Psychometric Properties of an English Writing Self-Efficacy Scale: Aspects of Construct Validity

• In Event: Validity Evidence in Writing and Science Assessments

Abstract

This study provided validity evidence of the Questionnaire of English Writing Self-Efficacy within Messick’s framework of construct validity. The validity was further supported using Item Response Theory. Items demonstrated close alignment with self-regulation theories, providing evidence of the content aspect validity. The substantive aspect was evaluated with consistencies of participants’ responses. Confirmatory factor analysis revealed a satisfactory global fit of the four-factor model, providing evidence of the structural aspect of validity. Significant correlations between writing self-efficacy, writing proficiency, and self-regulated learning supported the external aspect of validity. IRT analyses suggested that measurement precision was greatest at the lower range of the latent trait scale, providing justification for the primary use of the instrument for identifying students with low writing self-efficacy.

Authors

• Ting Sun, Northwestern University

• Chuang Wang, University of Macau

• Stella Yun Kim, University of North Carolina - Charlotte