1. To change the outcomes, change the system

Administrative data is about real people and their lives. So in the first episode of Connecting Society, we’re diving into the big picture behind the vision of ADR UK (Administrative Data Research UK): why does using administrative data for public good research matter? How can it make a meaningful difference to society? And is your data being stored and used securely?

Joining us to unpack these questions are Dr Emma Gordon, ADR UK’s Director, and Ally McAlpine, the Scottish Government’s Chief Statistician and Co-Director of ADR Scotland.

This episode explores the state of the data landscape before ADR UK’s work began, the goals we’re moving towards, and the benefits this work brings to the public. Our guests will also share their personal career journeys, what motivates them, and why they believe this work is so important.



Wondering what administrative data is? Visit https://www.adruk.org/our-mission/administrative-data/.

If we used any terms you're not familiar with, check out ADR UK's glossary at https://www.adruk.org/learning-hub/glossary/.

Read an article from Emma which provides more information on the background of ADR UK: https://www.globalgovernmentforum.com/to-change-the-outcome-of-a-process-you-have-to-be-prepared-to-change-the-system-five-minutes-with-public-service-data-live-speaker-administrative-data-research-uk-director-emma-go/. 

Connecting Society is brought to you by ADR UK (Administrative Data Research UK). Find out more about ADR UK on our website, or follow us on X (formerly Twitter) and LinkedIn. This podcast builds on a pilot series known as DataPod, produced by ADR Scotland.

Transcript

Shayda: Hello, and welcome to Connecting Society, a podcast about how everyday data can shape our lives. I'm Shayda Kashef, Senior Public Engagement Manager for Administrative Data Research UK, or ADR UK to me and you.
 
Mark: And I'm Mark Green, Professor of Health Geography at the University of Liverpool. We are your co-hosts and guides around the wonders of administrative data.

Shayda: In this podcast, we are exploring all the different ways in which the information that is collected about our everyday lives—from interactions with health services, voting behaviours, police and crime reporting, educational achievements, and more—is used by researchers and policymakers to make better decisions, support society, and make the world a better place.

Mark: Well, this is very exciting. Our first episode. I'm a bit nervous. How are you feeling?

Shayda: I'm an avid podcast listener. It's interesting to be on this end of the mic, though. How have you been preparing for today?

Mark: Well, I did a sit-up because I thought that might be useful, but you never know.

So Connecting Society. That's our snazzy title. I was a bit disappointed that some of my names were rejected, though: Data Data—you know, like day to day but with "data" shoehorned in—Behind the Data, Seize the Data, and my personal favourite, Admin-mazing Data.

Shayda: Admin-mazing Data. Wow. I'd expect nothing less from someone whose day job is to actually analyse data. Mark, you're our resident data scientist. Can you give us a quick definition of what administrative data even is?

Mark: Gosh! You're starting me with the hard ones there, aren't you?

For me, administrative data is the information that is generated every day when people interact with public services—schools, courts, welfare, or even the NHS. All these interactions are recorded by the government.

While these public bodies keep such information to help the way their service runs, this information can also help researchers find valuable insights about society and identify where change might be needed.

For example, the NHS records all diagnoses of diabetes so that it can know who needs help and plan for investment in treatments and medicines. By knowing how many people have diabetes, we can start to look at the patterns of which demographic characteristics or social groups have more cases of diabetes than others. This helps us identify what might be causing diabetes to occur and better target our responses to prevent it.

Shayda, this is your first role in the data space. Did you know anything about administrative data before joining ADR UK?

Shayda: No, to be honest, I never imagined that the words administrative data would involve anything exciting.

But the research is so interesting, as our listeners will find out as they continue with the series. I joined ADR UK five years ago and never looked back.

How about you, Mark? When did you first come across administrative data, and how was it different from the other types of data you worked with previously?

Mark: Well, I started my PhD using mortality records, so, you know, I'm an OG admin data person.

I think for me, what excites me the most is the size and detail of what you can get from administrative data.

For researchers, if we set up an experiment or collect a survey, we only ever get a snapshot into the lives of people. Administrative data allows us to see how society behaves at a large scale. It gives us more granular insights to help us make much better decisions.

Shayda: We could talk about this all day, but let’s get on to today’s podcast, where we’ll introduce the big picture behind ADR UK’s mission. Why should people care about the use of admin data? How can it make a real-world difference to society? 

We’ll talk about how the system worked before, what ADR UK is doing differently, and what this means for you, the listener.

Mark: And to help us do that, we have two admin-mazing guests with us today—Emma Gordon and Ally McAlpine. Welcome to the show!

Emma: Hi, great to be here.

Ally: Thank you.

Shayda: Can you please tell us a little bit about you, where you're from and what you do? But very importantly, we have one rule for this podcast, which is no technical jargon. Emma we'll start with you.

Emma: Thanks very much, Shayda. And yeah, I'm the Director of the Admin Data Research UK programme, like you said. And my role is basically a big coordination role. And so it's talking to data owners within government about how we could work with them. It's talking to researchers about the outputs from their research, and it's generally making things happen and really championing what we're doing. And so I speak at a lot of conferences and public events and I'm really, really passionate about what we're doing.

Shayda: Amazing. Ally?

Ally: So I'm Ally McAlpine, I work at the Scottish Government and I'm Chief Statistician. Now if that job's not big enough, I'm also Co-Director of ADR Scotland. So that's part of the ADR UK family. And I guess the best way to describe my job when I'm working with administrative data is trying to link that research to the policy needs that we've got coming up in the Scottish Government.

Shayda: Incredible.

Mark: OK, we're going to get the most important questions straight out of the way first. So we're going to ask you both to respond to a series of questions. Starting off, is it day-ta or dah-ta?

Emma: Most definitely, day-ta.

Shayda: What about the choice between a pie chart and a fruit pie?

Emma: Every day, during my time when I was working at ONS it drilled into me how bad pie charts were at representing data. 

Ally: I totally agree. I would get rid of pie charts altogether. 

Shayda: Let's make a motion to swap all pie charts for fruit pies.

Mark: What's your favourite statistic, Emma? Do you wanna go first?

Emma: I was really struck by a statistic from a recent report published by one of our researchers that was 33% of care-experienced children received a youth justice, caution or conviction compared to 4% of those without care experience.

And I didn't know that before. I bet there's a lot of people that didn't know that before. And even if we had an inkling about that that was the case, we wouldn't have known how great that risk was for these children. And we do know that now. So now governments and policymakers can think about how to change that situation and how important it is for those individuals as well as society for improving the lives of those people.

Mark: That's a good one. I saw that recently and I was quite intrigued by the whole thing. It's fascinating. Ally, do you have a favourite statistic?

Ally: So here's my favourite statistic, 58% of new manure is kept in heaps and is not covered. And you think, well, that's quite, why am I interested in manure and 58%? But the link there is that manure left uncovered is actually quite a big greenhouse gas emitter and contributes quite a lot to the UK's emissions for methane. And while it's losing all of that gas, it's actually losing its nutritional value as well.

So there's some statistics that first of all help mitigate climate change, but there's a cost attached to doing that as well. That's my favourite at the moment.

Mark: Excellent. I never thought we'd get into manure in this podcast, but you know, here we go.

Shayda: Before we jump into learning more about both of you, I'd like to start with the top line. So if listeners could take one thing away from this podcast, what would it be? Emma, you first.

Emma: For me, I've been passionate about data for a very long time, but it's only recently that it's really dawned on me the value of what we're doing to society. And that's because usually social and economic researchers are basing their research on survey data. And obviously, survey data is absolutely brilliant, but it's only ever as good as the people who answer surveys. And we've known for a very long time that there is really important sections of society that generally don't answer surveys and therefore aren't represented in the data.

And quite often these are the people who can have quite chaotic lives for various reasons through no fault of their own. But it's generally the most vulnerable people in society. And that's really, really important in terms of research and policymaking because we need to understand who those people are, not least so that we can improve the lives of those individuals, but also so that we can reduce the cost of services and the cost of services that need to speak to each other and for these people. 

And we generally think of people in terms of how they access individual services. So you're either a pupil or a pensioner, or a patient or a prisoner, but what if you're one or more more than one of these things? And what if you're all of these things? And suddenly your life becomes quite challenging in terms of how all these services are interacting or not interacting and not speaking to each other. And that has a really high cost burden to society in terms of those services not working efficiently because they're not dealing with the individual at the centre of that maelstrom of decisions that need to be taken about that person.

And so it's really dawned on me quite recently that linked administrative data gives you a view about who these vulnerable people are in society much, much better than any other form of data. And therefore the people that have been really missed from discussions in policymaking and research findings as suddenly we can drill down, we can find those things. So it's inexcusable now not to consider these people within policymaking.

Shayda: Yeah, I really like what you have to say about how administrative data can capture multiple identities and experiences of people. I think this is also what interested me in it too. Is that because when you link together the data sets from different departments, so for instance, education with health or education with crime or education with health and crime, you are getting a sort of more like well-rounded view of the sort of experience rather than and kind of a sort of population experience rather than a small number of people who have been designed to represent the experiences of a larger number of people.

Emma: Yeah, exactly. 

Shayda: Ally, if there's one thing our listeners could take from this podcast, what would it be?

Ally: I think what Emma just said makes a lot of sense. And I think for me as Chief Statistician, what I'm trying to do is make sure that evidence is embedded within anything that we're trying to deliver within Scottish Government or anywhere else in the public sector as well. And effectively, the work that we're doing is spending taxpayer money. So we need to be as efficient as we possibly can be. And I think that's where data helps.

And certainly, you know, you could look at statistics across the whole of Scotland, notice a problem, put a policy in place. But I think what Emma is talking about there, you know, around about linking the data, that gives us even more evidence. And it makes sure that when we're when we're trying to put out a policy that's actually reaches the parts of society that will benefit most from it. The flip side to that, further down the line, is that we can then evaluate whether it worked or if it benefited the parts of society we hoped it would.

Shayda: You're right—we are financially investing in these services through our taxes. By using data that better represents the population, we can make decisions based on evidence that includes people who engage with society. I suppose that makes sense to me.

Now that we've gotten the serious questions out of the way, let's get to know you both a bit more. Emma, let’s start with you. You used to work at the Bacardi Bat Sanctuary, which sounds very exciting. How did you go from there to being the Director of ADR UK?

Emma: I ended up going to university, picking my favourite subject at school, and then studying that as a degree. My favourite subject was biology—I loved animals—so I did zoology at university. I really wanted to do a PhD, and to do a PhD in zoology, you need fieldwork and experimental experience. I volunteered to get involved in other people's research studies. That’s how I ended up helping with a research study based at the Bacardi Bat Institute in Gainesville, Florida, which worked on megabats—absolutely giant fruit bats. I also got involved in research on bats in Costa Rica, which was absolutely fascinating.

From there, I did a PhD in bird flight. During my PhD, I had my daughter, and I was eight months pregnant with my son when I had my viva. It was really interesting going back to work after completing my PhD. I felt strongly that if I was going to leave my kids with someone else, it had to be to do something really meaningful. Although my PhD research was intellectually interesting and energising, I didn’t feel it had a broader purpose for society.

Long story short, after many random jobs and roles at universities, I did a part-time master’s in public health, nutrition, and physical activity. From there, I transitioned from being a zoologist trained to study animal systems to being an epidemiologist trained to study human population systems. I worked on a fantastic postdoc involving a longitudinal study based in Bristol. This study examined how children grow up and the impact of factors like family environment, diet, and health on their lives. 

From there, I went to work at the Office for National Statistics, where I eventually became Head of Health Analysis. My favourite part of the job was interacting with researchers who wanted to use data for research. I came to understand all the barriers they faced as gatekeepers of that data. Now, I’m in the amazing position of being part of the solution—supporting the government in finding ways to engage with academics, granting them secure access to data, and enabling researchers to contribute to policymaking decisions and discussions.

Shayda: Incredible. Ally, Were you once a chef in a previous life?

Ally: Oh, someone has told you stories! Yes, I was a chef, though not a very good one. I describe myself as a “Poundland Gordon Ramsay” and probably wasn’t even that. I worked in bars and restaurants. Before that, I went to university to study chemistry—it was the only subject I was good at in school, or at least the only one I thought I was good at. But I ended up dropping out in my third year. The only course I passed that year was on quantum chemistry, which was essentially statistics.
 
So after my stint in cheffing, I decided I needed to go back to university. I studied maths and economics and then landed a job working for the police in Scotland. At the time, it didn’t seem like a dream job, but it turned out to be one of the best I’ve ever had. It was amazing to see how incoming data could influence how policing was carried out.

After leaving the police, I worked on the Scottish Index of Multiple Deprivation (SIMD), an index identifying the most deprived areas in Scotland. That work reignited my passion for using data to make a difference. We changed how we thought about and used the SIMD data.

From there, I worked in agricultural statistics, and now, as Chief Statistician, the big challenge we’re tackling is sorting out our messy data. Preparing it for new techniques and improving access to data for linkage projects are essential for conducting more exciting and impactful research.

Mark: I’m starting to feel like my career has been quite boring in comparison. We’ve got police work, cheffing, and megabats—who wouldn’t want to work with megabats? I feel like I’m on the edge of a midlife crisis here!

Ally: Mark I'm going to cut across you and ask, when you say Bacardi bats, do you mean Bacardi as in the rum?

Emma: Yes, exactly. If you think about a bottle of Bacardi rum, the cap has a picture of a bat on it. Louis Bacardi loved animals and built his own private zoo. After his death, many of the animals were given to other zoos, but his collection of megabats was preserved. It’s an incredibly important scientific collection because of the breadth of species and the excellent conditions in which the bats were kept. 

As a researcher studying Louis Bacardi’s bat collection, you also got to stay in his house. The most extraordinary part of it was the shower, which was designed like a rock formation, with water cascading from various points as if you were standing under a waterfall.

Mark: This is a good time to say we are not sponsored by Bacardi!

Emma: Definitely.

Mark: So something you were kind of picking up there along the end, but I'd like to ask to both of you really is what do you think is the hardest part about turning administrative data research into public good outcomes?

Ally: I think the main thing is that we get researchers closer to policymakers. Policymakers generally have good links with the communities they’re trying to benefit, and researchers generally have that as well. But I think the link between researchers and policymakers, when you understand the ambition behind the policy you're trying to implement, that's when you can really link the research and policy much closer together. As a researcher, you can then be properly informing and influencing how the policy is taken forward.

Emma: I couldn’t agree more, Ally. What we’ve found with UK government departments is that each one has its own culture. Really breaking down those barriers to say, "Look, it's really good to involve academics in these conversations," and making sure that academics have access to the right data so they can complete the research that produces insightful outcomes that can feed into policymaking.

Breaking those barriers down within policymaking circles is crucial so that they're not just asking the analysts within the government departments. Those government analysts should also be talking to researchers. It then becomes a virtuous circle, but it has to be done individually within each department for that culture change to happen across the board.

Mark: I think that’s something I’m very much signed up to. Academics have a lot of value to bring to these kinds of policy decisions, especially on the analytical side, and that’s what I want to do more of. I'm sure I'm not alone on that side of things either.

Emma: Yeah, civil servants misunderstand academics a lot of the time. They tend to think of academics as this amorphous group of people who all think and act the same way, which clearly isn’t the case. You've got people who want to do primary research that feeds into public good decision-making, and you've got academics who want to do different types of research. And that's absolutely brilliant. But we can be the conduit, linking those groups up who want to talk to each other and form those really constructive working relationships.

Ally: I’d add to that, especially in Scotland, where we're focused on child poverty. Not just right now but all the time, trying to make a difference. We held an event with local authorities, people supporting local authorities, ONS academics—everyone coming together in one room. It was great to see—it was a melting pot of ideas and ambition. This is part of the ADR programme, which gives us a vehicle to have these conversations and focus on specific topics, trying to solve problems.

Mark: So, Emma, could you talk a bit about what ADR UK is doing with admin data and how that links to evidence and policymaking? 

Imagine you're explaining it to my dad—he’s in his mid-60s and often describes growing up on the mean streets of Gainsborough. He’s not someone who is degree educated like myself.

Emma: Yeah, sure! The way I explain it is that every government service—whether it’s health, education, or anything else—collects data about people using that service. That’s great as far as it goes. But at ADR UK, we have this phrase: the "missed use of data." If that data stays within the department’s silo, it’s not being fully utilised. Our approach really comes into its own when we link data from one department to another. 

For example, we have decades of education outcomes data—from primary school to university. If you link that to Police National Computer data, you can understand who’s being cautioned and convicted for crimes. This helps us better understand how to prevent those crimes, especially by identifying groups getting caught up at an early age. Without this insight, policy decisions would continue to be random and less effective.

The Institute for Fiscal Studies published an evaluation of the Sure Start programme, which was introduced in the early 90s by the Labour government. It provided targeted support in deprived communities for parents of young children, it was health advice, education advice, training advice. Decades later, we can assess the outcomes for children from those areas. The benefit of the programme was particularly greater for children in deprived areas, showing that helping them lead better lives can have lasting effects on education outcomes—especially by age 16, which is a key milestone. Admin data allows us to evaluate programmes like Sure Start and learn what works over time.

Shayda: You mentioned something really interesting, Emma. Government departments collect data when people interact with services, but that data often just stays there. 

When I talk to my friends about this or do public engagement, it’s almost presumed that all departments are already talking to each other and working together, but they just want more information. Can you explain what the real picture is now, and what it looked like before ADR UK and similar programmes started?

Emma: I’ll separate two things. When government services collect data and use it for operational purposes, they’re making decisions about individuals. But when researchers access linked data across departments for research purposes, it’s a whole different thing. It’s legally important to note that researchers don’t get identifiable information. They can’t see individual names, addresses, NHS numbers, or anything that directly identifies someone. Research is done on much larger groups, analysing patterns in the data rather than focusing on individuals.

By linking data, like ethnicity from the census (which is self-reported), to police data (where ethnicity is recorded by officers), we can do meaningful research that wouldn’t be possible using data from individual services alone.

Ally: When I first joined the police, I thought they knew everything about everyone. But once I started working with the data, I realised that wasn’t the case. 

Researchers are more interested in patterns and groups rather than individual data. And when it comes to deprived areas, we want to improve outcomes for everyone, but especially for those who face the most barriers in life. Whether it’s health, crime, housing, or service access—addressing these barriers allows individuals to reach their full potential. That’s the real goal.

Shayda: I love how you mentioned the public. That’s why we’re here. Can you help us understand why everyday people should care about this?

Emma: Absolutely. The data we use is about people’s lives, and we have to engage the public to maintain the social contract. People need to trust that we’re using their data responsibly for research and public good. Public engagement has shown us that feedback challenges us to be better at what we do, resulting in better outcomes, a better website, and more meaningful conversations with the public. We have to continue this engagement.

Shayda: Ally, can you give a real-world example of how this is making a difference?

Ally: We’ve been focused on improving outcomes, especially for those who might otherwise be left behind. And that’s really what it’s about—ensuring that no one is left behind. At the end of the day, it’s about improving lives efficiently, using taxpayers' money responsibly to get the maximum benefit.

Shayda: Thank you so much, Ally and Emma. That's it for today's episode. Mark, how do you think it went?

Mark: Well, I think if you cut out all my speaking bits, it probably went OK, so not too bad.

Shayda: You’re too hard on yourself! You’re going to be a bigger personality than me by the time this series is over.

Well, we hope you enjoyed listening. You can find out more about ADR UK's work in the shownotes.

Mark: On the next episode, we're going to dive into the impact of research using data from the Ministry of Justice - which is my second favourite ministry after the Ministry of Sound.

Until then, stay curious about how your everyday data might shape society.