Date of Award

June 2017

Degree Type


Degree Name

Doctor of Philosophy (PhD)


Instructional Design, Development and Evaluation


Tiffany A. Koszalka


Career Development, High School Longitudinal Study of 2009, Intersectionality, Social Cognitive Career Theory, STEM, Structural Equation Modeling

Subject Categories



This research examined high school students’ STEM career development using a Social Cognitive Career Theory framework. Data used in this study were from High School Longitudinal Study of 2009. Intersectional approaches were employed to gain an in-depth understanding of student characteristics, as well as identify potential differences in students’ STEM behaviors. Further, examinations of the STEM career development process were conducted using structural equation modeling statistical techniques. Findings suggest that prior learning experiences (i.e., math aptitude, informal STEM learning experiences, and math and science identity) and environmental supports and barriers (e.g., informal STEM exposure) are significant influences on students’ STEM career development. Additionally, when considering the entire student population, students’ math and science self-efficacy, outcome expectation, and interest are significant predictors of STEM career intentions and STEM major selection. However, multi-group structural equation modeling analyses, particularly with regard to race/ethnicity and socio-economic status, indicate substantial between group differences in students’ STEM career development. When examining race, the proposed model was most predictive for White students and least predictive for Black students. STEM career intention was significantly influenced by math interest and math outcome expectation for White and Asian students, but these factors were not predictive for Latino and Black students. Additionally, self-efficacy was predictive of STEM major selection for all racial/ethnic groups, except Black students. Finally, outcome expectation was shown to significantly influence STEM major selection for White students, but not for any of the other racial/ethnic groups. Similar trends emerged when analyzing the proposed model by students’ socio-economic status—the model was most predictive of STEM career development for students in the highest socio-economic quintiles, and least predictive for those in the lowest.


Open Access

Included in

Education Commons