ABSTRACT: This paper describes action research involving the preparation of a test of English writing ability to be included in a placement examination. The examination, which included tests of reading, grammar, vocabulary, and error recognition in addition to writing, needed to be limited to one hour to fit the examination day schedule. Based on the observation of test takers in previous years--primarily variation in reading speed--it was concluded that at least 45 minutes would be needed for the non-writing portions of the test. Thus, 15 minutes was allocated for writing. Another factor was that the writing test had to be short enough to be scored quickly by a limited number of scorers. With these points in mind, the research questions became the following: (1) Can a 15-minute test of writing ability tell us enough about the students’ ability to distinguish between high, medium, and low ability writers? (2) Can such a short test provide information about writing ability beyond sentence level grammar and syntax? That is, can it indicate, for example, coherence, unity, and organizational competence? To answer, these and related questions, a pilot study was conducted with four groups of Japanese learners of English in Japan: two groups of second-year university students (English majors and non-majors), and one group each of first- and second year junior college students (English majors). It was predicted that there would be significant differences between first- and second-year English major students, between junior college and university students, and between English major and non-English major students. The paper discusses the complexity of the findings and the factors that contributed to the decision to modify the time limit for the writing test.