This chapter moves from symmetrical to asymmetrical in addressing the logit log‐linear model, which Knoke and Burke characterized as an analog of ordinary least squares (OLS) regression. It begins with a review of studies that have used logit log‐linear analysis, and it then covers the fundamental components of logit models. Many of these components resemble those in the general log‐linear model and can be interpreted in line with that technique. Notably, one of the primary differences between the logit log‐linear model and logistic regression analysis involves the treatment of continuous measures; in loglinear analyses, such measures must be treated as covariates, but logistic regression models accommodate interval‐level explanatory measures. The logit log‐linear model accommodates more than one response measure. Logit log‐linear analyses can be conducted in SPSS, which offers both general and logit programs, and an SPSS add‐on module facilitates correspondence analysis and data visualization.
Logit Log‐linear Analysis
07.10.2016
29 pages
Aufsatz/Kapitel (Buch)
Elektronische Ressource
Englisch
CHAID, LOGIT, and log-linear modeling
TIBKAT | 1990
|Wiley | 1993
|Mixed Logit (or Logit Kernel) Model: Dispelling Misconceptions of Identification
Online Contents | 2002
|Elsevier | 1989
|Mixed Logit (or Logit Kernel) Model: Dispelling Misconceptions of Identification
Transportation Research Record | 2002
|