How accurately can learners discriminate their comprehension of texts? A comprehensive meta-analysis on relative metacomprehension accuracy and influencing factors

Over the last three and a half decades, much research has been conducted on how accurately learners can discriminate their comprehension of texts, that is, relative metacomprehension accuracy. The evidence suggests that this skill is usually not well-developed among learners. To assess the average level of relative metacomprehension accuracy and to explore which factors might influence this level, we conducted a comprehensive meta-analysis. The results obtained from 145 independent subgroups of 94 empirical studies indicate that relative metacomprehension accuracy is only moderate (r = .24, 95% CI [.21, .26]). At the same time, the level of relative metacomprehension accuracy was found to be moderated by a number of factors related to the learners (e.g., education level), texts (e.g., practice texts), judgments (e.g., time point of predictions), and tests (e.g., question format).