Building better survey data on violence

 

By Sally McManus, VISION co-Deputy Director

Good policy depends on good data. While administrative data sources provide key insight, only surveys of the general population can evidence the extent to which violence is experienced, how it has changed in prevalence over time, and whether some groups are more affected – or less likely to get support – than others.  

The VISION consortium is working to improve the measurement of violence in the UK. Here we highlight examples of how VISION has advanced survey methods and generated new survey datasets to improve the evidence base on violence across the population. 

Ensuring the questions on violence get asked 

VISION research has challenged the historical reluctance to ask survey participants about their experiences of violence, with Dr Lizzie Cook arguing that the topics and people excluded from research cannot be counted or represented. VISION has worked with partners and ethics committees to improve and extend survey measurement so that it more faithfully and fully reflects the realities survivors face.  

Developing new questions on violence at work 

VISION researcher Prof Vanessa Gash has been working with the UK Household Longitudinal Study (UKHLS) to pilot an entirely new set of questions on experiences of violence at work. Workplace violence has historically been poorly captured in national surveys. These new questions will make it possible to examine who is affected, in which sectors, and, because UKHLS follows the same people over time, how experiences change within individuals across their working lives. 

Building a clearer picture of economic abuse 

In a project supported by VISION, Rosa Garwood and Surviving Economic Abuse developed and administered a detailed set of survey questions on economic abuse. This work has deepened understanding of the different component parts of economic abuse and revealed important inequalities in who experiences it. There is now a national dataset available for download from the UK Data Service. It means that researchers can now examine the structure and breadth of economic abuse rather than relying on narrow proxy measures of it. 

Capturing the links between types of violence 

A new approach to the measurement of violence and abuse was developed by VISION researcher Prof Sally McManus for the 2024 Adult Psychiatric Morbidity Survey (APMS). Rather than treating different forms of violence in isolation, the new approach spans a range of types of violence and abuse in a way that allows data users to examine how they interact, including co-occurrence and their combined relationship with mental health outcomes. The new dataset is now available for researchers to use, downloadable from the UK Data Service. 

Evidencing the long-term impacts of violence 

Dr Polina Obolenskaya co-ordinates VISION’s responses to consultations, including on survey content. We are keen to encourage the UK’s birth cohort studies and other longitudinal surveys to ask about violence and abuse. Where adopted, such questions have helped generate some of the first longitudinal data on violence and abuse, opening up new research questions about trajectories, risk, and recovery across the life course. 

Adding depth through open-text responses 

Survey approaches can be augmented through the collection of open-text responses, enabling qualitative insight to be generated at scale alongside structured measures. Research by Dr Nadia Aghtale and Fatemeh Babakhani with VISION researcher Dr Ladan Hashemi used an anonymous online survey and open-ended questions to capture women’s narratives of violence and their proposed solutions, revealing perspectives and forms of harm that are often missed in closed survey items. Their use of online data collection methods has shown how survivors from marginalised or hidden populations – often underrepresented in traditional survey methods – can also be reached via online platforms. 

Supporting researchers with designing and using surveys 

VISION has produced tools and guidance to help researchers design and analyse survey data. Dr Alexandria Innes created the VISION Risk of Bias Toolkit to inform understanding of the potential biases in different data sources, and Dr Hannah Manzur critiqued standardised measurement of ethnicity in national survey data proposing changes to the way surveys ask about and code ethnicity. Dr Niels Blom and Prof Vanessa Gash’s examined the strengths and weaknesses of different violence victimisation measures; while Prof Sian Oram and Dr Vish Bhavsar reviewed violence perpetration measures; and Dr Ladan Hashemi and Maryam Ghasemi’s make recommendations for improved measurement of adverse childhood experiences. To help those analysing survey data, Dr Niels Blom generated and archived code that helps users of the Crime Survey for England and Wales to merge multiple years of data, for better examination on lower prevalence groups in society. 

Why it matters 

Survey data isn’t perfect. But surveys provide one key part of the evidence landscape. Together, these developments contribute to a shift in the survey landscape. Violence and abuse have been excluded, undercounted or narrowly defined in national statistics. These new resources help give researchers better tools, and policymakers better evidence, to respond to how violence is patterned in the general population 

For further information about and to access the new survey datasets on violence, please contact: 

  • Vanessa Gash about the new workplace violence module on UKHLS, which will be archived for users in the future 

Illustration from Adobe Stock subscription

Blog