Abstract
The factors responsible for sex related effects are examined by means of an analysis of methods frequently used to study them. Field observations, field experiments, and laboratory techniques used in the investigation of sex and gender are discussed. It is suggested that the distinction between reactive and nonobtrusive measurement may be particularly useful in predicting when sex related differences may be found. It is also suggested that there is no one universal sexual reality. Sex related effects may be constructed by the social cognitions of the participants in a study as well as by the conceptual assumptions underlying the methods themselves.

This publication has 16 references indexed in Scilit: