Suchergebnisse

Measurement errors in predictor variables bias multiple regression coefficients, and the problem is particularly complex with nonlinear models containing product variates. This article presents formulas and computing algorithms for correcting data before regression analyses in order to eliminate biases due to measurement errors in models with second- and third-order product variates (e.g., YZ and XYZ). The method requires independent estimates of error variances and covariances for the first-order variates (e.g., X, Y, and Z). An analysis of empirical data illustrates the approach. Errors in estimates are examined through Monte Carlo analyses.

A model is presented in which attitudes have effects on behaviors, behaviors are compared against norms to define deviance, and deviance has a return, corrective effect on attitudes through mechanisms of social control. Social interactions link these individual processes to form a single interpersonal system within a group. It is demonstrated that in a system of this kind, an attitude-behavior correlation is a function of many different social psychological effects, and its value conceivably could be positive, zero, or even negative. Results of a preliminary empirical study using two-stage least squares to estimate system parameters show that the proposed system is viable with some modifications. In particular, empirical evidence in support of an attitude-behavior effect is found.

In a variety of problems, it is desirable to have a single coefficient summarize the causal effects of a set of variables when other variables are controlled. The "sheaf coefficient" presented here does this and can be employed meaningfully in the context of path analysis models.

Against purely rationalist explanations of social movement participation -- particularly the resource mobilization perspective -- explored here is the role of emotional factors, specifically, shame & loneliness, & their transformation into pride & solidarity as a result of participation. Social aspects of individual emotional experiences are discussed via a review of previous theory & research. Shame & pride are discussed as "social emotions" that can be induced by particular behaviors. Ways that social movements create & utilize "emotional capital" to mobilize participants & motivate them to engage in collective action are described. A model is developed to depict the processes of emotion transformation involved in the identity politics of social movements; the gay pride movement is offered as illustration. 1 Figure, 49 References. K. Hyatt Stewart