This chapter covers topics which are frequently found in another category of textbooks directed to a different audience than that for which this textbook is written. Indeed, there are numerous and excellent “second-level” textbooks on advanced regression analysis (Dielman 2005; Draper and Smith 1998; Hocking 2003; Mendenhall and Sincich 2003; Ryan 2009); my purpose is not to compete with them in this single chapter. My purpose is limited to introducing the reader to two advanced regression applications frequently used in anthropology: multiple and logistic regression analysis.

The multiple regression model

The transition from simple to multiple regression analysis is not terribly difficult but it is certainly fun. In a multiple regression model we have one dependent variable and several independent variables which are potential predictors of the former. The art of regression model building resides in the analyst's particular choice of independent variables and their transformations into a model that is satisfactory. I call it an art because there is much room for individual choice. It is always fun to ask students to explain why they chose a model, and to hear different proposals for a better model. Not always is there an obvious best choice; sometimes there is considerable room for disagreement.