| The purpose of this study was to examine how selected linguistic components (including consistency of relational terms and extraneous information) impact performance at each stage of mathematical word problem solving (comprehension, equation construction, and computation accuracy) among students with different levels of computation achievement.;The performance of two groups of students was examined: individuals who scored in the upper quartile (Q4 subjects) and lower quartile (Q1 subjects) among their same-grade peers on a standardized measure of computation achievement. Subjects completed three experimental tasks. Both the Word Problem Solving Task and the Word Problem Comprehension Task measure word problem solving; however, the latter evaluates the understanding of the stages associated with word problem solving through comprehension questions. The Computation Task measures the performance on addition and subtraction computations of varying levels of difficulty.;In the area of word problem comprehension, the results demonstrated that inconsistent relational terms and extraneous information cause difficulties in word problem solving for both successful and less successful achievers. Further, Q1 subjects have increased difficulty with word problems that include these factors.;In the area of equation construction, the results indicated that if the proper equation is not developed, it is highly unlikely that the final computation will be accurate. Further, Q1 subjects have significantly more difficulty than Q4 subjects in determining a proper equation, as problems become increasingly more complex.;In the area of computation, Q1 subjects performed significantly better on addition problems than subtraction problems, while Q4 subjects typically did not reveal this difference. Further, while subjects in general were significantly more accurate on problems that did not require regrouping as compared to problems that did require regrouping, Q1 subjects were significantly more prone to this pattern of difficulty than Q4 subjects.;Although the groups were formed based on computation achievement, Q4 subjects demonstrated significantly higher proficiency in word problem solving than Q1 subjects. Further, as problem complexity increased, differences between the groups increased. Therefore, instructional strategies for subjects with low computation achievement should include instruction in which computations are both within and outside of the context of word problems. |
