Modelling Tabular Data with an Ordered Outcome

A large amount of data that is considered within sociological studies consists ofcategorical variables that lend themselves to tabular analysis. In thesociological analysis of data regarding social class and educational attainment,for example, the variables of interest can often plausibly be considered ashaving a substantively interesting order. Standard log-linear models do not takeordinality into account, thereby potentially they may disregard usefulinformation.Analyzing tables where the response variable has ordered categories through modelbuilding has been problematic in software packages such as GLIM (Aitken et al.,1989). Recent developments in statistical modelling have offered newpossibilities and this paper explores one option, namely the continuation ratiomodel which was initially reported by Fienberg and Mason (1979). The fitting ofthis model to data in tabular form is possible in GLIM although not especiallytrivial and by and large this approach has not been employed in sociologicalresearch. In this paper I outline the continuation ratio model and comment uponhow it can be fitted to data by sociologists using the GLIM software. Inaddition I present a short description of the relative merits of such anapproach.Presenting this paper in an electronic format facilitates the possibility ofreplicating the analysis. The data is appended to the paper in the appropriateformat along with a copy of the GLIM transcript. A dumped GLIM4 file is alsoattached.