Psicothema was founded in Asturias (northern Spain) in 1989, and is published jointly by the Psychology Faculty of the University of Oviedo and the Psychological Association of the Principality of Asturias (Colegio Oficial de Psicólogos del Principado de Asturias).
We currently publish four issues per year, which accounts for some 100 articles annually. We admit work from both the basic and applied research fields, and from all areas of Psychology, all manuscripts being anonymously reviewed prior to publication.

  • Director: Laura E. Gómez Sánchez
  • Frequency:
         February | May | August | November
  • ISSN: 0214-9915
  • Digital Edition:: 1886-144X
  • Address: Ildelfonso Sánchez del Río, 4, 1º B
    33001 Oviedo (Spain)
  • Phone: 985 285 778
  • Fax: 985 281 374

Deciding on Null Hypotheses using P-values or Bayesian alternatives: A simulation study

Ana María Ruiz-Ruano García and Jorge López Puga

UCAM Universidad Católica de Murcia

Background: The p-value is currently one of the key elements for testing statistical hypothesis despite its critics. Bayesian statistics and Bayes Factors have been proposed as alternatives to improve the scientific decision making when testing a hypothesis. This study compares the performance of two Bayes Factor estimations (the BIC-based Bayes Factor and the Vovk-Sellke p-value calibration) with the p-value when the null hypothesis holds. Method: A million pairs of independent data sets were simulated. All simulated data came from a normal population and different sample sizes were considered. Exact p-values for comparing sample means were recorded for each sample pair as well as Bayesian alternatives. Results: Bayes factors exhibit better performance than the p-value, favouring the null hypothesis over the alternative. The BIC-based Bayes Factor is more accurate than the p-value calibration under the simulation conditions and this behaviour improves as the sample size grows. Conclusions: Our results show that Bayesian factors are good complements for testing a hypothesis. The use of the Bayesian alternatives we have tested could help researchers avoid claiming false statistical discoveries. We suggest using classical and Bayesian statistics together instead of rejecting either of them.

Decisiones sobre hipótesis nulas usando p-valores o alternativas Bayesianas: un estudio de simulación. Antecedentes: el p-valor es hoy en día, pese a las críticas, uno de los elementos clave del contraste de hipótesis. La estadística Bayesiana y los factores de Bayes han sido propuestos como alternativas para mejorarlo. Este estudio compara la ejecución de dos factores de Bayes con el p-valor cuando la hipótesis nula es la más plausible. Método: se simularon un millón de pares de conjuntos de datos independientes procedentes de poblaciones normales y se consideraron diferentes tamaños muestrales. Se calcularon los p-valores para comparar las medias muestrales para cada par de muestras, así como las alternativas Bayesianas. Resultados: los factores de Bayes muestran mejor ejecución que el p-valor, favoreciendo la hipótesis nula frente a la alternativa. El Factor de Bayes basado en el BIC funciona mejor que la calibración del p-valor bajo las condiciones simuladas y su comportamiento mejora a medida que el tamaño de la muestra aumenta. Conclusiones: nuestros resultados muestran que los factores de Bayes son buenos complementos para el contraste de hipótesis. Su utilización puede ayudar a los investigadores a no caer en falsos descubrimientos estadísticos y nosotros sugerimos el uso conjunto de la estadística clásica y Bayesiana.


Impact factor 2021:   JCR (WOS): 4.104 (Q1)   |   SJR (Scopus) : 1.308 (Q1)    |  CiteScore 2020: 5,3