Ian Simpson of NatCen explains some of the findings published in a recent report looking at the public's views of UK official statistics.
As part of its annual British Social Attitudes (BSA) survey, NatCen Social Research measured views on UK official statistics and the organisation that produces them, the Office for National Statistics (ONS). The questions were also asked in the 2014 BSA survey, allowing for comparison of results.
With the EU referendum campaign and its aftermath dominating the news agenda throughout 2016, we wondered whether trust in official statistics may have taken a hit (all BSA interviews took place after the referendum). Statistics were used by both sides to make their case, which could have caused people to lose faith in statistics generally. After all, with data being used to justify diametrically opposed views, how much faith should one have in them?
In addition, the perception of polling errors at the 2015 general election and EU referendum could also have knocked people’s faith in statistics. Meanwhile, Michael Gove’s comment that 'people in this country have had enough of experts' appeared to strike a chord, at least in some quarters. Would this backdrop translate into reduced trust of statistical experts and the statistics they produce?
Given that the EU referendum received such prominence in 2016, we asked the British public whether the EU referendum campaign had affected their view of official statistics. While one third (33%) said the referendum campaign made them trust official statistics less, over half (54%) reported that the referendum had no effect on their level of trust in official statistics. Of course, it’s difficult to discern the impact of an event on our own attitudes; much of the interaction between experiences and attitudes is likely to be subconscious.
So perhaps a more enlightening way of measuring whether there has in fact been any change in trust in official statistics over the last couple of years, is to compare the answers to the questions that were asked in both 2014 and 2016. The ability to compare fluctuations in attitudes over time is one of the key strengths of BSA. So, did we find any drop in trust in official statistics in 2016? The answer is no. Indeed, we found some evidence of small increases in positive attitudes towards official statistics.
Despite Gove’s comments about experts, there continues to be a widespread belief that official statistics are important for understanding Britain (92% of those giving an opinion agreed) and that statistics produced by ONS are free from political interference (70% of those giving an opinion). Both these figures have remained stable over the past couple of years.
When it comes to whether people believe that official figures are generally accurate, there has actually been a modest, but statistically significant, increase from 73% in 2014, to 78% in 2016.
And trust in the ONS has also remained high. Of those able to give an opinion, 90% trust the ONS and 85% trust the statistics produced by ONS, which also represent no statistically significant change on 2014.
So it’s good news for statisticians and their production of statistics, but bad news for the government and newspapers. Only about a quarter (26%) think the government presents official figures honestly and even fewer (18%) think newspapers do so. Again, these figures are very stable compared to 2014.
Although the fallout from the EU referendum will continue to have major ramifications in many areas, there is no evidence that it has dented trust in the production of official statistics, or indeed altered the public’s already low opinions of how the government and newspapers present this official data.
Read the report Public Confidence in Official Statistics.