Showing papers on "Ordinal regression published in 1978"
TL;DR: Boyle et al. as discussed by the authors proposed a multivariate analysis of Ordinal Variables for path analysis and found that it is possible to reduce error interpretation for the squared generalized multiple, partial, and multiple partial correlation coefficients and their special cases.
Abstract: Boyle, Richard P. 1970. \"Path Analysis and Ordinal Data.\" American Journal of Sociology 75 (January): 461-80. Coleman, James S. 1964. Introduction to Mathematical Sociology. New York: Free Press. Hawkes, Roland J. 1971. \"The Multivariate Analysis of Ordinal Measures.\" American Journal of Sociology 76 (March): 908-26. Land, Kenneth C. 1969. \"Principles of Path Analysis.\" Pp. 3-37 in Sociology Methodology, 1969, edited by Edgar F. Borgatta. San Francisco: Jossey-Bass. Ploch, Donald R. 1974. \"Ordinal Measures of Association and the General Linear Model.\" Pp. 343-98 in Measurement in the Social Sciences, edited by H. M. Blalock, Jr. Chicago: Aldine. Reynolds, H. T. 1971. Making Causal Inferences with Ordinal Data. Chapel Hill, N.C.: Institute for Reaserch in Social Science. Smith, Robert B. 1974. \"Continuities in Ordinal Path Analysis.\" Social Forces 53 (December): 200-229. . 1977a. \"Proportional Reduction in Error Interpretations for Daniel's r2 and Its Special Cases.\" Social Forces 55 (June): 1067-75. . 1977b. \"Proportional Reduction in Error Interpretations for the Squared Generalized Multiple, Partial, and Multiple-Partial Correlation Coefficients and Their Special Cases.\" Social Forces 56 (December): 688-702. -. Forthcoming. \"The Use of Spearman's Pb in Causal Models of Ordinal Data.\" Somers, Robert H. 1962. \"A New Asymmetric Measure of Association for Ordinal Variables.\" American Sociological Review 27 (December): 799-811.
27 citations
TL;DR: A large number of ordinal measures of association are available today for use by sociologists as mentioned in this paper, however, while these measures may be similar in terms of a common property (e.g. expression of bivariate monotonic relationship), they may differ in other respects, such as the treatment of ties.
Abstract: A large number of ordinal measures of association are available today for use by sociologists. However, while these measures may be similar in terms of a common property (e.g. expression of bivariate monotonic relationship), they may differ in other respects (e.g. the treatment of ties). Moreover, the strength of the monotonic association, as indicated by the magnitude of the measure which is used in the analysis, may vary substantially depending on the choice of the statistic to summarize the relationship. The sociologist, then, faces the problem as to which measures to choose so that his hypotheses are properly tested. An additional problem derives from the tendency of some sociologists to resort to parametric statistics for data which are at most ordinal. On the one hand, there is an increasing and accumulating evidence that often the assumption, that the data can be analysed as if they were measured on an interval scale, is an acceptable assumption (e.g. Labovitz, 1970; Bohrnstedt and Carter, 197 l)yk .the other hand, however, there are serious objections to such an approach (e.g. Wilson, 1971) and also some limitations on the use of parametric statistics for ordinal data, admitted even by the proponents of such approach (e.g. Labovitz, 197 1). The present paper deals with both the question as to which ordinal measure to use for various purposes and also the substitutability of a particular parametric statistic, Pearson’s product-moment correlation, for these ordinal measures of association.
7 citations
TL;DR: In this paper, two measures of individual dispersion are considered and their distributions derived in the null case, and the relation of these estimates to the variance of the individual propensity distribution is derived.
Abstract: For ordinal measurement the concept of an individual propensity distribution is developed. For any given individual the mean of this distribution is his true score, for which estimation procedures are discussed. Two measures of individual dispersion are considered and their distributions derived in the null case. These measures are shown to be counterparts at the individual level of Kendall's tau and Spearman's rho. Estimation of the two dispersion measures from sample data is investigated, and the relation of these estimates to the variance of the individual propensity distribution is derived.
5 citations
TL;DR: In this article, the authors examined scaling via conjoint measurement as an alternative to the usual multiple linear regression in the bootstrapping model and found that scaling is superior to MRL as a method of approximating past data.
4 citations
4 citations
TL;DR: In this paper, an equal-interval, pure variance measure of ordinal dispersion is proposed and compared to Leik's D. The measure is defined as the difference between the random variable's expected values above and below the median.
2 citations