Given the challenges of working with and interpreting real-world data, learners in science classrooms too often are given simplified data sets with reduced variability, and are provided too few opportunities to apply conceptual understanding to interpreting highly variable real-world data. This may cause students to form misconceptions about what real datasets resemble and to reject scientific explanations based on, from their perception, “messy” data. We describe a curriculum that integrates the affordances of mobile broadband devices (e.g., smartphones) and environmental probeware to support students’ first-hand data collection and exploration of data variability in their local environment during a field trip to a pond near school.
The mobile technology-enabled curriculum was used in a design-based research context with one teacher and four classes of seventh grade students (n = 71) from a suburban school district in the northeastern U.S. Students used a multi-user virtual environment (MUVE) during a two-week period, followed by a one-week mobile intervention consisting of training sessions with mobile devices and environmental probeware, a 1.5 hour field trip to a local pond, and post-field trip discussion during a single 50-minute class period.
Learning outcomes were compared using pretest and posttest scores. Worksheet responses and log file data from the field trip, as well as transcripts from classroom discussions, were qualitatively analyzed to investigate changes in students’ thinking processes, such as their ability to describe data, understand variability, and the quality of their scientific explanations.
Results/Conclusions
This intervention enhanced students’ ability to understand variability and support claims with evidence. Analysis of the pre-post survey results indicated that students made significant gains on two of the three sub-measures. Students’ ability to describe data did not differ significantly between the pre and post survey. The sub-measure indicating students’ understanding of variability showed a statistically significant increase pre to post (Mpre = 21.5 (± 0.42), Mpost = 22.4 (± 0.37), T72 = -2.28, p = 0.025), yet the effect size was modest (Cohen’s d = 0.26). Students’ ability to support claims with evidence also increased significantly pre to post (Mpre = 15.9 (± 0.49), Mpost = 17.4 (± 0.47), T66 = -2.61, p = 0.011), with a moderate effect size of 0.37. These tools and resources enabled students to develop and demonstrate fairly sophisticated descriptions of highly variable environmental data and helped them to integrate data interpretation and explanation with personally meaningful experiences in their local environment.