Teaching Statistics Trust prize for 2021 special issue

Teaching Statistics are happy to announce that the Teaching Statistics Trust has awarded a prize for the best paper in the 2021 special issue, Teaching Data Science and Statistics: foundation and introductory, to Anna Fergusson and Chris Wild for their paper On traversing the data landscape: Introducing APIs to data-science students. Statistics and data science and their teaching are intrinsically linked. This is seen not only in the increasing inclusion of technology in teaching statistics, but also in the data and contexts considered, and the broadening of statistical issues, explorations, presentations, and discussions at introductory levels, whether school, undergraduate or postgraduate/workplace in other disciplines. The intent of the special issue is to provide impetus and inspiration to all readers and authors in furthering this progress, and to celebrate the new subtitle of the journal, in the increasing awareness of what data science is, and how statistics and data science work together in tackling real and complex datasets and problems involving complex data. Data science is much more than a new set of tools it opens doors to whole new ways of thinking about information, explanation, and action, and the special issue demonstrates what an extraordinarily rich field this is and just how much challenge and opportunity there are that could, and should, be considered by the statistical and data science community. Amongst the excellent papers illustrating a wide variety of approaches and offering some very rich examples for teaching in this emerging space, the special issue editors, after much debate, chose the winning paper because of the importance of harvesting the vast amounts of data now available combined with authentic student engagement in enquiry-based learning in a fun and universally appealing context. The pedagogic approach is an excellent demonstration of the long-time advocacy of leading statisticians and statistical educators of students learning of technical tools and statistical thinking via graduated needs arising in the tackling of a real data investigation that piques student curiosity and exploration. The proposal, using APIs, is unique and cutting edge, but is explained in an extremely clear way. It centers on the importance of the data gathering phase in data science (at least when it comes to data scraping), and mastery of this skill not only empowers students, but teaches them that the internet really is just organized data. However such approaches cannot succeed without careful scaffolding, preparation and deep understanding of student needs in learning about data. Students move from immersion in a search activity (for photos) to URL hacking and GUIdriven tools, to thinking of variables and then to API’s. Graphical explorations are then encouraged to at least partially discuss some of the questions that have arisen during a student’s personal journey in the investigation. The approach is simple, well written, directly usable, asks questions students will engage with, and readers will tend to want to try out the activities for themselves. The activities described can be used across the curriculum and with a variety of age groups, and hacking APIs could appeal to a variety of students. As well as influencing curricula, statistics courses could use this approach as an extra project, or computer science classes could be enticed to pay more attention to data. Since its inception in 1979, Teaching Statistics has always aimed to emphasize good practice in teaching statistics and statistical thinking in any context. As long advocated by professional statisticians and leading statistical educators, good practice in teaching statistics should reflect the practice of statistics in the fullest sense, integrating the principles and practice of data investigations with statistical literacy in congruence with current statistical developments and usage in real world contexts and problems. Statistics has long been both a user and driver of computing technology, and to tackle increasingly vast amounts of data and progressively more complex problems in ever more diverse contexts, the statistical sciences have been rapidly developing, and been involved in, more and better methods and technologies. We congratulate the authors for their excellent paper, and congratulate all authors in the special issue for their valuable contributions to this critically important area of teaching data science and statistics.