Abstract
dc:description.abstractModerated regression is used when there are continuous variables and categorical variables in the regression model. The purpose of moderated regression is to explain the differences between the levels of the categorical variables. This study introduces a set of regression models with and without moderating effects. To evaluate the effectiveness of those models, a simulation study was used. Type I error rates, power, and identification rates for the generating models are reported. Results show that a small difference between degrees of freedom for two models is related to inflated Type I error rates and sample size 500 is usually associated with adequate power and identification of the generating model. A large difference of R values between the focal group and the reference group is related to adequate power.
Author and committee
dc:creator, dc:contributor.*- Author dc:creator
-
- Kang, Yi
Rights
- Language dc:language.iso
- en
Identifiers
dc:identifier.*- Dc Identifier Citation
- Kang, Yi. (2024). Models of Moderation in Regression. Retrieved from the University Digital Conservancy, https://hdl.handle.net/11299/269658.