James P. Scanlan, Attorney at Law

Statistical Significance (sketch)

(May 15, 2010; rev. Nov. 29, 2010)

Eventually this page will give attention to a number of issues related to statistical significance.  These will include:

1.  The conflation of statistical significance with the strength of an association.  See item 4 of Comment on Bhopal and Section B of Comment on Schneider JAMA 2001.

2.  The treatment in scientific journal of nonsignificant results as evidence of the absence of an association. See Comment on Saxena BMJ 2002.

3.  The potential of measures of significance to provide a measure of association unaffected by the prevalence of an outcome.  See Statistical Significance SR sub-page of Scanlan’s Rule page of jpscanlan.com

4.  The treatment of statistical significance in the Datapoint case.  See note v in The Sears Case page of jpscanlan.com.