Substantive Variables Research Articles

Second Order Partials as a Means to Control for Diffusion

This paper supports Loftin's conclusion that first order partials as cur rently used in linked pair tests are inappropriate. It points out that for each cross-cultural correlation there are not one but at least two control traits: namely one constructed from each substantive variable. This sug gests that not first but second order partials are appropriate diffusion control measures. The paper applies this suggestion to Loftin's hypothetical data and discusses the results.

Ethnographer Bias in Cross-Cultural Research: An Empirical Study

We direct attention in this paper to the problem of ethnographer bias (i.e. systematic errors occurring in the ethnographic reporting process) in cross-cultural research, and therefore in ethnographic fieldwork itself. Using multiple regression and other multivariate statistics, we assess the influence of ethnographer bias on the correlation between traits in the cross-cultural survey component of Rohner's Rejection-Acceptance Project (RAP). These procedures suggest a systematic ethnographer error, "the bias of romanticism," in anthropological research. Overall, however, the relationships among substantive variables in this research cannot be ex plained by this bias or by other forms of systematic error plaguing cross- cultural research. Thus in the absence of a successful competing theory, we conclude that all but one of the universal causal-functional relation ships postulated in Rohner's theory are validated.

Premarital Sexual Intercourse: A Substantive Test of the Contingent Consistency Model

A theoretical model of Contingent Consistency relative to the attitudes-behavior dilemma was constructed and substantively tested using premarital intercourse as a dependent variable. Using Coleman 's multivariate analysis technique it was found that general attitudes (religiosity on three dimensions) accounted for substantially higher proportions of variation for males than females. Personal attitudinal permissiveness (a behavior-specific attitude) accounted for significant proportions of variation regardless of sex. Using Kemper's typology from reference group theory it was found that perceived norms of permissiveness in the respondents' comparison reference groups were of little influence on sexual behavior for either sex while perceived normative reference group standards were of substantial influence for males and of little influence for females. Overall, significant interaction effects were so rare as to suggest that these substantive variables operate additively rather than in interaction. The average random shocks value was about 12 per cent for both males and females indicating that substantial amounts of variation in the dependent variable were accounted for by these variables. In other words, there was considerable evidence to support an additive Contingent Consistency model.