Year: 1992
Type: article
Abstract: Many researchers are interested in providing causal interpretations of the statistical relationships they find in their one-shot health survey data. However, survey studies do not exert controls over independent variables while observing the dependent variable, as in experimental studies. Nor do they, as in longitudinal studies, define time order of variables (i.e., one event may occur prior to another). Survey studies are usually unable to isolate independent and dependent variables, making the causal interpretation of statistical relationships untenable. Nevertheless, these limitations should not discourage researchers from attempting the causal relationships by using a causal modeling approach. Causal modeling is a method that determines predictor variables that may have the potential to influence the criterion variable (the effects), and then analyzes the direct and indirect contributions made by each predictor variable to the effects. The two basic steps in causal modeling are: causal theorizing and causal analysis.
Author: J. Mark Eddy
Language: English
Subfield: Sociology and Political Science
Field: Social Sciences
Domain: Social Sciences
Open Access status: green
OpenAlex ID: W1916785252