The aim of this study is to compare PhD students’ performance with respect to genderusing a number of matching methods. The data consists of fine-grained information aboutPhD-students at the Institute of Clinical Research at the University of Southern Denmark.Men and women are matched controlling for sub-disciplinary affiliation, education, year ofenrolment and age. Publications and citations are identified in Web of Science.Our study shows that the average total number of publication is slightly higher for menthan for women. Excluding the “other” group of publications from the analyses reveals thatthere is a negligible difference between men and women in terms of published articles.A substantial proportion of women is on maternity leave during the time period analysedand thus we would expect their productivity to be considerably lower. Similarly, we havefound very little difference between the citation impact of men and women.We find matching methods to be a promising set of methods for evaluating productivityand impact of individuals from various sub-fields, universities and time periods as we areable to discard some of the underlying factors determining the results of analyses of genderdifferences in productivity and citation impact.