The goal of the research project was to assess the effect of a written intervention on the state anxiety of new mathematics teachers. Twenty-eight beginning mathematics teachers in Washington, Oregon, and California were randomly assigned to treatment or control groups. Treatment was a survival guide, Green Broke, carefully designed to meet the needs of beginning mathematics teachers. A Delphi panel determined topics for the guide, beginning with suggestions from the literature. The panel included mathematics teachers in grades 7 to 12 from all three Pacific states, and both urban and rural areas. School administration and college mathematics education were also represented on the panel. A subset of the panel and a writer edited the guide. Trait anxiety of subjects was determined prior to the 1989-90 school year using the trait anxiety portion of Spielberger's State-Trait Anxiety Inventory. The state anxiety inventory was administered at that same time and again three times during the school year. All subjects also completed a teacher report twice during the year regarding their teaching experiences. In addition, the treatment group was asked to complete two questionnaires regarding the effectiveness of Green Broke. Subjects perceived that they were part of two studies, one involving the guide and one involving anxiety inventories. Analysis of covariance for repeated measures was used to assess differences between groups in state anxiety, using trait anxiety as covariate. Level of significance was set at .05. There was no significant main effect, but a significant interaction effect was found. The nearly linear relationship between state and trait anxiety found in the control group disappeared for the group having the guide. Mini case studies, using responses to the teacher reports, were done for seven subjects, including those with very high state anxiety. In addition, a subgroup of the treatment group, for which a linear relationship between state and trait anxiety did exist, was examined for common characteristics. Locus of control was hypothesized to be a variable that would differentiate this group. Other suggestions for further study include use of audio-visual forms of the survival guide, extension to other subject areas, and replication of the study. / Graduation date: 1991