Ensuring trustworthiness, quality and value in all our work

Ensuring trustworthiness, quality and value in all our work
by Allan LEONARD for FactCheckNI
12 April 2018

The Northern Ireland Statistics and Research Agency (NISRA) hosted an information seminar at Belfast City Hall, for staff from government organisations in Northern Ireland who have responsibility for statistics in their work (creating and/or applying).

The audience was welcomed by Siobhan Carey (Chief Executive and Registrar General, NISRA), who explained the rationale for the UK Statistics Authority revising the Code of Practice for Statistics:

“We live in a new world with supercomputers in our pockets. There is fake news and contested statistics,” Carey said.

She was recognising how fast information — and misinformation — can spread via the Internet, which most of us access constantly with our smartphones. Carey wants to make statistics available for public debate, and this value will be explicitly included in the revised code.

Ed Humpherson (Director General for Regulation) began with a humourous anecdote of him committing a statistical fallacy of thinking Belfast was a sunny place, because it was always sunny when he visited. Alas today’s overcast skies was a regression to the mean.

Humpherson outlined the three pillars of the revised code: trustworthiness, quality, and value.

  1. Trustworthiness is about having confidence in the people and organisations the produce statistics and data
  2. Quality is about using data and methods that produce assured statistics
  3. Value is about producing statistics that support society’s needs for information

Penny Babb (Office for Statistics Regulation) discussed the application of the code, and she had us consider a case study of migration statistics, which caused controversy covered in the local press, including the Belfast Telegraph.

This prompted my colleague, Orna Young, and I to double-check our fact-checked claim article, “What is EU to Northern Ireland net migration?”, published during the EU referendum campaign in 2016 — had we misinterpreted different statistics products?

Thankfully we were redeemed in our analysis. But we noticed that we used an ONS bulletin that included several statistics products — its Migration Statistics Quarterly Report is derived from : long-term international migration (LTIM), National Insurance number allocations to adult overseas nationals from the Department for Work and Pensions, and visa data from the Home Office.

And that is the point. The dimension of value is measured not only by a citizen’s entitlement to public information (collected data and statistics), but its practical utility to the public debate.

Here I responded to the audience by explaining that at FactCheckNI we accuse no one of lying, because we don’t know the intent of the person making a claim. Often claims are made by politicians who repeat what they heard or what they were told by a press officer or policy advisor; any three of these actors could simply be simply mistaken. Yet by having a statistical product that can clarify the various sources of official data from various government departments, there should be less cause to repeat misleading information, or at least easier to correct it .

NISRA is assisting with their NI: IN PROFILE product, which is intended to provide a high-level statistical summary of life in Northern Ireland, and be useful for policy makers in the public and private sector, students and academics, schools and the general public.

My presentation was an introduction to the work of FactCheckNI, going through the classic journalist’s 5Ws and How:

Of particular relevance was my explanation of the code of principles established by the International Fact-Checking Network (IFCN), one which FactCheckNI was an early signatory.

I argued how the IFCN code coincides with the three pillars of the UK Statistics Authority’s revised code.

Trustworthiness is reflected in our commitment to nonpartisanship and fairness; to transparency of sources of information; to transparency of funding and organisational details.

Quality is reflected in our commitment to transparency of methodology; and to open and honest corrections.

Value is reflected in FactCheckNI producing fact-checked articles so that individuals, citizens, and voters can have the most accurate information possible for their analysis and decision making.

We share the code’s belief that nonpartisan and transparent fact-checking can be a powerful instrument of accountability journalism, and the converse can increase distrust and pollute the public debate.

At FactCheckNI, we would also add that fact-checking can be part of a peacebuilding process. At the time of our launch, I was interviewed on the Good Morning Ulster radio programme and asked how would we know if we were being effective. Instead of reciting metrics agreed with our funder, I replied that the next time there’s a viral meme of a hoax riot about to take place, it won’t happen, because someone we’ve trained will have done a simple reverse image search and reply with a corrective tweet or posting.

And that is the two parts of the solution — online tools that are easy to use, but backed with practical training and to establish the importance and value of critical thinking.

A final perspective is that we should consider ourselves fortunate that we operate in an environment where the professionalism and independence of our work is respected and defended. The fact that we can have a code of practice that will elevate our standards. We meet other fact-checkers from other parts of the world where basic journalism is physically threatened and where government agencies are thoroughly corrupted.

It is crucial to ensure trustworthiness, quality, and value in all our work; we want the public to demand this from us. It is the best way forward for our democracy.