Item Writing and Dyslexic Students

Promoting Best Practices

Abstract

In language testing, item writing is an important process which can be creative, rewarding but also challenging and sometimes frustrating. Quality items are fundamental for test validity and item writers can be formally trained in order to improve what they write (Rossi and Brunfaut 2019).
The present study analyses some items specifically written for dyslexic students as part of a high-stakes, Internet-based B1 English test administered to undergraduate students in an Italian university, namely the University of Modena and Reggio Emilia, where the number of dyslexic students enrolled has increased exponentially in recent years. The aim was to investigate whether the accommodations suggested by the item reviewer succeeded in removing unnecessary barriers which represent an unintended bias, while preserving the test construct and consequently its validity (Pelleriti 2018; Kormos and Smith 2012).
For the present investigation, a questionnaire was devised, and four certified dyslexic students were interviewed on a voluntary basis. The data collected during the semi-structured interviews confirmed that some of the suggested accommodations would be beneficial, whereas others would be detrimental to candidates with dyslexia. As a consequence, some conclusions have been drawn, with the aim of sharing best practice among the language testing community—in particular language testers, test developers, and item writers. Nevertheless, this research has also confirmed that item writing undoubtedly deserves more scholarly attention, in an attempt to shed light on this pivotal aspect, which is at times neglected despite being one of the pillars in language testing.