Although researchers have investigated the effects automated writing evaluation (AWE) and teacher corrective feedback have on second language (L2) writing performance, these studies have conflated two different variables—feedback source and perceptions of feedback source. Using a quasi-experimental design, two groups of second language English writers—perceived teacher feedback group (N = 88) and perceived automated feedback group (N = 73)—received feedback on four essays written over an 18-week academic semester. While both groups received AWE and a marked rubric as feedback, the perceived teacher feedback group was under the impression that the AWE was given by their teacher. Results showed the perceived automated feedback group outperformed the perceived teacher feedback group on essay 2 and essay 3; however, the reverse was found for essay 4. Furthermore, the perceived automated feedback group showed statistically significant differences in writing performance; they improved from essay 2 to essay 3 but regressed from essay 3 to essay 4. Perceptions of the feedback source may have caused these results, suggesting that it should be worthwhile to inform L2 student writers of the feedback source. Before providing feedback to L2 student writers, teachers can inform students on how their writing performance can benefit from AWE and afterwards the teachers can meet students’ needs by following up with the delivery of teacher feedback.

尽管研究人员调查了自动写作评估 (AWE) 和教师纠正反馈对第二语言 (L2) 写作表现的影响，但这些研究混淆了两个不同的变量——反馈源和对反馈源的看法。使用准实验设计，两组第二语言英语作家——感知教师反馈组（N  = 88）和感知自动反馈组（N = 73)——在 18 周的学术学期内收到四篇论文的反馈。虽然两个小组都收到了 AWE 和一个标记的评分量规作为反馈，但感知到的教师反馈组的印象是 AWE 是他们的老师给的。结果显示感知自动反馈组在论文 2 和论文 3 上的表现优于感知教师反馈组；然而，论文 4 的情况正好相反。此外，感知自动反馈组在写作表现上表现出统计学上的显着差异；他们从论文 2 改进到论文 3，但从论文 3 退化到论文 4。对反馈来源的看法可能导致了这些结果，这表明将反馈来源告知 L2 学生作者应该是值得的。在向 L2 学生作家提供反馈之前，