COMPARATIVE STUDY ON RELIABILITY BETWEEN ONLINE AND PAPER-BASED VERSIONS OF A TEST FOR READING IN ENGLISH AS A FOREIGN LANGUAGE

The aim of this study is to examine how reliable the results of the internet-based version of a reading test are, with a view to replacing paper tests with the online versions. This is of increasing importance as universities focus on improving efficiency and supporting SDGs by going paperless. The study was also suggested by the need to test reading comprehension of larger numbers of students across the university and deliver meaningful results on which to base intensive programs of instruction quickly. The Extensive Reading Foundation's online reading test and the (now discontinued) Edinburgh Project for Extensive Reading's placement test (paper-based), were administered to university students under controlled conditions, and the data was analyzed for possible relationships. An initial one-way ANOVA analysis of the results suggested little evidence of a relationship between online and paper-based test scores. However, further analysis using other measures found evidence of interaction between them, and a second ANOVA analysis, only of scores for students who had completed all versions of the test found a significant relationship. Familiarity with both versions of the test was considered as a possible factor. Although this is only a small-scale study, the findings help to support the argument for adopting the online version of the test, with its various potential benefits to schools and educators.