Supporting Latino High School Students' Science Motivation: Examining the Unique and Collective Contributions of Family, Teachers, and Friends

Authors: Sandra D. Simpkins, Yangyang Liu, Gabriel Estrella
Presented at 2017 AERA

​Abstract: While many US students aspire to pursue a career in STEM (i.e., science, technology, engineering, and math) at some point during their education, approximately 45% of 10th grade students lose that desire by the end of high school (Aschbacher et al., 2010). Of these students, females and Latinos are particularly at risk of forgoing a future in STEM. According to the Eccles’ expectancy-value model, family, friends, and teachers can support adolescents’ science motivation and subsequent academic achievement (Wigfield et al., 2015). However, whose support is most consequential for improving Latino adolescents’ motivation in science? Various theories argue for the relative importance of each source. Friends, for example, become increasingly important as youth age into and through adolescence. Yet, Latino culture’s emphasis on familism suggests family members are critical. Moreover, because many Latino parents have lower levels of education and some lack familiarity with the US school system, older siblings or cousins often provide support that is typical of parents in US-born mainstream American families. The goal of this study is to examine the contribution of families’, teachers’, and friends’ support in predicting adolescents’ motivational beliefs, ability, and classroom engagement. One key aspect of this question is to examine the unique contribution of each source of support as well as how these sources work together to predict adolescents’ science outcomes. Longitudinal data on 103 Latino adolescent students were collected in 9th and 10th grade (the 2012-2013 and 2013-2014 school years). Adolescents reported on (a) the support they received from their family (averaged across parents and an older sibling/cousin), science teacher, and friends (15 items per scale, α =.92-.94), (b) their science ability self-concept (12 items, α =.92), and (c) science value (15 items, α =.93-.95). We used multiple indicators of adolescents’ support: (a) overall support which was calculated as the average of family, friend, and teacher support; and (b) support from each these three separate sources. Overall support in 9th grade positively predicted adolescents’ concurrent and 10th grade science self-concepts and values, but did not predict subsequent changes in motivational beliefs from 9th to 10th grade. Analyses including the three different sources of support suggest that friend support was a strong predictor of adolescents’ concurrent motivation. Although sometimes only marginally significant, family support predicted adolescent motivational beliefs in 9th and 10th. Interestingly, teacher support was never a statistically significant predictor of adolescents’ motivational beliefs. We use pattern-centered approaches to test whether particular patterns of support across the three sources uniquely predict adolescents’ science motivational beliefs. These findings underscore the importance of continued social support in high school from friends and families.