One of the most significant contributions of the study is the demonstration that post-editing offers a consistent accuracy advantage over human translation, even in early stages of training. At the same time, the study underlines the importance of explicit terminology training, as persistent difficulties in distinguishing between terms and general vocabulary suggest a gap between theoretical knowledge and its practical application, particularly in the post-editor group. The findings also point to the risks of over-reliance on MT, emphasising the need for structured post-editing instruction to develop critical evaluation skills.

Moreover, it was found that both translation and post-editing performance improved between the two data collection waves, with human translators showing significant gains in accuracy, while post-editors improved in general performance and fluency. A key finding is that post-editing performance improves as a result of traditional translation training, suggesting that core translation skills transfer to post-editing even in the absence of explicit PE instruction.

Finally, the study reveals that students’ beliefs about translation mature over time and become increasingly correlated with performance. In general, a shift from linguistic competence as the primary predictor of success can be observed to a more complex interrelation of factors affecting translation and post-editing performance.