Effects of CAPES and CNPQ research evaluation systems on the publication patterns of the health science researchers in Brazil

Introduction: despite the increasing role of research evaluation and funding systems in evaluating the performance of researchers, there is a lack of empirical studies that analyze whether these systems influence their publication patterns. Objective: to analyze whether the evaluation criteria used by the Coordination for the Improvement of Higher Education Personnel and the National Council for Scientific and Technological Development influence the publication patterns of Health Sciences researchers in Brazil. Methodology: a semi-structured questionnaire is applied to a random stratified sample made up of two homogeneous strata: (a) researchers who are permanent professors in graduate programs and research productivity fellows; and (b) researchers who are not permanent professors or scholarship holders. Results: criteria more directly related to the evaluations, such as the score received by the articles in the evaluations, the demanded number of articles, the indexing of the journals in the main databases, or the value of the journal's bibliometric indicators, influence the choices of the evaluated researchers more than that of the non- evaluated ones. Factors not directly related to the evaluations, such as the importance of the articles for increasing the researcher's reputation, the degree of dissemination of this type of publication, the journal's prestige, or the quality peer review of the journal, also influence the choices of evaluated researchers. Conclusions: Researchers' responses to assessments depend on several factors; therefore, it is not possible to establish a deterministic cause-and-effect relationship between the evaluation criteria more directly related to the evaluations and the publication patterns of the researchers.