WHAT KINDS OF QUESTIONS DO UNDERGRADUATES GENERATE WHILE EXPLORING DATA VISUALIZATIONS PERTAINING TO SEA LEVEL RISE AND CLIMATE CHANGE?

Question-asking in science is a necessary step towards generating hypotheses, making decisions, and solving problems. However, research on student question-asking behaviors has been sparse. In this study, undergraduates engaged with geoscience data visualizations and generated questions about what they were seeing. Visualizations came from the Polar Explorer app (https://thepolarhub.org/project/polar-explorer), which offers numerous data maps pertaining to changing sea level and global climate. Experimental conditions allowed comparison between: (a) question-generating vs. non-question generating; (b) use of the app vs. use of a paper atlas with the same visualizations; (c) exposure to a map set about causes of sealevel change vs. a map set about who is vulnerable to sealevel change; and (d) a prompt to generate as many questions as possible vs. to write down questions you would like to ask the scientist who collected the data. Pre- and post-activity questions probed knowledge and belief about climate change and scientific data.

From the questions that students generated, we developed a taxonomy of question types, with the following major categories: Non-question; Questions about the app; Questions about data or data representation; Questions about the Earth. Students asked more and deeper questions about the Earth, a moderate number of questions about the data and its representations, and few questions about the app. The "generate many questions" prompt elicited more questions than the "Ask a scientist" prompt. Within "Questions about the Earth," major subcategories are Causality, Consequences, Other Data, and Mitigation/Adaptation/Intervention. Causality is by far the most abundant. Within “Causality,” the most sophisticated questions queried an apparent discrepancy between an aspect of the mapped data and the students’ mental model, suggested a process or mechanism that may have influenced the mapped phenomenon, or asked about the cause of a temporal or spatial trend or pattern that the student had noticed. When asked, post-activity, how likely they would be to engage in future climate-related activities, student in the question-generating condition were more likely to "seek out data maps when I have questions about the Earth’s climate" than were non-question-generating participants.