Impact case study database
Changing national policy on measuring and recording domestic violence in England and Wales, and catalysing new campaigns to prevent football-related domestic violence
1. Summary of the impact
Francis’s investigation into the reporting of violence data in the UK has:
transformed Office for National Statistics (ONS)/UK Government policy on the reporting of series victimisations in the Crime Survey for England and Wales crime statistics, prompting a recalculation of historical estimates of series victimisations by the ONS dating back to 1991 (which showed violent crime to be between 6% and 32% more prevalent than previously estimated).
increased public awareness of the link between domestic violence and FIFA World Cup tournaments, and prompted advertising campaigns and policies among domestic violence charities and police forces throughout the UK and beyond.
increased reporting of incidents to domestic violence charities during the 2018 FIFA World Cup (eg 19.6% increase in reporting to the National Centre for Domestic Violence (NCDV)).
2. Underpinning research
As a member of the Centre for Crime, Law and Justice (CCLJ), Brian Francis (an expert criminological statistician), Sylvia Walby and Jude Towers (Sociology) investigated how series victimisations were reported in the Crime Survey for England and Wales (CSEW). This annual survey was initiated in 1982 and allows for crime reports to be made by either reporting individual crimes or reporting repeated crimes carried out by the same person under the same circumstances (series victimisations). When making series victimisation reports, the person reporting the crime is asked how many times the repeated crime occurred within the 12-month survey period.
Since the CSEW began, series victimisation crime counts were capped to 5 by the ONS, to reduce the volatility of the series [S3]. Volatility is important to the ONS, as their aim is to reduce year-to-year changes in estimates. Even if 20 series victimisations occurred within the survey period, only 5 (the capped figure) were included in the official statistics report. Thus, data were adjusted thereby seriously underestimating the incidence of crime in England and Wales.
In a series of ground-breaking publications between 2014 and 2017, and with ESRC funding [G1], Francis and colleagues explored the level of bias the cap introduces into the ONS methodology for the reporting of different forms of violent crime using CSEW data from 2011/12 [R1-R4]. They found that:
capping affected the estimates of trend lines and resulted in the overall level of crime being underestimated;
the cap on series victimisations disproportionately affected the reporting of domestic violence and violence against women;
analysis of all crimes indicated that the capping reduced the estimated number of domestic violence crimes against women after 2008 by 40% and the number of acquaintance crimes after 2008 by 50% [R2], [G1].
In response to these findings, Francis proposed a statistical approach for reducing volatility while also avoiding the bias problems of capping. The approach uses a moving average technique, where the reported number is based on data from a small number of recent years instead of only the current year. This method removes the bias, while keeping the year-to-year volatility at the same level as the previous method [R4].
In further research on domestic violence, Francis and Stuart Kirby (from the Lancaster University Law School) carried out the first statistical analysis of domestic violence reports received by Lancashire Police over 3 FIFA World Cup tournaments (2002, 2006, and 2010) [R5]. This analysis focused on whether the number of domestic violence incidents reported to the police increased when the England national team were playing in a FIFA World Cup match; whether there were differences in the number of reports of incidents based on whether England won, lost or drew; and whether there was a carry-over effect to the next day. Negative binomial regression was used to control for both day of the week and year, as domestic violence calls tend to be higher at the end of the week and police are becoming increasingly efficient at recording and labelling such calls. The results showed an increased risk of domestic violence when England played and an even larger risk when they lost. More specifically:
the risk of domestic violence rose by 26% when England won or drew compared to that on a non-match day;
there was a 38% increase in the risk of domestic violence when England lost compared to that on a non-match day;
there was an 11% increase in the risk of domestic violence on the day after an England match, irrespective of the match outcome, suggesting a carry-over effect.
Francis and Kirby’s research has advanced the understanding of football-related domestic violence in England by demonstrating an important connection between fans’ responses to the England playing and the increased risk of the occurrence of domestic violence, whatever the result of the match [R5].
3. References to the research
[R1] Walby, S., Towers, J. and Francis, B. (2014) Mainstreaming domestic and gender-based violence into sociology and the criminology of violence, The Sociological Review, 62(S2), 187-214. DOI: 10.1111/1467-954X.12198. Peer-reviewed publication. Ranked 36 out of 1,282 outputs (the top 3% for this journal) by Altmetric Attention Score in Context.
[R2] Walby, S., Towers, J., and Francis, B. (2014) The decline in the rate of domestic violence has stopped: Removing the cap on repeat victimisation reveals more violence. Research briefing for ESRC project detailed in [G1].
[R3] Walby, S., Towers, J. and Francis, B. (2016) Is violent crime increasing or decreasing? A new methodology to measure repeat attacks making visible the significance of gender and domestic relations. British Journal of Criminology, 56, 1203-1234. DOI: 10.1093/bjc/azv131. Peer-reviewed publication. Ranked 11 out of 1,180 outputs (the top 1% for this journal) by Altmetric Attention Score in Context.
[R4] Walby, S., Towers, J., Balderston, S., Corradi, C., Francis, B., Heiskanen, M., Helweg-Larsen, K., Mergaert, L., Olive, P., Palmer, E., Stöckl, H., and Strid, S. (2017) The Concept and Measurement of Violence against Women and Men. Policy Press. Held at HEI. Funded by [G1] and other external funders. Downloaded 2,936 times from the university research repository and cited 77 times.
[R5] Kirby, S., Francis, B. and O’Flaherty, R. (2013) Can the FIFA World Cup football (soccer) tournament be associated with an increase in domestic abuse? Journal of Research in Crime and Delinquency, 51(3), 259-276. DOI: 10.1177/0022427813494843. Peer-reviewed publication.
Peer-reviewed grants
[G1] Francis (Co-I) (with Walby S (PI)), Is the rate of domestic violence decreasing or increasing? A re-analysis of the British Crime Survey, ESRC: (2013-14) GBP158,849.
Quality of research
The quality of the research is reflected by the Royal Statistical Society’s award of the Howard Medal for 2021 to Francis for outstanding contributions to the development and application of social statistics.
4. Details of the impact
In the year up to March 2020, there were 2.3 million cases of adults aged 16 to 74 who experienced domestic abuse in England and Wales; of these, 1.6 million were women (source: ONS). The adoption of the recommendations and new methodology designed by Francis and colleagues by the ONS has resulted in changes to the reporting of crime statistics, a less biased reporting system, and more accurate reporting of domestic violence incidents across the UK. The ONS have also corrected historical estimates of violent crime in England and Wales from 1991 onwards. In addition, the pioneering research into the relationship between domestic violence and FIFA World Cup football tournaments has revealed a hidden problem, catalysing NGO and police campaigns, raising public awareness of the issue for the first time, and leading to increased reporting of domestic violence during the 2018 FIFA World Cup.
(a) Transforming ONS/UK Government policy on the reporting of series victimisations in CSEW crime statistics, prompting a recalculation of historical estimates of series victimisations by the ONS dating back to 1991
Research on the effects of the cap on serial victimisation crimes and the development of a new reporting methodology was disseminated through a briefing paper [R2] published on the website of the Violence and Society UNESCO Centre at Lancaster, and in peer-reviewed publications [R1, R3, R4]. The briefing paper was sent to the ONS and domestic violence charities, and The Guardian and The Telegraph published articles on the underestimation of violent crimes against women, discussing the research [June 2015, combined potential reach: 3.52million, source: Kantar, S1]. Following the popular and practitioner attention given to the research findings, and citing [R2], the ONS subsequently sought the advice of the Government Statistical Service Methodological Advisory Committee (GSS MAC), which agreed that the ONS’s methodology of capping crime volume in the Crime Survey should be reviewed [S2, S3]. The research was the impetus for the ONS’s 2016 public consultation on their methodology for reporting high-frequency repeat victimisations in the CSEW [R2, S4]. The results of the consultation were published in November 2016, and the ONS committed to changing their methodology for counting incidents of high-frequency repeat victimisation and accepted Francis and colleagues’ original criticisms [R3, S5]. Following the press reports, the issue of counting domestic violence events was discussed on the More or Less Radio 4 programme in September 2017 (3.28 million listeners, source: Kantar) [S6], further increasing public awareness of the issue.
As well as changing its methodology for counting incidents of high-frequency repeat victimisation from that time on, the ONS also decided to apply the new methodology to historical data. Drawing on the research recommendations [R2], in 2017 the ONS agreed to recalculate all crime estimates, initially from 2002 onwards and now going back to 1991 [S7, p.5]. For violent crimes, the strict cap of 5 was relaxed to 12, which substantially reduced the bias in ONS estimates. The Deputy Director, Crime, Income & Wealth Division, Public Policy Analysis Directorate at the ONS stated that the research “ played an important contribution to the revision of this methodology” [S3]. The adoption of Francis's methodology [R2] led to new crime volume estimates, published in January 2019 [S7], which were higher than those initially reported. “The most substantial effect was found in relation to violent incidents”, with between 6% and 32% more violent instances being counted [S7, p.2, p.5]. In one specific year, for instance, “ *the year ending March 2018 … the new methodology has increased the total number of violent incidents by 167,000 from 1.3 million to 1.4 million incidents, an increase of 13.3%*” [S7, p.5].
(b) Increasing public awareness of the link between domestic violence and FIFA World Cup tournaments, and prompting advertising campaigns and policies among domestic violence charities and police forces throughout the UK and beyond
Francis and Kirby’s research raised public awareness of the link between domestic violence and the FIFA World Cup leading up to the 2014 tournament [R5]. Between September 2013 and June 2014, the research was reported in, for example, The Times of India, Algemeen Dagblad in The Netherlands, and the Medical Daily website and The Cut in America, with a combined potential readership of 8.34million (sources: Kantar, Agility and Wikipedia). The research played a key role in advertising strategies for domestic violence charities. For example, in June 2014, BMB (a London-based advertising agency) released a public service announcement video for a London-based charity, Tender Education and Arts. It highlighted the finding that a 38% increase in domestic violence is expected when England lose in a FIFA World Cup [R5, S8]. The video was viewed over 600,000 times on YouTube during the 2014 FIFA World Cup and had been viewed over 1,000,000 times at the end of 2020).
(c) Wunderman Thompson
Prior to and during the 2018 FIFA World Cup, the research gained further traction and renewed attention [R5]. In June and July 2018, media coverage included the CNN and BBC websites, French newspaper Liberation and Elle magazine (combined potential reach: 9.27million, sources: Agility; Kantar). The research also featured in The Conversation and South Africa’s EyeWitness News website in July 2018, receiving 68,490 views across the UK, Australia, the USA, South Africa, Canada, India, France, Norway, Germany and Ireland (116,767 views, source: The Conversation). The potential value of the research for domestic violence charities was again harnessed. As a prominent example, the Creative Team at Wunderman Thompson (WT), a London-based advertising agency, drew on the statistical findings to create the ‘Not-So-Beautiful Game’ campaign for the NCDV as the 2018 FIFA World Cup approached. The NCDV helps provide advice and protection to thousands of domestic violence victims and survivors throughout England and their research-informed campaign sought to reach and support victims when they were at particular risk of danger [S9]. The campaign ran for the duration of the tournament between 14th June and 15th July 2018, featuring the image below with the embedded caption reading ‘If England get beaten, so will she. Domestic Violence increases 26% when England play. 38% if they lose’. With the FIFA World Cup rapidly approaching, the campaign was made a priority for the NCDV, with the research findings being key to the “ incredible response” and support that WT were able to secure from key contributors that allowed them to go from concept to artwork in eight days [S9]. WT’s Creative Director stated: “ [Francis’s] statistics enabled us to confront people with the appalling reality of this issue, create awareness and breed a culture of disgust, not acceptance, when it comes to domestic violence. In using [Francis’s] figures, we aimed to provoke much-needed conversations and force the public to rethink what the World Cup means for domestic abuse victims at exactly the moment when they are most vulnerable. We also wanted to ensure that victims and their friends/families were aware of the free legal advice they could access via the NCDV, and the support this organisation offers in terms of life-saving injunctions” [S9].
The power of the statistical findings contributed to the campaign’s success, which “made the headlines across 100+ news platforms in over 18 countries” [S9] (sample images can be seen here). The NCDV noted: “ Better informed and more accurate statistics form an important gauge on the scale of the problem of domestic abuse as well as providing milestones for progress towards the fulfilment of our mission” [S10]. WT also produced a video using the research as part of the campaign, which was entered into the Cannes Lions Film Festival 2019 and won the Gold Award in the Sports Category [S11]. The video was displayed on digital screens at main railway stations across the UK, and finishes with the following quote from a survivor of domestic violence: “As a survivor Thank You so much for making hard to view campaigns that really explain what its like”. Similar, successful campaigns based on the research [R5] were launched by other organisations to coincide with the 2018 FIFA World Cup competition, including the Pathway Project (PP), a domestic and sexual abuse service based in Lichfield & Tamworth, Staffordshire. The PP’s CEO confirmed that research statistics from the research [R5] were embedded in its 'Show Violence the Red Card' campaign. He also reported that the “ …research also had a huge impact on our marketing…This was a real eye-opener as we had previously relied on flyers and posters, and it changed our marketing strategy going forward. We now use social media in other campaigns…to raise awareness and education people about the issues around domestic violence” [S12]. Furthermore, in 2018, Full Fact checked and verified the 38% figure and its provenance [S13]. Alongside domestic violence charities’ campaigns, police authorities across the UK, including in Cleveland, Devon and Cornwall, Humberside and Lancashire, drew on the research findings of [R5] in support of the ‘Give domestic violence the red card’ poster campaigns. In addition, citing the 38% increase in domestic violence statistic, Hampshire Constabulary committed to deploying dedicated response cars to attend domestic abuse incidents following England World Cup matches and safeguard victims.
(c) Increased reporting of incidents to domestic violence charities during the 2018 FIFA World Cup
The “extraordinarily powerful and successful” NCDV research-informed promotional campaign [S10] resulted in a substantial increase in social awareness of the charity, generating thousands of organic social media shares and comments: “Social awareness of the NCDV and the services they offer also increased by 1250%. This generated hundreds of thousands of shares and comments, and sparked millions of much needed conversations worldwide. We targeted mass reach and shares on social media and individuals with large followings spontaneously shared the posts, enabling us to reach many more people organically. A single post helped us reach around 4.5 million people, while another was shared 40,000 times and generated 30,000 comments” [S9]. Referral calls to the NCDV increased by 19.6% while the campaign was live, the largest sudden spike (until the country first went into lockdown in 2020) [S9, S10]. The Pathway Project similarly received an increase in calls to the charity following their research-informed campaign [S12], [R5].
The reach and significance of the impact detailed above is recognised by the statement from the Royal Statistical Society in bestowing the Howard Award on Francis that “ The quality and importance of his work can be seen by its impact … on how the ONS report crime figures; and being used in an extensive campaign to reduce domestic violence during the football world cup” [S14].
5. Sources to corroborate the impact
[S1] Gayle D, ‘Violent crime against women massively understated, statistics agency told’, The Guardian, 9 June 2015; Barrett D, ‘Million violent crimes a year “cut out” of official figures’, The Telegraph, 9 June 2015
[S2] Paper presented to GSS MAC by the ONS seeking advice (November 2015) (see pp.18-30)
[S3] Testimonial from Deputy Director, Crime, Income & Wealth Division, Public Policy Analysis Directorate, ONS (February 2021)
[S4] ONS, Review of methodology for addressing high frequency repeat victimisation in Crime Survey for England and Wales estimates (July 2016)
[S5] ONS, Response to the ONS consultation on the methodology for addressing high-frequency repeat victimisation in Crime Survey for England and Wales estimates (September 2016)
[S6] BBC Radio 4 - More or Less: Behind the Stats, Uber; EU passports; counting domestic violence, 29 September 2017
[S7] ONS, Improving victimisation estimates derived from the Crime Survey for England and Wales (January 2019)
[S8] Tender Education and Arts, ‘Woman’s reaction to England World Cup knockout’, public service announcement video, 24 June 2014
[S9] Testimonial from Creative Director of Wunderman Thompson (formerly J Walter Thompson) (January 2021)
[S10] Testimonial from CEO of National Centre for Domestic Violence (February 2021)
[S11] NCDV, The 'Not-So-Beautiful Game" campaign video, 20 June 2019
[S12] Testimonial from CEO of Pathway Project (March 2020)
[S13] Full Fact independent fact check, ‘World Cup: does domestic abuse spike when England lose?’ 19 June 2018
[S14] Howard Award letter from the Royal Statistical Society to Brian Francis (February 2021)
Additional contextual information
Grant funding
Grant number | Value of grant |
---|---|
ES/K002899/1 | £158,849 |