Anonymous Examinations at Uppsala University – Do the Names Matter?
Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
This study investigates gender, ethnic and social class bias in examinations at Uppsala University, Sweden. I employ the difference-in-differences approach to evaluate the blind marking reform and assess whether grading bias is taking place in the university. I find no statistically significant grading bias for the selected sample of courses. Although interaction terms on gender appear significant when dividing the subjects into the "objective" and "subjective" (technical and non-technical) sciences, these effects do not pass the robustness checks. Descriptive statistics support previous literature as differences in results between genders and natives/immigrants are significant. The difference between holders of high social class surnames and other surnames is not statistically significant. This study also supports some previous literature as there is an overall decrease in grades after the introduction of anonymous examinations.
Place, publisher, year, edition, pages
IdentifiersURN: urn:nbn:se:uu:diva-203610OAI: oai:DiVA.org:uu-203610DiVA: diva2:637072
Engström, Per, Lektor
Elinder, Mikael, Biträdande lektor