Adjudicating Agency.

Agency is a core concept in the discipline of sociology yet it suffers from at least two weaknesses. First, debates over the nature of agency are curiously abstract and of only tangential importance for empirical researchers. Second, discussions about the nature of individual agency rarely draw on the substantive subfield most suited for study of the individual, social psychology. Drawing on a nationally representative, longitudinal data set, we propose an empirical model of agency that allows us to begin to adjudicate among various theoretical and empirical approaches to the topic and lays the groundwork for future work situating agency more fully within social contexts across the life course. Empirical work on agency tends either to focus on self-efficacy or planful competence as measures of the concept; we find that both constructs load strongly on a second-order latent construct of agency, and that future study of the topic profitably will include both aspects. We find that autonomy, as measured here, is significantly less important as a factor in agency. We conclude with suggestions for future directions of empirical inquiry that will serve to situate the concept more fully in sociological processes and determine the utility of the concept for life course research. [ABSTRACT FROM AUTHOR]