Do modality and frequency matter in predicting reading proficiency from vocabulary knowledge? A comparison of meaning recognition and meaning recall

Vocabulary knowledge is a strong predictor of reading ability, yet findings remain inconclusive regarding how different modalities of vocabulary knowledge, such as meaning recognition and recall, predict reading comprehension. This study compared the predictive strength of these modalities for reading ability across different frequency levels, and examined whether these effects differed between higher- and lower-proficiency readers and extended to learners whose L1s are linguistically distant from English (i.e., non-Indo-European languages) with no shared cognates. Results showed that (a) both vocabulary tests correlated significantly with reading comprehension but meaning recognition emerged as a superior predictor of reading ability, (b) combining recognition and recall did not substantially improve the prediction of students' reading abilities, (c) tests covering the first three 1000-word levels showed similar correlations, but at 4 K and 5 K levels recognition emerged as the superior predictor of reading ability to a substantive and statistically significant degree, (d) reading proficiency did not significantly change these tendencies, indicating that meaning recognition's stronger sensitivity to expanded frequency coverage held across both higher- and lower-proficiency readers, and (e) these patterns also held for learners whose L1s are linguistically distant from English. These findings suggest that item type (recall or recognition) significantly influences the ability of vocabulary tests to predict reading comprehension, and that meaning recognition formats may be particularly effective for predicting reading when lower-frequency words are included in the test.