Let me begin by giving a warm welcome to Conchita Kleijweg, who takes over from Ada as Director of the ISI on 1 July. We wish Conchita every success and will do all we can to effect a smooth transition.
It has been good to get back out again to an international conference – this time the IAOS conference in Kraków, Poland. Having given several talks to various conferences virtually over the last six months, it was strange but pleasing to get on a plane again knowing that I would be meeting up with some old friends and familiar faces.
This IAOS conference had special meaning for me, as I was President of IAOS ten years ago during which time we hosted the 2012 IAOS conference in Kyiv, Ukraine. And I was pleased to meet again with Oleksandr Osaulenko, who was head of the Ukraine Statistical Service at the time.
Kraków is only a few hundred kilometres from the Ukraine border so there was a lot of discussion, both informally and in the programme about the situation there.
Sometimes when we hold international conferences we take the opportunity to hold a special workshop on a topic of common interest. This was the case this time when some of us met to discuss the UN Fundamental Principles of Official Statistics, and the misuse of statistics. The Fundamental Principles have been around since the early 1990s, but were adopted by all governments through General Assembly resolution in 2014. They set out good practice for official statistics offices and help guarantee their independence on statistical issues. Clearly there are issues of compliance with these Principles in some countries, and the ISI is active in working with the UN on this. The more interesting discussion though was on issues around the use of non-official statistics. Statistics from other providers, some reputable, others of unknown provenance, also populate the public space. How are discerning members of the public meant to know which statistics to trust?
The answer probably lies in three directions. Firstly, a need for official statisticians and others to promote good practice – which includes published and verifiable methods, and quality standards – among wider publishers of statistics which inhabit the public policy space. Secondly, a need to improve the statistical literacy of citizen and public decision makers, so that they are better able to decide which statistics best match their needs. And thirdly, a need for fact checking organisations to continue and expand the excellent work they do, to call out rogue statistics and their producers. Having poor quality statistics being used in public debate leads to poor decision making and reduces the credibility of all statisticians.
All these approaches, and others, we should be promoting through the ISI; and the upcoming conferences in Zambia (IAOS April 2023) and in Ottawa (WSC July 2023) give us the opportunity to develop our thinking and promote solutions. I look forward to seeing session proposals!
Steve
Championing official statistics – methods, standards, and as a career destination for trained statisticians – might have lost its morale-boosting polish as national offices compete for relevance with alternative data sources – non-official statistics – in the market for policy relevant evidence. The answer now may not be to teach policymakers, and the public, to be good consumers via the three-step mantra above, valuable though this is. Some answers may lie in the way the production of official statistics is distributed across jurisdictions. Once a national office was the flagship institution for the design, operation, processing, storage, and dissemination of official statistics, with resources reflecting national pride and joint yoking of statistics with progress. Now policy frontiers do not fit necessarily into constitutional jurisdictions. Evidence-driving policy builds authority not from national (or international) commission but from the globalised quality of policy challenges. This shift may have been in play for some time, although papered over by a landscape of agreements and conventions, that govern the flows, sources, and sinks of human activity across conventional boundaries. It might help to reconfigure official statistics in this global context, no less buckled by the POS, but without the comfortable supports of stand alone NSOs. Global statistics now feed policy on many topics once contained within national charters, but without a natural home. So this is the challenge for IAOS members, to step outside their roles in government offices, and consider what architecture is needed to build statistics demanded in this changed policy context. It should not be left to physical or social scientists alone to project a programme for a more sustainable existence for all of us.