The College Outcome Measures Project (COMP) was established to develop, validate, and implement measures of the knowledge and skills that undergraduate students are expected to acquire and that are important to effective adult functioning. The COMP Measurement Battery yields a "communicating" score based in part on a speaking assessment and a writing assessment, each composed of three role-playing tasks. A high content validity for the scales has been established due to input from communication experts at a diverse group of institutions, the use of standardized role-playing tasks rather than experience-based writing topics, the sensitivity of the assessment to the levels of proficiency normally acquired by undergraduate students, and the establishment of the importance to effective adult functioning of the skills defined by the COMP rating scales. Three studies of interrater reliability for the COMP measures were conducted. In the first study, the correlation between two independent ratings of 74 speaking samples was .83. In the second study, the correlation between two independent ratings of 61 writing samples was .70. In the final study, two raters who had been trained by a refined set of training materials achieved .89 correlation in rating 50 writing samples. The studies show that speaking and writing samples can be measured reliably. (GT)