The course is designed for those with some previous experience in regression analysis who seek to refresh and enhance their current knowledge and extend their basic proficiency in applying the technique. In addition, it provides the necessary background in statistical methods for those seeking to advance onto courses dealing with factor analysis and structural equation modelling (SEM), as well as the use of associated software packages, such as LISREL and AMOS. The main focus of the course will be on the practical side of regression in terms of when it is an appropriate tool of analysis to use for a given problem, and how to interpret the results that it produces.
Beginning with a short review of the principles of multiple regression, the course then examines a series of specific issues and problems that arise from its application. These issues include:
• the appropriate use of categorical (“nominal level”) variables, including dummy variables
• how to treat instances of missing data
• establishing the existence of non-linear relationships and identifying the transformation procedures necessary to apply to the data
• how to incorporate and interpret interactive effects between variables within regression models
• how to form a scale and index for inclusion in regression analysis
• analysis and interpretation of residuals to detect potential violation of OLS assumptions
• how to identify the presence of multicollinearity and what to do about it
• how to estimate models with simultaneity (“two-way” causation).
Data will be provided; however, participants will have an opportunity to analyse their own data and discuss the output. Data analysis will be done using SPSS software; however, no previous knowledge or experience of SPSS is required.
David John Gow is a consultant in research methods and statistics and their application in the social sciences. He has taught in many ACSPRI Summer and Winter Programs
Knowledge of elementary statistical techniques and some knowledge of the principles of multiple regression at a level comparable to that provided by the ‘Fundamentals of Multiple Regression’ course.
The instructor’s course notes, which will be distributed to all participants, serve as the course text.
Nearly all social statistics texts treat regression analysis and thus constitute suitable reference material. The following short monographs provide short, clear and technically sound coverage.
• Lewis-Beck, M., Applied Regression: An Introduction, Sage, 1980.
• Achen, C., Interpreting and Using Regression, Sage, 1982.
• Berry, William and Stanley Feldman, Multiple Regression in Practice, Sage, 1985.
• Hardy, M., Regression with Dummy Variables, Sage, 1993.
In addition the following texts are of relevance:
• Karhane, L.H. Regression Basics, Sage, 2001
• Pedhazur, E.J., Multiple Regression in Behavioral Research, 3rd ed., Holt, Rinehart & Winston, 1997.