Date of Award

2009

Degree Type

Thesis

Degree Name

Master of Science

Program

Biostatistics

Supervisor

Dr. Neil Klar

Second Advisor

Dr. Yun-Hee Choi

Abstract

The purpose of this research was twofold. The first objective was to assess the quality of modern cross-sectional ecological studies with a bibliometric review. The second objective was to investigate via simulation study the reliability of common ecological regression models for analysing count data.

The bibliometric review found that the quality and areas of application of the ecological literature is quite diverse. However, a large proportion of studies exhibited poor statistical practice and provided insufficient amounts of justification and information.

Linear, weighted linear, Poisson, and negative binomial regression were included in the simulation study based on their prevalence in the bibliometric review. The Poisson and negative binomial models had overly-liberal Type I error rates when faced with overdispersion or small samples respectively. Linear and weighted linear regression had highly robust Type I error rates. For all models, power decreased primarily as a function of overdispersion.

Share

COinS
 
 

To view the content in your browser, please download Adobe Reader or, alternately,
you may Download the file to your hard drive.

NOTE: The latest versions of Adobe Reader do not support viewing PDF files within Firefox on Mac OS and if you are using a modern (Intel) Mac, there is no official plugin for viewing PDF files within the browser window.