Intervention and replicability programs: considerations on their Evaluation of in Psychology Programas de intervención y replicabilidad: consideraciones sobre su evaluación en psicología

Abstract

In this work a review is made about what evaluation of intervention programs is, as well as the procedure and purpose that it has been sought to provide from psychology; and the procedure used to carry out the replication of these studies, especially in psychology, in order to give context to the so-called "crisis of confidence" in psychology. This in order to propose that replicability of intervention programs could not be satisfactorily achieved without first carrying out an evaluation of the original study. For this, three parameters are proposed, Bayes Factor, Effect Size and Confidence Intervals; which have shown their usefulness in English-speaking literature, but, nevertheless, it has not been widely used in Spanish-speaking literature. It is proposed to achieve a wider dissemination of Bayesian statistics to solve the problematics caused by the frequentist statistics of the Null Hypothesis Significance Test, since one of the purposes of Bayes' theorem is the accumulation and continuous updating of the knowledge obtained from replications, without leave aside the evaluation of the degrees of certainty of the results obtained. Thus, this work is aimed at new and veteran professionals in the field of evaluation, who are beginning to delve into statistics as a way of evaluate research and intervention programs, which is why we have sought to explain the parameters proposed in the most clear and concise way as possible.