Archives

VISION Policy Briefing: The impact of intimate partner violence on job loss and time off work in the UK

 

 

Key research findings

The latest research by VISION colleagues, Niels Blom and Vanessa Gash, finds serious negative effects of intimate partner violence and abuse (IPVA) on labour market outcomes, with 4% of those who experienced intimate partner violence losing their jobs because of the abuse. Furthermore, 1 in 8 of those who experienced intimate partner violence took a period of leave from work, with 1 in 4 of those who took leave needing to take a month or more off work.

Based on a large statistically representative sample for England and Wales, this research is one of the first to examine different types of IPVA, with five categories distinguished in the analysis.

The report examines differences between those who experienced; (1) physical abuse, (2) sexual abuse, (3) stalking, (4) coercive or controlling behaviour, as well as those who were (5) threatened with abuse by a current or former intimate partner.

There were strong differences in prevalence of IPVA by sex, with women disproportionately exposed to threats and to sexual violence. Additionally, compared to men, women were more likely to report multiple types of violence and abuse.

IPVA experiences are strongly related to employment status for both men and women; 5.2% of employed women experienced IPVA in the past year compared to 11.7% of unemployed women. IPVA was experienced by 2.8% of employed men and by 5.3% of unemployed men.

Job loss is associated with all five forms of IPVA, and the risks were highest for those who experienced: stalking, sexual violence as well as physical threats by an intimate partner.

The research, compiled in the latest VISION Policy Briefing, The Impact of Intimate Partner Violence on Job Loss and Time Off Work, also includes qualitative findings from those with lived experience of IPVA and abuse. Participants noted an ongoing stigmatisation of victims of abuse, which had serious impacts on disclosure. Victim-survivors noted their fear of being declared ‘unfit for work’ and of becoming a ’marked person’ should they disclose their abuse to relevant managers.

Figure 1. Types of IPVA experienced by female and male victim-survivors.

 

Policy implications

  • Though IPVA was found to have significant effects on victims’ experiences at work, those with lived experience noted a reluctance to disclose IPVA to relevant managers.
  • Employers may therefore want to consider enhanced IPVA and DA support systems for employees in the workplace.
  • While we can expect enhanced support to improve job retention and productivity, we currently lack the appropriate data to directly examine these effects

Updated version: key changes from version published in March 2024

Key changes are that we now include the survey collected in 2022/23, do not include 2004/05, and we now exclude people who had missing information on job classification (NS-SEC), or who lost their most recent employment more than a year before the interview. We now also include IPVA estimates by employment status.

Original VISION Policy Briefing

Short-report-Labour-market-consequences-of-IPVADownload

 

For further information, please contact: Dr Niels Blom at niels.blom@manchester.ac.uk or Prof Vanessa Gash at vanessa.gash.1@citystgeorges.ac.uk

Upcoming webinar – Left behind: People without support after experiencing violence

 

Thursday 4 June 2026, 13:00  – 14:30, online

Join VISION for a free webinar exploring groups who can be overlooked by health services, policing, and specialist support systems after experiencing violence.

Register here: TicketTailor – 4 June VISION webinar 

Many people affected by violence do not receive the help they need, for a variety of reasons. At VISION, we’ve analysed data sources such as the Crime Survey for England and Wales to better understand these gaps. In some cases, individuals do not seek medical care from hospitals or GPs for violence-related injuries, while others choose not to report incidents to the police. There are also those indirectly affected—such as people whose loved ones have experienced serious assault—who frequently go unsupported. In addition, a significant but less visible group includes victims of intimate partner violence and serious sexual assault in England and Wales who do not disclose their experiences, particularly to specialist services.

This research offers fresh insights into the risk factors, lived experiences, inequalities, and consequences of violence among those who neither seek nor receive support—the left behind.

After the short presentations, there will be a ’roundtable’ discussion with all present to look deeper into each presentation and talk about the barriers and opportunities. We want to better identify these missing populations and underst and their behaviours for not seeking help and conversely for those that are looking for support but the services aren’t necessarily there.

We welcome anyone working in government, police, healthcare, academia, specialist services, education and the community and voluntary sector interested in and / or working in violence prevention and support for those affected.

Programme

Discussant: Professor Ravi Thiara, VISION co-Investigator, University of Warwick

Healthcare inequalities following violence: analysis of the Crime Survey for England and Wales 2010-2024, Dr Anastasia Fadeeva, VISION Research Fellow, City St George’s University of London

  • Although healthcare is key to supporting victims of physical violence, some do not receive it despite injuries. The present research used the Crime Survey for England and Wales (combined waves 2010-2024) to identify which victims of physical violence were less likely to receive healthcare. Despite the presence of injuries, in almost a half of the incidents, victims receive no healthcare. We examined individual and violence-related factors that were associated with not receiving healthcare following violence victimisation. 

Indirect victims of violence: Mental health and the close relatives of serious assault victims in England, Professor Sally McManus, VISION co-Deputy Director, City St George’s University of London and Dr Elizabeth Cook, VISION co-Investigator, City St George’s University of London

  • Violence does not just harm direct victims; its effects ripple out through families. Drawing on a representative survey of adults in England, this study found that one in twenty adults were closely related to a victim of serious assault, and that these relatives carry a disproportionate burden of poor mental health. Even after accounting for their own histories of violence, adversity, and disadvantage, close family members face significantly higher levels of depression, anxiety, and feeling unsafe: evidence that policy must recognise, and victim services be resourced to respond to, the needs of families too. 

Reporting of violence victimisation to the police in England and Wales, Dr Polina Obolenskaya, VISION Research Fellow, City St George’s University of London and Dr Annie Bunce, VISION Research Fellow, City St George’s University of London

  • Who reports violence to the police, and under what circumstances, remains a critical but under‑examined question in England and Wales. Although national victimisation surveys consistently show that more than half of violent incidents never come to the attention of police, existing research is fragmented, often focused on single forms of violence (e.g., intimate partner or sexual violence), based on small studies or non-UK contexts. By mapping multiple routes through which violence does or does not come to the attention of the criminal justice system, this research advances an understanding of the “justice gap” and offers evidence with implications for policy, prevention, and victim‑survivor support. 

Disclosure of Intimate Partner Abuse and Sexual Violence to Formal Agencies and Specialist Services: Comparing Inequality Patterns, Victim Profiles, and Harms by Disclosure, Dr Hannah Manzur, VISION Research Fellow, City St George’s University of London and Dr Annie Bunce

  • Our study examines the hidden population of victims of intimate partner violence (IPV) and serious sexual assault (SSA) in England and Wales who report non-disclosure of their victimisation, particularly to specialised services. Whilst evidence-building largely relies on victim-survivors’ disclosure through help-seeking pathways and interventions, the experiences and inequality patterns for victim-survivors outside of these pathways are significantly missing from evidence and support provision. In particular, specialised services support some of the most marginalised and invisible victims of violence, yet barriers to disclosure and resource limitations pose significant challenges for both data collection and support access for these groups. The nationally representative Crime Survey for England and Wales offers a unique opportunity to analyse data on IPV and SSA victim-survivors who have not contacted specialised services or disclosed to any other formal agency (inc. The police and health services). Using pooled data (2004-2019) on past-year IPV and lifetime SSA, we compare inequality patterns (by gender, ethnicity, and migrant-status) and victim profiles (including risk-factors, victimisation characteristics, and harms) of victim-survivors based on disclosure (CSEW only, formal agency, or specialised services). Here, we reveal new insights into the risk-factors, experiences, inequalities, and impacts of violence against otherwise hidden violence victims, particularly those excluded from specialised services support.

Join us at this free webinar on 4 June, 13:00 – 14:30. To book your place please register here: TicketTailor – 4 June VISION webinar 

Photograph provided via Adobe Stock subscription

Using AI to investigate publicly available documents on violence prevention

 

Artificial intelligence (AI) systems are increasingly applied in public health, yet their use for analysing fragmented, multi-sectoral policy landscapes remains underdeveloped. Many applications have focused on service delivery, such as AI-powered chatbots, data surveillance and monitoring, and tracking social media interactions for emerging risks, with less attention paid to how AI might support policy analysis. This is especially true for the violence prevention sector, where AI is gaining traction as a solution for triaging help-seeking calls, detecting threatening messages, predicting conflict and improving police data, but not for understanding the policy landscape.

Policy responses to violence are undergoing scrutiny in the UK, coinciding with the recent publication of an updated cross-government strategy addressing violence against women and girls. This renewed focus places increased demands on researchers and policymakers to rapidly synthesise large and fragmented bodies of policy evidence spanning multiple sectors and both local and national government. Traditional approaches to policy review formed around a wholly manual approach may struggle to meet these demands within policy-relevant timeframes.

This research, an exploratory, proof-of-concept case study, aimed to describe the development and preliminary exploration of an AI-enabled tool designed to synthesise evidence from violence-related policy documents in the UK. The team was led by VISION Research Fellow Dr Darren Cook and inlcuded several members from the wider VISION consortium, Dr Elizabeth Cook, Kimberly Cullen, Professor Sally McManus, Professor Gene Feder and Professor Mark Bellis. 

For their article, Artificial intelligence in critical synthesis of public health responses to violence: A novel application to UK violence prevention policy, the team compiled a corpus of publicly available UK policy and strategy documents on violence (N = 343) through expert review, manual searches of government and third sector organisation websites, and automated web scraping.

Then, they used the corpus to train an existing AI framework and deployed it through a question-answer interface. Stakeholders working in violence prevention (academics, practitioners in specialist services and government officials) were invited to pose natural-language questions about violence policy and consider the system’s utility and the usefulness of its outputs. Their feedback indicated that the AI generated reports were well-grounded in the underlying source documents. Syntheses aligned closely with the documents in the tool, and the inclusion of document references and page-level citations supported credibility assessments. Corpus coverage statistics were considered particularly helpful when judging the robustness of responses. 

This research contributes by documenting the early application of an AI-enabled tool designed to support exploratory policy analysis. The team illustrates an emerging analytic capability and its potential role within policy-oriented research workflows. By demonstrating how a document-grounded, closed-domain AI system can be used to interrogate policy framings and identify potential siloes, this work addresses a gap in current public health applications of AI, specifically in the context of violence prevention.

To access the VISION AI tool to ask your own questions about violence prevention: VISION: Violence, Health & Society  

To download the paper: Artificial intelligence in critical synthesis of public health responses to violence: A novel application to UK violence prevention policy

To cite: Cook, D., Cook, E., Cullen, K., Zachos, K., McManus, S., Feder, G., Bellis, M., Maiden, N. Artificial intelligence in critical synthesis of public health responses to violence: A novel application to UK violence prevention policy. Science Direct (2026). https://doi.org/10.1186/s40163-026-00272-2

Illustration from Adobe Photo Stock subscription

Computational text analysis on unstructured police data: A scoping review

 

Police reports made following attendance at events such as crashes, domestic violence and theft often contain rich contextual details including indicators of mental health issues or abuse types, and persons/entities involved and their relationships, which are not typically captured in structured administrative data, interviews or official statistics. However, the sheer volume of information along with strict data access protocols render manual analysis impractical. Computational text analysis methods offer a feasible and effective approach to automatically process this underutilized data source.

The research team led by Dr Wilson Lukmanjaya (University of New South Wales) included VISION Research Fellow Dr Darren Cook. The team conducted an overview of studies using computational text analysis (e.g., text mining, natural language processing (NLP)), on unstructured police data, serving as a guide for researchers interested in employing similar methodologies. 

Their article, Computational text analysis on unstructured police data: A scoping review, was conducted in accordance with the PRISMA-SCR guidelines, following the two screening processes (title/abstract and full text screening) and the development of a pre-defined protocol. A search was conducted across seven electronic databases covering the past 20 years.

After removing duplicate entries and screening titles/abstracts and full-text publications, 61 studies met the inclusion criteria. Included studies were published between 2004 and 2024, with most from the United States, Australia and the Netherlands.

The scoping review indicates applications of computational text analysis on unstructured police data have moderate to high performance. Common limitations included variable data quality, with reliability depending on the level of detail provided by the police report’s author, and failure to report ethical implications or methodological limitations.

Computational text analysis can extract key information from unstructured police data. However, future research should clearly report ethics approvals and implications, and methodological limitations. 

Recommendation

  1. Establishing a structured data-sharing framework between law enforcement and researchers is crucial to facilitate access and support high quality, impactful research in this field.

To download the paper: Computational text analysis on unstructured police data: A scoping review

To cite: Lukmanjaya, W., Halmich, C., Butler, T., Cook, D., Karystianis, G. Computational text analysis on unstructured police data: a scoping review. Crime Sci (2026). https://doi.org/10.1186/s40163-026-00272-2

Photograph from Adobe Photo Stock subscription

Perpetration of intimate partner violence and suicide attempt, suicidal ideation, and non-suicidal self-harm: a cross-sectional secondary analysis using the Adult Psychiatric Morbidity Survey

Intimate partner violence (IPV) victimization is associated with suicidal behaviour. Suicidal behaviour may also be raised among those who perpetrate IPV compared to those who do not; general population-based evidence is, however, lacking.

The research team, led by Dr Sophie Carlisle (Nottingham University Hospitals NHS Trust) with VISION researchers Professor Sally McManus, Professor Louise Howard, and Dr Vishal Bhavsar and others, aimed to investigate the associations between using violence against an intimate partner with suicidal thoughts, suicide attempt and non-suicidal self-harm in the past year.

In contrast to previous research focusing on those in contact with criminal justice or health services or with IPV perpetrator programmes, this study presents the first examination of the association between IPV perpetration and suicidality in a recent UK general population sample, which can contribute to the development of a national picture of this association and inform population level strategies to address both suicide and IPV perpetration.

The research team analysed data from the 2014 Adult Psychiatric Morbidity Survey. Logistic regressions estimated associations between IPV perpetration and suicide attempt, suicidal ideation, and self-harm. Associations were estimated for men and women separately, and the team explored interaction in estimates by IPV victimization.

There were greater odds of suicidality and self-harm among self-reported perpetrators of IPV compared to the general population. Many of these associations were accounted for by non-IPV life adversities, IPV victimization and substance use. Improving the identification and management of IPV perpetration, and developing targeted safety planning and interventions for this group could reduce suicide for perpetrators and victims of IPV.

Future research generating adequately powered evidence on differences in these associations based on age or ethnic group, could inform targeted prevention/intervention strategies. Future work assessing the impact of increasing severity, or frequency, of IPV perpetration on risk of suicidality could also be helpful in informing future intervention strategies. Finally, further work should also consider the relevance of suicidality to a variety of harmful behaviours perpetrated within IPV. There remains limited evidence for interventions to reduce suicidality for perpetrators of IPV, including perpetrators who are also IPV victims.

Recommendation

Targeted identification and support for perpetrators of IPV could positively impact responses to suicidality and non-suicidal self-harm.

To download the paper: Perpetration of intimate partner violence and suicide attempt, suicidal ideation, and non-suicidal self-harm: a cross-sectional secondary analysis using the Adult Psychiatric Morbidity Survey

To cite: Carlisle S, Whyte R, Saunders K, McManus S, Oram S, Howard L, Bhavsar V. Perpetration of intimate partner violence and suicide attempt, suicidal ideation, and non-suicidal self-harm: a cross-sectional secondary analysis using the Adult Psychiatric Morbidity Survey. Epidemiol Psychiatr Sci. 2026 Mar 26;35:e16. doi: 10.1017/S2045796026100559. PMID: 41883282.

Illustration from Adobe Stock subscription

Gendered violence and epistemic injustice in Iran: women’s civic aspirations for justice

Iranian women perceive themselves as active participants in overcoming barriers that have historically silenced them. Their perspectives on eliminating violence against women (VAW), with a particular focus on how they challenge the structural and epistemic injustices that underpin such violence, are analysed in a recent publication, Gendered violence and epistemic injustice in Iran: women’s civic aspirations for justice.

In their article, Dr Nadia Aghtaie (University of Bristol), Dr Ladan Hashemi (VISION Senior Research Fellow at City St George’s University of London), and Fatemeh Babakhani (Mehre Shams Afarid, Non-Governmental Domestic Violence Organisation for Women and Children, Urmia, Iran), draw on a qualitative, purposive voluntary sample via an anonymous Instagram survey, chosen for its accessibility and reach among women and girls in Iran. A total of 453 respondents aged 16–59 answered the open-ended question, “Write your views and suggestions regarding violence against women and how to eliminate it,” and their responses were thematically analysed. While this method enabled wide participation, it also introduced likely sampling bias towards internet users—particularly younger, urban, and tech-savvy participants.

Participants identified a wide range of violence, including emotional, physical, sexual, coercive control, and street harassment. However, the most prominent themes that emerged were the need for raising awareness, among both women and men, about what constitutes violence, and the demand for comprehensive legal reforms to address and prevent VAW.

Many responses indicated a desire to reshape cultural and religious norms that have historically contributed to women’s marginalisation. The participants’ narratives highlighted how women’s experiences of violence are frequently dismissed, minimised, or rendered unintelligible in dominant public discourses. By articulating their understandings of violence and proposing solutions, these women actively resisted such injustice and asserted themselves as credible knowers.

Overall, respondents acknowledged the intersecting structural, cultural and religious norms that perpetuate VAW in Iran. Yet their responses were not solely diagnostic; they were also future-oriented and hopeful. They strongly believed that education, awareness-raising and legal reforms are catalysts for change and emphasised the right to be heard and valued as credible sources on their views on VAW.

To download the paper: Gendered violence and epistemic injustice in Iran: women’s civic aspirations for justice

To cite: Aghtaie, N., Hashemi, L. & Babakhani, F. Gendered violence and epistemic injustice in Iran: women’s civic aspirations for justice. Humanit Soc Sci Commun (2026). https://doi.org/10.1057/s41599-026-06952-3

For further information: Please contact Ladan at ladan.hashemi@citystgeorges.ac.uk

Illustration from Adobe Stock subscription

VISION researcher receives funding for secondary data analysis

Dr Annie Bunce, Research Fellow at VISION, received funding from the British Association for Counselling and Psychotherapy for her application, Exploring resilience, self-empowerment and wellbeing outcomes of women referred to specialist domestic abuse counselling services.

With the support of Dr Estela Capelas Barbosa, VISION co-Deputy Director, and in collaboration with Sarah Davidge, Head of Membership, Research and Evaluation at Women’s Aid, Annie will investigate whether and how receiving counselling from a specialist domestic abuse (DA) support service is associated with change in wellbeing.

She will analyse quantitative data from national DA charity, Women’s Aid, which includes information on various aspects of victim-survivors’ wellbeing at the start, during, and end of accessing services. Data analysis will reveal whether victim-survivors who receive counselling experience greater improvements in their wellbeing than those who receive other community-based services.

Annie will also examine whether counselling may be associated with greater wellbeing gains for some groups than others, and whether change in wellbeing is associated with the type/s of abuse experienced and other services received.

The analysis will show which factors influence the effect of counselling on changes in wellbeing the most, and which wellbeing indicators are most improved following counselling.

Findings will be shared via an academic report, blog, policy briefing, webinar and conference presentations.

The research will help to improve understanding of the relationship between counselling and wellbeing in the context of DA, feed into Women’s Aid’s ongoing work to ensure they are measuring the things most important to victim-survivors when it comes to their wellbeing and promote consistency in measuring wellbeing-related outcomes across DA services more widely.

Please contact Annie at annie.bunce@citystgeorges.ac.uk for further information.

Image from Adobe Stock subscription.

Non-intimate femicide in England and Wales: A ‘continua’ approach

A key pledge in the Labour Party’s 2024 election manifesto was to halve violence against women and girls (VAWG) over the next ten years. It is well known that violence and abuse experienced by women and girls incorporates a multitude of contexts outside of (as well as within) family and intimate relationships, therefore, any strategy aimed at significantly reducing VAWG will need to extend beyond the domestic sphere, including lethal violence. Despite important advances in domestic homicide (DH) prevention in recent years, 55 per cent of adult women (16+ years) killed across England and Wales were not categorized as DH.

Dr Caroline Miles (University of Manchester) and VISION Co-Investigator Dr Elizabeth A Cook (City St George’s University of London) specifically address the killing of women and girls outside of family and intimate relationships, referred to here as ‘non-intimate femicide’ (NIF), in their recently published article, Non-intimate femicide in England and Wales: A ‘continua’ approach.

There have been numerous high-profile killings of women and girls by male strangers in the UK over the past few year (for example, Sarah Everard, Sabina Nessa, sisters Nicole Smallman and Bibaa Henry, and three girls, Bebe King, Elsie Dot Stancombe and Alice Dasilva Aguiar killed in Southport). These cases attracted high levels of public attention as is often the outcome of intense media interest in particular femicides. Attracting less media attention are the killings of women and girls by men with whom they are acquainted but not intimately connected to (or in some cases, not recognized as such). There is currently a dearth of data, knowledge and policy aimed at preventing NIF, a problem which Caroline and Elizabeth strive to redress.

The research underpinning this article derives from the first exclusive study of NIF in England and Wales, presenting a statistical analysis of the victim, suspect and incident characteristics for all cases involving women who were killed by non-intimate partners or family members between 2002 and 2022. Using Homicide Index data for England and Wales (2002–2022), the researchers provide original insight into the victim, perpetrator and incident characteristics in NIF cases, and reveal important differences between intimate and NIF, as well as high levels of missing or poorly recorded data. They argue for a more accurate recording of NIF, alongside a ‘continua thinking’ approach to femicide research, which documents the killing of all women and girls across a range of intimate and non-intimate contexts. Caroline and Elizabeth write that by adopting a ‘continua of violence’ approach to femicide, which recognizes how ‘gender links violence at different points on a scale’, a nuanced and inclusive understanding of femicide can be developed that is not restricted to those categorized as ‘intimate’ or ‘domestic’.

If the current UK Labour Government are to succeed in their 2024 pledge to halve VAWG over a ten-year period, it is crucial that they focus on the whole continuum of lethal VAWG, including those killed by strangers and people known to them in some capacity who are not intimate partners or family members. NIF accounts for substantial proportions of female homicide victimization and although a key focus in recent years has been on learning more about the contexts of DH, the pathways leading to and circumstances surrounding the killing of women outside of intimate and kin relationships remain to a large extent unknown. 

Recommendation

Recent work to improve the recording of femicide and measure sex/gender motivations acknowledges some non-intimate contexts of femicide; however, in order to fully understand the gendered contexts of NIF, it is essential to improve the quality of data recording for all forms of femicide and to mainstream sex/gender motivation data collection across the whole continuum of femicide.

To download the paper: Non-intimate femicide in England and Wales: A ‘continua’ approach

To cite: Caroline Miles, Elizabeth A Cook, Non-intimate femicide in England and Wales: A ‘continua’ approach, The British Journal of Criminology, 2026;, azag005, https://doi.org/10.1093/bjc/azag005

For further information: Please contact Elizabeth (Lizzie) at elizabeth.cook@citystgeorges.ac.uk

Illustration from Adobe Stock subscription

Anastasia Fadeeva shares thoughts on Data Impact Fellowship placement studying healthy ageing

 

Dr Anastasia Fadeeva

VISION researcher and Data Impact Fellow, Dr Anastasia Fadeeva, has written a personal blog, Reflections from being a Data Impact Fellow: a placement in Japan, about her time in the country visiting universities and discussing healthy ageing.

In the blog, Anastasia reflects on her short-term placement at Chiba University and Kyoto University, meeting fellow researchers interested in population health and a focus on studying the ageing population and promoting healthy ageing.

As a Data Impact Fellow, Anastasia is researching the issues of violence in older age, the long-term impacts of violence on mental health, and the lack of reliable data. The placement to Japan is one component of the fellowship.

For further information, please see VISION member awarded Data Impact Fellow to study violence and mental health in older age to find out more about her fellowship or contact Anastasia at anastasia.fadeeva@citystgeorges.ac.uk

Top photo supplied through Adobe Stock subscription and bottom photo supplied by Dr Anastasia Fadeeva.

Improving police recorded crime data with natural language processing

Understanding and preventing Domestic Violence and Abuse (DVA) is compounded by long-standing data quality issues in police records. Accurate police-recorded crime data is vital for responding to DVA, yet it often contains missing values and inaccuracies.

Across all crime types, the quality of police data in England and Wales has been a concern. While there have been improvements in overall crime data recording since 2014, individual police forces still encounter difficulties adequately recording instances of DVA in police-recorded crime datasets. 

Correcting poorly recorded or missing data at this scale is non-trivial and beyond the capabilities of manual intervention alone. Fortunately, the increasing availability of computational solutions and machine learning algorithms such as text mining and natural language processing (NLP) can augment, and to a degree, offset much of this processing. NLP is supported by a growing body of interdisciplinary research, which shows that valuable information can be automatically extracted from unstructured data such as crime reports and case summaries through technology.

However, automated prediction systems are not without risk, particularly when applied in sensitive domains such as policing. Data inherently reflects societal biases that poorly designed AI solutions can amplify, and in the context of DVA, these biases may stem from underreporting of marginalized demographic groups or inconsistencies in police recording practices.

In their recent study, Improving police recorded crime data for domestic violence and abuse through natural language processing, VISION researchers Dr Darren Cook and Dr Ruth Weir (City St George’s University of London) and Dr Leslie Humphries (University of Lancashire), evaluated the capability of supervised machine learning models to automatically extract victim–offender relationship information from free-text crime notes in DVA cases.

Both models demonstrated that such tools could serve as cost-effective and efficient alternatives to manual coding, accurately classifying relationship type in around four out of five cases. The incorporation of a selective classification function improved precision for the most challenging cases by abstaining from low-confidence predictions, though at the cost of reduced coverage. This research represents a meaningful step toward addressing concerns about the completeness and reliability of police-recorded crime data.

Recommendation

Given that police-recorded crime lost its status as an accredited official statistic in 2014 due in part to weaknesses in data collection and processing, the application of data science methods to reliably impute missing values offers a promising route to restoring confidence in these records. Police constabularies are encouraged to use the available technology and implement text mining and NLP solutions to extract valuable information from unstructured data such as crime reports and case summaries.

For further information: Please contact Darren at darren.cook@citystgeorges.ac.uk

To cite: Cook D, Weir R, Humphries, L. Improving police recorded crime data for domestic violence and abuse through natural language processing. Front. Sociol., 24 November 2025, Sec. Medical Sociology Volume 10 – 2025 https://doi.org/10.3389/fsoc.2025.1686632

Photograph from Adobe Photo Stock subscription