Environmental data-driven inquiry and exploration (Project EDDIE): A model to engage students in quantitative reasoning and scientific discourse
Scientists are increasingly using sensor-collected, high-frequency and long-term datasets to study environmental processes. To expose undergraduate students to similar experiences, our team has developed six classroom modules that utilize large, long-term, and sensor-based, datasets for science courses designed to: 1) Improve quantitative skills and reasoning; 2) Develop scientific discourse and argumentation; and 3) Increase student engagement in science. A team of ten interdisciplinary faculty from both private and public research universities and undergraduate institutions have developed flexible modules suitable for a variety of undergraduate courses. These modules meet a series of pedagogical goals that include: 1) Developing skills required to manipulate large datasets at different scales to conduct inquiry-based investigations; 2) Developing students’ reasoning about statistical variation; and 3) Fostering desirable conceptions about the nature of environmental science. Six modules on the following topics are being piloted during the 2014-15 and 2015-16 academic years prior to broad dissemination including 1) Temporal stream discharge evaluation using USGS data; 2) Temporal stream nutrient loads and eutrophication risk using USGS and Long Term Ecological Research (LTER) data; 3) Climate change using NOAA weather and Vostok ice core data; 4) Lake ice-off dates using Global Lake Ecological Observatory Network (GLEON) data; 5) Thermal dynamics in lakes using GLEON data; and 6) Lake metabolism dynamics using GLEON data.
To assess achievement of the pedagogical goals, we used pre/post questionnaires and video-recordings of students working on modules. Questionnaires contain modified items from the Experimental Design Ability Test (Sirum & Humberg 2011), the Views on the Nature of Science questionnaire (Lederman et al. 2001), and a validated instrument to measure students’ ideas about variation (Watson et al. 2003). Information gained from these assessments and recordings will allow us to determine whether our modules are effective at engaging students and increasing their quantitative skills. Feedback will also be used by the faculty to revise the modules before they are posted online for widespread dissemination in 2016. Our initial results suggest that students appreciate the value of high-resolution and long-term data, and that working with large datasets cements the “real world” application of basic ecological and environmental concepts. This project is funded by an National Science Foundation Transforming Undergraduate Education in STEM grant.