| Abstract: | 111 average and above-average 5th grade students read 1 of 7 short, nonnarrative passages and wrote answers to a set of either textually explicit (TE), textually implicit (TI), schema-based (SB), or text-irrelevant (control) questions. One week later Ss were given a written free-recall test. The results of the analysis of the S's responses to the treatment questions indicate that TE and TI questions promoted text-based question–answer interactions, whereas SB questions promoted both text-based and SB interactions. The recall data indicate that TI question–answer interactions resulted in the generation of larger proportions of text-based inferences than the other types of interactions, without any loss in the amount of explicit information, text-based inferences, and SB inferences. Finally, SB question–answer interactions and interactions with text-irrelevant questions both resulted in larger proportions of SB inferences than either TE or TI interactions. These results are consistent with a schema–theoretic view of reading comprehension. (14 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved) |
