“I was attracted to an organisation that cares about data and evidence”: An interview with RSS Executive Director Hetan Shah

Hetan Shah has held the position of Executive Director of the Royal Statistical Society since October 2011. He is responsible for the overall strategy and management of the Society. With an educational background in economics and politics, his previous role was Chief Executive of the national charity Think Global. His interests include evidence based policymaking, Big Data, international development and sustainability.

 

Hetan is also a visiting senior research fellow at the Kings Policy Institute, Kings College London and a trustee of the Friends Provident Foundation and a non executive director of the Higher Education Statistics Agency. He was appointed as an Honorary Vice-President of the Geographical Association in 2011.

Statistics Views talks to Mr Shah about the General Election and the Society’s Data Manifesto,  the getstats campaign,  benefits of RSS membership, Big Data, social media and how the Society is meeting the needs of the statistical community.

1. There has been much hype about Big Data with a lot of attention in the media and talks held including the RSS’s own recent Big Data Debate. Do you think such attention has helped or hindered the profile of statistics as a discipline?

I think it has helped. It is partly for statisticians to step up to the plate and own that agenda, which we have done at the RSS on behalf of the profession. I think, for me, the term ‘Big Data’ is best understood as a mood term. It is mainly about the ubiquity of data and that is how it translates into the public discourse. Obviously there are more technical definitions but in terms of the public debate, I think it is about the ubiquity of data and statisticians should be at the centre of the debate.

In a sense, Big Data provides an interesting challenge back to the statistical community – moving towards working with large, unstructured datasets is something that perhaps challenges the traditional statistical tools, so there is something for statisticians to think about in terms of their computing skills, etc. In turn, statistics has a lot to challenge back to the Big Data discourse which has been traditionally led by the computer scientists and where the focus has perhaps been on the size of the dataset , not so much on the analytics, and that is somewhere where statisticians have a great deal to add.

2. Many of the news items from StatsLife cover updates from former President, now National Statistician John Pullinger and Andrew Dilnot on letters that are written to the government when a statistic has more often than not been misrepresented. We know that governments and learned societies need to do more to raise awareness of statistics and stop people being afraid of numbers. What do you think the RSS can do to help raise public awareness of statistics and prevent the misuse and abuse of data?

There is a lot going on in this space now. As you say, the UK Statistics Authority is active in issuing reprimands when necessary, particularly to politicians. We’ve also seen a rise in fact-checking organisations – Full FactChannel 4’s Fact Check – and they are checking on a daily basis what has been said in the media.

In terms of what the RSS can do, we have decided that fact checking is not the right space for us given that there are already organisations working on this. What we are trying to do is build good practice; in particular our award for the good use of statistics in journalism has been a very positive way of building pride amongst the journalists’ profession in terms of doing a good job as opposed to just rapping people on the knuckles for doing a bad job. Last year we had a hundred applications so it has become quite a prestigious award.

3. Current RSS President, Peter Diggle recently said in a StatsLife interview “I see myself as a statistical scientist rather than a statistical mathematician. Almost everything I do, I couldn’t do on my own, and I like to think that the people I do it with couldn’t do it as well if they didn’t have me. The statistics are integral to the science rather than a separate discipline.” Sir David Cox also made a similar comment when I interviewed him last year – ““I would like to think of myself as a scientist, who happens largely to specialise in the use of statistics”. Do you agree that statistics is less of a discipline in its own right, and more of an integral part of science as a whole? What does this mean for the Society and the work of the Society?

I think it is both. I think there is clearly important work being carried out that is feeding into statistical methodology and there is an enormous amount of application work which is much more in the variety of what Professor Diggle was talking about in terms of statistics as part of science.

Our Series B journal focuses on the deep methodological side, whereas the other journals consider applications, so we do cater towards both in our Society.

One of the interesting issues it raises is how the research councils are structured – at the moment, statistics is really seen as being held by EPSRC, who in turn understand the methodology very well but not so much, the applications. The other research councils do find a lot of statistics – ESRC or MRC – and they understand the applications but there sometimes can be a disjoint in terms of the lead research council perhaps not always understanding the research application side of statistics. There is something to think through about the structure of research councils and how they respond to the different kinds of statistics that exist.

The Independent wrote a news story on Perils of Perception and it was the fourth most tweeted news story that they had that year.

4. The RSS conference is due to take place this September in Exeter and has attracted some big name speakers in recent years. What will the focus be on this year? Who should we look out for?

We have got some great speakers this year – Scott Zeger, Peter Hall, Julia Slingo who is the chief scientist at the UK’s Met Office and Alberto Nardelli – the data editor of the Guardian – who is giving the Significance lecture. The thing to remember is that the RSS Conference is the place where you can really keep up with developments across the statistical field, so not just keeping in touch with your own part in statistics but to see what else is happening in the wider field.

This year we are trying to stream the conference in a more clear way – e.g. data science, sports statistics, official statistics, biostatistics, social statistics, etc. – and we hope people can then follow a stream of specialism if they wish to, but at the same time, step out and attend different talks too.

5. You have a strong educational background in economics and politics with degrees from Oxford and Birkbeck College London. You were previously Chief Executive at Think Global. What attracted you to the role of Executive Director and what would you say have been the highlights of your time in the role so far?

I was attracted to an organisation that cares about data and evidence. My particular interest is in public policy – how do you make evidence-based decisions in terms of policy. If you go back to the founding statements of the RSS, it was all about using data to cast light on solving problems in society, which is very important and that was also an attraction.

I feel privileged to be working with a strong staff team and amazing volunteers here. To name two highlights, the research that we did for ‘Perils of Perception’ really cut through to the public and created a debate about statistics, perceptions and misperceptions that perhaps the RSS had not been able to achieve before; the other is probably the Data Manifesto that we created. It was good to be able to pull everything together that the RSS believes into one place and give voice to that we can now quickly and clearly assert our position on different issues.