Abstract

FARJI-BRENER, Alejandro G. The (significant) biological significance of statistical non-significance. Ecol. austral [online]. 2006, vol.16, n.1, pp.79-84. ISSN 1667-782X.

Rejecting incorrect biological hypotheses as a consequence of statistically non-significant results is commonly undervalued as a step in the growth of our knowledge. Previous arguments against this incorrect belief had been largely based on the negative consequences associated with not publishing papers with statistically non-significant results rather than on the intrinsic epistemological merits to discarding erroneous ideas. In this essay, I will discuss how classical statistical methods and the epistemological approach from which these tools are derived are largely based on the elimination of falsehood rather than on the discovery of truth. The rejection of researches with statistically non-significant results denies the main advantage of classical hypothesis-testing methods. In fact, the rejection of an incorrect biological hypothesis based on high-quality research is one of the most powerful ways to understand nature.

Keywords : Biological and statistical hypotheses; Hypothesis testing; Non-significant results.

        · abstract in Spanish     · text in Spanish     · Spanish ( pdf )