A multifactor perspective on writing self-efficacy was examined in 2 studies. Three factors were proposed-self-efficacy for writing ideation, writing conventions, and writing self-regulation-and a scale constructed to reflect these factors. In Study 1, middle school students (N = 697) completed the Self-Efficacy for Writing Scale (SEWS), along with associated measures. Confirmatory factor analysis (CFA) showed SEWS data fit the proposed 3-factor model well. In Study 2, a second CFA of data from 563 students from 2 high schools likewise showed good model fit. Scores based on the 3 writing self-efficacy factors were examined in relation to students' liking writing, self-reported writing grades, and statewide writing assessment (SWA) scores. Results showed writing ideation and self-regulation self-efficacy to be significantly more strongly related to liking writing than conventions self-efficacy but less related than conventions self-efficacy to SWA scores. All 3 writing self-efficacy dimensions showed moderate positive correlations with self-reported writing performance. Further analyses showed higher levels for all 3 dimensions of writing self-efficacy for students in more advanced English/language arts classes. Overall, results from the studies were interpreted as supporting multifactor models of writing self-efficacy and the utility of closer ties between self-efficacy measures and domains being assessed.