Making graphs with academic software tools (SPSS, Stata and R) Paul Lambert, University of Stirling Presentation to the Scottish Civil Society Data Partnership Project (S-CSDP), Webinar 6 on ‘Methods of data analysis: Approaches to visualisation’ 7 Apr 2016
Making graphs with academic software tools Graphical summaries of statistical data… – ‘Classical’ or routinised tools of graphical summary (e.g. scatterplot, bar chart, line plot) – Integration of quantitative data with graphical summary via analytical software – Used in analysis and/or publication Recent developments in graphical options via standard academic software – Innovations in presentation modes (e.g. multiple ‘trellis’ plots; representations of uncertainty) – Analytical innovations with their implications for displaying results (e.g. Bayesianism) – Online appendix options (e.g. interactive graphs) Standard academic software doesn’t cater to all aspects of visualisation (cf. artwork) S-CSDP, 7 Apr 20162
Academic software and ‘documentation for replication’ Ideally work with ‘command scripts’ or ‘syntax’ For graphs, benefit is reproducibility, adaptability, and collaboration capacity S-CSDP, 7 Apr Next: Some brief examples of making selected graphical displays in Stata, SPSS & R: scsdp_webinar6.{sps/do/R}
Stata scsdp_webinar6.do; [Mitchell 2012] Innovative Stata: e.g. trellis plots; summaries of summary statistics; predicted values… S-CSDP, 7 Apr Classical Stata: crisp display options with considerable control over formating
S-CSDP, 7 Apr Classical SPSS: accessible defaults, with editing options for colours, patterns, text SPSS scsdp_webinar6.sps Innovative SPSS: e.g. of model-oriented interactive graphics
Innovative R: plausible simulated regression lines (using simulation functions and ‘qqplot’ extension library) S-CSDP, 7 Apr Classical R: combined plots; matrix plots R scsdp_webinar6.R;
Summary: Graphing with academic tools 1)Wide range of statistical summary graphs available across software packages 2)Attractions of syntax-based graph construction 3)Academic software tools are not comprehensive (cf. popular and specialist visualisations) S-CSDP, 7 Apr References cited Bulmer, M., Gibbs, J., & Hyman, L. (Eds.). (2010). Social Measurement through Social Surveys: An Applied Approach. Aldershot: Ashgate. Dorling, D. (2013). The Population of the UK, 2nd Edition. London: Sage. Huff, D. (1954). How to Lie with Statistics. London: Gollancz. Mitchell, M. N. (2012). A Visual Guide to Stata Graphics, Third Edition. College Station, Tx: Stata Press.