A pledge to acknowledge the "post-p <.05" era

Although p-values have been used in numerous studies, a longstanding criticism has existed over the misuse and abuse of p <.05.For example, in a correlation coefficient test (i.e., test of non-correlations), as a sample size becomes larger, its p-value becomes significant even if the correlation coefficient is small.Conversely, if the sample size is small, no matter how large the correlation coefficient is, the p-value may not be significant.It must be judged instead based on the correlation coefficient, which is the effect size.This is one very obvious example.In 2015, Basic and Applied Social Psychology (BASP) declared the null hypothesis significance testing procedure is invalid and banned the use of p-values (Trafimow & Marks, 2015).Even the prestigious journal Nature published a paper in 2019 that asserted "We should never conclude there is 'no difference' or 'no association' just because a p-value is larger than a threshold such as .05or, equivalently, because a confidence interval includes zero" (Amrhein, Greenland, & McShane, 2019).When reporting p-values, do not use adornments such as stars or letters to denote statistical significance and use sensitive precision (for example, p =.021 or p =.13).Moreover, in 2016, the American Statistical Association (ASA) published the "ASA statement on statistical significance and p-values," indicating six principles underlying the proper use and interpretation of the p-value and mentioned the use of Bayes factors instead of relying only on p-value (i.e., test of statistical significance) (Wasserstein & Lazar, 2016).Furthermore, in 2019 the assertion was made that, "We conclude, based on our review of the articles in this special issue and the broader literature, that it is time to completely dispense with use of the term 'statistically significant.'Nor should variants such as 'significantly different,' 'p <.05,' and 'nonsignificant' survive whether expressed in words, by asterisks in a table, or in some other manner" (Wasserstein, Schirm, & Lazar, 2019).Since the 6 th edition issued in 2009, the "Publication Manual of the American Psychological Association," which is often used in the nursing field, requires that p-values be expressed as specific numerical values, and that effect size and confidence interval be specified, among others.Under these circumstances, the Journal of International Nursing Research (JINR) hereby declares the transition to the "post-p <.05" era.As a side note, I think I would also like to mention that JINR does not necessarily guarantee that papers with statistically significant results are more likely to be accepted.