Examining the Impact of Inference Instruction on the Literal and Inferential Comprehension of Skilled and Less Skilled Readers: A Meta-Analytic Review
Mastering reading is pivotal for success in school. A new study released in early 2017 examines strategies for teaching reading comprehension, one of the five essential instructional practices necessary for improving reading achievement, and also offers educators practical information on how to increase the effectiveness of reading comprehension instruction. The 2000 National Reading Panel report on reading identified comprehension along with phonemic awareness, phonics instruction, fluency, and vocabulary as requisite skills for effective reading. This new meta-analytic review evaluates the impact of various comprehension instructional practices and suggests a strong relationship between inference generation and reading comprehension. It found that inference instruction was effective for increasing general comprehension (*0.58), inferential comprehension (*0.68), and literal comprehension (*0.28). Another interesting finding was the difference in performance between skilled readers and less skilled readers. Less skilled readers improved in both inferential comprehension and literal comprehension, whereas skilled readers mostly improved in inferential comprehension. These findings suggest that all students can increase their inference ability and that less skilled readers experience the extra benefit of increased literal comprehension. The study concludes that inference ability is not a product of comprehension, but rather a plausible cause of reading comprehension performance.
* = Effect Size
Elleman, A. M. (2017). Examining the impact of inference instruction on the literal and inferential comprehension of skilled and less skilled readers: a meta-analytic review. Journal of Educational Psychology.
Additional commentary of interest on this study is available from Daniel Willingham.