Archives


Abandon Statistical Significance?

[From the abstract of a recent working paper by Blakeley McShane, David Gal, Andrew Gelman, Christian Robert, and Jennifer Tackett.] “In science publishing and many areas of research, the status quo is a lexicographic decision rule in which any result is first  required to have…

Read More

If At First You Don’t Succeed, Change Alpha

In a recent working paper, posted on PsyArXiv Preprints, Daniel Benjamin, James Berger, Magnus Johanneson, Brian Nosek, Eric-Jan Wagenmakers, and 67 other authors(!) argue for a stricter standard of statistical significance for studies claiming new discoveries.  In their words: “…we…

Read More

SCHEEL: When Null Results Beat Significant Results OR Why Nothing May Be Truer Than Something

[The following is an adaption of (and in large parts identical to) a recent blog post by Anne Scheel that appeared on The 100% CI .] Many, probably most empirical scientists use frequentist statistics to decide if a hypothesis should be rejected…

Read More

Reasons for Loving Null Results

In a great blog (“Why we should love null results”) posted at The 100% CI, Anne Scheel gives some reasons why we should love statistically insignificant findings. Her reasons include: — “We should love null results to counter our tendency…

Read More