ABSTRACT
The primary aims of this mixed method study were to (a) examine the effectiveness of a brief inference intervention, (b) compare the types of knowledge-based inferencing errors less skilled middle grade readers make, and (c) evaluate if self-reported cognitive load relates to inferencing. Participants (n = 17) were randomly assigned to a graphic organizer-inference intervention (GO-Inference) (n = 9) or business as usual (BAU) condition (n = 8), and differences between groups were explored for each study purpose. Quantitative and qualitative results suggested that while less skilled readers in the GO-Inference condition made modest progress in forming knowledge-based inferences, they continued to struggle to distinguish relevant versus irrelevant information from text and/or retrieve knowledge necessary to form inferences. Students in the BAU condition were more likely to make errors such as providing irrelevant information or failing to respond. Additionally, students in the GO-Inference condition reported lower cognitive load during inference-making tasks.
ABSTRACT
We investigated differences in knowledge-based inferencing between rural, middle grade monolingual English-speaking students and English learners. Students were introduced to facts about an imaginary planet Gan followed by a multi-episode story about Gan. Participants were tested on the accuracy of fact recall and inferences using this knowledge at three time points (i.e., immediate, one-week, and one-month follow-up). Results show that monolingual English-speaking students significantly outperformed English learners on the inference task. Both subgroups made elaborative inferences more accurately than coherence. Students' ability to recall knowledge base facts was the strongest predictor of their ability to accurately make inferences using this knowledge at each time point.
