Why to use Poisson regression for count data analysis in consumer behavior research

Feihong Xia ()
Additional contact information
Feihong Xia: University of Rhode Island

Journal of Marketing Analytics, 2023, vol. 11, issue 3, No 9, 379-384

Abstract: Abstract Count data are often encountered in consumer behavior research. Normal regression, or ordinary least squares, has been used predominantly to analyze count data in experimental studies, while the appropriate models for count data analysis such as Poisson regression have not been fully embraced in consumer behavior research. The fact that only a small fraction of published papers in consumer behavior research with count data have used Poisson regression calls for a push to rethink the common approach of using normal regression for count data analysis. To demonstrate the importance and value of using Poisson regression for count data, we first discuss the parametric forms and properties of both normal regression and Poisson regression, and then show readers through large-scale simulated experiments that Poisson regression is the appropriate model to use for count data, not only because of better model fit but also because of lower error rates in hypothesis testing in various experimental settings, which is critical for consumer behavior researchers.

Keywords: Consumer behavior research; Count data; Poisson regression; Hypothesis testing; False negative; False positive (search for similar items in EconPapers)
Date: 2023
References: View references in EconPapers View complete reference list from CitEc
Citations: View citations in EconPapers (2)

Downloads: (external link)
http://link.springer.com/10.1057/s41270-022-00166-7 Abstract (text/html)
Access to the full text of the articles in this series is restricted.

Related works:
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:pal:jmarka:v:11:y:2023:i:3:d:10.1057_s41270-022-00166-7

Ordering information: This journal article can be ordered from
http://www.springer. ... gement/journal/41270

DOI: 10.1057/s41270-022-00166-7

Access Statistics for this article

Journal of Marketing Analytics is currently edited by Maria Petrescu and Anjala Krishnen

More articles in Journal of Marketing Analytics from Palgrave Macmillan
Bibliographic data for series maintained by Sonal Shukla () and Springer Nature Abstracting and Indexing ().