One hundred seventy-nine students enrolled in a first-year college level composition course were subjects in a study of the effects of four levels of audience specification on writing anxiety, performance, and sensitivity to audience. Subjects completed the Writing Apprehension Test, which determined levels of writing apprehension, and then responded to one of four writing tasks that provided different amounts of audience information: assignment 1 made no reference to audience; assignment 2 included a statement that reminded subjects that they were to address an audience but did not specify the audience; assignment 3 specified the receptiveness, level of comprehension, and prior knowledge of the audience; and assignment 4 specified the characteristics given in assignment 3 and indicated that a response would be received from the audience. Subjects also completed the Situational Writing Anxiety Test and an audience questionnaire. Among the findings were the following: (1) high apprehensives who received assignment 2 were less anxious than those who received assignment 4, (2) moderately apprehensive writers produced better essays in response to assignment 3 than they did to any of the other assignments, and (3) assignments 2 and 4 were more likely to sensitize writers to audience and to influence them to consider a variety of readers. (HOD)