In online labor markets such as Uber, Lieferando, and Etsy, ratings determine income, visibility, and career opportunities—often more quickly and definitively than in the traditional job market. The ECDF Award for Digitalization and Diversity 2025 recognizes work that shows how strongly such ratings can be influenced by gender-specific expectations: The award goes to Diana Tran Nhat, doctoral candidate and research assistant in the Digital Service Engineering department at the Technical University of Berlin, for her publication “Expectation vs. reality: How stereotypes and expectation disconfirmation affect job evaluations in online labor markets” (PLOS ONE, 2025). The co-author is ECDF professor Timm Teubner.
Digital labor markets and gender bias: How expectations distort evaluations
In the award-winning paper, she conducts an online experiment in which the gender, field of work, and performance level of workers are systematically varied. Although gender bias in such reputation systems is well documented empirically, the underlying mechanisms have not yet been sufficiently researched. Theoretically, the study is based on expectation disconfirmation theory and role congruity theory. It examines whether deviations between expected and actual performance (positive or negative “disconfirmation”) are evaluated differently depending on the stereotypical fit between gender and field of activity.
A key finding of her research: “Contrary to common assumptions, gender-stereotypically ‘appropriate’ constellations do not automatically lead to higher expectations or more positive evaluations. Only evaluators with strong stereotypical ideas, i.e., people who clearly associate certain areas of work with a particular gender, show significant effects,” explains Diana Tran Nhat. In these cases, workers in stereotypically matching fields are expected to perform better, and their failure to meet these expectations is judged more leniently. The study thus reveals how subtle and context-dependent discrimination mechanisms can be on digital platforms. It shows why identical performances are evaluated differently depending on gender stereotypes, expectations, and their (non-)fulfillment—with direct consequences for income and career opportunities.
Diana Tran Nhat has been a doctoral candidate and research assistant in the Digital Service Engineering department since August 2022. She previously studied Industrial Economics (M.Sc.) at the Technical University of Berlin and Economics (B.Sc.) at the University of Hohenheim. In her research, she examines reputation systems on online platforms, in particular how ratings are created, perceived, and can reinforce social inequalities.
About the ECDF Award for Digitalization and Diversity
With the ECDF Award “Digitalization and Diversity,” the ECDF annually honors outstanding work that addresses issues of gender, diversity, and social inequality in the context of digital transformation. The aim is to highlight both theoretical-conceptual and empirical and practical contributions. Researchers, employees, and students in the context of the ECDF are eligible to apply. In addition to traditional publications, experimental, artistic, or innovative research formats are also expressly welcome. The deadline for submissions is October 31. //Further information.