Spreadsheets, spreadsheets, spreadsheets: With the public sector still powered by office stalwart Excel, SPC’s Director of Statistics for Development Peter Ellis has advocated for greater use of ‘R’—open-source statistical software with Pacific roots and potential.
The Pacific Community (SPC)’s Director of Statistics for Development, Mr Peter Ellis, has given a talk ‘Making R work in government’ at New Zealand’s University of Auckland in a thought-leadership series of public lectures on the programming language.
R is free, open-source software for statistical computing, data analysis and visualisation. Mr Ellis’ talk was part of the university’s public Ihaka Lecture Series, named for Associate Professor Ross Ihaka who, along with Dr Robert Gentleman, co-created R in New Zealand in the 1990s.
Now the language of choice for many of the world’s statisticians, data analysts and researchers, R allows users to perform complex analyses and manipulate data efficiently—with no licensing costs, a vast ecosystem of add-ons and an engaged community of contributors.
These factors make it especially useful in the public sector—including in a Pacific statistics context—argues Mr Ellis. Managed well, R can be a critical component of a transformation of the effectiveness and efficiency of an analytical team in government, he says.
His lecture addressed three key themes. First, that analytics in government is “often a mess”. Second, that “R is a great tool but doesn’t solve everything by itself.” Third, that “change is hard but possible.”
Mr Ellis’ talk touched on the common scenario of a “lone genius who understands the spreadsheet” in an organisation, with data fragmented and siloed across the org chart, processes and files. Loss of institutional knowledge through staff turnover is a compounding factor, he says.
He also talks to the role of R within an ecosystem of other office programs and open-source software. He covered practical adoption tips to greater use of R, advocating new workflows that incorporate “teamwork, transparency, reproducible analytical pipelines, peer review, and home-grown R packages and rules for use.”
On change management in organisations of all sizes, he advocates for “lighting a fire, rather than filling a bucket” when it comes to staff training.
“Doing this successfully is difficult,” he says. “It 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.”
R in Pacific statistics
SPC’s Statistics for Development Division plays a leadership role in developing statistical skills and systems in the Pacific—providing both ‘hands-on’ assistance to national statistics offices, as well as capacity development.
One of the case studies in his lecture, Vanuatu Bureau of Statistics uses R in the production of their merchandise trade statistics. An ‘innovative experiment’ on data science driving innovation in climate change and natural disasters will also use R in its toolchain, funded by the World Bank as part of the PACSTAT project—fostering statistical innovation and capacity-building in the Pacific Islands.
R is one of the principal tools used for analyses in SPC’s Fisheries, Aquaculture and Marine Ecosystems Division, including the use of dashboards and automated reporting for fisheries management. Advanced skills in R are a key benefit for participants in the Pacific Islands Fisheries Professionals one-year placement programme. Short workshops support fisheries officers in SPC’s members to transition from tools like Excel towards R for data manipulation, analyses, creating tables and figures, and reporting.
SPC is sponsoring Pacific representation and participation in the OceaniaR Hackathon in Melbourne this week, a chance to collaborate in-person on R-focused projects for social good with some of the leading developers in the R community.
R can also be used to manipulate datasets available from the SPC-managed Pacific Data Hub, such as exploring indicators held in the indicator database, or analysing national census and household income and expenditure survey microdata—supporting evidence-informed policymaking by Pacific institutions and leaders.
Replay the lecture
Professional biography
Mr Ellis is an Accredited Statistician with the Statistical Society of Australia. Before SPC, he was 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 New Zealand’s Social Investment Agency, Manager Sector Performance at New Zealand’s Ministry of Business, Innovation and Employment, and Director Program Evaluation for the Australian aid programme.
Contact: Ben Campion, Communications Adviser, Statistics for Development, Pacific Community (SPC) | [email protected]