Two studies were conducted to investigate the influence of self-monitoring ability and gender on leader emergence. In Study I, groups composed of a male high self-monitor, male low self-monitor, female high self-monitor, andfemale low self-monitor worked on a salary allocation task. At the end of the task, subjects completed questionnaires that asked them to select one group member as their leader. The amount of influence each group member exerted during the discussion was also assessed. Analyses indicated that high self-monitors emerged as leaders more frequently than did low self-monitors and men emerged as leaders more frequently than did women. Study II replicated these results in 9 volunteer organizations. Implications of the findings for managerial practice and future research are discussed.

The present research examined the effects of employee age on managers' attributions and actions toward subordinates. The major prediction was that managers' attributions would mediate the effects of age on managerial actions. One-hundred and twenty-four undergraduate management students participated in a laboratory experiment in which subordinate age (30 versus 60 years) and performance (high versus low) were manipulated. A laboratory study was conducted so that subordinate credentials and performance levels could be controlled. Analyses indicated that poor performance by older subordinates was attributed more to stable factors than similar performance by younger subordinates. In addition, job simplification was rated as more appropriate and training was recommended as less appropriate for older subordinates. A series of hierarchical regression analyses indicated that the differences in managerial actions were a function of subjects' attributional patterns. Implications of the findings for practice and future research are discussed.

The present study compared rating accuracy of logical and illogical raters on mixed standard scales. A laboratory methodology was used so that the true level of ratee performance would be available and Cronbach's (1955) accuracy components could serve as dependent measures. Analyses indicated that illogical raters were less accurate than were logical raters, but only on differential evaluation. Implications for personnel practice and future research are discussed.