1 online resource : illustrations (black and white, and colour)
Summary
This work introduces readers to logistic regression, often simply called logit. This technique allows researchers to evaluate whether a dichotomous dependent variable is a function of one or more independent variables. The logit model is most commonly estimated via maximum likelihood estimation (MLE). This example uses a subset of data from the 2012 American National Election Study. It presents an analysis of whether survey respondents reported voting for Barack Obama or Mitt Romney for U.S. President in 2012 and whether that vote choice can be predicted by several factors, including a respondent's race/ethnicity and how they feel about the Democratic and Republican Parties. An analysis like this allows researchers to test theories regarding what types of factors predict voting behaviour
Audience
Specialized
Notes
Description based on online resource; title from home page (viewed on May 18, 2015)