D. STEPHEN LINDSAY, University of Victoria
I will describe the background and rationale for the
following ten recommendations: 1. Sign on to the Transparency
and Openness Promotion guidelines; 2. Be wary of papers that
report a single underpowered study with surprising findings; 3.
Encourage preregistration; 4. Invite revisions with preregistered
replications; 5. Encourage sharing data and materials with
reviewers (and more widely after publication); 6. Ensure that
at least one reviewer has fierce stats chops; 7. Require finegrained
graphical presentations of data; 8. Require authors to
address known/anticipated constraints on generality; 9. Use
tools such as StatCheck to detect errors; and 10. Consider
inviting submissions that propose Registered Reports (pre data
?collection). The overarching principle is to value replicability
and importance over flashiness. Although the talk is pitched to
journal editors, authors may find the information useful too.
Email: D. Stephen Lindsay, slindsay@uvic.ca
That was the plan. In fact I gave 19 recommendations.
Steve