A tutorial in logistic regression

This article discusses some major uses of the logistic regression model in social data analysis. Using the example of personal happiness, a trichotomous variable from the 1993 General Social Survey (n = 1,601), properties of the technique are illustrated by attempting to predict the odds of individuals being less, rather than more, happy with their lives. The exercise begins by treating happiness as dichotomous, distinguishing those who are not too happy from everyone else. Later in the article, all three categories of happiness are modeled via both polytomous and ordered logit models.