Visualization technology, no matter how well it is designed, is of little educational value unless it engages learners in an active learning activity.
Overview of Best Practices: Provide resources that help learners interpret the graphical representation, adapt to the knowledge level of the user, provide multiple views, include performance information, include execution history, support flexible execution control, support learner-built visualizations, support custom input data sets, support dynamic questions, support dynamic feedback, and complement visualizations with explanations.
Results from surveys conducted through this research show that many visualization tools are not designed for the content, but more for the visualization/animation. It is recognized that algorithm simulation promotes learning, but it is not clear whether it is from exercises/automatic assessment, or from the visual form itself.
The six different forms of learner engagement with visualization technology are (not in ordinal scale): No viewing, Viewing, Responding, Changing, Constructing, and Presenting. There is possible overlap (of any combination) between the last five, with “Viewing” forming the universe the last four must occur.
Viewing is the core form of engagement, with the largest number of variations. It is the most passive, and includes auralization.
Responding’s key activity is answering questions presented by the system. Learners are asked questions including topics on prediction, coding, efficiency, and debugging based on the visualization, which may lead to further viewing activities.
Changing entails modifying the visualization, such as changing input.
Constructing allows for learners to construct their own visualizations of the algorithms in either direct generation (map a program/algorithm to a visualization or manipulate a visual representation to simulate the algorithm) or hand construction (learners use a drawing/animation editor to construct their own visualizations, which may or may not be executed).
Presenting entails presenting a visualization (possibly made by the learner) for feedback and discussion.
Studies on effectiveness of visualizations depend on a learner’s depth of understanding, which in order are: knowledge level, comprehension level, application level, analysis level, synthesis level, and evaluation level. Other factors include the learner’s progress, drop-out rate, learning time, learner satisfaction, learning style, familiarity with using visualization technology, learning orientation, and other background information.
The authors hypothesize (based on the six forms of engagement) that, viewing results in equivalent learning outcomes to no viewing, that each successive form by itself is better than the prior, and that a mix of several forms is even better.
Link to PDF