New way of predicting domestic abuse hotspots

An expert in crime mapping has come up with a way of accurately predicting domestic abuse hotspots so preventative and support services can be targeted where they are needed most.

Ruth Weir, who is studying for a PhD in criminology, hopes her findings will help reduce domestic violence by making it easier for victims to report the crime and receive the support they need.

It is estimated one in four women and one in six men will experience domestic abuse in their lifetime – with two women a week dying as a result of their abuse, at an estimated cost to society of £15.7 billion a year. It is also one of the most under-reported crimes, with the Crime Survey in England and Wales (CSEW) estimating that only 21 per cent of abuse is reported to the police.

Therefore having the ability to predict and identify abuse earlier and target resources and services to the right areas is fundamental in reducing its impact. Previous research has concentrated on who is most likely to be abused – using known risk factors such as sex, age and income. Now for the first time, the research looks at where the abuse is likely to take place.

Ruth, who has previously analysed crime patterns for both local and national government, used an innovative interdisciplinary approach to her research, combining geographical methodology with social theory. She mapped the whole of Essex, dividing the county into small sections, each with a population of around 1,500 people.

Looking at police and other data she found vast differences in abuse rates in different areas. For example in one area of Frinton, there were 11 incidents of domestic abuse in three years but, just six miles away in an area of Clacton, there were 536 reported incidents in the same period.

“I found that police-reported domestic abuse can be predicted accurately at the neighbourhood level using structural and cultural variables. The model, which was produced using spatial statistics, was found to explain 81 per cent of the variation in the data, with income and anti-social behaviour the biggest predictors in the model.

“It also showed that the predictors varied across the geographical area, which allows policy makers to design very focused and targeted services to support victims. Essex County Council has already used the research and other authorities and agencies are interested in adopting the model,” said Ruth.

Director for Integration and Partnerships at Essex County Council, Peter Fairley added: “ Ruth’s work and research is extremely exciting and ground-breaking. We at ECC are privileged to be working with Ruth developing our own domestic abuse risk model. Ruth’s input has meant that our work links well to current academic thinking and approaches and provides a welcome challenge to our thinking.”

In the next part of her study, she looked at the impact community facilities and support services, such as churches, sports clubs, village halls and food banks, have on reporting levels and the help available to abuse victims. She looked at 10 differing areas in more detail, visiting each in turn and talking to people who live there.

She found:

• in some areas there are good support services, but they are not always known about or well-used
• the most successful support services are in areas where there is a more settled population and a good community spirit
• where there are support services, people are less likely to report abuse to the police

“Although I only looked at a small number of places in detail, it was clear there is some good support work going on, but provision is patchy and there is a real need for all agencies to work together to share information and best practice,“ added Ruth.