دانلود رایگان مقاله اثرات استراتژی تست آنلاین دانش آموز بر ساخت دانش

Abstract

The differential effects of online student-constructed tests (SCT) and student-generated questions (SGQ) strategies on knowledge construction in term of the breadth, depth, interconnectivity and elaboration of knowledge were examined via a quasi-experimental research design. Two classes of undergraduates (N = 65) participated and were randomly assigned to two different treatment groups. An online system supporting the associated learning activities for the respective groups was developed. The results from the ANCOVA showed that students who engaged in SCT activities generated questions that covered significantly more concepts, involved significantly more levels of subordinate concepts, and built significantly more links between clusters of study topics, as compared to those in the SGQ group. Moreover, significantly more students in the SCT group engaged in item revision behavior than those in the SGQ group, and a majority of students in the SCT group exhibited item sequencing behavior, both of which are deemed indicative of knowledge elaboration. Suggestions and implications for instructional implementation, online system development and future studies are offered.

4. Discussion and conclusions

Based on the results of the data analysis, it was found that concepts covered in questions generated by the SCT group enclosed significantly more nodes, reached significantly more hierarchical levels, and created significantly more cross-overs, as compared to those of the SGQ group. In other words, the students who engaged in the SCT activities tended to generate questions that not only covered significantly more concepts with regard to the study material, but also involved significantly more levels of specific, subordinate connected concepts, and these students built significantly more links among different clusters of topics, in comparison to their counterparts in the SGQ group. As suggested by a number of researchers (e.g., Hoeft et al., 2003; McClure et al., 1999; Novak & Gowin, 1984), cross-links between concepts, including the hierarchical links between subordinated concepts and links between different segments or domains of knowledge in concept-maps, could be a valid and reliable indicator for students’ integrative and in-depth understanding of the learning content. It can thus be reasonably inferred that the SCT promoted better knowledge integration and integrative reconciliation of the meaning of the content, as compared to SGQ. Moreover, it was found that a significantly larger proportion of students in the SCT group engaged in item-revision behavior as compared to those in the SGQ group, and that a majority of students in the former utilized the built-in itemsequencing function to re-sequence questions for their inclusion in a test. Because both item-revision and re-sequencing behavior involve various types and extents of cognitive adjustments and fine-tuning to meet the task criteria (e.g., assessing and clarifying the completeness, correctness, appropriateness, similarity, and difficulty of different pieces of knowledge contained in the study material, and making amendments accordingly, which are indicators of knowledge elaboration), SCT was shown to encourage students to be more involved in the refinement of their knowledge.