Investigating the use of text-to-speech assistive technologies in the language testing of students with Specific Learning Difficulties

Language testers involved in assessing the language skills of Italian university students with Specific Learning Difficulties (SpLDs) in mandatory language tests are faced with the dilemma created between the need to ensure that such students have fair access to tests and the application of the accommodations made available under Italian educational legislation. In particular, the use of text-to-speech software in testing settings raises serious concerns regarding construct irrelevance and construct under-representation, especially in tests of reading comprehension. The small case study reported on here sets out to investigate how Italian university students with SpLDs interact with a reading text using text-to-speech software. It specifically explores to what degree such students orient to a reading text on computer video monitor while using text-to-speech assistive technology. The study also examines whether text length, text format, and the language of the text (L1 or L2) influence the use of the software. The findings of the study suggest that there is evidence that students with SpLDs read while using text-to-speech software, which, if confirmed by further investigation, may reduce the concern for weakened test validity. The study also finds that students make varied use of the software to deal with different text lengths, formats and languages. This adds to evidence informing Italian universities of test item formats that are most accessible to students with such difficulties. Information from stimulated verbal recall and interviews also provides insights into the strategies students with SpLDs use to compensate for their individual learning difficulties