The study aimed (a) to estimate the overall average effect size of the relationship between writing self-efficacy and writing achievement for first language (L1) and second language (L2) writers in English; and (b) to examine how writing in English as a L1/L2 moderates the relationships based on a meta-analysis of published journal articles and dissertations/theses. Data included 565 effect sizes from 76 studies through a rigorous process of literature searches, screening, and data coding. A two-level meta-regression model was constructed to estimate the average effect size and to examine the moderating effects of the covariates. Results revealed a medium effect size (r =.29) with both L1 and L2 writers, which indicated approximately 9% of the variability in English writing achievement was associated with variability in students’ self-efficacy. Furthermore, writing in English as a L1/L2 was found to moderate the relationship between writing self-efficacy and writing achievement, with the effect size estimated with L2 learners (r =.441) being statistically significantly larger than that yielded with L1 learners (r =.233), after controlling for the covariates of sample size, gender, grade, statistical procedures, and publication type. Results also revealed that statistical procedure moderated effect size estimates. This meta-analysis has practical implications for heightening the awareness of English teachers in developing L2 students’ self-efficacy in the writing classroom setting. It also cautions researchers against the misinterpretation and misuse of effect sizes calculated by different statistical procedures.