Testing audiovisual comprehension tasks with questions embedded in videos as subtitles: a pilot multimethod study

摘要

Listening, watching, reading and writing simultaneously in a foreign language is very complex. This paper is part of wider research which explores the use of audiovisual comprehension questions imprinted in the video image in the form of subtitles and synchronized with the relevant fragments for the purpose of language learning and testing. Compared to viewings where the comprehension activity is available only on paper, this innovative methodology may provide some benefits. Among them, it could reduce the conflict in visual attention between watching the video and completing the task, by spatially and temporally approximating the questions and the relevant fragments. The technique is seen as especially beneficial for students with a low proficiency language level.The main objectives of this study were to investigate if embedded questions had an impact on SFL students’ audiovisual comprehension test performance and to find out what examinees thought about them. A multimethod design (Morse, 2003) involving the sequential collection of three quantitative datasets was employed. A total of 41 learners of Spanish as a foreign language (SFL) participated in the study (22 in the control group and 19 in the experimental one). Informants were selected by non-probabilistic sampling. The results showed that imprinted questions did not have any effect on test performance. Test-takers’ attitudes towards this methodology were positive. Globally, students in the experimental group agreed that the embedded questions helped them to complete the tasks. Furthermore, most of them were in favour of having the questions imprinted in the video in the audiovisual comprehension test of the final exam. These opinions are in line with those obtained in previous studies that looked into experts’, SFL students’ and SFL teachers’ views about this methodology (Casañ Núñez, 2015a, 2016a, in press-b). On the whole, these studies suggest that this technique has potential benefits for FL learning and testing. Finally, the limitations of the study are discussed and some directions for future research are proposed.