Youth activity involvement has been operationalized and analyzed using a wide range of approaches. Researchers face the challenges of distinguishing between the effects of involvement versus noninvolvement and intensity of involvement in a particular activity, accounting simultaneously for cumulative effects of involvement, and addressing multiple unique effects of individual activities. In the present work, we review and illustrate the conceptual and empirical implications of these issues using data from a study of activity involvement and successful development in early adolescence (N=537; M age=11.56, 52% female). An integrative solution is introduced based on a latent composite variable (LCV) model (Bollen & Lennox, 1991), which can be used to address all three issues simultaneously. Using this approach, we show that of the aggregate indices examined, breadth of involvement was uniquely and positively associated with multiple indices of successful development. Of the individual activities, a dichotomous score and residual frequency rating for involvement in out‐of‐school clubs were both uniquely associated with less positive development indicators. We concluded that an LCV approach provides a novel method for addressing several fundamental operational and analytic issues facing researchers who investigate youth activity involvement as a context for positive development.
%G %I John Wiley and Sons %@ 1050-8392 %U http://dx.doi.org/10.1111/j.1532-7795.2010.00652.x