Regression analysis for social sciences

by Alexander von Eye

Regression analysis is the most widely used method in applied statistics. The method has many facets and provides many options. Introductory textbooks typically cover only standard ordinary least-squares regression. Regression Analysis for Social Sciences covers many more options, including robust regression, curvilinear regression, symmetrical regression, ridge regression, piecewise regression, regression for longitudinal data, and the partial interaction method for dealing with interaction problems. Sample applications are presented and sample command files are included for SYSTAT and S+. Results of analyses and characteristics of solutions are illustrated in over 50 figures.