The Difference Between “Significant” and “Not Significant” is not Itself Statistically Significant.

One problem with declarations of statistical significance or nonsignificance is that changes in statistical significance are often not themselves statistically significant. Not only is any particular threshold arbitrary, but even large changes in significance levels can correspond to small, nonsign...

Descripción completa

Detalles Bibliográficos
Publicado en:American Statistician Vol. 60; no. 4; pp. 328 - 332
Autores principales: Gelman, Andrew, Stern, Hal
Formato: Artículo
Publicado: American Statistical Association November 2006
Acceso en línea:Ver este registro en EBSCOhost
Descripción
Sumario:One problem with declarations of statistical significance or nonsignificance is that changes in statistical significance are often not themselves statistically significant. Not only is any particular threshold arbitrary, but even large changes in significance levels can correspond to small, nonsignificant changes in the underlying quantities. The writers present theoretical and applied examples to illustrate this error of interpretation, stressing that students and practitioners must be made more aware of this ubiquitous statistical error.