This website describes the vignette to accompany the K12 data science tools paper (see link at bottom).
These materials were created by Daniel Pimentel, Nicholas J. Horton, and Michelle H. Wilkerson to accompany one of the papers commissioned for the NASEM Foundations of Data Science for Students in Grades K-12 workshop.
Opinions and statements included in the paper are solely those of the individual authors, and are not necessarily adopted, endorsed, or verified as accurate by the Workshop Planning Committee on Foundations of Data Science for Students in Grades K-12, the Board on Science Education, the Board on Mathematical Sciences and Analytics, the Computer Science and Telecommunications Board, or the National Academy of Sciences, Engineering, and Medicine.
The NASEM K-12 Data Science workshop was supported by the Valhalla Foundation; additional support for MHW was made possible by National Science Foundation grant IIS-1900606.
To explore how the selection of data analysis tools can fundamentally shape what is highlighted in an investigation, we conducted a comparative analysis using free, popular tools from distinct genres: R (scripting), CODAP (visual), Google Sheets (spreadsheet), and Tableau. Links to the vignette, dataset, and source files are provided below.
Title: Tools to Support Data Analysis and Data Science in K-12 Education
Abstract: There has been a proliferation of tools for teaching data analysis and data science at the middle and high school levels. While a few frameworks for systematically exploring the affordances and constraints of such tools exist, most work has only explored one or a few tools at once, or has not focused on K-12 usage. In this paper, we blend first-hand comparative analysis methods and supplemental literature review to conduct a systematic analysis of several common data analysis software packages in use at the K12 level. Using an adaptation of a framework proposed by McNamara (2019), we grouped the tools into related genres. Spreadsheets, while familiar and accessible to many, lacked many desirable features. Visual tools (e.g., CODAP, Social Explorer, iNZight) lower the barrier for data exploration, but may not easily support more advanced statistical tests. Scripting tools (e.g., Python, Pyret, R) provide great flexibility but with increased degree of difficulty. Looking across tools and genres, our analysis suggests that these genres boast complementary strengths depending on students’ developmental and investigative needs. We make recommendations for the design and use of tools, notably highlighting the importance of working across different tool types as a part of data practice.
Last updated November 8, 2022