论文信息 - The insignificance of statistical significance

The insignificance of statistical significance

KEY MESSAGE(S): - Medical research will not improve if we simply interpret our results using the clear-cut difference between statistical significance and non-significance - Medical researchers should be more interested in the size of the observed result than in its statistical significance - Confidence intervals provide more information than P-values.

[1] Julian P T Higgins,et al. Systematic reviews and metaanalysis , 1997 .

[2] A. Akobeng,et al. Confidence intervals and p‐values in clinical decision making , 2008, Acta paediatrica.

[3] G. Hommel,et al. Confidence interval or p-value?: part 4 of a series on evaluation of scientific publications. , 2009, Deutsches Arzteblatt international.

[4] That confounded P-value revisited , 2011 .

[5] Sherry A. Greenberg,et al. Geriatric Depression Scale (GDS). , 2007, Occasional paper.

[6] D. C. Howell. Statistical Methods for Psychology , 1987 .

[7] J. Fethney. Statistical and clinical significance, and how to use confidence intervals to help interpret both. , 2010, Australian critical care : official journal of the Confederation of Australian Critical Care Nurses.

[8] H. J. Arnold. Introduction to the Practice of Statistics , 1990 .

[9] Seishi Terada,et al. [Geriatric Depression Scale (GDS)]. , 2011, Nihon rinsho. Japanese journal of clinical medicine.

[10] Jonathan A C Sterne,et al. Sifting the evidence—what's wrong with significance tests? , 2001, BMJ : British Medical Journal.

[11] J. Yesavage,et al. Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. , 1986 .

[12] T. Sterling. Publication Decisions and their Possible Effects on Inferences Drawn from Tests of Significance—or Vice Versa , 1959 .

[13] A Statistics Primer , 1996, The American journal of sports medicine.