the variables defining proximal context are dichotomous, the mean also represents the percentage of each variable. For example, while fathers of autistic students were less likely to be employed in STEM fields compared to their non-autistic peers, autistic students' mothers showed a higher likelihood of STEM professions. With this, autistic students were more prone to discussing STEM courses with their parents, but, conversely, they were less inclined to have such conversations with their peers. It 's noteworthy that the only statistically significant difference between the two groups observed was the frequency of discussions about STEM courses between autistic and non-autistic students with their friends.</p> <p>Table 2. STEM Pathway Components Among College Students by Autism Status</p> <p>Graph</p> <p></p> <p> <ephtml> <table><colgroup><col align="left" /><col align="left" /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /></colgroup><thead><tr><th align="left" /><th align="left" /><th align="left" /><th align="left" /><th align="left" colspan="2">Total</th><th align="left" colspan="2">ASD</th><th align="left" colspan="2">Non-ASD</th></tr><tr><th align="left" /><th align="left" /><th align="left">Min</th><th align="left">Max</th><th align="left">X̄</th><th align="left"><italic>SE</italic></th><th align="left">X̄</th><th align="left"><italic>SE</italic></th><th align="left">X̄</th><th align="left"><italic>SE</italic></th></tr></thead><tbody><tr><td>PC</td><td>DadSTEM</td><td>0</td><td>1</td><td>.152</td><td>.006</td><td>.119</td><td>.044</td><td>.167</td><td>.006</td></tr><tr><td /><td>MomSTEM</td><td>0</td><td>1</td><td>.168</td><td>.005</td><td>.172</td><td>.055</td><td>.153</td><td>.005</td></tr><tr><td /><td>DadTalk</td><td>0</td><td>1</td><td>.447</td><td>.008</td><td>.490</td><td>.101</td><td>.411</td><td>.008</td></tr><tr><td /><td>MomTalk</td><td>0</td><td>1</td><td>.583</td><td>.009</td><td>.599</td><td>.090</td><td>.551</td><td>.009</td></tr><tr><td /><td>FriendTalk</td><td>0</td><td>1</td><td>.269</td><td>.006</td><td>.249</td><td>.112</td><td>.322*</td><td>.007</td></tr><tr><td>SE</td><td>SEM1</td><td>−2.92</td><td>1.62</td><td>.260</td><td>.081</td><td>.367</td><td>.163</td><td>.153***</td><td>.020</td></tr><tr><td /><td>SES1</td><td>−2.91</td><td>1.83</td><td>.276</td><td>.068</td><td>.436</td><td>.133</td><td>.116***</td><td>.018</td></tr><tr><td /><td>SEM2</td><td>−2.50</td><td>1.73</td><td>.086</td><td>.070</td><td>.064</td><td>.139</td><td>.108***</td><td>.015</td></tr><tr><td /><td>SES2</td><td>−2.47</td><td>1.64</td><td>.149</td><td>.098</td><td>.215</td><td>.194</td><td>.084***</td><td>.019</td></tr><tr><td>OE</td><td>OEM1</td><td>−3.51</td><td>1.31</td><td>.055</td><td>.052</td><td>.121</td><td>.103</td><td>−.010**</td><td>.017</td></tr><tr><td /><td>OES1</td><td>−3.10</td><td>1.69</td><td>.148</td><td>.073</td><td>.245</td><td>.143</td><td>.052**</td><td>.022</td></tr><tr><td /><td>OEM2</td><td>−3.94</td><td>1.21</td><td>−.029</td><td>.070</td><td>−.112</td><td>.140</td><td>.062***</td><td>.017</td></tr><tr><td /><td>OES2</td><td>−.23</td><td>.1</td><td>.007</td><td>.008</td><td>.008</td><td>.016</td><td>.006***</td><td>.002</td></tr></tbody></table> </ephtml> </p> <ulist> <item>3 <emph>Note.</emph> PC = proximal context, SE = self-efficacy, SEM = self-efficacy in math, SES = self-efficacy in science, OE = outcome expectations, OEM = outcome expectations in math, OES = outcome expectations in science. Source: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics, High School Longitudinal Study of 2009 (HSLS:09).</item> <item>4 *.05, **.01 ***.001.</item> </ulist> <p>Regarding STEM self-efficacy, distinct differences were observed across all variables between the two groups. In ninth grade, autistic students displayed greater confidence in both math and science compared to their non-autistic peers. By 11<sups>th</sups> grade, although their science self-efficacy remained higher, their math self-efficacy decreased below that of the non-autistic students.</p> <p>The outcome expectations presented a nuanced picture similar to that of self-efficacy. In ninth grade, autistic students perceived the utility of math courses in college, jobs, and life as higher than their non-autistic peers. Yet, by 11<sups>th</sups> grade, this perception not only decreased but also fell below that of their non-autistic counterparts. Conversely, their expectations for the utility of science remained higher in both ninth and 11th grades.</p> <p>A distinct pattern emerged when examining both self-efficacy and outcome expectations: in ninth grade, autistic students displayed higher math self-efficacy and outcome expectations than their non-autistic peers. However, by 11th grade, both metrics had declined for autistic students to the point where their scores were lower than those of non-autistic peers. Conversely, science self-efficacy and outcome expectations for autistic students consistently outpaced those of their non-autistic peers across both periods. It 's essential to interpret these findings with caution, as the sample is drawn from college attendees and may not reflect the broader autistic population.</p> <hd id="AN0183813417-15">From STEM Self-Efficacy to Major Selection</hd> <p>We examined SEM models to discern the pathways connecting proximal context, self-efficacy, and outcome expectations to the final selection of college major. The omnibus model returned a CFI of.995 and an RMSEA of.029, suggesting a satisfactory fit as per [<reflink idref="bib26" id="ref52">26</reflink>]. In subsequent multi-group models (CFI =.963, RMSEA =.03), the shared variance of latent variables exhibited acceptable measures. While some original variables exhibited factor loadings below 0.3 (e.g., mother's STEM occupation for the autistic group, and math outcome expectation in 11th grade for non-autistic groups), they were retained in the model due to their theoretical significance ([<reflink idref="bib23" id="ref53">23</reflink>]).</p> <p>The analysis of construct/composite reliability (CR) and average variance extracted (AVE) reaffirmed the reliability and validity of the low factor loading items. CR, which measures the consistency of latent constructs through multiple original variables, showed that our constructs were reliably measured, with a threshold deemed satisfactory at.7 or above ([<reflink idref="bib23" id="ref54">23</reflink>], p. 619). AVE, indicating the proportion of variance a construct captures relative to the variance from measurement error, affirmed convergent validity with a threshold of.5 or higher (Fornell & Larcker, [<reflink idref="bib19" id="ref55">19</reflink>], p. 46). These measures validated our choice to maintain these items in the model, as detailed in Table 3. Each construct demonstrated satisfactory discriminant validity as per the Fornell-Larcker criterion, wherein the square root of the AVE for each construct should be greater than its correlations with all other constructs in the model, ensuring the distinctiveness of each construct (refer to Table 4).</p> <p>Table 3. Convergent Validity</p> <p>Graph</p> <p></p> <p> <ephtml> <table><colgroup><col align="left" /><col align="left" /><col align="left" /><col align="char" char="(" /><col align="char" char="(" /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /></colgroup><thead><tr><th align="left" /><th align="left" /><th align="left" /><th align="left" colspan="2">Estimates (SE)</th><th align="left" colspan="2">CR</th><th align="left" colspan="2">AVE</th></tr><tr><th align="left" /><th align="left" /><th align="left" /><th align="left">ASD</th><th align="left">NASD</th><th align="left">ASD</th><th align="left">NASD</th><th align="left">ASD</th><th align="left">NASD</th></tr></thead><tbody><tr><td>PC</td><td>→</td><td>DadSTEM</td><td>.427 (0.114)</td><td>.408 (0.74)</td><td>.946</td><td>.915</td><td>.797</td><td>.702</td></tr><tr><td /><td>→</td><td>MomSTEM</td><td>.294 (0.135)</td><td>.332 (0.72)</td><td /><td /><td /><td /></tr><tr><td /><td>→</td><td>DadTalk</td><td>.759 (0.203)</td><td>.745 (0.138)</td><td /><td /><td /><td /></tr><tr><td /><td>→</td><td>MomTalk</td><td>.63 (0.125)</td><td>.471 (0.059)</td><td /><td /><td /><td /></tr><tr><td /><td>→</td><td>FrndTalk</td><td>.403 (0.13)</td><td>.362 (0.074)</td><td /><td /><td /><td /></tr><tr><td>SE</td><td>→</td><td>SEM1</td><td>.721 (0.213)</td><td>.741 (0.114)</td><td>.902</td><td>.924</td><td>.715</td><td>.764</td></tr><tr><td /><td>→</td><td>SES1</td><td>.462 (0.17)</td><td>.385 (0.072)</td><td /><td /><td /><td /></tr><tr><td /><td>→</td><td>SEM2</td><td>.278 (0.144)</td><td>.406 (0.071)</td><td /><td /><td /><td /></tr><tr><td /><td>→</td><td>SES2</td><td>.451 (0.298)</td><td>.499 (0.094)</td><td /><td /><td /><td /></tr><tr><td>OE</td><td>→</td><td>OEM1</td><td>.611 (0.091)</td><td>.423 (0.054)</td><td>.938</td><td>.950</td><td>.798</td><td>.850</td></tr><tr><td /><td>→</td><td>OES1</td><td>.627 (0.162)</td><td>.418 (0.034)</td><td /><td /><td /><td /></tr><tr><td /><td>→</td><td>OEM2</td><td>.615 (0.164)</td><td>.141 (0.032)</td><td /><td /><td /><td /></tr><tr><td /><td>→</td><td>OES2</td><td>.715 (0.2)</td><td>.709 (0.031)</td><td /><td /><td /><td /></tr></tbody></table> </ephtml> </p> <p>5 <emph>Note</emph>. CR (composite/construct reliability) assesses the consistency of latent constructs as measured by multiple indicators. AVE (average variance extracted) measures the amount of variance that a construct captures as opposed to the variance stemming from measurement error. Source: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics, High School Longitudinal Study of 2009 (HSLS:09).</p> <p>Table 4. Discriminant Validity</p> <p>Graph</p> <p></p> <p> <ephtml> <table><colgroup><col align="left" /><col align="center" /><col align="char" char="." /><col align="char" char="." /><col align="char" char="." /></colgroup><thead><tr><th align="left" /><th align="left" /><th align="left">PC</th><th align="left">SE</th><th align="left">OE</th></tr></thead><tbody><tr><td>ASD</td><td>PC</td><td><bold>.893</bold></td><td /><td /></tr><tr><td /><td>SE</td><td>.252</td><td>.<bold>846</bold></td><td /></tr><tr><td /><td>OE</td><td>.348</td><td>.232</td><td>.<bold>893</bold></td></tr><tr><td>NASD</td><td>PC</td><td>.<bold>838</bold></td><td /><td /></tr><tr><td /><td>SE</td><td>.325</td><td>.<bold>874</bold></td><td /></tr><tr><td /><td>OE</td><td>.040</td><td>.149</td><td>.<bold>922</bold></td></tr></tbody></table> </ephtml> </p> <p>6 <emph>Note</emph>. Bold figures represent the square root of the AVE. Non-bold figures represent the correlation coefficients between different constructs. Source: U.S. Department of Education, Institute of Education Sciences, National Center for Education Statistics, High School Longitudinal Study of 2009 (HSLS:09).</p> <p>Figure 3 shows the varied STEM major selection pathways for autistic students compared to their non-autistic counterparts. There were two confounding variables that explain those paths (income was not depicted in the figure as it was not significant). Being NHW positively influenced proximal context, outcome expectations, and major for both groups. Sex, on the other hand, displays a nuanced role. It directly yet negatively connects with STEM outcome expectations and major, and, notably, being female explains outcome expectation only for the non-autistic group. Interestingly, the figure does not show direct associations between NHW or Sex and self-efficacy. However, an indirect link from NHW to self-efficacy via proximal context is present but exclusive to the non-autistic group.</p> <p>Graph: Figure 3. STEM Major Choice Pathways by Autism Status</p> <p>The non-autistic group's STEM major choices are influenced by proximal context both directly and indirectly via self-efficacy. The role of self-efficacy in shaping both outcome expectations and ultimate major selection is evident in both groups, with a more pronounced impact observed in autistic students. This impact is underlined by the strong effects from self-efficacy—gauged by students' confidence in excelling in exams and assignments, comprehending complex textbook content, and mastering skills taught in courses—to major. However, outcome expectations do not seem to have a significant direct influence on STEM major choices. As such, the selection of a STEM major is determined by proximal context and self-efficacy, with NHW and Sex serving as key confounding variables.</p> <p>In summary, the impact of proximal context on majors—whether directly or indirectly through self-efficacy—was found to be significant only for the non-autistic group. Self-efficacy emerges as a pivotal element in these STEM pathways, influencing both outcome expectation and major for all students. On the other hand, direct influence of outcome expectations on major selection remains undisclosed in this model.</p> <hd id="AN0183813417-16">Discussion</hd> <p>This study explored the role of contextual supports, self-efficacy, and outcome expectations, as outlined by [<reflink idref="bib3" id="ref56">3</reflink>] and [<reflink idref="bib31" id="ref57">31</reflink>], and its influence on STEM major selection, particularly among autistic students. The results provide potential explanations for their inclination towards STEM fields. Recent explorations—such as the works of [<reflink idref="bib7" id="ref58">7</reflink>])—provide a glimpse into the potential of autistic individuals in STEM domains, emphasizing the need for more nuanced and inclusive educational strategies and interventions tailored to their unique needs and potentials. This discussion offers a preview of the key areas to target, including proximal context, self-efficacy, and outcome expectations, which exert significant influences on the choice of STEM majors. It also examines the intricate interplay between those constructs, along with the implications of these factors within the context of postsecondary transition within the U.S. education framework.</p> <hd id="AN0183813417-17">Preferences in STEM Among Autistic Students</hd> <p>Consistent with prior research ([<reflink idref="bib65" id="ref59">65</reflink>]), this investigation also observed a heightened inclination towards STEM fields in autistic students, even though their college attendance rates were lower compared to their non-autistic counterparts. One potential explanation suggested by our findings is the significant effect of STEM self-efficacy on major choice. Notably, autistic students displayed higher levels of self-efficacy in their STEM abilities compared to non-autistic students. While the connection between autism and STEM-related strengths—particularly as it relates to mathematics—has been extensively discussed, recent literature has presented diverging viewpoints on mathematical prowess among autistic individuals (e.g., [<reflink idref="bib45" id="ref60">45</reflink>]; [<reflink idref="bib60" id="ref61">60</reflink>]). STEM self-efficacy, therefore, might be a compelling factor in explaining the pronounced STEM preferences seen in those on the autism spectrum. However, note that the self-efficacy scores were reflective of only approximately 55% of autistic students who eventually pursued higher education in this survey.</p> <hd id="AN0183813417-18">The Intersectionality of Race, Sex, and Autism</hd> <p>[<reflink idref="bib31" id="ref62">31</reflink>] initially proposed that Person Inputs—such as race and gender—could influence career decisions like STEM major selection, mediated by the proximal context. In this study, race and income served as confounding variables, consistent with a significant body of literature that highlights disparities in STEM related to race and gender. These disparities are evident in several domains, from interactions with leading figures in STEM ([<reflink idref="bib37" id="ref63">37</reflink>]) and self-efficacy beliefs ([<reflink idref="bib72" id="ref64">72</reflink>]), to outcome expectations ([<reflink idref="bib30" id="ref65">30</reflink>]) and choices of STEM disciplines ([<reflink idref="bib39" id="ref66">39</reflink>]). The results of this study provided nuanced explanations for the social cognitive model by incorporating these personal factors.</p> <p>Specifically, the influence of NHW status was significant for both autistic and non-autistic groups, albeit with some variance. This suggests that White male students, irrespective of their autism status, were more likely to have guiding figures in STEM, stronger STEM outcome expectations, and a preference for STEM majors. These findings align with consistent statistical evidence pointing to the overrepresentation of White males in STEM (e.g., [<reflink idref="bib1" id="ref67">1</reflink>]; [<reflink idref="bib41" id="ref68">41</reflink>]). [<reflink idref="bib49" id="ref69">49</reflink>] research offers potential explanations for this disparity, such as sufficient STEM preparation in high school. Their results indicated that White males were more likely to exhibit confidence in math—mirroring the math self-efficacy measures in this study—and were better prepared academically for STEM, as reflected in their course levels and math test scores. The dynamics between race and gender in STEM have been extensively researched, yet research on the intersection with autism is in the nascent stage.</p> <p>Moreover, there is a recognized overrepresentation of White males in the autistic population. Distinctly, female autistic children of color often exhibit more severe symptoms, including co-occurring intellectual disabilities ([<reflink idref="bib18" id="ref70">18</reflink>]) and notable deficits in executive functioning ([<reflink idref="bib29" id="ref71">29</reflink>]). Given that autistic students are predominantly White males, these variations in autism profiles might partly elucidate the disparities in the links found between NHW status and proximal context or major choices, and between sex and major choices among autistic students, compared to their non-autistic peers. While those differences were not statistically significant, it underscores the need for deeper exploration in future research.</p> <hd id="AN0183813417-19">Effects of Influential Others on STEM Major Choice</hd> <p>In our study, the proximal contexts, including parents' STEM occupations, were associated with self-efficacy solely within the non-autistic group. Although the connection between proximal contexts and the STEM major for the autistic group was not explicitly illustrated in the model, the differences between the groups were noteworthy. For instance, autistic students reported fewer fathers in STEM occupations but more mothers in these fields than their non-autistic counterparts. Over the past 2 decades, the overrepresentation of autistic children in Silicon Valley has gained significant attention, often termed "The Geek Syndrome." This phenomenon, which links autistic children to their parents—particularly those in STEM fields—has garnered media and high-tech community interest, with prominent figures like Temple Grandin echoing similar observations ([<reflink idref="bib14" id="ref72">14</reflink>]; [<reflink idref="bib54" id="ref73">54</reflink>]). Further, [<reflink idref="bib70" id="ref74">70</reflink>] identified that mothers of autistic children were slightly more inclined to have engineering jobs compared to mothers of non-autistic children, although such a distinction was negligible for fathers. The connection with mothers can be traced from early autism research to the present, as researchers consistently attempt to unveil the underlying reasons for this association. Our study's findings also mirrored these patterns related to parents' professions. However, note that these differences in this study were not statistically significant, and the ramifications of parents' STEM backgrounds were beyond the scope of this study. Still, these patterns necessitate further investigation, especially the relationship between parents' occupational backgrounds and STEM self-efficacy in autistic students.</p> <p>In terms of influences of significant others, our data indicates that autistic students engaged significantly less in STEM-related discussions with their peers. It remains uncertain if this reflects a blend of STEM inclinations and the characteristic social communication challenges of the autistic students. However, this finding underscores the need to reevaluate how educational settings can leverage the effect of STEM dialogue to bolster autistic students' aspirations to pursue STEM fields. For instance, peer-mediated interventions might not only enhance STEM competencies (e.g., [<reflink idref="bib24" id="ref75">24</reflink>]; [<reflink idref="bib56" id="ref76">56</reflink>]); they might also cultivate an environment where autistic students can openly express their interests, form genuine friendships, and possibly bridge these connections to their higher education pursuits. Moreover, as our data shows that some autistic students do not participate in special education programs, focusing solely on supports for students with IEPs may overlook the broader spectrum of the autistic population. This diversity calls for enhanced interagency collaboration among families, local STEM communities, and educational institutions to support transition-aged students in both formal and informal educational settings ([<reflink idref="bib50" id="ref77">50</reflink>]; [<reflink idref="bib25" id="ref78">25</reflink>]). Such universal approaches to partnerships can facilitate access to resources, mentorship, and STEM enrichment opportunities outside traditional special education settings, ensuring that autistic students receive the genuine support needed to foster their STEM efficacy based on their interests rather than their disability category.</p> <hd id="AN0183813417-20">STEM Self-Efficacy and Autism</hd> <p>Self-efficacy, defined by ([<reflink idref="bib3" id="ref79">3</reflink>]) as an individual's perceived capacity to perform tasks to achieve desired outcomes, is instrumental in forecasting persistence and performance in STEM fields. Extensive studies (e.g., [<reflink idref="bib35" id="ref80">35</reflink>]; [<reflink idref="bib58" id="ref81">58</reflink>]) have investigated disparities in self-efficacy in STEM, focusing predominantly on gender and race/ethnicity, but research relating to students with an IEP—especially autistic students—remains limited. Notably, findings indicate that autistic students who pursued college education experienced a decline in math self-efficacy from ninth to 11<sups>th</sups> grade, to levels significantly lower than their non-autistic peers. [<reflink idref="bib20" id="ref82">20</reflink>] observed that "high-functioning students with ASD" aged 13 to 17 exhibited slightly elevated self-efficacy beliefs in mathematics achievement goals compared to their "typically developing" (p. 100) counterparts. Additionally, the American Association of University Women (1991) revealed a particular decline in math and science confidence among girls transitioning from elementary to high school. Yet, a longitudinal examination of changes in STEM self-efficacy among autistic students, including the mechanisms driving such decreases, remains unaddressed. In this regard, this study broadens the discourse on disparities in self-efficacy trends and highlights the significant impact of STEM self-efficacy on major selection, thus providing a potential avenue to explore potential disparities in STEM self-efficacy.</p> <p>Recent studies, such as those by [<reflink idref="bib7" id="ref83">7</reflink>], spotlight the inherent abilities of autistic students in STEM, leading to increasing demands for refined instructional approaches. Evidence-based interventions and methodologies to foster specific self-efficacy for transition-aged students in general are well-studied, revealing potential strategies to develop tailored approaches for autistic students. For instance, given the reciprocal relationship between self-efficacy and academic performance, a curriculum focused on providing mastery experiences can improve STEM self-efficacy (e.g., [<reflink idref="bib17" id="ref84">17</reflink>]; [<reflink idref="bib51" id="ref85">51</reflink>]) which, in turn, can influence their decision to pursue STEM majors. While the exploration of STEM career pathways and related intervention studies for autistic students is still in its early stages, an increasing number of studies are experimenting with the effects of hands-on activities to enhance STEM outcomes for autistic students. For instance, [<reflink idref="bib40" id="ref86">40</reflink>] found that their inclusive maker program was proven effective in elevating STEM self-efficacy among autistic students, offering insights to enhance their involvement in STEM careers. Given the observed decrease in math self-efficacy for autistic students during their 11<sups>th</sups> grade, this insight underscores the potentials of creating targeted STEM self-efficacy interventions for those with neurodiverse learners, in line with the suggestions of [<reflink idref="bib16" id="ref87">16</reflink>] for promoting diversity and inclusivity in self-efficacy interventions.</p> <hd id="AN0183813417-21">Effect of Outcome Expectations on STEM Major Selection</hd> <p>Although our model does not depict the route linking self-efficacy to major selection via outcome expectations, recent studies provide an optimistic glimpse into the potential of further research on this construct (e.g., [<reflink idref="bib40" id="ref88">40</reflink>]; [<reflink idref="bib64" id="ref89">64</reflink>]). For autistic students in both ninth and 11<sups>th</sups> grades, there is a consistent demonstration of higher outcome expectations in both math and science, except for 11<sups>th</sups> grade math, where both self-efficacy and outcome expectations exhibit a decline. This trend necessitates further investigation about the underlying causes and potential interventions to address these declines during their high school years, particularly in mathematics.</p> <hd id="AN0183813417-22">Implications for Transition Education</hd> <p>The emphasis of current transition education lies in catering to the individual preferences, interests, needs, and strengths of students (Individuals with Disabilities Education Act [IDEA], 2004). Given the recognition that enhancing self-efficacy can positively influence autistic students' pursuit of STEM careers, it becomes imperative for educators to leverage student strengths in ways that boost their self-efficacy. Although research specifically targeting autistic students in this domain is in its early stages, the growing body of literature demonstrates interventions that can improve self-efficacy and career prospects. For example, [<reflink idref="bib53" id="ref90">53</reflink>] discovered that their Motivational Enhancement Group Intervention provided high school students with disabilities opportunities to explore career options and improved vocational skills, self-efficacy, self-determination, and outcome expectations among these students. Additionally, substantial research has documented methods to enhance STEM self-efficacy among high school students in the general population, illustrating potential applications for autistic students aspiring to pursue STEM careers. This includes the effects of performance accomplishments ([<reflink idref="bib34" id="ref91">34</reflink>]) and the impact of writing about the utility of mathematics ([<reflink idref="bib9" id="ref92">9</reflink>]). The analysis also highlighted how the intersection of autism and other characteristics influence college readiness, aligning with findings from other studies, such as those by [<reflink idref="bib32" id="ref93">32</reflink>]. These insights support the potential benefits of implementing school-wide college readiness systems, particularly those based on the Multi-Tiered Systems of Support (MTSS) framework, as discussed by [<reflink idref="bib42" id="ref94">42</reflink>]. This framework emphasizes how schoolwide support within secondary schools can be tailored to meet diverse learning needs, thereby enhancing students' preparation for postsecondary education. In addition to the exiting research, this study's insights into unique pathways for education provide critical information for educators and policymakers, advocating for a heightened focus on initiatives aimed at strengthening self-efficacy and shaping career aspirations among neurodiverse students.</p> <hd id="AN0183813417-23">Limitations</hd> <p>The inherent limitation of a secondary analysis of a large national survey is the reliance on datasets collected for broader fields, which may not always support subgroup analysis due to limited sample sizes. This limitation is also evident in our study, as our sample size constraints made it unfeasible to identify specific STEM majors, such as computer science, or include variables related to STEM goals and interests. Furthermore, the disparities in standard errors between groups, as highlighted in the tables, underscore the constraints imposed by the smaller sample sizes within the autistic group. Additionally, comprehensive analysis incorporating various learning environment variables among autistic students using HSLS:09 data has been limited. Given the recent emphasis on diversifying the STEM workforce, including individuals with disabilities, one potential solution to overcome this limitation would be to oversample minoritized populations in the survey. Another survey, the National Longitudinal Transition Study, which specifically focuses on students with disabilities, also has the potential to provide valuable insights into the career pathways of autistic students. With the anticipated release of upcoming outcome data, particularly NLTS2012—expected in 2024—there may be another opportunity to utilize the more recent dataset for a more in-depth examination of the specific STEM majors and career trajectories of autistic students.</p> <hd id="AN0183813417-24">Summary</hd> <p>This study presents the postsecondary STEM pathways of autistic students. Despite a lesser inclination to attend college, autistic students are more likely to select STEM majors. Possessing higher self-efficacy and its significant influence on final major choice may contribute to this heightened preference for STEM majors among autistic students. Recognizing and leveraging the unique skills and proficiencies of autistic students could bring in more meaningful career opportunities and economic benefits. Thus, these findings suggest the potential outlets for adjustments in future classroom designs to better support college-bound autistic students in planning for STEM careers.</p> <ref id="AN0183813417-25"> <title> References </title> <blist> <bibl id="bib1" idref="ref26" type="bt">1</bibl> <bibtext> American Society for Engineering Education. (2016). Engineering by the numbers: Asee retention and time-to-graduation benchmarks for undergraduate engineering schools, departments and programs. Brian l. Yoder.</bibtext> </blist> <blist> <bibl id="bib2" idref="ref27" type="bt">2</bibl> <bibtext> Arbuckle J. L. (2021). IBM® SPSS® Amos<sups>TM</sups> 25 User's Guide.</bibtext> </blist> <blist> <bibl id="bib3" idref="ref18" type="bt">3</bibl> <bibtext> Bandura A. (1978). Self-efficacy: Toward a unifying theory of behavioral change. Advances in Behaviour Research and Therapy, 1(4), 139–161. https://doi.org/10.1016/0146-6402(78)90002-4</bibtext> </blist> <blist> <bibl id="bib4" idref="ref23" type="bt">4</bibl> <bibtext> Bandura A. (1986). Social foundations of thought and action: A social cognitive theory. Prentice-Hall, Inc.</bibtext> </blist> <blist> <bibl id="bib5" idref="ref20" type="bt">5</bibl> <bibtext> Bandura A., Barbaranelli C., Caprara G. V., Pastorelli C. (1996). Multifaceted impact of self-efficacy beliefs on academic functioning. Child Development, 67(3), 1206–1222. https://doi.org/10.2307/1131888</bibtext> </blist> <blist> <bibl id="bib6" idref="ref2" type="bt">6</bibl> <bibtext> Baron-Cohen S. (2009). Autism: The empathizing-systemizing (E-S) theory. Annals of the New York Academy of Sciences, 1156(1), 68–80. https://doi.org/10.1111/j.1749-6632.2009.04467.x</bibtext> </blist> <blist> <bibl id="bib7" idref="ref5" type="bt">7</bibl> <bibtext> Baron-Cohen S., Lombardo M. V. (2017). Autism and talent: The cognitive and neural basis of systemizing. Dialogues in Clinical Neuroscience, 19(4), 345–353. https://doi.org/10.31887/DCNS.2017.19.4/sbaroncohen</bibtext> </blist> <blist> <bibl id="bib8" idref="ref30" type="bt">8</bibl> <bibtext> Bollen K. A. (1989). Structural equations with latent variables. John Wiley & Sons. https://doi.org/10.1002/9781118619179</bibtext> </blist> <blist> <bibl id="bib9" idref="ref31" type="bt">9</bibl> <bibtext> Brisson B. M., Dicke A.-L., Gaspard H., Häfner I., Flunger B., Nagengast B., Trautwein U. (2017). Short intervention, sustained effects: Promoting Students' math competence beliefs, effort, and achievement. American Educational Research Journal, 54(6), 1048–1078. https://doi.org/10.3102/0002831217716084</bibtext> </blist> <blist> <bibtext> Brown T. A. (2015). Confirmatory factor analysis for applied research, 2nd ed. The Guilford Press.</bibtext> </blist> <blist> <bibtext> Chen X. (2013). Stem attrition: College students' paths into and out of stem fields. Statistical Analysis Report. NCES 2014-001. In National Center for Education Statistics. National Center for Education Statistics. https://eric.ed.gov/?id=ED544470.</bibtext> </blist> <blist> <bibtext> Cheryan S., Ziegler S. A., Montoya A. K., Jiang L. (2017). Why are some STEM fields more gender balanced than others? Psychological Bulletin, 143(1), 1–35. https://doi.org/10.1037/bul0000052</bibtext> </blist> <blist> <bibtext> Cohen J. (1988). Statistical power analysis for the behavioral sciences (2nd Ed.). Lawrence Earlbaum Associates.</bibtext> </blist> <blist> <bibtext> Davenport M., Mazurek M., Brown A., McCollom E. (2018). A systematic review of cultural considerations and adaptation of social skills interventions for individuals with autism spectrum disorder. Research in Autism Spectrum Disorders, 52, 23–33. https://doi.org/10.1016/j.rasd.2018.05.003</bibtext> </blist> <blist> <bibtext> Eccles J. (2009). Who am I and what am I going to do with my life? Personal and collective identities as motivators of action. Educational Psychologist, 44(2), 78–89. https://doi.org/10.1080/00461520902832368</bibtext> </blist> <blist> <bibtext> Falco L. D., Summers J. J. (2019). Improving career decision self-efficacy and STEM self-efficacy in high school girls: Evaluation of an intervention. Journal of Career Development, 46(1), 62–76. https://doi.org/10.1177/0894845317721651</bibtext> </blist> <blist> <bibtext> Feldhausen R., Weese J. L., Bean N. H. (2018). Increasing student self-efficacy in computational thinking via STEM outreach programs. Proceedings of the 49th ACM Technical Symposium on Computer Science Education, 302–307. https://doi.org/10.1145/3159450.3159593</bibtext> </blist> <blist> <bibtext> Fombonne E. (2009). Epidemiology of pervasive developmental disorders. Pediatric Research, 65(6), 591–598. https://doi.org/10.1203/PDR.0b013e31819e7203</bibtext> </blist> <blist> <bibtext> Fornell C., Larcker D. F. (1981). Evaluating structural equation models with unobservable variables and measurement error. Journal of Marketing Research, 18(1), 39–50. https://doi.org/10.2307/3151312</bibtext> </blist> <blist> <bibtext> Georgiou A., Soulis S.-G., Rapti D. (2018). Motivation in mathematics of high functioning students with autism spectrum disorder. Psychology Research, 8(3), 96–106. https://doi.org/10.17265/2159-5542/2018.03.002</bibtext> </blist> <blist> <bibtext> Gillespie-Lynch K., Kapp S. K., Brooks P. J., Pickens J., Schwartzman B. (2017). Whose expertise is it? Evidence for autistic adults as critical autism experts. Frontiers in Psychology, 8, 438. https://doi.org/10.3389/fpsyg.2017.00438</bibtext> </blist> <blist> <bibtext> Grandin T., Panek R. (2013). The autistic brain: Thinking across the spectrum. Houghton Mifflin Harcourt.</bibtext> </blist> <blist> <bibtext> Hair J. F., Black W. C., Babin B. J., Anderson R. E. (2009). Multivariate data analysis (7th Ed.). Pearson.</bibtext> </blist> <blist> <bibtext> Heinrich S., Collins B. C., Knight V., Spriggs A. D. (2016). Embedded simultaneous prompting procedure to teach STEM content to high school students with moderate disabilities in an inclusive setting. Education and Training in Autism and Developmental Disabilities, 51(1), 41–54.</bibtext> </blist> <blist> <bibtext> Hendricks D. R., Wehman P. (2009). Transition from school to adulthood for youth with autism spectrum disorders: Review and recommendations. Focus on Autism and Other Developmental Disabilities, 24(2), 77–88. https://doi.org/10.1177/1088357608329827</bibtext> </blist> <blist> <bibtext> Hu L., Bentler P. M. (1995). Evaluating model fit. In Hoyle R. H. (Ed.), Structural equation modeling: Concepts, issues and application (pp. 77–99). SAGE Publications.</bibtext> </blist> <blist> <bibtext> Ingels S. J., Prattt D. J., Herget D. R., Burns L. J., Dever J. A., Ottem R., Rogers J. E., Jin Y., Leinwand S. (2011). High School Longitudinal Study of 2009 (HSLS:09): Base-Year Data File Documentation (NCES 2011-328). U.S. Department of Education. National Center for Education Statistics.</bibtext> </blist> <blist> <bibtext> Kline R. B. (2023). Principles and practice of structural equation modeling (Fifth Edition). Guilford Press.</bibtext> </blist> <blist> <bibtext> Lemon J. M., Gargaro B., Enticott P. G., Rinehart N. J. (2011). Executive functioning in autism spectrum disorders: A gender comparison of response inhibition. Journal of Autism and Developmental Disorders, 41(3), 352–356. https://doi.org/10.1007/s10803-010-1039-2</bibtext> </blist> <blist> <bibtext> Lent R. W., Brown S. D., Hackett G. (2000). Contextual Supports and Barriers to Career.</bibtext> </blist> <blist> <bibtext> Lent R. W., Brown S. D., Schmidt J., Brenner B., Lyons H., Treistman D. (2003). Relation of contextual supports and barriers to choice behavior in engineering majors: Test of alternative social cognitive models. Journal of Counseling Psychology, 50(4), 458–465. https://doi.org/10.1037/0022-0167.50.4.458</bibtext> </blist> <blist> <bibtext> Lombardi A. R., Rifenbark G. G., Hicks T. A., Taconet A., Challenger C. (2022). College and career readiness support for youth with and without disabilities based on the national longitudinal transition study 2012. Exceptional Children, 89(1), 5–21. https://doi.org/10.1177/00144029221088940</bibtext> </blist> <blist> <bibtext> Lombardi A. R., Rifenbark G. G., Taconet A. V. (2023). Quality indicators of secondary data analyses in special education research: A pre-registration guide. Exceptional Children, 89(4), 397–411. https://doi.org/10.1177/00144029221141029</bibtext> </blist> <blist> <bibtext> Luzzo D. A., Hasper P., Albert K. A., Bibby M. A., Martinelli Jr E. A. (1999). Effects of self-efficacy-enhancing interventions on the math/science self-efficacy and career interests, goals, and actions of career undecided college students. Journal of Counseling Psychology, 46(2), 233–243. https://doi.org/10.1037/0022-0167.46.2.233</bibtext> </blist> <blist> <bibtext> MacPhee D., Farro S., Canetto S. S. (2013). Academic self-efficacy and performance of underrepresented STEM majors: Gender, ethnic, and social class patterns. Analyses of Social Issues and Public Policy, 13(1), 347–369. https://doi.org/10.1111/asap.12033</bibtext> </blist> <blist> <bibtext> Maenner M. J. (2023). Prevalence and characteristics of autism spectrum disorder among children aged 8 years—Autism and Developmental Disabilities Monitoring Network, 11 Sites, United States, 2020. MMWR. Surveillance Summaries, 72(2), 1–14. https://doi.org/10.15585/mmwr.ss7202a1</bibtext> </blist> <blist> <bibtext> Malcolm S. M., Hall P. Q., Brown J. W. (1976). The double bind: The price of being a minority woman in science. American Association for the Advancement of Science.</bibtext> </blist> <blist> <bibtext> Malcom S. M., Malcom-Piqueux L. E. (2013). Critical mass revisited: Learning lessons from research on diversity in STEM fields. Educational Researcher, 42(3), 176–178. https://doi.org/10.3102/0013189X13486763</bibtext> </blist> <blist> <bibtext> Mann A., DiPrete T. A. (2013). Trends in gender segregation in the choice of science and engineering majors. Social Science Research, 42(6), 1519–1541. https://doi.org/10.1016/j.ssresearch.2013.07.002</bibtext> </blist> <blist> <bibtext> Martin W. B., Yu J., Wei X., Vidiksis R., Patten K. K., Riccio A. (2020). Promoting science, technology, and engineering self-efficacy and knowledge for all with an autism inclusion maker program. Frontiers in Education, 5, 1–18. https://<ulink href="http://www.frontiersin.org/articles/10.3389/feduc.2020.00075">www.frontiersin.org/articles/10.3389/feduc.2020.00075</ulink></bibtext> </blist> <blist> <bibtext> Mau W.-C. J. (2016). Characteristics of us students that pursued a STEM major and factors that predicted their persistence in degree completion. Universal Journal of Educational Research, 4(6), 1495–1500. https://doi.org/10.13189/ujer.2016.040630</bibtext> </blist> <blist> <bibtext> Morningstar M. E., Lombardi A. R., Test D. (2018). Including college and career readiness within a multi-tiered systems of support framework. AERA Open, 4(1), 1–11. https://doi.org/10.1177/2332858418761880</bibtext> </blist> <blist> <bibtext> Morsanyi K., Primi C., Handley S. J., Chiesi F., Galli S. (2012). Are systemizing and autistic traits related to talent and interest in mathematics and engineering? Testing some of the central claims of the empathizing-systemizing theory. British Journal of Psychology, 103(4), 472–496. https://doi.org/10.1111/j.2044-8295.2011.02089.x</bibtext> </blist> <blist> <bibtext> Newman L., Wagner M., Knokey A.-M., Marder C., Nagle K., Shaver D., Wei X., with Cameto R., Contreras E., Ferguson K., Greene S., Schwarting M. (2011). The post-high school outcomes of young adults with disabilities up to 8 years after high school. A Report From the National Longitudinal Transition Study-2 (NLTS2) (NCSER 2011-3005). SRI International. Available at <ulink href="http://www.nlts2.org/reports/">www.nlts2.org/reports/</ulink>.</bibtext> </blist> <blist> <bibtext> Oswald T. M., Beck J. S., Iosif A.-M., McCauley J. B., Gilhooly L. J., Matter J. C., Solomon M. (2016). Clinical and cognitive characteristics associated with mathematics problem solving in adolescents with autism spectrum disorder. Autism Research, 9(4), 480–490. https://doi.org/10.1002/aur.1524</bibtext> </blist> <blist> <bibtext> Pajares F. (1996). Self-efficacy beliefs in academic settings. Review of Educational Research, 66(4), 543–578. https://doi.org/10.3102/00346543066004543</bibtext> </blist> <blist> <bibtext> Pan N., Auyeung B., Wang X., Lin L.-Z., Li H.-L., Zhan X.-L., Jin C.-K., Jing J., Li X.-H. (2022). Empathizing, systemizing, empathizing-systemizing difference and their association with autistic traits in children with autism spectrum disorder, with and without intellectual disability. Autism Research, 15(7), 1348–1357. https://doi.org/10.1002/aur.2766</bibtext> </blist> <blist> <bibtext> Reiter J. P., Raghunathan T. E., Kinney S. K. (2006). The importance of modeling the sampling design in multiple imputation for missing data. Methodology, 32(2), 143–149.</bibtext> </blist> <blist> <bibtext> Riegle-Crumb C., King B. (2010). Questioning a white male advantage in STEM: Examining disparities in college major by gender and race/ethnicity. Educational Researcher, 39(9), 656–664. https://doi.org/10.3102/0013189X10391657</bibtext> </blist> <blist> <bibtext> Schall C., Wehman P. (2009). Understanding the transition from school to adulthood for students with autism. In Wehman P., Smith M. D., Schall C. (Eds.), Autism and the transition to adulthood: Success beyond the classroom (pp. 1–14). Paul H. Brookes Publishing Co.</bibtext> </blist> <blist> <bibtext> Schlegel R. J., Chu S. L., Chen K., Deuermeyer E., Christy A. G., Quek F. (2019). Making in the classroom: Longitudinal evidence of increases in self-efficacy and STEM possible selves over time. Computers & Education, 142, 103637. https://doi.org/10.1016/j.compedu.2019.103637</bibtext> </blist> <blist> <bibtext> Schunk D. H., Pajares F. (2009). Self-efficacy theory. In Handbook of motivation at school (pp. 35–53). Routledge/Taylor & Francis Group. https://doi.org/10.4324/9780203879498</bibtext> </blist> <blist> <bibtext> Sheftel, et al. (2014). Motivational enhancement career intervention for youth with disabilities. Advances in School Mental Health Promotion, 7(4), 208–224. https://doi.org/10.1080/1754730X.2014.949061</bibtext> </blist> <blist> <bibtext> Silberman S. (2001, December 1). The Geek Syndrome. Wired. https://<ulink href="http://www.wired.com/2001/12">www.wired.com/2001/12</ulink>.</bibtext> </blist> <blist> <bibtext> Singer J. (1998). NeuroDiversity the birth of an idea. Bolton.</bibtext> </blist> <blist> <bibtext> Smith B. R., Spooner F., Wood C. L. (2013). Using embedded computer-assisted explicit instruction to teach science to students with autism spectrum disorder. Research in Autism Spectrum Disorders, 7(3), 433–443. https://doi.org/10.1016/j.rasd.2012.10.010</bibtext> </blist> <blist> <bibtext> Soper D. S. (2024). A-priori sample size calculator for structural equation models [Software]. Available from https://<ulink href="http://www.danielsoper.com/statcalc">www.danielsoper.com/statcalc</ulink>.</bibtext> </blist> <blist> <bibtext> Tellhed U., Bäckström M., Björklund F. (2017). Will I fit in and do well? The importance of social belongingness and self-efficacy for explaining gender differences in interest in STEM and HEED majors. Sex Roles, 77(1), 86–96. https://doi.org/10.1007/s11199-016-0694-y</bibtext> </blist> <blist> <bibtext> Thissen D., Steinberg L., Wainer H. (1988). Use of item response theory in the study of group differences in trace lines. In Wainer H., Braun H. (Eds.), Test validity (pp. 147–169). Erlbaum.</bibtext> </blist> <blist> <bibtext> Tonizzi I., Usai M. C. (2023). Math abilities in autism spectrum disorder: A meta-analysis. Research in Developmental Disabilities, 139, 104559. https://doi.org/10.1016/j.ridd.2023.104559</bibtext> </blist> <blist> <bibtext> U.S. Department of Education. (2004). Individuals with Disabilities Education Act (IDEA), 20 U.S.C. § 1400 et seq. https://sites.ed.gov/idea/</bibtext> </blist> <blist> <bibtext> Walker C. M. (2011). What's the DIF? Why differential item functioning analyses are an important part of instrument development and validation. Journal of Psychoeducational Assessment, 29(4), 364–376. https://doi.org/10.1177/0734282911406666</bibtext> </blist> <blist> <bibtext> Wei X., Christiano E. R. A., Yu J. W., Blackorby J., Shattuck P., Newman L. A. (2014). Postsecondary pathways and persistence for stem versus non-stem majors: Among college students with an autism spectrum disorder. Journal of Autism and Developmental Disorders, 44(5), 1159–1167. https://doi.org/10.1007/s10803-013-1978-5</bibtext> </blist> <blist> <bibtext> Wei X., Yu J. W., Shattuck P., Blackorby J. (2017). High school math and science preparation and postsecondary STEM participation for students with an autism spectrum disorder. Focus on Autism and Other Developmental Disabilities, 32(2), 83–92. https://doi.org/10.1177/1088357615588489</bibtext> </blist> <blist> <bibtext> Wei X., Yu J. W., Shattuck P., McCracken M., Blackorby J. (2013). Science, technology, engineering, and mathematics (STEM) participation among college students with an autism spectrum disorder. Journal of Autism and Developmental Disorders, 43(7), 1539–1546. https://doi.org/10.1007/s10803-012-1700-z</bibtext> </blist> <blist> <bibtext> Westland J. C. (2010). Lower bounds on sample size in structural equation modeling. Electronic Commerce Research and Applications, 9(6), 476–487. https://doi.org/10.1016/j.elerap.2010.07.003</bibtext> </blist> <blist> <bibtext> White I. R., Royston P., Wood A. M. (2011). Multiple imputation using chained equations: Issues and guidance for practice. Statistics in Medicine, 30(4), 377–399. https://doi.org/10.1002/sim.4067</bibtext> </blist> <blist> <bibtext> Wigfield A., Eccles J. S. (2000). Expectancy–value theory of achievement motivation. Contemporary Educational Psychology, 25(1), 68–81. https://doi.org/10.1006/ceps.1999.1015</bibtext> </blist> <blist> <bibtext> Wigfield A., Eccles J. S. (2002). The development of competence beliefs, expectancies for success, and achievement values from childhood through adolescence. In Development of achievement motivation (pp. 91–120). Academic Press. https://doi.org/10.1016/B978-012750053-9/50006-1</bibtext> </blist> <blist> <bibtext> Windham G. C., Fessel K., Grether J. K. (2009). Autism spectrum disorders in relation to parental occupation in technical fields. Autism Research, 2(4), 183–191. https://doi.org/10.1002/aur.84</bibtext> </blist> <blist> <bibtext> Zablotsky B., Black L. I., Maenner M. J., Schieve L. A., Danielson M. L., Bitsko R. H., Blumberg S. J., Kogan M. D., Boyle C. A. (2019). Prevalence and trends of developmental disabilities among children in the United States: 2009–2017. Pediatrics, 144(4), e20190811. https://doi.org/10.1542/peds.2019-0811</bibtext> </blist> <blist> <bibtext> Zeldin A. L., Pajares F. (2000). Against the odds: Self-efficacy beliefs of women in mathematical, scientific, and technological careers. American Educational Research Journal, 37(1), 215–246. https://doi.org/10.3102/00028312037001215</bibtext> </blist> <blist> <bibtext> Zimmerman B. J. (2000). Self-efficacy: An essential motive to learn. Contemporary Educational Psychology, 25(1), 82–91. https://doi.org/10.1006/ceps.1999.1016</bibtext> </blist> </ref> <ref id="AN0183813417-26"> <title> Footnotes </title> <blist> <bibtext> The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.</bibtext> </blist> <blist> <bibtext> Ethical approval for this study was obtained from the Institutional Review Board (IRB) at Binghamton University (Approval Number: STUDY00002606).</bibtext> </blist> <blist> <bibtext> The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Binghamton University, (grant number CCPA Research Excellence Awards).</bibtext> </blist> <blist> <bibtext> This work has been funded by the CCPA Research Excellence Awards at Binghamton University. The authors declare no competing financial interests or personal relationships that could be perceived to influence the outcomes and findings presented in this manuscript.</bibtext> </blist> <blist> <bibtext> Hyejung Kim https://orcid.org/0000-0002-4635-9292 Manuscript received November 2023; accepted December 2024.</bibtext> </blist> <blist> <bibtext> The term "neurodiversity," emphasizing differences rather than disorders, has been the preferred terminology for describing autism and other neurological conditions since the late 1990s ([21]; [55]).</bibtext> </blist> <blist> <bibtext> In this article, we use identity-first (e.g., "autistic students") language rather than person-first (e.g., "students with autism") to reflect current preferences within the autism community.</bibtext> </blist> <blist> <bibtext> The quality indicator checklist has been added to our OSF repository, accessible at https://osf.io/u9sn4/?view_only=835a13c917dc4972bd877d2e7fabeb28.</bibtext> </blist> <blist> <bibtext> The raw sample sizes (<emph>n</emph>) have been rounded to the nearest 10 to adhere to the National Center for Education Statistics data reporting guidelines. The count for autistic students has excluded fewer than 10 cases with a zero weight.</bibtext> </blist> <blist> <bibtext> For a complete list of the variables used in this study, please visit the Open Science Framework (OSF) repository.</bibtext> </blist> <blist> <bibtext> For specific weights associated with each research question, please refer to the OSF repository.</bibtext> </blist> <blist> <bibtext> Refer to the OSF repository for the original variables associated with these latent variables.</bibtext> </blist> </ref> <aug> <p>By Hyejung Kim; Mack Ottens; Matthew Jacob and Xingye Qiao</p> <p>Reported by Author; Author; Author; Author</p> </aug> <nolink nlid="nl1" bibid="bib36" firstref="ref1"></nolink> <nolink nlid="nl2" bibid="bib65" firstref="ref3"></nolink> <nolink nlid="nl3" bibid="bib63" firstref="ref4"></nolink> <nolink nlid="nl4" bibid="bib22" firstref="ref7"></nolink> <nolink nlid="nl5" bibid="bib15" firstref="ref8"></nolink> <nolink nlid="nl6" bibid="bib43" firstref="ref10"></nolink> <nolink nlid="nl7" bibid="bib47" firstref="ref11"></nolink> <nolink nlid="nl8" bibid="bib44" firstref="ref15"></nolink> <nolink nlid="nl9" bibid="bib31" firstref="ref16"></nolink> <nolink nlid="nl10" bibid="bib46" firstref="ref19"></nolink> <nolink nlid="nl11" bibid="bib52" firstref="ref21"></nolink> <nolink nlid="nl12" bibid="bib73" firstref="ref22"></nolink> <nolink nlid="nl13" bibid="bib68" firstref="ref24"></nolink> <nolink nlid="nl14" bibid="bib69" firstref="ref25"></nolink> <nolink nlid="nl15" bibid="bib33" firstref="ref29"></nolink> <nolink nlid="nl16" bibid="bib71" firstref="ref32"></nolink> <nolink nlid="nl17" bibid="bib10" firstref="ref33"></nolink> <nolink nlid="nl18" bibid="bib27" firstref="ref35"></nolink> <nolink nlid="nl19" bibid="bib11" firstref="ref36"></nolink> <nolink nlid="nl20" bibid="bib67" firstref="ref37"></nolink> <nolink nlid="nl21" bibid="bib48" firstref="ref38"></nolink> <nolink nlid="nl22" bibid="bib12" firstref="ref41"></nolink> <nolink nlid="nl23" bibid="bib38" firstref="ref42"></nolink> <nolink nlid="nl24" bibid="bib13" firstref="ref44"></nolink> <nolink nlid="nl25" bibid="bib57" firstref="ref45"></nolink> <nolink nlid="nl26" bibid="bib66" firstref="ref46"></nolink> <nolink nlid="nl27" bibid="bib62" firstref="ref47"></nolink> <nolink nlid="nl28" bibid="bib59" firstref="ref48"></nolink> <nolink nlid="nl29" bibid="bib28" firstref="ref51"></nolink> <nolink nlid="nl30" bibid="bib26" firstref="ref52"></nolink> <nolink nlid="nl31" bibid="bib23" firstref="ref53"></nolink> <nolink nlid="nl32" bibid="bib19" firstref="ref55"></nolink> <nolink nlid="nl33" bibid="bib45" firstref="ref60"></nolink> <nolink nlid="nl34" bibid="bib60" firstref="ref61"></nolink> <nolink nlid="nl35" bibid="bib37" firstref="ref63"></nolink> <nolink nlid="nl36" bibid="bib72" firstref="ref64"></nolink> <nolink nlid="nl37" bibid="bib30" firstref="ref65"></nolink> <nolink nlid="nl38" bibid="bib39" firstref="ref66"></nolink> <nolink nlid="nl39" bibid="bib41" firstref="ref68"></nolink> <nolink nlid="nl40" bibid="bib49" firstref="ref69"></nolink> <nolink nlid="nl41" bibid="bib18" firstref="ref70"></nolink> <nolink nlid="nl42" bibid="bib29" firstref="ref71"></nolink> <nolink nlid="nl43" bibid="bib14" firstref="ref72"></nolink> <nolink nlid="nl44" bibid="bib54" firstref="ref73"></nolink> <nolink nlid="nl45" bibid="bib70" firstref="ref74"></nolink> <nolink nlid="nl46" bibid="bib24" firstref="ref75"></nolink> <nolink nlid="nl47" bibid="bib56" firstref="ref76"></nolink> <nolink nlid="nl48" bibid="bib50" firstref="ref77"></nolink> <nolink nlid="nl49" bibid="bib25" firstref="ref78"></nolink> <nolink nlid="nl50" bibid="bib35" firstref="ref80"></nolink> <nolink nlid="nl51" bibid="bib58" firstref="ref81"></nolink> <nolink nlid="nl52" bibid="bib20" firstref="ref82"></nolink> <nolink nlid="nl53" bibid="bib17" firstref="ref84"></nolink> <nolink nlid="nl54" bibid="bib51" firstref="ref85"></nolink> <nolink nlid="nl55" bibid="bib40" firstref="ref86"></nolink> <nolink nlid="nl56" bibid="bib16" firstref="ref87"></nolink> <nolink nlid="nl57" bibid="bib64" firstref="ref89"></nolink> <nolink nlid="nl58" bibid="bib53" firstref="ref90"></nolink> <nolink nlid="nl59" bibid="bib34" firstref="ref91"></nolink> <nolink nlid="nl60" bibid="bib32" firstref="ref93"></nolink> <nolink nlid="nl61" bibid="bib42" firstref="ref94"></nolink>
Header DbId: eric
DbLabel: ERIC
An: EJ1466407
AccessLevel: 3
PubType: Academic Journal
PubTypeId: academicJournal
PreciseRelevancyScore: 0
IllustrationInfo
Items – Name: Title
  Label: Title
  Group: Ti
  Data: Examining STEM Preferences in Autistic Students: The Role of Contextual Support, Self-Efficacy, and Outcome Expectations
– Name: Language
  Label: Language
  Group: Lang
  Data: English
– Name: Author
  Label: Authors
  Group: Au
  Data: <searchLink fieldCode="AR" term="%22Hyejung+Kim%22">Hyejung Kim</searchLink> (ORCID <externalLink term="https://orcid.org/0000-0002-4635-9292">0000-0002-4635-9292</externalLink>)<br /><searchLink fieldCode="AR" term="%22Mack+Ottens%22">Mack Ottens</searchLink><br /><searchLink fieldCode="AR" term="%22Matthew+Jacob%22">Matthew Jacob</searchLink><br /><searchLink fieldCode="AR" term="%22Xingye+Qiao%22">Xingye Qiao</searchLink>
– Name: TitleSource
  Label: Source
  Group: Src
  Data: <searchLink fieldCode="SO" term="%22Exceptional+Children%22"><i>Exceptional Children</i></searchLink>. 2025 91(3):303-320.
– Name: Avail
  Label: Availability
  Group: Avail
  Data: SAGE Publications. 2455 Teller Road, Thousand Oaks, CA 91320. Tel: 800-818-7243; Tel: 805-499-9774; Fax: 800-583-2665; e-mail: journals@sagepub.com; Web site: https://sagepub.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="%22High+Schools%22">High Schools</searchLink><br /><searchLink fieldCode="EL" term="%22Secondary+Education%22">Secondary Education</searchLink><br /><searchLink fieldCode="EL" term="%22Higher+Education%22">Higher Education</searchLink><br /><searchLink fieldCode="EL" term="%22Postsecondary+Education%22">Postsecondary Education</searchLink>
– Name: Subject
  Label: Descriptors
  Group: Su
  Data: <searchLink fieldCode="DE" term="%22High+School+Students%22">High School Students</searchLink><br /><searchLink fieldCode="DE" term="%22Undergraduate+Students%22">Undergraduate Students</searchLink><br /><searchLink fieldCode="DE" term="%22Autism+Spectrum+Disorders%22">Autism Spectrum Disorders</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="%22Majors+%28Students%29%22">Majors (Students)</searchLink><br /><searchLink fieldCode="DE" term="%22Course+Selection+%28Students%29%22">Course Selection (Students)</searchLink><br /><searchLink fieldCode="DE" term="%22Longitudinal+Studies%22">Longitudinal Studies</searchLink><br /><searchLink fieldCode="DE" term="%22Academic+Achievement%22">Academic Achievement</searchLink><br /><searchLink fieldCode="DE" term="%22Students+with+Disabilities%22">Students with Disabilities</searchLink><br /><searchLink fieldCode="DE" term="%22Self+Efficacy%22">Self Efficacy</searchLink><br /><searchLink fieldCode="DE" term="%22Student+Educational+Objectives%22">Student Educational Objectives</searchLink><br /><searchLink fieldCode="DE" term="%22Context+Effect%22">Context Effect</searchLink><br /><searchLink fieldCode="DE" term="%22Occupational+Aspiration%22">Occupational Aspiration</searchLink>
– Name: DOI
  Label: DOI
  Group: ID
  Data: 10.1177/00144029241312777
– Name: ISSN
  Label: ISSN
  Group: ISSN
  Data: 0014-4029<br />2163-5560
– Name: Abstract
  Label: Abstract
  Group: Ab
  Data: Over recent decades, there has been a significant increase in postsecondary STEM education among autistic individuals. Using data from the High School Longitudinal Study of 2009, this study examined the STEM pathways of autistic students, emphasizing key determinants like proximal context, self-efficacy, and outcome expectations within the framework of social cognitive theory. The results revealed that despite a lower college attendance rate, autistic students displayed a pronounced inclination for STEM majors, particularly in the fields of science, engineering, and mathematics. Notably, autistic students who pursue higher education tend to exhibit increased levels of self-efficacy and anticipate more positive outcomes within STEM disciplines. However, the levels of both constructs in mathematics had decreased by the 11th grade. Nonetheless, STEM self-efficacy played a significant role in influencing outcome expectations and major choices, with this relationship being more pronounced among autistic students. For autistic students, their choice of a STEM major was influenced by their self-efficacy, as well as factors like race and gender. On the other hand, for non-autistic students, their proximal context was an additional determinant in their decision. Insights gained from this research can inform educational strategies aimed at facilitating the participation of autistic individuals in postsecondary STEM education and related career paths.
– Name: AbstractInfo
  Label: Abstractor
  Group: Ab
  Data: As Provided
– Name: Note
  Label: Notes
  Group: Note
  Data: https://osf.io/u9sn4/?view_only=835a13c917dc4972bd877-d2e7fabeb28
– Name: DateEntry
  Label: Entry Date
  Group: Date
  Data: 2025
– Name: AN
  Label: Accession Number
  Group: ID
  Data: EJ1466407
PLink https://search.ebscohost.com/login.aspx?direct=true&site=eds-live&db=eric&AN=EJ1466407
RecordInfo BibRecord:
  BibEntity:
    Identifiers:
      – Type: doi
        Value: 10.1177/00144029241312777
    Languages:
      – Text: English
    PhysicalDescription:
      Pagination:
        PageCount: 18
        StartPage: 303
    Subjects:
      – SubjectFull: High School Students
        Type: general
      – SubjectFull: Undergraduate Students
        Type: general
      – SubjectFull: Autism Spectrum Disorders
        Type: general
      – SubjectFull: STEM Education
        Type: general
      – SubjectFull: STEM Careers
        Type: general
      – SubjectFull: Majors (Students)
        Type: general
      – SubjectFull: Course Selection (Students)
        Type: general
      – SubjectFull: Longitudinal Studies
        Type: general
      – SubjectFull: Academic Achievement
        Type: general
      – SubjectFull: Students with Disabilities
        Type: general
      – SubjectFull: Self Efficacy
        Type: general
      – SubjectFull: Student Educational Objectives
        Type: general
      – SubjectFull: Context Effect
        Type: general
      – SubjectFull: Occupational Aspiration
        Type: general
    Titles:
      – TitleFull: Examining STEM Preferences in Autistic Students: The Role of Contextual Support, Self-Efficacy, and Outcome Expectations
        Type: main
  BibRelationships:
    HasContributorRelationships:
      – PersonEntity:
          Name:
            NameFull: Hyejung Kim
      – PersonEntity:
          Name:
            NameFull: Mack Ottens
      – PersonEntity:
          Name:
            NameFull: Matthew Jacob
      – PersonEntity:
          Name:
            NameFull: Xingye Qiao
    IsPartOfRelationships:
      – BibEntity:
          Dates:
            – D: 01
              M: 04
              Type: published
              Y: 2025
          Identifiers:
            – Type: issn-print
              Value: 0014-4029
            – Type: issn-electronic
              Value: 2163-5560
          Numbering:
            – Type: volume
              Value: 91
            – Type: issue
              Value: 3
          Titles:
            – TitleFull: Exceptional Children
              Type: main
ResultId 1