The report proposes to complete the validation and refinement of a new domain referenced testing technology designed to assess literal comprehension ability in students in grades 1-12. The domain referenced measures in this technology, along with other more traditional measures of reading comprehension, literal and non-literal, are subsequently intended to be used in part in large scale studies of productivity in school reading programs. To date, studies of productivity in reading instruction have had little influence on educational decision-making due to serious methodological problems, one of the major problems being the lack of adequate measures of program output. The report further proposes to solve a number of important instructional management problems created by the use of the inadequate information available from traditional measures of reading comprehension. The new domain referenced measures of reading comprehension will have an improved basis for scaling students on comprehension ability, and abilty scores from this scale will be referenced to an additional scale defining an individual or group's ability to read in several domains of written discourse. These scaling features will allow for the assignment of students to specific levels of reading materials in specific instructional or content domains, a procedure not possible with existing measures of reading comprehension. (Author)