Managed well, R can be a critical component of a transformation of the effectiveness and efficiency of an analytical team in government. R’s competitor in government is not Julia or Python or even SAS, but overwhelmingly Excel. The keys to making the most of R are not the latest and fanciest R packages, but integrating it into a new workflow. That workflow also uses Git and (probably) SQL. It breaks down micro-silos and the “lone genius who understands the spreadsheet”, replacing them with teamwork, transparency, reproducible analytical pipelines, peer review, and home-grown R packages and rules for use.
Doing this successfully is difficult and depends on process changes, firm direction from management, and a nuanced understanding of public sector incentives and risk aversion. It means challenging assumptions that public servants, non-IT contractors and management consultants don’t write code; and changing recruitment and professional development. In this talk I’ll draw on experiences in several countries and very different environments to explore these issues; and see if we can identify the secret sauce to making R bloom in the potentially difficult soil of a public sector bureaucracy.
Peter Ellis is the Director of the Statistics for Development Division at the Pacific Community (SPC), where his team of around 30 statisticians, data scientists and analysts help Pacific Island country and territory statistical offices collect, process, analyse and disseminate data for official statistics. He is an Accredited Statistician with the Statistical Society of Australia. He was previously the Chief Data Scientist at Australian-headquartered management consultancy Nous Group where he led a transformation of its approach to analytics based on R, SQL and Git. Prior roles included the Principal Data Scientist at Stats NZ, General Manager Evidence and Insights at the Social Investment Agency, Manager Sector Performance at Ministry of Business, Innovation and Employment, and Director Program Evaluation for the Australian aid program.
His blog is at https://freerangestats.info/.