Data generated by the method of pair-comparisons can be analyzed by the least squares technique. A system of normal equations is derived by which the original objects are scaled. Among the advantages of this method of scaling is no requirement that all object pairs be compared. Several examples are provided demonstrating the application of the least squares equations.
Get full access to this article
View all access options for this article.
References
1.
Bechtel, G. G.The multivariate analysis of variance and multi-attribute scaling. Oregon Institute Research Bulletin, 1968, 8(2).
2.
Bechtel, G. G.A covariance analysis of multiple paired comparisons. Psychometrika, 1971, 36, 35-55.
3.
Greenberg, M. G.A method of successive cumulations for the scaling of pair-comparisons preference judgments. Psychometrika, 1965, 36, 441-448.
4.
Guilford, J. P.Psychometric methods. New York: McGraw-Hill1954, Chapter 7.
5.
McCormick, E. J. and Bachus, J. A.Paired comparison ratings: I. The effect of ratings of reduction in the number of pairs. Journal of Applied Psychology , 1952, 36, 123-127.
