There is widespread evidence of different outcomes for people from different ethnic groups in the British criminal justice system, but studies attempting to reveal discrimination by documenting different outcomes for offenders who are similar except for ethnicity have suffered from the difficulty of controlling for all possibly relevant characteristics. This article uses an alternative methodology: by reviewing British studies by the authors and others in which risk-assessed offenders have received similar disposals, it shows that in almost all cases the minority ethnic offenders have been assessed as presenting, on average, a lower score than white British offenders on various risk–need instruments. In most studies, this difference is significant. This suggests that members of minority ethnic groups are likely to receive, on average, more severe sentences than white offenders assessed as presenting similar levels of risk and criminogenic need, and therefore provides support for studies that have used the conventional methodology, as well as suggesting that risk–need assessment data can help social workers, probation officers and other professionals to monitor discrimination.