Cookies on this website
We use cookies to ensure that we give you the best experience on our website. If you click 'Continue' we'll assume that you are happy to receive all cookies and you won't see this message again. Click 'Find out more' for information on how to change your cookie settings.

Sir Andrew Dilnot of the UK Statistics Authority drew gasps and laughs as he exposed internal and media biases with public policy statistics at the Oxford local chapter of the Royal Statistical Society, emphasising the crucial role of statisticians and social scientists in public policy.

We need to think as hard about the words as we do about the numbers - Andrew Dilnot

Speaking to a full house of students, researchers, and retired members on a warm May afternoon, Sir Andrew Dilnot reminded the Oxford local branch of the Royal Statistical Society that public policy needs a solid base of numbers and statistics. It is the job of statisticians and social scientists to make this information available.

Sir Andrew, Warden of Nuffield College at the University of Oxford, is Chair of the UK Statistics Authority, which exists to promote and safeguard the production and publication of official statistics that serve the public good. Sir Andrew emphasised, as he so often has done on the BBC radio show More or Less, how statistics can be used and abused, to either illuminate or mystify. Statistics can only form a solid base for public policy if they measure something sensible, are properly contextualised, and are explained in a clear, impartial manner.

Know your measurement

“It’s very important to know things. But it’s more important to know what it is that you know. You can’t just assume that you are measuring what you want to be measuring,” said Sir Andrew. “Think about what it is you are measuring. That helps you to think about what you should be measuring.”

Know your context

Statistical analysis is at its very best when it is forcing us to think clearly - Andrew Dilnot

Even when you know exactly what it is that you are measuring, imprecise statistics can lead to misunderstandings. The British media was recently in a furore at the announcement that the number of people aged 85 years and older had risen by 25 times since the early 1900s. Reading the headlines, you would have been forgiven for expecting "British society to have been overrun by gangs of marauding grannies and completely collapsed".

The trouble here is a lack of context: although there are 25 times more 85 year olds, the age group still makes up less than 3% of the total population. If a dataset shows a worrying change, you need more data to contextualise that change. Contextualising information gives the full picture.

Know your story

Statistics are just a way of telling a story. It’s all in the narrative - Andrew Dilnot

No matter how accurate and well-contextualised statistics are, how they are presented affects how they are perceived. In his work with the UK Statistics Authority, Sir Andrew has come across widespread misinformation about our society, generated by how our world is represented in the media around us. The roomful of Oxford statisticians were shocked to find that their impressions of income distribution and equality in Britain, gleaned from the media and other sources, were much more pessimistic than objective measures really show.

“The media is not holding up a mirror to our society; or if it is, it’s a very cracked mirror!”

Know things!

Well thought out and collected statistics, placed in context, and reported fully and impartially: this is the solid base of information on which public policy should rest. And it is social scientists and statisticians – the members of the Royal Statistical Society – who Sir Andrew charged with the responsibility to go out and know things, to help decision-makers to know things, and to ensure a data-driven public policy.

The Oxford local branch of the Royal Statistical Society meets around six times a year. All those with an interest in statistics, whether academic or non-academic, are welcome to attend these meetings. Read more about the group and upcoming events

Similar stories

Patients set to benefit from new guidelines on Artificial Intelligence health solutions

Patients could benefit from a faster and more effective introduction of artificial intelligence (AI) innovations to diagnose and treat disease – thanks to the first international standards for reporting of clinical trials for AI.

Six-month outcomes after treatment for COVID-19 on intensive care units in England

Researchers at the University of Oxford are investigating the long-term health outcomes for patients who have been treated for severe COVID-19 disease in intensive care.

Patients with rheumatic pain increasingly turning to opioids

Reporting at EULAR, researchers found that opioid use by patients with rheumatic and musculoskeletal pain was on the rise in Europe.

Could do better: clinical trial reporting fails to live up to the mark

The statistical analysis and reporting of treatment effects in reports of randomised trials with a binary primary endpoint requires substantial improvement, suggests NDORMS research published in BMC Medicine.

COVID-19 prognosis and prediction models for medical decision-making are flawed, say researchers

The modelling and approach to tackle the hard medical decisions associated with the spread of the COVID-19 virus may be based on weak and overly-optimistic evidence from studies that are biased and unreliable, suggests research published by The BMJ today.

Sponsored place available for the Randomised Controlled Trials Course

The inaugural Doug Altman scholarship will be offered to an LMIC (lower-to-middle-income country) applicant or a student.