The properties of Quade's "index of matched correlation"are explored in the
context of causal analysis. Quade's approach is tested on data with specified
characteristics generated through a computer simulation. Although nonparametric partial correlation must be used with caution, it supplies an easily interpreted and applied method for analyzing categorized data.
Get full access to this article
View all access options for this article.
References
1.
Blalock, H.M. (1970) "A causal approach to nonrandom measurement error."Amer. Pol. Sci. Rev.64 (December): 1099-1111.
2.
--- ( 1965) "Some implications of random measurement error for causal inference."Amer. J. of Sociology71 (July): 39-42.
3.
--- ( 1960) Social StatisticsNew York: McGraw-Hill.
4.
Davis, J.A. (1967) "A partial coefficient for Goodman and Kruskal's gamma."J. of Amer. Stat. Assn.62 (March): 189-193.
5.
Hyman, H. (1954) Survey Design and AnalysisNew York: Free Press.
6.
Kendall, M.G. (1955) Rank Correlation MethodsLondon : Charles Griffin.
7.
Labovitz, S. (1970) "The assignment of numbers to rank order categories."Amer. Soc. Rev.35 (June): 515-525.
8.
--- ( 1967) "Some observations on measurement and statistics."Social Forces46 (December): 151-160.
9.
Lazarsfeld, P.F. (1955) "Interpretation of statistical relations as a research operation," pp. 115-125 in P. F. Lazarsfeld and M. Rosenberg (eds.) The Language of Social Research.New York: Free Press.
10.
Quade, D. (1974) "Nonparametric partial correlation," in H. M. Blalock, Jr. (ed.) Measurement in the Social Sciences.New York: Aldine.
