The effects of trait label and order of presentation on observers' ratings of performance anxiety

In two experiments, the effects of different stimulus presentation orders and trait labels on observers' ratings of performance anxiety were examined. In Experiment 1, 60 college students viewed, in one session, ten 3-min videotapes of a man speaking while displaying a variety of the behavioral manifestations of anxiety. The tapes varied linearly in the percentage of time during which anxious behaviors were emitted. Anxiety ratings were made along either a “calm” or an “anxious” dimension. Subjects were randomly assigned to one of three orders of presentation conditions (with the frequencies of anxious behaviors ascending, descending, or randomized across tapes) and one of two trait-label conditions, calm or anxious. In Experiment 2, 60 additional subjects viewed these same tapes and a contrasting series designed to display a consistent rate of anxiety behaviors. These tapes were viewed in 10 sessions over a 2- to 5-week period. Results replicated prior studies showing that the order of presentation can significantly affect observers' ratings. Additionally, the results suggested that, under certain conditions, subjects may not perceive a conceptually dichotomous variable (e.g., calm vs. anxious) in the same way. These results are discussed in terms of the complexity of observer behavior and potential problems in comparing outcome studies that use dichotomous traits.