We were puzzled by Dr Haghighat's criticism of our development of a stigma scale and would like to respond to his points. First, ours is a self-report measure of perceived stigma and we do not claim otherwise. Perceived stigma is a valuable construct that may have a greater impact on mental and social well-being (including relationships and occupation) than so-called objective acts of discrimination. This is also true of social support. Second, we agree that the relationship between perceived stigma and low self-esteem is potentially confounded by low mood. However, our sample contained a heterogeneous group of participants from a range of settings and thus it is unlikely that a sizeable proportion were depressed at the time of the study. In addition, Dr Haghighat overlooks the complexity of any putative association between stigma and depressive symptoms. Perceived stigma may cause or maintain depressive episodes.

Third, it is important to avoid invalidating reports of perceived stigma by dismissing them as depressive or paranoid epiphenomena. Fourth, Dr Haghighat claims that our instrument has no validity. In fact, as we made clear in our paper, it is based firmly on the views and experiences of people with mental illness who were interviewed in depth in a previous study (Reference Dinos, Stevens and SerfatyDinos et al, 2004), and thus it has greater validity than many scales used in the field of mental health. Fifth, we do not understand Dr Haghighat's reference to randomisation, which has no role here. If he means random selection of people to participate, then our method closely approximates to this in that potential participants were not selected on any predetermined basis. Naturally, participation depends to some degree on participants’ abilities and personal inclinations but that is true whether selected randomly or not.

Finally, participants in our earlier qualitative study (Reference Dinos, Stevens and SerfatyDinos et al, 2004) emphasised that positive outcomes may arise from experiencing mental illness and thus such items were included in our scale. We reversed their scores to indicate that stigma might be greater when such positive aspects were lacking. This is not the same thing as assuming mental illness has only negative aspects. In parallel fashion the opposite of risk is not protection, it is lack of risk.