The purpose of this study was to develop a testing method for the assessment of various types of writing at the elementary school level that would meet acceptable standards for educational measurement instruments as well as standards of utility and feasibility within a given educational system. The study was conducted within the framework of an Israeli national survey of sixth graders' writing performance. The direct assessment method chosen involved a variety of writing tasks, including school newsletter announcements, letters, and explanatory essays. The method distinguished among practical, expressive, and school writing. Five writing tasks were developed and field tested for each type of writing. The pilot study also included classroom observation and interviews with teachers and students. Scoring guides were developed for each writing task. The tasks were administered to a nationally representative sample of 2,590 sixth graders enrolled in a total of 57 schools. Approximately 800 responses were received for each writing task. A team of 12 lay scorers was trained to conduct the scoring. Data are now being analyzed to determine interrelationships between writing types and scoring methods and between test scores and other variables. Preliminary results indicate that the method may have curricular validity but not instructional validity. (TJH)