ABSTRACT
Statistical P value and its threshold have been controversial worldwide for a while. Recent heated debate was triggered by two practical issues: unexplainable high false positive rate in biomedical research, and global misunderstood of "statistical significance" in scientific community. Thus, part of scientists suggests applying more stringent significance level (from 0.05 to 0.005), or even giving up the use of significance level. We believe that they are throwing the baby out with the bath water. These suggestions will not contribute to any improvement of this unfavorable situation but will lead the scientific decision-making to a more difficult and subjective corner. Scientists should use statistical P value and threshold only if they correctly understand the soul of statistics-uncertainty. Statistical significance is neither sole nor dominant criterion to measure the scientific value, but an honest assistant. Scientific decision-making should initiate from the scientific experimental design, followed by rigorous implementation and transparent analysis, and synthesize a variety of information to reach a tenable conclusion.
ABSTRACT
RESUMEN Tras varios decenios de críticas a las técnicas inferenciales basadas en las pruebas de significación estadística orientadas al rechazo de la llamada "hipótesis nula" y, a pesar del notable consenso alcanzado entre los estadísticos profesionales, este recurso se mantiene vigente tanto en las publicaciones biomédicas, entre ellas las de Salud Pública, como en cursos introductorios de estadística. Entre las muchas deficiencias señaladas por los más prominentes especialistas se destacan tres por ser las más obvias y fáciles de comprender: que no contribuyen a cumplimentar la encomienda de la ciencia, que se conocen de antemano las respuestas a las preguntas que se encaran por su conducto y que los resultados que producen dependen de un elemento ajeno a la realidad estudiada: el tamaño muestral. El artículo discute en detalle tales limitaciones, ilustra su perniciosa presencia en la investigación actual y valora las razones para la subsistencia de la sinrazón en esta materia.
ABSTRACT After decades of criticism against inferential techniques based on statistical significance tests, which mainly reject the so-called "null hypothesis", and in spite of the remarkable consensus among professional statisticians, this resource remains prevalent in both biomedical publications (including public health journals) and introductory statistics courses. Among the many problems identified by the most prominent specialists, three of them are the most obvious and easy to understand: that these tests do not contribute to the actual enterprise of science, that the answers to the questions that are addressed are known in advance and that their results depend critically on an element that is external to the domain that is being studied: sample size. This paper discusses in detail these limitations, illustrates their pernicious presence in current research and evaluates the reasons for the survival of the senselessness in this matter.
RESUMO Trás vários decênios de críticas as técnicas inferenciais baseadas nas provas de significação estatística orientadas ao rejeito da chamada "hipótese nula" e, embora do notável consenso alcançado entre os estatísticos profissionais, este recurso se mantem vigente tanto nas publicações biomédicas, entre elas as de Saúde Pública, como nos cursos introdutórios de estatística. Entre as muitas deficiências assinaladas pelos mais proeminentes especialistas se destacam três por ser as mais obvias e fácies de compreender: que não contribuem a complementar a encomenda da ciência, que se conhecem de antemão as respostas ás perguntas que se encaram pelo seu conduto y que os resultados que produzem depende dum elemento alheio á realidade estudada: o tamanho amostral. O artigo discute em detalhe tais limitações, ilustra a sua perniciosa presença na investigação atual e valora as razões para a subsistência da sem-razão em esta matéria.