Students and teachers generally acknowledge that word problems are difficult to solve. The National Assessment of Educational Progress mathematics assessments of 1973 and 1978 revealed that "students typically did better on computational skill exercises than on word problems using the same numbers" (NAEP, 1979, p. 14). Unsuccessful encounters with word problems early in their elementary school experience may well have lasting negative effects on the mathematical development of many students. In fact, initial work with word problems could be the source of mathematics anxiety for many people (Tobias, 1978, p. 129). Recently public concern has deepened for the teaching of basic skills (the so-called back-to-basics movement). The National Council of Supervisors of Mathematics (1978) takes the position that a too-narrow interpretation of basic skills tends to de-emphasize the importance of teaching for mathematical understanding. Word problems provide one avenue for the development of understanding while at the same time providing the opportunity to develop computational skills. Basic research in the teaching of word problems could restore some of the emphasis on understanding by finding better ways to combine computational practice with application to real life or simulated problems. Word problems can be thought of as applications requiring the use of reading and computational skills. Reading and computation are widely recognized as factors important to success in solving word problems (Balow, 1964; Cohen & Stover, 1981; Glennon & Callahan, 1968, pp. 94-