Impact of patterns of language use and socio-economic status on a constructed response Situational Judgment Test (SJT).

Abstract

The study explored the impacts of patterns of language use (PLU) and socio-economic status (SES) on Casper, a constructed-response situational judgment test (SJT). 10,266 applicants from two U.S. medical schools were grouped into self-reported balanced bilinguals, unbalanced bilinguals, English monolinguals, and English as a Second Language (ESL) students. A multicomponent SES composite was used to assess the degree of socioeconomic disadvantage (DSD). Results from a hierarchical regression analysis showed that after accounting for demographic variables, both PLU and DED were significant factors on applicants' Casper performance. Bilingualism was associated with better Casper performance compared to English monolinguals and ESL students. No significant effect of speaking English as a native language was found on applicants' Casper performance. English monolinguals and ESL students performed equivalently on Casper. Finally, high DSD was associated with better Casper performance than low DSD, and the impact of DSD on Casper held the same across all four language groups. These findings provide evidence that socio-cultural factors, such as PLU and DSD have important impacts on SJT performance. Further research is needed to understand the role of differences in language construction across socio-cultural factors on constructed-response SJT performance.