Abstract
A meta-analysis of 20 experiments examining incidental word learning during normal reading shows that students learn around 15% of the unknown words they encounter A test of homogeneity indicates that study outcomes diverge, although their heterogeneity is relatively small (19%). An exploratory multi-level analysis of the variability in the results suggests that several factors affect the probability of learning an unknown word while reading: pretest sensitization, students' grade level, students' level of reading ability, the sensitivity of assessment methods to partial word knowledge, and the amount of text surrounding the target words. A model that contains students' grade level and assessment methods' sensitivity to partial word knowledge predicts 66% of the systematic variance in the effect sizes. Implications for research and instruction are discussed.
Original language | English |
---|---|
Pages (from-to) | 261-285 |
Number of pages | 25 |
Journal | Review of Educational Research |
Volume | 69 |
Issue number | 3 |
DOIs | |
Publication status | Published - 1999 |
Externally published | Yes |
Keywords
- CONTEXT
- VOCABULARY
- MEANINGS
- COMPREHENSION
- INSTRUCTION