As maker spaces, engineering design curricula, and other hands-on active learning tasks become more popular in science classrooms, it is important to consider what students are intended to take away from these tasks. Many teachers use engineering design tasks as a means of teaching students more general science principles. However, few studies have explored exactly how the design of these activities can support more generalized student learning and transfer. Specifically, research has yet to sufficiently investigate the effects of task design components on the learning and transfer processes that can occur during these kinds of tasks.
This dissertation explores how various task manipulations and focusing processes affect how well students can learn and transfers science concepts from an engineering design task. I hypothesized that learning goals that focus students on the deep structure of the problem, and contrasting cases that help students notice that deep structure, would aid learning and transfer. In two experimental studies, students were given an engineering design task. The first study was a 2x2 between subjects design where goal where goal (outcome or learning) and reflection (on contrasting cases or the engineering design process) were manipulated. A subsequent second study then gave all students contrasting cases to reflect on, and only the goal manipulation was manipulated. Results showed that learning goals improved student performance on a transfer task that required students to apply the deep structure to a different engineering design task. In the second study, learning goals improved student performance on a transfer test. Transfer performance in both studies was predicted by the ability to notice the deep structure during the reflection on contrasting cases, even though noticing this structure did not differ by goal condition. Students with a learning goal valued the learning resources they were given more during the engineering design activity, and this perceived value of resources was linked to greater learning.
A qualitative case study analysis was then conducted using video data from the second study. This case study investigated noticing processes during the building process, partner dialogue, and resource use. This analysis showed how high transfer pairs were better able to focus on the deep structure of the problem. Results suggest that what students noticed didn’t differ much between the various pairs. However, high transfer pairs were better able to focus on the deep structure through establishing a joint understanding of the deep structure, sustaining concentration on that deep structure during the cases reflection, referencing resources to identify features to test, and then systematically testing those features to identify their relevance. These processes are discussed in relation to how they differ in low transfer pairs.
This dissertation consists of four chapters: an intro, two standalone journal articles, and a conclusion. The first chapter provides a conceptual framing for the two journal articles, and discusses the findings from these articles in conversation. The second chapter describes the two empirical studies investigating how task goals and contrasting cases affect learning, and transfer from an engineering design task. The third chapter describes the comparative case study of how mechanisms of focusing on the deep structure differ between high and low transfer pairs. Finally, the fourth conclusion chapter discusses the implications of the work from both of these papers.
| Identifer | oai:union.ndltd.org:columbia.edu/oai:academiccommons.columbia.edu:10.7916/D8418DHB |
| Date | January 2018 |
| Creators | Malkiewich, Laura Jane |
| Source Sets | Columbia University |
| Language | English |
| Detected Language | English |
| Type | Theses |
Page generated in 0.0084 seconds