Abstract:
Poor performance in mathematics word problems continue to cause concern to educators and the general Kenyan public. Although some interventions have been undertaken, little has changed. It was against that background that the need for this study was envisaged. The study was undertaken in kakamega municipality. Using stratified random sampling; six schools were selected from public and private schools. Sixteen grade 4, 5 and 6 mathematics teachers from sampled schools were interviewed. The teachers selected were all trained teachers who had taught for more than three years. Simple random sampling was used to select 30 percent of the students in each grade (primary 4, 5 and 6) from the sampled schools. They were required to fill 22 close- ended questions in a questionnaire. Expost-facto research design was adopted. The Study examined four independent variables namely: Semantic structure, grade level, type of school and gender. On the other hand the dependent variable was: ability to identify correct operation. Data obtained was analyzed using both descriptive and inferential statistics. The findings of the study indicated that students related better to additive word problems than to multiplicative word problems and that they solved problems that did not involve relational statements better. The study also showed that the student‟s ability to solve word problems increases with grade level, but the relative difficulty of each problem type is grade independent. Thirdly, the study found that the type of school attended by the students significantly influence the ability to identify the correct operation required to solve arithmetic word problems. Finally, the study revealed that the students‟ gender does not significantly influence the ability to identify the correct operation required to solve arithmetic word problems. From the findings it was recommended that mathematics instructors should help students to comprehend the relations embedded in arithmetic word problem and hence deduce the solution operation that correspond to their semantic structure. Besides, mathematics instructors and teachers should formulate arithmetic word problems representing the whole range of semantic structure. Thirdly, Kenya institute of education (K.I.E) should make revisions in mathematics curriculum based on a continuum of difficulty as predicted by semantic structures and other task variables. Fourthly ministry of education should mount regular in-service training (INSET) to update teachers on new instructional techniques and philosophy of arithmetic word problems of various semantic structures.
