The proportional odds with partial proportionality constraints model for ordinal response variables.

The proportional odds assumption in ordered logit models is a restrictive assumption that is often violated in practice. A violation of the assumption indicates that the effects of one or more independent variables significantly vary across cutpoint equations in the model. In order to relax this assumption for the cumulative odds model, researchers may use either a "partial" model that relaxes the assumption for a subset of variables or the "generalized" model that relaxes the assumption for every independent variable. In this paper, we propose a relatively new and under-utilized third alternative, the proportional odds with partial proportionality constraints (POPPC) model, which allows the effects of a subset of variables to vary across cutpoint equations by a common factor. We improve upon an earlier formulation of the POPPC model by offering an additional conceptual justification for the model and an estimation method that does not require the use of person-threshold data. We illustrate the POPPC model with two examples from the 2008 General Social Survey.

[1]  P. McCullagh.  Regression Models for Ordinal Data , 1980 .

[2]  John Aitchison,et al.  THE GENERALIZATION OF PROBIT ANALYSIS TO THE CASE OF MULTIPLE RESPONSES , 1957 .

[3]  R. Brant.  Assessing proportionality in the proportional odds model for ordinal logistic regression. , 1990, Biometrics.

[4]  Robin Plackett.  The analysis of categorical data , 1974 .

[5]  J. S. Long,et al.  Regression models for categorical dependent variables using Stata, 2nd Edition , 2005 .

[6]  James K. Yarnold,et al.  The Minimum Expectation in X 2 Goodness of Fit Tests and the Accuracy of Approximations for the Null Distribution , 1970 .

[7]  A. Agresti.  Analysis of Ordinal Categorical Data , 1985 .

[8]  D. Rubin,et al.  Multiple Imputation for Nonresponse in Surveys , 1989 .

[9]  A. Agresti,et al.  Categorical Data Analysis , 1991, International Encyclopedia of Statistical Science.

[10]  A. McCright,et al.  Environmental Concern and Sociodemographic Variables: A Study of Statistical Models , 2007 .

[11]  Richard Williams.  Generalized Ordered Logit/Partial Proportional Odds Models for Ordinal Dependent Variables , 2006 .

[12]  David Andrich,et al.  A model for contingency tables having an ordered response classification , 1979 .

[13]  A. Goldberger,et al.  Estimation of a Model with Multiple Indicators and Multiple Causes of a Single Latent Variable , 1975 .

[14]  Darren E. Sherkat,et al.  Race, Religion, and Opposition to Same‐Sex Marriage , 2010 .

[15]  D. R. Heise.  Employing Nominal Variables, Induced Variables, and Block Variables in Path Analyses , 1972 .

[16]  P. J. Green,et al.  Probability and Statistical Inference , 1978 .

[17]  C. I. Bliss,et al.  THE METHOD OF PROBITS--A CORRECTION. , 1934, Science.

[18]  Traversing the flexible turn: US workers’ perceptions of job security, 1977–2002 , 2007 .

[19]  V. Fu.  Estimating generalized ordered logit models , 1999 .

[20]  Stephen E. Fienberg,et al.  Three centuries of categorical data analysis: Log-linear models and maximum likelihood estimation , 2007 .

[21]  Jeffrey M. Wooldridge,et al.  Solutions Manual and Supplementary Materials for Econometric Analysis of Cross Section and Panel Data , 2003 .

[22]  Robert M. Hauser,et al.  1. Another Look at the Stratification of Educational Transitions: The Logistic Response Model with Partial Proportionality Constraints , 2006 .

[23]  P. Schmidt,et al.  Limited-Dependent and Qualitative Variables in Econometrics. , 1984 .

[24]  W. Greene,et al.  Modeling Ordered Choices: A Primer , 2010 .

[25]  Bradford S. Jones,et al.  Reexamining Racial Attitudes: The Conditional Relationship Between Diversity and Socioeconomic Environment , 2005 .

[26]  Stephen R. Cole,et al.  Regression models for unconstrained, partially or fully constrained continuation odds ratios. , 2001, International journal of epidemiology.

[27]  R. McKelvey,et al.  A statistical model for the analysis of ordinal level dependent variables , 1975 .

[28]  D. J. Finney.  Probit analysis; a statistical treatment of the sigmoid response curve. , 1947 .

[29]  J. Gurland,et al.  Polychotomous Quantal Response in Biological Assay , 1960 .

[30]  F. Harrell,et al.  Partial Proportional Odds Models for Ordinal Response Variables , 1990 .

[31]  S. Cole,et al.  Estimation of cumulative odds ratios. , 2004, Annals of epidemiology.

[32]  Strother H. Walker,et al.  Estimation of the probability of an event as a function of several independent variables. , 1967, Biometrika.

[33]  P. G. Sankaran,et al.  On proportional odds models , 2008 .

[34]  K. Fujimoto.  From women’s college to work: Inter-organizational networks in the Japanese female labor market , 2005 .

[35]  E. Snell,et al.  A Scaling Procedure for Ordered Categorical Data , 1964 .

[36]  Roderick J. A. Little,et al.  Statistical Analysis with Missing Data , 1988 .

[37]  Andrew S. Fullerton.  A Conceptual Framework for Ordered Logistic Regression Models , 2009 .

[38]  J. S. Long,et al.  Regression Models for Categorical and Limited Dependent Variables , 1997 .

[39]  Ji-Hyun Kim,et al.  Assessing practical significance of the proportional odds assumption , 2003 .

[40]  W. G. Cochran.  Some Methods for Strengthening the Common χ 2 Tests , 1954 .

[41]  A. Goldberger,et al.  The Treatment of Unobservable Variables in Path Analysis , 1971 .

[42]  Andrew S. Fullerton,et al.  Generational Conflict Or Methodological Artifact?Reconsidering the Relationship between Age and Policy Attitudes in the U.S., 1984–2008 , 2010 .

[43]  A. Cameron,et al.  Microeconometrics: Methods and Applications , 2005 .

[44]  C. I. Bliss,et al.  THE METHOD OF PROBITS. , 1934, Science.

[45]  P. McCullagh,et al.  Generalized Linear Models , 1972, Predictive Analytics.

[46]  J. Anderson.  Regression and Ordered Categorical Variables , 1984 .