Search for: All records

The data sets used in statistics education have changed over time, from mathematically “well‐behaved” ones that facilitated computation, to more context‐rich sources and now, with the increasing influence of data science practices, to “found” data, often from open data sites. As data sources change, it is important for educators to take a fresh look at the ways we engage students in thinking about the processes that generated the data they encounter. The use of already collected data requires particular attention because many of the decisions that went into the processes of obtaining the data are hidden. Students need to learn to ask “Who, When, How, Where, and Why?” data were collected and to wonder if the data really measure what needs to be measured. Our advocacy in this paper is to deepen the educational treatment of data production to better reflect the current and future practice of statistics and data science.

What happens when a diverse group of youth ages 11 through 14 are introduced to data science using authentic, public, multivariate data in an out-of-school context assuming no special prerequisite knowledge? We designed three 10-hour Data Club modules in which real-world data and the questions students asked of such data drove the learning process. Each module was grounded in a topic that youth connected with at a personal level. Youth learned how to use a free online data platform that made it easy to rearrange, group, filter, and graph data. Within the progression of the module, we used youths’ own questions, data moves, and data visualizations to engage them in critical inquiry and foster productive habits of mind for working with data. Our goal was for youth to emerge from the Data Clubs experience feeling empowered to interact with, ask questions of, and reason about and from data.