STEM Intention of High School Students: Its Development and Antecedents

• In Event: Understanding Individual and Contextual Predictors of Students' STEM Major Intention and Promoting Diversity in STEM

Abstract

Objectives. Research has focused on improving students’ persistence and reducing the high attrition in the STEM fields (Graham et al., 2013). However, fewer studies have examined the supply side of the STEM pipeline to address the shortage of students pursuing STEM disciplines. Students’ STEM pursuance is directly predicted by their early intent to STEM-majors before attending to postsecondary education (e.g., Perez et al., 2019; Wang, 2013). As one of the critical first steps is to attract students into STEM-majors, it is crucial to have a comprehensive understanding of factors predicting students’ early STEM- intention (Kurban & Cabrera, 2020). We aimed to conduct a longitudinal study over a three-year course of high-school to address the research questions: (1) what is the developmental trajectory of high-school students’ STEM-intention over time; (2) how does math and science self-efficacy predict the initial STEM-intention and its change over time; (3) how do individual (e.g., self-expectation and academic performance) and social influences (e.g., parent-expectation, teacher-expectation, and school activities) promote students’ STEM-intention?
Theoretical Framework. The study is based on Bandura’s social cognitive theory (1986), which posits that individual, social agents and motivation have important influences on student academic choice and decision.
Method. The data were collected from a large southern Charter School System in United States. The current sample included 1,456 students (747 girls at Grade 10 at Time1) who completed the yearly online survey from 2017 to 2019. STEM-intention as the dependent variable was measured with one dichotomous item. Time-level predictors included time (three time-waves), math and science self-efficacy (αs = .92-.93). STEM-activities (STEM-club and STEM-AP), educational expectations (self-expectation, teacher-expectation, and parent-expectation), and academic performance (self-reported GPA). Individual-level predictors included gender, race, lunch status, and initial STEM-activities experiences (STEM-camp and STEM-internship). A two-level linear growth model was conducted using HLM 6.8 to analyze a data structure where students (individual-level/level-2) were repeatedly measured across three time-points (Time-level/level-1).
Results. Results are presented in Table 1. High-school students’ intention of choosing a STEM major significantly declined over their high school years. Math and science self-efficacy positively influenced students’ STEM intention respectively. That is, students with stronger confidence in their math and science abilities were more likely to choose a STEM-major (ps< .001). When students perceived higher expectations from their parents (p< .05), they had stronger intention to choose a STEM-major. Attending STEM-relevant clubs at school positively influenced students’ STEM-intention (p< .001). Female students expressed lower STEM-intention compared male students (p< .01) but showed a lower decline rate over the high-school years (p< .05).
Significance. Using a diverse sample of participants, the study employed a longitudinal framework to examine influences of a wide range of individual and social factors on high-school students’ STEM-intention. The findings provide evidence that students’ STEM-intention declined over high-school years (Kennedy et al., 2020; Sadler et al., 2012). The study also identified the influential factors (math and science self-efficacy, parent expectation and school STEM-club participation) to help improve students’ STEM-intention in postsecondary education. Taken together, findings provide important implications to educators seeking to increase STEM enrollment.

Authors

• Weihua Fan, University of Houston
  Presenting Author

• Daijiazi Tang, University of Houston
  Non-Presenting Author

• Adam Sahin, Harmony Public Schools
  Non-Presenting Author