Sustaining the integrity of the threatened self: A cluster-randomised trial among social assistance applicants in the Netherlands

Stereotypes and stigma associated with living on welfare or a low income can be a psychological threat that hampers performance and undermines aspirations. Our paper explores the potential of a novel self-affirmation intervention to mitigate such adverse impacts. The intervention comprises a verbal self-affirmation exercise for applicants during their first meeting with a caseworker. We conduct a cluster-randomised trial among a sample of 174 applicants for social assistance benefits in a Social Services office in Maastricht, the Netherlands. We measure outcomes on feelings of self-worth, stress, societal belonging, job search behaviour self-efficacy and cognitive performance immediately after the meeting. In our full sample, the intervention has a negative impact on feelings of societal belonging, but no effect on other outcomes. Effects, however, vary by subgroups. Our treatment increases negative feelings of self-worth and negatively affects societal belonging, but also improves cognitive performance among the group that had paid work in the previous two years. By contrast, self-affirmation positively impacts job search behaviour self-efficacy and cognitive performance for individuals who face increased challenges to (re)integrate into the labour market, proxied by lower levels of education or social assistance receipt in the previous two years. Since our intervention gives rise to testing more than one null hypothesis, we control the false discovery rate using the Benjamini-Hochberg approach. Our findings are sobering. Effects only remain significant for negative feelings of self-worth and improved cognitive performance for one particular subgroup: individuals with paid work in the past two years. This suggests self-affirmation may have reminded them of the time they still had a job, hence creating a backlash effect on feelings of self-worth. At the same time, they may have felt a need to distinguish themselves from others on social assistance benefits resulting in better cognitive performance. These interpretations are consistent with theory and empirical evidence on social identity and self-categorisation. We discuss the implications of our results and outline avenues for future work.