What differences exist in sexual assault prevalence across ethnic groups in the UK?
This fact-check may be outdated. Consider refreshing it to get the most current information.
Executive summary
ONS Crime Survey for England and Wales (CSEW) provides the most reliable prevalence estimates by ethnicity: CSEW tables (notably Table 4 and Table 2/5 in the Sexual offences prevalence and victim characteristics releases) show sexual assault prevalence broken down by ethnic group for adults and for selected year ranges, most recently up to year ending March 2022 and in other datasets up to year ending March 2025 for some age bands (16–59) [1] [2] [3]. ONS and other UK official releases repeatedly warn that disclosure differences, survey coverage and incomplete perpetrator ethnicity recording limit direct comparison between groups and between victim and offender data [2] [4] [5].
1. What the official victim‑survey data say — ethnicity differences in prevalence
The Office for National Statistics publishes CSEW estimates showing the prevalence of sexual assault in the last year by personal characteristics, including ethnicity; these are available in Table 4 of the Sexual offences prevalence and victim characteristics dataset and in longitudinal tables going back to 2005 for adults aged 16–59 [2] [1] [3]. Those tables provide percentage rates of adults reporting sexual assault in the last year by ethnic group and sex [5]. The ONS release most recently referenced for ethnicity breakdowns is the year ending March 2022 tables, with plans or notes about later year releases such as year ending March 2025 for some CSEW outputs [1] [6] [3].
2. Why numbers vary and why comparisons are fraught
ONS cautions that several methodological issues complicate comparisons across ethnic groups: disclosure rates differ between communities, sample sizes for some ethnic groups are small (raising statistical uncertainty), and the survey covers those aged within specific bands (often 16–59 or 16–74 depending on table), which affects comparability [2] [5]. The ONS also notes police recorded data are not consistently broken down by victim ethnicity in headline tables and that offender ethnicity is incompletely recorded — limiting any direct match between victimisation patterns and perpetrator characteristics [1] [4].
3. Victim vs offender data — different questions, different limits
ONS and the Ministry of Justice (MoJ) collect distinct streams: CSEW measures victimisation experience (who reports being a victim), while police and MoJ data record offences, arrests and convictions (who is recorded or prosecuted). ONS points readers to MoJ tools for offender/outcome breakdowns but emphasises it does not hold comprehensive defendant/offender ethnicity tables for recent years — and that published perpetrator characteristics derived from victims’ reports are limited to specific datasets [4] [2]. This means a straightforward claim that "Group X commits more sexual assaults because victims of Group Y report being assaulted" is not supported by a single authoritative dataset [4] [2].
4. What official sources say about relative prevalence by major ethnic groups
ONS aggregated reporting has shown differences in prevalence across broad ethnic categories in past CSEW tables: for example, other ONS bulletins and related releases have presented higher overall prevalence of certain types of abuse in the White group compared with Asian and Black groups in some surveys [7]. However, the ONS emphasises cultural and disclosure effects may influence those patterns and that detailed, comparable year‑on‑year ethnicity breakdowns are provided only in specific tables [1] [2].
5. Perpetrator ethnicity: large gaps and cautious findings
Available ONS material states perpetrator ethnicity is not comprehensively recorded in police or justice datasets and that where offender characteristics exist they often come from victims’ descriptions rather than a full administrative breakdown; the MoJ retains more detailed offender‑by‑ethnicity tools that researchers are directed to for convictions and outcomes [4] [3]. The ONS therefore warns against overinterpreting partial offender ethnicity counts as representative across all sexual offences [4].
6. Political narratives and the evidence base
Media and political debate sometimes frame sexual offending in ethnic or immigration terms; reporting sources note contested interpretations and potential for racialised narratives. Independent coverage highlights concerns about over‑representation claims being influenced by policing bias, incomplete data and selective use of statistics [8]. The ONS and MoJ signpost that careful, contextual use of multiple datasets (CSEW for victimisation; policing and MoJ for offenders/outcomes) is required [3] [2].
7. Key limitations readers must accept
Official sources repeatedly state limitations: CSEW sample sizes and age bands, disclosure variation across cultures, lack of comprehensive offender ethnicity recording, and differences between victim‑reported perpetrator characteristics and administrative offender records [2] [4] [5]. These constraints mean definitive, policy‑grade rankings of sexual assault prevalence by fine ethnic subgroups are not available in a single, up‑to‑date public dataset [2] [4].
8. Where to look next (and what those sources will show you)
For prevalence by ethnicity consult ONS Sexual offences prevalence and victim characteristics tables (Table 4 and Table 2/5 referenced above) and CSEW annual supplementary tables (Table A11c for 16–59 age band through year ending March 2025) [1] [3] [2]. For offender/conviction breakdowns use the MoJ "Outcomes by Offence" tool and MoJ quarterly stats; ONS FOI guidance points users there for convictions by ethnicity [4] [9].
Limitations and alternative viewpoints are explicit in those sources: available data show differences across broad ethnic groups but also underline that disclosure, coverage and recording gaps make causal or policy conclusions from raw comparisons unreliable [2] [4] [5].