An Optimal Scaling of Murdock's Theories of Illness Data—an Approach To the Problem of Interdependence

1 February 1988

journal article
research article
Published by SAGE Publications in Behavior Science Research

Vol. 22 (1-4) , 161-179
https://doi.org/10.1177/106939718802200112

Abstract

In Theories of Illness (1980) Murdock reported that theories of illness causation were not randomlydistributed with respect to world regions or language phyla. For example, witchcraft theories were prevalent among peoples of the Circum-Mediterranean and absent elsewhere. However, Murdock was unable to estimate the magnitude of either the regional or linguistic effect. In this paper Murdock's data are reanalyzed using the multivariate technique of optimal scaling, a method sensitive to interdependence among both cases and variables. The results of the optimal scaling are used to discuss the clustering of the theories of illness within world regions and language phyla.

Keywords

This publication has 8 references indexed in Scilit:

Cross-Cultural Surveys Today
Annual Review of Anthropology, 1987
Galton's Problem as network autocorrelation
American Ethnologist, 1984
Sexual Division of Labor in African Agriculture: A Network Autocorrelation Analysis
American Anthropologist, 1981
Analysis of Categorical Data
Published by University of Toronto Press Inc. (UTPress) ,1980
Theories of Illness
Published by JSTOR ,1980
World Distribution of Theories of Illness
Ethnology, 1978
Historical Inference from Cross‐Cultural Data: The Case of Dowry
Ethos, 1973
Standard Cross-Cultural Sample
Ethnology, 1969