Workplace Sexual Harassment in Denmark: A Methodological Critique

Mohammed Abou El-Kheir & Wasan El-Hallak

Student thesis: Master thesis


In this interdisciplinary study, we investigate and review how contemporary Danish studies on workplace sexual harassment measure the phenomenon and what possible limitations there exists in their applied methodologies. Withal, we assume there exists certain limitations and discrepancies among surveys in the field, which we believe, harnesses inaccurate and underreported estimates on the prevalence of sexual harassment. Our analysis is compiled of five selected Danish studies from the National Research Center of Working Environment along with studies from a select number of labor unions, namely FOA, FH and Epinion, 3F and HK. We conduct a methodological review of these surveys respectively and compare them in terms of their survey designs and methodological framework, which we relate to their findings and discoveries on the subject matter. Through our analysis we highlight the limitations of the question formulations, frames of references and sampling strategies, among other things, in producing adequate and realistic estimates. We show that providing respondents with general and broad questions without any elaboration on how the sexual harassment was expressed, nor any definition of the construct, is associated with lower prevalence levels. Similarly, we emphasize how frame of reference and the sample composition largely impacts and increases the chances of recall bias and sampling errors. Moreover, we briefly assess how qualitative modes of inquiry can harness descriptions and facets of the incident(s) that are not otherwise captured by the surveys in question, such as cases of quid pro quo sexual harassment. In addition, we reveal tendencies across the five studies that serve as potential limitations in properly assessing severity, by the exclusion of variables that account for other personal characteristics, such as ethnicity and sexual orientation, but certainly also how the relationship between the offender and victim affects experienced severity. Finally, we address the shortcomings of not including questions and items that exhibit the consequences that the incident(s) has on victims’ mental health and wellbeing.

EducationsMSc in International Business and Politics, (Graduate Programme) Final Thesis
Publication date2021
Number of pages138
SupervisorsChristoph Houman Ellersgaard