Measuring the effects of computer-managed instruction (CMI) on the teaching of and student writing about literature involves more than having students write and then evaluating their performance. Measurement is made difficult by the fact that the computer technology used in instruction is in a state of flux. Variation of computer technology, whether in the form of the operating system, platform, or instructional software, is profitable for its designers. The use of computers in literature instruction can mean: (1) greater creativity, because instruction is freed from the two-dimensional text; (2) more interconnectivity to the text, as a result of greater accessibility; (3) closer analysis of the text, as the mechanics of textual analysis are simpler; and (4) greater fluency in the manipulation of symbols. Many different types of assessment can be applied over a period of years. Teachers can use "process logs" to track student interaction with literature and the new technology. The "blind" reading of student essays can be replaced by a consensual assessment model, by which multiple scorers/readers assess a given essay simultaneously and anonymously. Among the dangers of CMI are that the technology will overshadow the writing and that the system will aggravate the inability of many students to concentrate. (SG)