NotesFAQContact Us
Collection
Advanced
Search Tips
ERIC Number: ED510703
Record Type: Non-Journal
Publication Date: 2009-Apr-16
Pages: 40
Abstractor: As Provided
Reference Count: 41
ISBN: N/A
ISSN: N/A
A Model Comparison for Count Data with a Positively Skewed Distribution with an Application to the Number of University Mathematics Courses Completed
Liou, Pey-Yan
Online Submission, Paper presented at the Annual Meeting of the American Educational Research Association (San Diego, CA, Apr 16, 2009)
The current study examines three regression models: OLS (ordinary least square) linear regression, Poisson regression, and negative binomial regression for analyzing count data. Simulation results show that the OLS regression model performed better than the others, since it did not produce more false statistically significant relationships than expected by chance at alpha levels 0.05 and 0.01. The Poisson regression model produced fewer Type I errors than expected at alpha levels 0.05 and 0.01. The negative binomial regression model produced more Type I errors at both 0.05 and 0.01 alpha levels, but it did not produce more incorrect statistically significant relationships than expected by chance as the sample sizes increased. (Contains 19 tables and 7 figures.)
Publication Type: Reports - Evaluative; Speeches/Meeting Papers
Education Level: Higher Education
Audience: N/A
Language: English
Sponsor: N/A
Authoring Institution: N/A