None

Why should we believe you? The problem with unpublished evidence

9th Mar 2017 | Claire Milne

One of Full Fact’s aims is to improve the quality and accountability of public debate for everyone. We believe no one should make a factual claim without being able to point to their evidence.

Things get difficult when we don’t have the information to factcheck a claim

In November last year we wrote about some figures published by Age UK in a press release. It said that 1.2 million older people in England were not getting the social care they need to carry out everyday tasks. Of these people around 700,000 got no help at all and around 500,000 didn’t get enough help (although they did get some).

That all seems quite specific and simple to factcheck, doesn’t it?

As it turns out, it wasn’t.

Although the first press release was published in November, Age UK didn’t publish the report to back up its findings until February.

We’ve asked Age UK why there was such a gap between the two publications.

We think it’s important to be as clear as possible when explaining any analysis and now that we’ve been able to look at Age UK’s report it could be clearer in explaining how the 1.2 million figure was reached.

No one should expect others to blindly trust what they say. Whether you’re a politician, a journalist, or a charity, claims made in public debate should be backed up with evidence.

The problem of unpublished sources isn’t just about Age UK though

It’s not the first time we’ve encountered the issue of not having enough information to factcheck a claim. We’ve encountered a Prime Minister using figures on immigration that hadn’t yet been published and the Department of Health quoting numbers of A&E doctors that weren’t publicly available. Then there was the claim about the number of weekend deaths in hospitals, where the evidence was not published at the time. It’s something that crops up too often.

All of these instances contradicted a key principle of the Code of Practice for Official Statistics, which says that official statistics should be made “equally available to all, subject to statutory provisions for pre-release access. Some weren’t actually official statistics, but the principle stands.

We’ve written about the results of a poll published recently by campaign group Leave.EU which had looked at how people were expected to vote in the recent Stoke Central by-election. But there was no information about how the poll was carried out, how representative it was and what questions people were asked, which contradicts the rules set out by the British Polling Council to its members. We’ve asked for more information but haven’t heard back.

The BBC also used an unpublished forecast to claim that they would make £150 million less from the licence fee in 2016/17 than planned because of the iPlayer ‘loophole’. It has since responded to our freedom of information request, five months late. We can update our factcheck but the news has moved on long since.

In all of these circumstances figures were published without the data behind them to back them up. This means people who use them, hear them, or read them don’t have the information available to properly assess the claim being made, or know if it’s being made correctly.

No one should be faced with the choice between blind faith and blind cynicism. Help us keep pushing for publication of evidence to ensure we can place informed trust by making a donation.


We aim for our factchecks to be as accurate and up-to-date as possible. If you think we've made an error or missed some relevant information, please email team@fullfact.org.