Data Analytics (DA) is a field that merges quantitative methods in statistics, data mining, and computer science to discover and summarize information in data. In practice, a good application of DA requires comprehensive critical thinking skills that (1) extend beyond the application of quantitative statistical or computational methods and (2) include qualitative forms of thought, such as formalizing potential biases, communicating personal judgment, exploring multiple solutions, assimilating new information with old, and assessing implications of discoveries. Unfortunately, current methods in teaching DA focus primarily on its quantitative aspects; only after students master quantitative theory and methods do they have an opportunity to think critically about applications. Students who fail to complete current DA courses neither develop comprehensive DA skills nor experience how complex quantitative summaries may advance knowledge.
The investigators are attempting to revolutionize current practice in teaching introductory DA to first- and second-year college students. Specifically, they are developing a new course, "Critical Thinking with Data Visualization" (CTDV), which synchronizes teaching methods in DA with critical thinking and makes DA accessible to diverse learners. CTDV includes four modules, each of which is motivated by a different real-world case study and focuses on one DA method (e.g., multidimensional scaling). CTDV uses novel interactive data visualization software (and techniques therein) as a platform for students to build from what they know and construct their understanding of (1) how to think critically, (2) the role of data in problem solving, and (3) some mathematical and computational methods for summarizing complex datasets. The software is based on methods called Bayesian Visual Analytics (BaVA), which have been developed by the investigators (see NSF Award No. 0937071). BaVA enables domain experts--e.g., biologists or homeland security analysts--to assess complex datasets and incorporate their expert knowledge within quantitatively rigorous data characterizations, without technical training in statistics or computer science. For this reason, BaVA fosters creative, critical thinking and problem solving with data and is ideal for experiential learning of DA concepts in the classroom.
The project team is developing CTDV (including its objectives, curriculum, lesson plans, case studies, and course materials), designing the interactive data visualization software (including tests of usability), implementing CTDV (or modules within it), and assessing the success of CTDV. The assessment includes both quantitative (e.g., tests) and qualitative measures (e.g., surveys and interviews) that take place before, during, and after the students experience CTDV. In addition, the team is disseminating CTDV via a comprehensive Web site, papers, and conference workshops.
