A novel approach to assessing language proficiency in adults: A pre-pilot classification accuracy study. (June 2021)

• Record Type: Journal Article
• Authors: Schwarz, Amy Louise; Resendiz, Maria; Herrera, Laura Catarina; Gonzales, Maria Diana
• Is Part Of: International journal of bilingualism. Volume 25:Number 3(2021)
• Page Start: 812
• Page End: 832
• Publication Date: 2021-06
• Subjects: Bilingualism -- proficiency -- self-report -- language use questionnaire -- classification accuracy -- pre-pilot; Bilingualism -- Periodicals; Bilingualism -- Research -- Periodicals; Language acquisition -- Periodicals; 306.446072
• DOI: 10.1177/1367006921999452
• Languages: English
• ISSNs: 1367-0069

Abstract

Aims and objectives/purpose/research questions: Speech–language pathologists who speak more than one language and who are members of the American Speech–Language–Hearing Association self-identify in one yes/no question whether they have the proficiency level to be bilingual service providers. This research note describes a preliminary attempt to address the very practical issue of whether and in what circumstances Spanish–English bilinguals can accurately judge their proficiency levels in both languages. The research question is: Will bilingual adults accurately identify their first language and second language proficiency levels using a self-assessment when compared to a commonly used standardized norm-referenced test (SNRT) in both formal and informal contexts across the following outcome measures: (a) overall proficiency; (b) listening; (c) speaking; (d) reading; and (e) writing? Design/methodology/approach: Classification accuracy studies require at least 34 participants. Thirty-nine participants completed the commonly used Language Use Questionnaire (LUQ) self-assessment and the commonly used Woodcock–Muñoz Language Survey SNRT (WMLS-III). For this pre-pilot study, participants were Spanish–English bilingual university students. Data and analysis: Forty likelihood ratios (LRs) were calculated. Benchmarks for interpreting LRs for classification accuracy studies were applied to identify the likelihood of an individual being proficient or non-proficient in two languages. Findings/conclusions: For the overall proficiency and formal speaking proficiency outcomes in Spanish, positive LRs met the benchmark for strong agreement. Originality: The current study is the first to show that Spanish–English bilingual adults can accurately judge their Spanish proficiency levels for two specific outcome measures. Significance/implications: These results are important for two reasons. First, they suggest that Spanish–English bilingual adults can accurately judge their overall proficiency levels in Spanish. Second, they identify which outcome measures from the LUQ and WMLS-III should be considered in a future classification accuracy study.