Speculation Is Not Evidence: A Response to Conviction Is Not Commission

This response addresses a commentary by Orhan and Bal on our study of immigrant background and rape+ conviction in Sweden. We agree that conviction data do not capture all offending and that bias may occur at different stages of the criminal justice process. Our disagreement concerns the commentary's movement from the possibility of such bias to an interpretation in which unequal legal treatment is treated as an explanation for the associations observed in our study. The commentary does not measure police, prosecutorial, judicial, or legal-defense bias in our cohort and therefore cannot establish that these mechanisms account for our findings. Conviction remains a legally established individual-level outcome and an accepted source in criminological research, while alternative measures of offending also have important limitations. The associations in our study were observed across several immigrant-background categories and followed a clear gradient by duration of residence in Sweden. A general reference to possible discrimination does not explain this pattern or its magnitude. We also reject the claim that discussion of prevention or possible sociocultural mechanisms is inappropriate when such interpretations are presented cautiously and as hypothesis-generating. Research on sensitive subjects requires precise language, transparent methods, and explicit recognition of limitations. It should not, however, replace observed empirical associations with unmeasured alternative explanations or avoid discussing findings because they are socially contentious.