Weekend update--it's baaaaaack! Our paper explaining why N=55, 95% liberal, is not a valid sample for "replicating" our "motivated numeracy" study
Sunday, October 29, 2017 at 2:01PM
Dan Kahan

After a brief hiatus (primarily so we could reanalyze the data after using multiple imputation to handle missing data), our working paper responding to Ballarini & Sloman (2017) is back up at SSRN

As you likely will recall, B&S reported their "failure to replicate" our motivated numeracy study. Our response points out that B&S's N=55 student sample, which was 95% liberal (not a joke), had inadequate statistical power to replicate our study, which in addition to employing a design very different from B&S's used a large (N = 1100), nationally representative sample.

In addition to our paper, you can (re)read Mark Brandt's very reflective blog post on our paper and B&S's.

I'm still baffled about B&S's motivations for making such a weakly supported claim.  Very weird . . . .




Article originally appeared on cultural cognition project (http://www.culturalcognition.net/).
See website for complete article licensing information.