The U-Ranking’s predictive validity for postgraduate studies in the fields of Health Studies

Abstract

INTRODUCTION. In recent years, university rankings have attracted significant media attention. The information −that these rankings provide about the university activities in relation to quality of teaching, research and innovation and technological development− has an important social influence and it affects the reputation and prestige of universities. In addition, these rankings increase the university attractiveness for students, conditioning the decision-making process of potential students. This paper studies the predictive validity of the U-Ranking for graduates of the Faculties of Medicine, Pharmacy, Nursing and Psychology in Spain. METHOD. This paper analyzes how the scores obtained by different public and private Spanish universities may predict the level of performance achieved by their graduates in a national test for selecting health workers (using data from four calls 2012/13, 2013/14, 2014/15 and 2015/16). RESULTS. The results show that the position in the U-Ranking is not related to the universities’ success rates in these selective tests. DISCUSSION. The marked trend of these results reveals a validity problem of the U-Ranking, which could be associated with a systematic error in predicting performance criteria through features that are not relevant in relation to it. Consequently, specific predictions to certain groups or entities might show a major benefit for definite units, producing significant inconsistencies with reality. Likewise, it is also important to rethink how the university rankings are built, based on a precise definition of t