Review by Choice Review
Lieberson labels the paradigm underlying mainstream sociological research as essentially ``undoable'' given the domain of sociology, i.e., complex, historically embedded collectivities, institutions, and societies constituted through intersubjective symbolic processes. Although sociologists prefer to employ the experimental method of the ``hard'' sciences, this generally proves impossible. Researchers must use a variety of techniques and statistical methods to approximate the experimental model. Known as the quasi-experimental method, this model uses randomization, control of extraneous and intervening variables, analysis of variance, and causal modeling to explain relationships between independent and dependent variables that would have occurred in an experimental context. Lieberson argues that quasi-experimentation falls prey to various problems: unmeasured selectivity in control variables, contamination by independent variables, failure to distinguish between symmetrical and asymmetrical causality, and the proclivity to explain variance in the dependent variable in large data sets without explaining fundamental causal relationships between variables. Lieberson does little to reconstruct a workable paradigm; he demands that researchers clarify what can be explained, what cannot be done, and the errors and limitations of contemporary empirical sociology. Upper-division undergraduates and above.-J.H. Rubin, Saint Joseph College
Copyright American Library Association, used with permission.
Review by Choice Review
