2025 ASEE Annual Conference & Exposition

(paper type: ERM, Methods/Theory, research brief) Social Cognitive Career Theory (SCCT) is a dominant and established theoretical framework [1]. SCCT assumes that stronger self-efficacy, outcome expectations, and personal goals contribute to higher interest, choice, and performance [2]. While empirical studies generally support the SCCT model (i.e. mathematics self-efficacy is correlated positively with mathematics performance [3]), research with minoritized youth is inconsistent. For example, Black students had higher mathematics self-efficacy compared to White, Hispanic, or Asian peers, but that did not translate to performance [4], [5]. Using PISA 2003 data, researchers even found a significant negative relationship between mathematics self-efficacy and achievement for Black students [6]. A recent randomized study of urban mostly Black high school students in an algebra-for-engineering afterschool program revealed significant effects on STEM choice goals but not mathematics self-efficacy, STEM interest, or STEM outcome expectations comparing treatment and control groups [7]. Together, these studies point to a puzzling relationship between self-efficacy, choice, and performance. SCCT lacks qualitative research methods and mixed-methods studies exploring whether SCCT encompasses the experiences and structural inequalities shaping minoritized youths’ career development [8]. In this study, SCCT is revisited through the Phenomenological Variant of Ecological Systems Theory [9], [10].

Qualitative interviews sought to understand socio-environmental factors, post-secondary planning processes, and STEM career interests of primarily Black youth (n = 25) from a large urban district in the aforementioned algebra-for-engineering afterschool program between 2021 and 2024. The research question is “How do students describe mesosystem factors (STEM course quality, peer support, mentoring, family support) in relation to microsystem factors (STEM interest, STEM identity, math self-efficacy, and STEM choice goals)?” The first cycle of coding highlighted SCCT-derived constructs and the second cycle highlighted in vivo coding [11]. Codes were thematically analyzed [12] and to visually illustrate relationships between these factors [11], data were presented in a color-coded chart (Table 1).

Participants overall described high math self-efficacy, despite variable levels of mesosystem support, STEM choice goals, performance (self-reported grades), and persistence in the program. Many students with low-quality STEM course quality reported the highest self-efficacy (contrary to SCCT): “I don't even get instructions, honestly… I'm usually just letting them answer it and then I get the answers,” said Sylvia (pseudonym) who rated her mathematics confidence as “8 out of ten.” The authors argue that elevated mathematics self-efficacy may be a reactive coping process [9] in the absence of structurally and culturally responsive education [13], leading to overinflated but not realistic mathematics self-efficacy. Ultimately, this maladaptive process did not buoy performance and meta-cognitive insight [14] but instead led to STEM attrition (Figure 1). Prior research has found that high mathematics self-efficacy (measured in high school) was associated with enrollment in a four-year institution for young Black women, but this mathematics self-efficacy eroded over time and contributed to attrition [15]. While educators may want to grow students' confidence, grade inflation or passing students through (i.e. "How did I pass this and I didn't know any of it?”) produces unintended consequences. Minoritized youth voices shed light on SCCT’s shortcomings and implications for practice.