Measuring English Language Workplace Proficiency Across Subgroups: Using CFA Models to Validate Test Score Interpretation TOEIC
- Author(s):
- Yoo, Hanwook; Manna, Venessa F.
- Publication Year:
- 2017
- Source:
- Language Testing, v34 n1 p101-126, 2017
- Document Type:
- Article
- Page Count:
- 26
- Subject/Key Words:
- Workplace English, Workplace Literacy, Score Interpretation, Test of English for International Communication (TOEIC), Confirmatory Factor Analysis, Measurement Invariance, Score Reporting, TOEIC Listening and Reading Tests, English (Second Language), English Language Proficiency, Listening Comprehension Tests, Reading Tests, Second Language Learning
Abstract
SUMMARY: This study used a statistical technique called "factor analysis" to determine which statistical model best explained performance on the TOEIC Listening and Reading test. Researchers found that a model (two-factor model) in which reading and listening skills were represented as distinct abilities best accounted for performance, consistent with how scores are supposed to be interpreted. In addition, researchers examined whether the two-factor model held across different subgroups of test takers (e.g., gender, age, employment status) in order to investigate whether scores could be interpreted in the same manner for all groups of test takers. Results indicated that TOEIC test scores have the same meaning across the subgroups included in this study, which provides evidence of fairness.