The authors provide a systematic treatment of the major problems involved in using regression analysis. They clearly and concisely discuss the consequences of violating the assumptions of the regression model, procedures for detecting violations, and strategies for dealing with these problems.