Estimating dyslexia in the web

Abstract

In this study we present an estimation of texts containing English dyslexic errors in the Web. A classification of lexical errors is proposed and unique dyslexic errors are distinguished from other kind of errors due to spelling and grammatical errors, typos, OCR errors and errors produced when English is used as a foreign language. A representative sample of each kind of error is used to calculate a lower bound for the prevalence of dyslexia in the English Web. Although dyslexia has been studied in the context of Web accessibility, to the best of our knowledge, an estimation of Web texts containing dyslexic errors was unknown. Our results are useful to tackle future work in Web accessibility among dyslexic users focusing not only in the interface but also in the text content.