Loglinear analysis is best described as

• A. An extension of the chi-square test to analyze relationships between more than two categorical variables.
• B. A regression model for continuous outcomes.
• C. A time-series method for forecasting categorical data.
• D. A t-test for comparing two means.

Answer: (A)

Loglinear analysis examines relationships among several categorical variables by looking at cell counts in a contingency table and modeling the log of the expected counts as a function of the variables and their interactions. This approach extends the chi-square test to scenarios with three or more categorical variables, allowing formal tests of independence and interactions among them. It’s not a regression model for continuous outcomes, not a time-series method for forecasting categorical data, and not a t-test for comparing means.