The DfE makes a large amount of statistical data available to the public, for free and online. In this session, Miles introduces participants to the data set. He demonstrates how simple analysis can be done using pivot tables in Microsoft Excel, and how the statistical programming language R can be used to create visualisations, interactive graphics and maps of the data. He shows how regression analysis can be used to uncover statistically significant relationships. Along the way, participants will explore how the intake characteristics of schools can predict exam performance, as well as identifying the schools that significantly outperform those with similar intakes.
The session will be of interest for anyone interested in learning how qualitative techniques can be applied to open data at national scale, but would also be useful for those interested in using R and Excel to explore the trends and exceptions present in assessment data for an individual school.
Miles is principal lecturer in computing education at the University of Roehampton. Prior to joining Roehampton, he spent 18 years in four schools, much of the time as an ICT coordinator and most recently as a head teacher. He is a board member of Computing At School, the BCS Academy of Computing and the CSTA. He is a fellow of the BCS, RSA and HEA, and a member of the Raspberry Pi Foundation.
Over the years he has contributed to a number of computing related projects including the national curriculum computing programmes of study, Switched on Computing, Barefoot Computing, QuickStart Computing, CAS TV, Project Quantum, Hello World and Roehampton’s annual computing education report, TRACER.
Computing made it onto the national curriculum in 2014, replacing the much disparaged ‘ICT’, and take up of the GCSE qualification has increased rapidly over the last few years. The English computing education community are finding out for themselves what appears to make for effective practice in this subject, which is high risk, and quite exciting. Miles presents an overview of what pedagogic approaches might perhaps work in computing education, drawing on some of the research that’s been done into teaching generally, what happened with Logo programming back in the 70s and 80s, undergraduate CS education, and some of the recent work on teaching programming (and other aspects of computing) to children in school. He’ll also share Roehampton’s analysis of data from the National Pupil Database, as to how things are going so far for computing.