This paper describes a study that evaluated the reliability of a recently developed rating form designed to assess air traffic controller performance. Six supervisors from different radar approach control facilities nationwide viewed 20 video tapes of controllers working traffic from a previously recorded simulation study. The observer/raters used a new evaluation form that consisted of 24 different rating scales measuring specific areas of controller performance. An important part of this study was observer training. The training consisted of practice rating sessions followed by group discussions. In discussion, observers established mutual evaluation criteria for each performance area. Inter-rater reliability was assessed using intraclass correlations, and intra-rater reliability was assessed using Pearson product-moment correlations on repeated video tapes. In general, the reliability of the form was quite good, however, a few rating scales were much less reliable than the others. Reasons for the differences in rating scale reliability are discussed.
May I Interrupt? The effect of SPAM Probe Questions on Air Traffic Controller Performance
