Mental health-related anti-stigma strategies are premised on the assumption that stigma is sustained by the public's deficiencies in abstract professional knowledge. In this paper, we critically assess this proposition and suggest new directions for research. Our analysis draws on three data sets: news reports (N = 529); focus groups (N = 20); interviews (N = 19). In each social context, we explored representations of mental health and illness in relation to students' shared living arrangements, a key group indicated for mental health-related anti-stigma efforts. We analysed the data using term-frequency inverse-document frequency (TF-IDF) models. Possible meanings indicated by TF-IDF modelling were interpreted using deep qualitative readings of verbatim quotations, as is standard in corpus-based research approaches to health and illness. These results evidence the flawed basis of dominant mental health-related anti-stigma campaigns. In contrast to deficiency models, we found that the public made sense of mental health and illness using dynamic and static epistemologies and often referenced professionalised understandings. Furthermore, rather than holding knowledge in the abstract, we also found public understanding to be functional to the social context. In addition, rather than being agnostic about mental health-related knowledge, we found public understandings are motivated by group-based identity-related concerns. We will argue that we need to develop alternative anti-stigma strategies rooted in the public's multiple contextualised sense-making strategies and highlight the potential of engaging with ecological approaches to stigma.
Keywords: behaviour change; communication; culture; data science; mental health; mixed-methods; public health; stigma.