LAUSR

Abstract This study aimed to develop and validate an English writing strategy inventory (EWSI) that measures English writing strategies at the tertiary level. To achieve this goal, we focused our analysis primarily on construct and response validity, and took a mixed approach using a statistical technique and qualitative method, think-aloud protocol analysis . We constructed questionnaire items from a broad literature review and an analysis of 18 college or graduate students' reflection logs on their English writing experience. We administered the resulting questionnaire to 271 Korean EFL students. An exploratory factor analysis of the participants' responses revealed a set of eight factors—metacognitive, memory, cognitive, L1 use, revision, L2 use, social strategies, and compensatory strategies—each of which exhibited a high degree of internal consistency. The think-aloud protocol analysis with four students identified three main types of issues—writing context, writing tasks, and wording problems—that could leave respondents uncertain about how to answer and could therefore hamper the questionnaire's validity and reliability. This study presents possible solutions to these problems, and provides suggestions regarding the practical use of the EWSI to minimize the unfavorable impacts of these error sources during survey administration(191).