Big, Little, or Both? Exploring the Impact of Granularity on Learning for Students with Different Incoming Competence
- Guojing Zhou, North Carolina State University, Raleigh, North Carolina, United States
- Xi Yang, Computer Science, North Carolina State University, Raleigh, North Carolina, United States
- Min Chi, Computer Science, North Carolina State University, Raleigh, North Carolina, United States
AbstractWe explored the impact of three types of decision granularity on student learning by comparing three conditions: problem level (Prob), step level (Step), and both problem and step levels (Both). We first conducted an empirical study to directly compare the three conditions and then three subsequent studies to evaluate one or two of the three conditions. In a post-hoc analysis, all students were combined by their conditions. Overall, no significant difference was found among the three conditions either in the empirical study or in the post-hoc analysis. We further split students into High, Medium, and Low groups based on pre-test scores and our post-hoc analysis suggested that Prob can be effective for Low students but ineffective for High ones; on the other hand, Step could be effective for High students but ineffective for Low ones; finally, the Both condition seemed to be effective for both High and Low students.