Conclusion
Coleman's linear model for attributes parallels the classical regression model for real variables. The aim is to obtain, from the data, a set of numbers which serve as measures of the effect of a set of characteristics on an attitude. The probability of the attitude is assumed to be a linear function of the parameters.
We have distinguished the case where the distribution of the population into the different classes defined by the characteristics is random (model A) from the case where this distribution is given (model B). We have shown that, for large samples, these two models have very similar properties.
Coleman's method of estimation, a least-squares method, leads to very simple solutions, and this even in the general case. It has, however, certain shortcomings: it uses only proportions, and not the magnitude of the classes, and does not lead to a test of goodness-of-fit. We know, however, the asymptotic distribution of the estimates from which we get confidence regions and tests for definite values of the parameters.
We propose two other methods, the maximum likelihood and the minimum X 21 , which both lead to unbiased estimates with asymptotic minimum variances, and to tests of hypotheses of the X 2 type. The min. X 21 is particularly well suited in a linear context.
Section II is devoted to the case where all variables are dichotomies, and contains a numerical example.
Section III is devoted to the general treatment of the model and demonstrates its properties from basic results in probability and statistic theory, which are recalled in the Appendix.
Similar content being viewed by others
References
J. S. Coleman, Introduction to Mathematical Sociology, New York, Collier-MacMillan, 1964.
W. G. Cochran, “Annals of the Mathematical Society”, XI (1940), pp. 335–47.
J. H. Curtiss, “Annals of the Mathematical Society”, XIV (1943), pp. 107–22.
G. V. Dyke, H. D. Patterson, “Biometrics”, 8 (1952), pp. 1–12.
A. E. Maxwell, Analysing Qualitative Data, New York, Wiley, 1961.
K. R. Gabriel, “JASA” 58, (1963), pp. 1133–57.
J. Neyman, Berkeley Symposium on Mathematical Statistics and Probability, (1949), pp. 239–73.
V. P. Bhapkar, “Annals of the Mathematical Society”, XXXII (1961), pp. 72–83.
L. A. Goodman, “Annals of the Mathematical Society”, XXXV (1964), pp. 716–25.
R. S. Rodger, “The British Journal of Mathematical and Statistical Psychology”, XXII (1969), pp. 29–48.
C. R. Rao, Linear Statistical Inference and Its Applications, New York, Wiley, 1965.
O. Reiersøl, Tests of Linear Hypothesis Concerning Binomial Experiments, Stockholm, Skandinavisk Aktuerietidskrift, 1954, pp. 38–59.
O. Reiersøl, Identifiability, Estimability, Pheno-restricting specifications, and Zero Lagrange Multipliers in the Analysis of Variance, Stockholm, Skandinavisk, Aktuarietidskrift, 1963, pp. 131–142.
H. Thell, On the Estimation of Relationships Involving Qualitative Variables, “dAmerican Journal of Sociology”, LXXVI (1970), p. 103. — Reference [7] clarifies the problem of estimating paramenters by the unweighted least squares method (paragraphs II, 1 and III, 5). — Reference [8] uses for Coleman's problem the logic transformation which insures that the stimates of the probabilities are always in the [0.1] interval. (This we cannot insure here).
Rights and permissions
About this article
Cite this article
Feldman, J. A study of Coleman's linear model for attributes. Qual Quant 4, 255–297 (1970). https://doi.org/10.1007/BF00199566
Issue Date:
DOI: https://doi.org/10.1007/BF00199566