There is some concern that ordered responses on health questions may differ across populations or even across subgroups of a population. This reporting heterogeneity may invalidate group comparisons and measures of health inequality. This paper proposes a test for differential reporting in ordered response models which enables to distinguish between cut-point shift and index shift. The method is illustrated using Canadian National Population Health Survey data. The McMaster Health Utility Index Mark 3 (HUI3) is used as a more objective health measure than the simple five-point scale of self-assessed health. We find clear evidence of index shifting and cut-point shifting for age and gender, but not for income, education or language. © 2004 Elsevier B.V. All rights reserved.