Test Format and Test Performance. A Meta-Analysis

      Аннотация: Research on the efficacy of test formats has produced mixed results, despite reports of significant improvements in many studies. The aim of this study was to report a meta-analysis that synthesizes and determines the overall effects of test formats (cloze test versus c-test). The present meta-analysis was undertaken to synthesize the findings of all experimental studies that were published from 1990 to 2016. Based on the inclusion and exclusion criteria, a total of 22 study reports were identified to be included and, accordingly, these studies generated 39 effect sizes. Each study was then coded for substantive and methodological features. Study outcomes were arranged by Hedges' g. The results showed a medium random effect (g= 0.506) and a large effect size in the fixed effect model (g=0.607). Drawing upon the analysis of moderator variables, the effectiveness of the following variables on cloze test and c-test formats were found: deletion pattern, first language, number of items, participants’ gender, reliability, scoring method, test time, and author’s gender.

Ключевые слова: test format, test performance, Meta-Analysis, Cloze test, C-Test