Reception study: how young viewers perceive subtitles in different modalities – AI-generated, post-edited, and human translation

The aim of this study was to compare the reception of three different types of film subtitles: (1) human-only, (2) AI-generated, and (3) post-edited subtitles. The study made use of three questionnaires distributed among 120 participants divided into three groups. Each group watched the video with different subtitles - human translation (HT), machine translation (MT) or post-edited machine translation (PE) and filled in the appropriate questionnaire. The dataset includes the three questionnaires, labelled HT, MT and PE. In also includes a version of the questionnaire with coded answers. For the sake of convenience, the files are both in MS Word and Open Document formats. There are also two files with results - the raw results, which were transcibed from the paper questionnaires into an electronic from, and the results of statistical analyses in the final results file.