Partialling and Purging

Equivalencies between Log-Linear Analysis and the Purging Method of Rate Adjustment

Bibliographic Data

ID2331632
AuthorsLawrence L Santi (University of Wisconsin, corresponding author)
Year1989
Volume17
Issue4
Pages376-397
Publication date1989-05-01
Peer ReviewedYes
Open AccessYes
TypeARTICLE
VenueSociological Methods & Research (JOURNAL)
Journal identifiersISSN: 0049-1241 • E-ISSN: 1552-8294
PublisherSAGE Publications Inc (PUBLISHER)
DOI10.1177/0049124189017004004
OpenAlexW2066563138
LanguageEN
Citations received5
References cited7

This article explicitly demonstrates the relationship between the purging method of rate adjustment introduced by Clogg and his associates and the more general framework of log-linear analysis. After a review of the general multiplicative (or log-linear) model, two types of purging are discussed; these 'purged' rates are shown to be related to the parameters of this model. A parallel discussion of 'purged' odds follows, wherein either purging method is shown to produce summary odds whose ratios are identical to the corresponding odds ratios of an appropriate log-linear model. These various formal relationships are illustrated by means of a numerical example

Econometrics · Linear model · Log-linear model · Logistic regression · Mathematical analysis · Multiplicative function · Odds · Statistics · Advanced Statistical Methods and Models · Computer Science · Mathematics · Optimal Experimental Design Methods · Statistical Methods and Bayesian Inference

Tools

Open DOISci-Hub