Program Visualization: Effect Of Viewing Vs. Responding On Student Learning

Visualizations in computer science topics are known to have several benefits such as promoting conceptual and procedural understanding, improving prediction and reasoning abilities and helping learners construct mental models. This learning effectiveness has been found to be a function of students‟ engagement level with visualization. In the current study, we did a controlled field experiment to determine the effect of two different instructional strategies with visualization on procedural understanding of the topic of pointers in a 1st year programming classroom. These instructional strategies, operationalizing different engagement levels, were: prediction activity interleaved with instructor feedback using visualization (experimental), and simply viewing the visualization with parallel instructor commentary (control). We found significant difference in the relative rate of correct solution of the procedural questions on the post-test. However, there was no significant difference on the post-test scores. We also found a significant difference in classroom behavioral engagement between the two groups. We propose that there may be conditions, other than engagement level with visualization, such as learner characteristics or challenge level of assessment questions that may play a role in the determining learning effectiveness of visualizations.