Critical Points and Places in the STEM Pathway: The Case of High School-to-College Transition

• In Event: Poster Session 08
  In Poster Session: PS 08 Topic - Education, Schooling

Abstract

The Science, Technology, Engineering, and Mathematics (STEM) workforce is the forefront of innovation in the U.S. economy. However, only a small percentage of youth pursue degrees in STEM disciplines (Cannady et al., 2013). Previous STEM education research mostly focuses on critical STEM milestones in the educational system, notably, high school calculus course-taking (e.g., Adelman, 2006; Tyson et al., 2007). This study explored high school students’ various exposure and experiences and which ones contribute to their STEM progress into higher education. We consider several high school STEM precursors (e.g., final math/science achievement and its comparative advantage relative to reading/English) that may be associated with students’ post-secondary STEM outcomes, and how these alternative STEM precursors may be differentially associated with post-secondary STEM outcomes depending on youth’s parental financial and social capital (Crosnoe & Schneider, 2010).

Drawn from the Educational Longitudinal Study of 2002, a nationally representative longitudinal study, we limited our sample to 6,570 students who pursued post-secondary education. Post-secondary STEM outcomes were assessed in young adulthood and divided into three categories: no pursuits of a bachelor’s degree (a reference group selected by the largest number of observable significant mean differences among groups); a STEM bachelor’s degree; and a non-STEM degree. STEM precursors were drawn from the end of high school and consist of students’ highest math/science course level, math/science GPA, standardized math test scores, and the ratio of math to English score. Parental capital (e.g., in poverty, educational level and aspiration, and school involvement) and covariates, including individual and school characteristics, were drawn from the 10th grade. To control neighborhood characteristics, we integrated various neighborhood-level data. including the National Center for Charitable Statistics and the Bureau of Labor Statistics. The summary statistics for these variables by STEM outcomes are presented in Table 1.

With three levels of post-secondary STEM outcomes, we used a multi-level multinomial logit regression for the analysis. Furthermore, our sample restriction indicated potential selection biases. Thus, a two-stage modeling strategy called Heckman correction (Heckman, 1977) was adopted.

Preliminary regression results are presented in Table 2. The results suggest that youth’s likelihood of pursuing either a STEM or non-STEM bachelor’s degree was positively associated with their STEM precursors including high-level math/science course-taking, standardized math scores, and math/science course GPA. The results suggested a moderating role of family capital on youth’s math to reading score ratio and STEM pursuits. With high math to reading score ratio, youth with a non-English-speaking parent were less likely to pursue a STEM degree than youth with English-speaking parents, and youth with a college-educated parent were less likely to pursue a non-STEM degree than youth with a college-educated parent. The results are consistent with previous literature highlighting the importance of the high school-to-college transition in the STEM pathways (Simpkins et al., 2015). This project will provide insights into critical points and places in the STEM pathway that may facilitate or hinder students’ STEM educational trajectories.

Authors

• Sae Hyun Oh, Tufts University
  Presenting Author

• Tama Leventhal, Tufts University
  Non-Presenting Author