Methodological Issues and Further Research in the Study of Reading Comprehension with Different Levels of Knowledge

Some methodological issues in the study of levels of knowledge are reviewed, and needs for further research are explored, drawing on an analysis of 12 studies reported since the late 1970s. In the 12 studies, 16 quantitative experiments were conducted. These were assessed for internal and external validity. Analysis revealed some shortcomings in study design, some confounds in data collection, some threats to statistical procedures, and other issues related to external validity. For example, in 11 of 12 experiments, subjects were not randomly selected. None of the 16 experiments included a discussion of data assumptions for the statistical procedures researchers chose to use. In 8 of the 16 experiments, no information was given about the number of subjects assigned to each group. Several other issues were found that might have affected validity. In addition, none of the studies reported the effect size of significant results. Research in more natural settings is required, and further study of the compensatory effects of prior knowledge is required. Research into age differences and on the effects of prior instruction is needed. Two tables present study data. (SLD)