Interview with Sara Thornton

Can you briefly summarise your time at the Institute of Criminology?

I'm an odd case. I studied for the diploma when I was in the middle of my police career, as part of a leadership course. That was in 2000, and at the time there was an option to carry on for another year and turn it into a masters. But I was going to a new job, I was moving house, I had a family. I thought there's no way I can do that, so I said no.

But then, ten years later, by which time I was a Chief Constable, I was talking to Lawrence Sherman, and he said, “why don't you come back and complete the masters? Finish what you started”. I think I did have a sense of unfinished business from 10 years earlier. So, I went back to Cambridge as a Chief Constable and completed the part-time Masters.

The course in 2000 had changed enormously by the time it got to 2010. There was much more focus on application: what does your learning mean for your practice? How could you make policing more effective? How could you use data properly? How could you focus on evidence to lead to better outcomes? The course was pretty theoretical in 2000, but by 2010 it was much more applied and relevant.

I did the course as Chief Constable of Thames Valley Police, and for the next five years while I was chief, we sent three or four people on the Masters course every year because I was convinced of its value to people who were operationally competent leaders.

If you're going to introduce new ways of working, new initiatives, new operations- if you’re introducing these, you must be clear about why you've chosen these options. And how do you evaluate them in a way which delivers the best outcomes to the public?

What convinced you to study Criminology?

For my initial diploma in 2000, I was a police officer in the middle of my career. I was on a leadership course; we did a couple of sessions at Cambridge and there was an option to do a diploma. I guess it was an opportunity that was given to me, and I took it.

Why? Because I was a police practitioner, and I wanted to understand not just the narrow silo of my practice but understand the context in which I was operating.

Why did you decide to come to Cambridge?

Because the leadership course had a relationship with Cambridge.

Can you briefly describe your thesis?

I did a thesis on predicting domestic violence homicide. As a chief officer in Thames Valley Police, we had had a number of really horrifying domestic violence murders.

One in particular, where I had met with the family on several occasions, and they were rightly very concerned about the response that their sister had received prior to her murder. There had been some prior contact with the police, and in their view, it was predictable and therefore preventable. I had been very moved by that case and wanted to understand the ways in which it could be predicted. More broadly, I wanted to investigate which cases of domestic abuse are likely to escalate to the most lethal or near-lethal forms of attack.

The standard method in policing was to risk-assess the victims, and I did a lot of work looking at the effectiveness of that assessment in predicting homicide and serious assault on Thames Valley. I found a lot of false positives and a lot of false negatives.

I also went back and looked at the method by which the risk assessment was developed, and that was flawed. There was no control group, for example, and there were issues about hindsight fallacy and failing to think about base rates.  More generally the risk assessments had become another checklist. It was more about the process than the impact of the outcomes.

I discovered that a better way to look at these cases was not at the victim, but at the offender.  We might have a better chance of predicting accurately if we looked at the offenders. It wasn't just me that was saying that; it was a school of thought that was gaining ground. Look at what Lawrence Sherman did as the Director of Research in the Metropolitan Police more recently: focusing on their top 100 offenders and having great success with that.

So, my thesis challenged the way in which things were done. What was interesting from my perspective was that my ability to get that changed was really quite limited, even as a chief constable. This was partly because it was a national approach, but also because a lot of people were wedded to that approach. It's taken a long time for thinking to change, and in a way, I don't blame people for not wanting to get rid of the system if there wasn't something else to put in its place. Because if it's protecting even some people, it has some value.

What have you gone onto since graduating?

I left policing six years ago, and I now work in anti-slavery and anti-trafficking.

The Cambridge course gave me a better understanding of the world of academia. I went on to be the UK’s Independent Anti-Slavery Commissioner and now work as a consultant on slavery and trafficking. But I also work as a Professor of Practice at the University of Nottingham, so I’m working part time in the world of academia.

I think the time at Cambridge was a good introduction to the world of research, which has stood me in good stead. I mean, I'm a professor of practice, so I'm bridging between the university and the world of policy and practice. But I'm a co-investigator on several research projects. In that regard, I suppose it was a springboard into my later career.  It's a good example where sometimes in your career, you follow your interests, you take opportunities, and you're not quite sure where you're going to end up. But that doesn't matter, because sometimes it actually proves to be a really good foundation, and a really good experience and you didn't realise where you're going to end up.

Can you give some examples of how you have used learning from the course in your career?

I suppose the Cambridge course gave me some of the building blocks in terms of thinking about research. It also I think gave me confidence that the academy could be relevant to policy and practice. In my subsequent career I always tried to bridge the research and the knowledge that was being produced, and either the world of policy or the world of operational practice.  And now, as a professor of practice, that’s the essence of my role. I'm not somebody who's published 100 research papers like my colleagues, but I do work part time now in the private sector.

I've been very involved in policy issues and the world of NGOs, so I'm able to form that bridge. And Cambridge showed me that it was valuable, that there were many relevant things happening in the academy which had application.

I am interested in how we could get those two worlds - the academy, and policy and practice - to understand each other more. That's certainly an issue in my work in Nottingham. The researchers will do great work, there are great insights, and then at the end there's a scramble to work out the recommendations for policy or practice.

If you don't spend some time with the policy people or the practitioners thinking through findings, the research lands and the policy people and the practitioners say ‘well, these recommendations are naïve; they're not based in our operating context; they don't understand some of the policy constraints’. I think my ability to bridge and understand both of those worlds is the value I bring to Nottingham. And as I say, I think I would trace that back to Cambridge, particularly on the Masters course.

What was the most important thing criminology taught you?

I learned a lot about data. I did classes on statistics for the first time in my life, on understanding evidence. I did a thesis on predicting domestic violence homicide, and in doing so I tested the approach methodologically.  I found many false positives and false negatives, which prompted questions about the appropriateness of the approach the police had taken to predict serious harm.

Then I started to look at what was the methodology on which the approach was based, the research was flawed. It taught me to be very careful about the data, to avoid confirmation bias and to check the facts.  A statement is not evidence, and it's certainly not proof that something works.

Specifically on domestic violence, there’s the risk of hindsight fallacy, where things look so predictable in hindsight. The question is what do they look like with foresight?

Then there’s issues about base rates. It's all very well saying what the percentage is, but actually the base rate matters a lot. It might be that there’s a high-risk group with a 1% chance and a standard risk group with a 10% chance. But if there's only 10 people in the high-risk group and 1000 people in the standard group, then you're going to have a lot more people affected in the standard group.

I think it's imbued in me a healthy scepticism around data, research and evidence, and a need to go back to the source. It wasn't the specificity about what I was studying. It was about an approach to knowledge, which I think was invaluable for my work now.

April 2025