Knocking down some Assumptions about the Effects of Evaluation Systems on Publications

In 1989 the Spanish Government established an individual retrospective research evaluation system (RES) for public researchers. This system was voluntary but involved the incentive of a permanent salary bonus. Policy makers have associated the establishment of this evaluation system with the signifi cant increase in the volume of scientifi c publications attributed to Spain over the last decades. In a similar vein to the analyses of other country cases, some scholars have also claimed that the growth of Spain’s international scientifi c publications is a result of the establishment of the new evaluation system. In this paper, we provide a methodological revision of the validity threats in previous research, including some interrupted time series analyses and control groups to investigate the effects of this policy instrument on the number of papers produced by Spanish authors. In the years following the establishment of the evaluation system, the results indicate a considerable increase in the number of papers attributed to Spanish authors among those eligible for evaluation (the 'treated' group), but also in the control groups. After testing various alternative explanations, we conclude that the growth in Spanish publications cannot be attributed indisputably to the effect of the establishment of the RES, but rather to the increase of expenditure and number of researchers in the Spanish R&D system along with some maturation effects. We take this case as an example of the need to improve and refi ne methodologies and to be more cautious when attributing effects to research evaluation mechanisms at the national level.Research evaluation systems; Scientific production; Effects of evaluations; Quasi-experimental design; Spain; CNEAI