The 1970s and 1980s were times when communication behavior was a primary interest of many communication scholars. The aim of this book is to reignite some interest in and passion about how human communication behavior should be studied. It presents the best advice, techniques, cautions, and controversies from the 1970s and 1980s and then updates them. Several chapters also introduce statistical methods and procedures to allow readers to analyze behavioral data.
This book is a useful resource for communication scholars and graduate students to guide their study of communication behavior.
10 Log-Linear Models of Nominal Observational Data
| 268 →
Log-Linear Models of Nominal Observational Data
Most empirical scholars of human communication have a passing acquaintance with the analysis of categorical data, although it was likely gleaned from the final chapter of their statistics text that was otherwise devoted almost exclusively to the analysis of continuous data (for one of the guilty texts, see Bowers & Courtright, 1984). As a result, they are not intimately familiar with a variety of statistical procedures that are available to analyze categorical data. Log-linear modeling is certainly one of those procedures.
As we saw in Chapter 9, the sequential analysis of interaction data requires that the data be arrayed in contingency-like tables. Log-linear models can address questions about such data that are beyond the scope of the basic X2 analysis. Beyond that, however, log-linear models are particularly useful when observational data are arrayed in multidimensional tables (more than two dimensions), because they provide the researcher with a systematic strategy for analyzing all of the possible relationships in the table, including analyzing the table as a whole (cf. Green, 1988).1