Measuring Sexual Harassment in Higher Education: A Case Study of a Belgian University.

Higher education has been a focal point of research on sexual harassment for decades. Although studies have found this issue to be prevalent across universities, younger women from ethnic and/or sexual minority groups and university members belonging to medical disciplines are considered to be the most vulnerable (e.g. Bull & Page, 2019; Geldolf et al. 2021; Jussen et al. 2019; Lipinsky et al. 2022). Prevalence rates, however, differ greatly, ranging from 11 to 73 percent for women and 3 to 26 percent for men (Bondestam & Lundqvist, 2020). A potential cause of this remarkable disparity is methodology selection. One of the most widely used instruments to measure sexual harassment prevalence in higher education has been (an adapted version of) the Sexual Experiences Questionnaire developed by Fitzgerald et al. in 1988. This questionnaire enquires about respondents’ experiences with a predefined list of potentially transgressive behaviors. In general, prevalence rates measured by the SEQ are much higher than when a single-item question about self-labeled exposure to sexual harassment is employed. Through the lens of hermeneutical injustice, this study aims to examine whether this so-called ‘prevalence gap’ might be higher for certain social groups in higher education as opposed to others. A questionnaire including both a list of sexually transgressive behaviors and a single-item question was completed by a total of 2240 students and staff at a Belgian University. Preliminary results indicate that staff who were male, migrant, non-tenured and/or older than 50 were more likely to have inconsistent responses to both prevalence measures than their respective female, Belgian, tenured and younger colleagues. No significant differences were found in the student sample. These results indicate a discrepancy between how certain social groups within the University’s staff understand sexual harassment and how the research field measures it, risking a distorted representation of the issue.