False Positive Poisson

13 Pages Posted: 11 Nov 2018 Last revised: 30 Jun 2021

See all articles by William Ryan

William Ryan

University of California, Berkeley - Haas School of Business

Ellen Evers

UC Berkeley, Haas

Don A. Moore

University of California, Berkeley - Haas School of Business

Date Written: October 19, 2018

Abstract

When analyzing count data (such as number of questions answered correctly), psychologists often use Poisson regressions. We show through simulations that violating the assumptions of a Poisson distribution even slightly can lead to false positive rates more than doubling, and illustrate this issue with a study that finds a clearly spurious but highly significant connection between seeing blue and eating fish candies. In additional simulations we test alternate methods for analyzing count-data and show that these generally do not suffer from the same inflated false positive rate, nor do they result in much higher false negatives in situations where Poisson would be appropriate.

Keywords: poisson regression, false positives, Type I Errors, methodology, fish

Suggested Citation

Ryan, William and Evers, Ellen and Moore, Don A., False Positive Poisson (October 19, 2018). Available at SSRN: https://ssrn.com/abstract=3270063 or http://dx.doi.org/10.2139/ssrn.3270063

William Ryan (Contact Author)

University of California, Berkeley - Haas School of Business ( email )

545 Student Services Building, #1900
2220 Piedmont Avenue
Berkeley, CA 94720
United States

Ellen Evers

UC Berkeley, Haas ( email )

Haas School of Business
Berkeley, CA 94720
United States

Don A. Moore

University of California, Berkeley - Haas School of Business ( email )

545 Student Services Building, #1900
2220 Piedmont Avenue
Berkeley, CA 94720
United States

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
111
Abstract Views
819
rank
305,857
PlumX Metrics