By Regina Lisinker (University of Minnesota)
Information
This poster presents three activities for an introductory data visualization course designed to be implemented with a Productive Failure pedagogy (Kapur, 2008; Kapur & Bielaczyc, 2012). Productive Failure (PF) is a learning design where students tackle open-ended, complex problems, often leading to initial "failure", before receiving direct instruction. The three activities asked students to recreate detailed data visualizations with features they had and hadn't seen before. Typically in the course, students worked in groups of three to complete guided step-by-step activities where new features of graphs were introduced alongside code and examples using ggplot2 in RStudio. However, on days using the PF design, students were shown a graph and asked to re-create it, rather than following structured directions. Graphs varied as the class progressed and students were asked to re-create a bar chart, histogram, and scatter plot, each with up to five new features (e.g. changing title color, importing icons for scatter plot points). As a feature of PF, students were not provided with any guidance: questions were met with words of encouragement and told to keep trying. Each activity ended with a comparison of students’ code, resulting plots, and a classroom discussion of major challenges, failures, and successes. The activities will be shared along with examples of student work. Student comments from class discussions will be highlighted, as well as instructor reflections on classroom culture and student engagement. These activities were implemented in an introductory data visualization course of 42 students at a large midwestern university.