Evaluation of a self-report measure of Spanish vs. English language use preferences among middle school English Learners with reading difficulties

Evaluation of a self-report measure of Spanish vs. English language use preferences among middle school English Learners with reading difficulties

First Author: Kelly Macdonald -- University of Houston
Additional authors/chairs: 
Paul Cirino
Keywords: Bilingualism, Reader characteristics, language proficiency
Abstract / Summary: 

Understanding variability in language skills among bilinguals is a growing area of interest, particularly among at-risk populations. In the US, one at-risk population is English Learners (ELs) with reading difficulties. The extent to which language skills may be measured through self-report vs. performance-based measures is not well understood, particularly for young at-risk samples. Bilingual research with adults demonstrates moderate correlations between self-report and performance-based assessments, suggesting that these tools should be used together to provide a richer understanding of the bilingual experience. The purpose of our study was to evaluate the psychometric properties of a self-report measure of language use preferences among n = 158 ELs in middle school identified as struggling readers. We evaluated internal consistency, convergent validity with performance-based language assessments, and conducted exploratory factor analysis (EFA). We predicted that self-report measures would correlate moderately with performance-based measures such that a preference for English would be negatively related to performance on Spanish tests but positively related to performance on English tests. Results demonstrated that, on average, students in this sample had a balance between their Spanish and English use preferences. Evaluation of psychometric properties demonstrated acceptable reliability of the self-report measure (Cronbach’s α = .67). Self-reported English preferences correlated negatively with Spanish picture vocabulary (r = -.59, p < .001) and positively with English picture vocabulary (r = .38, p < .001). Finally, the EFA demonstrated a 2-factor structure consisting of a family use preferences factor and a friends use preferences factor. Results suggest that self-report and performance-based measures should be used together to characterize language skills in this population; further work should evaluate the ways in which these measures relate to reading.