This dataset is from a quasi-experimental study that evaluated two methods for teaching phonemic transcription to university students of English: (i) the transcription of auditory stimuli and (ii) an activity-based approach employing a phonemic version of the classic board game Scrabble®. Participants were recruited from three (parallel) "English Phonetics & Phonology" courses, where the final exam involves phonemic transcriptions. A pretest-posttest control group design was used, with group sizes of 9 (control group), 15 (audio group) and 12 (scrabble group). Participants signed up for two different time slots, which were then assigned to different methods of instruction. The training groups were offered five 90-minute sessions, which were identical for both groups in the first 45 minutes each week, but used different methods in the second half of each session. Transcription ability was assessed prior to and after training using the same (40-minute) test, which covered a wide range of segmental and suprasegmental features, with a maximum of 345 points attainable. The dataset also includes further information on the students, which was elicited using a questionnaire (e.g. whether (or how long) they had spent time abroad, whether they did their homework for the course regularly).

IBM SPSS Statistics, 14.0