In this study, researchers explored how reading and writing from several documents, which may be contradictory, affected how accurately students could assess and regulate their understanding of complex, multi-source material. After giving 163 students several historical documents to review, the researchers assessed their comprehension by instructing each participant to write an essay that integrated and summarized the texts.
Next the researchers randomly assigned 109 students to write three additional short responses in a forum-post format, while the remaining 54 students took a break lasting less than 5 minutes. Of the response writers, 57 were randomly assigned to receive a guided rubric in the format of a forum post written by an expert to serve as a model for self-evaluation. Each “expert post” was presented alongside an explanation describing why it was considered a strong response. The 54 students who took a break and did not write responses were also given the rubric to review. Then the researchers measured metacomprehension accuracy (how well students judge their own understanding) and regulation (how well they adjust their learning strategies) using several computational models.
Contrary to their hypothesis, the researchers found that response-writing did not improve metacomprehension accuracy; in fact, it unexpectedly lowered accuracy for dimensions related to reading multiple texts. Furthermore, the researchers found that self-evaluation did not improve when students generated responses based on their essays; however, self-evaluation did improve when students engaged in response-writing and then compared their work with the expert models. When the external standards were added, learners judged their essay performance more accurately than learners who received the rubric only. The researchers concluded that while generative tasks such as response writing can offer useful cues for learners, comparing to a model appears necessary for more accurate comprehension, self-evaluation, and self-regulation.
StepUp Note
Self-evaluation is the key to independent learning. In this research, we see that students were able to look at a model, then structure their own work to match the key concepts of the model. The ability to look at a model, and to independently “match” the critical details of the model, applies across many different kinds of learning. StepUp software programs require independent learning: watch and learn then think and do. Students need to tell themselves what to do and when to do it. Only an independent learner becomes an independent adult!
Note by Nancy W Rowe, MS, CCC/A