A Reliability Assessment of Participant Observational Measures of Leader Behavior in Natural Settings.

This study makes a reliability assessment of 88 trained participant observers who measured the behavior of 120 target leaders in 5 diverse organizational settings. Eight trained outside observers were used as agreement checks. Drawing from three methods of calculation, the interrater agreement was quite impressive. Other analysis techniques employed in the study support the value of the training given to the observers. The overall conclusion of the study is that, especially in light of the current dissatisfaction, observation may be an effective measurement alternative. (Author)