In the academic literature on the measurement of discriminations, several approaches have been developed and have turned out to complement each other: the analysis of discriminatory residues associated with a characteristic (most often gender and origin); testings; the measure of implicit bias against minorities; or experience reported in surveys. This article applies the latter approach using data from the Trajectories and Origins survey (INED, INSEE, 2008-2009). Various discrimination indicators were built - on representations regarding discriminations, on the self-reported experience of discrimination and on discrimination situations in various areas of social life. These indicators are analysed jointly with the repeated experience of referrals to origins and the feeling of not being seen as French, considers as measurements of otherness. The article shows that “ethno-racial” discriminations mainly affect immigrants and descendants of immigrants of African and Turkish origin. A comparison of the discrimination indicators shows that the (more subjective) self-reported experience provides a level of prevalence which is systematically lower than that calculated from the discriminatory situations encountered in the various areas of social life. The strong correlation between these two indicators nonetheless shows that self-reported experience is based on concrete fact. Discriminations felt and measurements of otherness are also strongly linked. Lastly, we observe a significant and autonomous impact of the Muslim religion on the risk of reporting discriminations. These results plead in favour of evaluating discriminations using an approach that combines these types of measurement.