Zooming into two measurement issues in writing self-efficacy: revision as a distinct dimension and the generality hypothesis in argumentative writing

Abstract

Self-efficacy plays a crucial role in writing, influencing what and how students write. However, issues related to assessing this construct need further examination. The purpose of this study was twofold. First, we assessed the dimensionality and reliability of two versions of a self-efficacy scale f3333or writing. One version assessed writing self-efficacy in a general manner, while the other version assessed self-efficacy for argumentative writing. Both versions of the scale assessed efficacy for writing ideation and self-regulation, and included efficacy for revising, a dimension rarely examined in studies of writing self-efficacy. Secondly, we tested the generality hypothesis, which posits that a writing self-efficacy measure that is more closely aligned to how writing is assessed is a better predictor of writing performance than a measure that is less closely aligned. A total of 426 middle and high school students were randomly assigned to complete either the general or argumentative specific writing self-efficacy scale. All students completed an argumentative writing task focused on a historical event in the United States, which was scored for quality of writing. Both the general and argumentative specific writing self-efficacy scales were found to be reliable measures with a multi-dimensional structure, including ideation, self-regulation, and revision. The revision subscale was validated as a distinct dimension, alongside ideation and self-regulation dimensions. The general and argumentative specific writing self-efficacy scales were similarly correlated with the quality of students’ argumentative writing on a historical topic. This study did not provide evidence to support the generality hypothesis. Educational implications were discussed.