Understanding Latina adolescents' science identities : a mixed methods study of socialization practices across contexts

Research on differences in STEM outcomes for females and students of color has been an ongoing educational research imperative, but Latinas continue to be under-represented in high school and college science classes and majors (National Science Foundation, 2011; Riegle-Crumb & King, 2010). The aim of this study was to investigate how Latina adolescents seek to establish themselves as future scientists within their environments and how others help sustain these developing identities. I used a mixed method procedure called an exploratory sequential design that starts with a qualitative stage followed by a quantitative stage (Creswell & Plano Clark, 2007). In the qualitative stage, 32 college-aged Latinas in science majors participated in focus groups with an additional 12 in interviews. Using Interactive Qualitative Analysis (Northcutt & McCoy, 2004), eight factors of science identity development were identified: home environment, teacher influences, school experiences, environmental factors, media influences, using your brain, emotions, and career planning. Participants saw the first four factors as drivers of their development, with media as an irregular contributor. These social factors were filtered through the individual factors of using your brain and emotions, with career planning as the outcome. The qualitative results were used to develop a survey given to middle school students in the next stage. The majority of the survey consisted of previously validated scales that corresponded in content to the qualitative factors. One new measure was developed to address science-related experiences. In the quantitative stage, 90 middle school Latinas from two central Texas school districts participated in the survey study. Univariate analysis showed differences in science-related experiences by demographic variables of parent occupation, parent nativity, first language spoken, and school district. Multivariate regression analysis found positive emotions about science to be the best predictor of science career related outcomes, and that emotions act as a mediator between science experiences and career outcomes. These results are discussed in light of current career theories.