Cognitive style and children's performance on measures of elementary science competencies

The purpose of this exploratory study was to determine the effect of Witkin's construct of cognitive style on children's performance on salient elementary science competencies. These competencies involved the ability to use science processes and the acquisition of specific attitudes. During the development of the study (see Appendix A), it was first necessary to determine the measurable objectives of the Elementary Science Study (E.S.S.). The Test of Science Processes was used to measure those E.S.S. objectives which pertained to science processes. In order to measure the attitudinal objective of the E.S.S. programme, the author developed four attitude scales, utilizing proper attitude measuring techniques. The four scales measured children's attitudes towards the following beliefs: children will feel that "Messing about in Science" is fun (Fun Scale); children will follow-up phenomena encountered during E.S.S. experiences (Pursue Scale); children will impose a structure on their play to find out more (Structure Scale); children will themselves initiate their own investigations (Independent Investigation Scale). The development of these scales is reported intact in Appendix B. Good reliability and factoral validity were established for these scales. It was hoped that the four attitude scales would prove to be useful tools for elementary science educators. A natural experiment in a small city school district was utilized to determine the effect of years of E.S.S. experience, the effect of Witkin's construct of cognitive style, and the interaction effect of years of experience with cognitive style -- on children's performance on measures of elementary science competencies. Utilizing a three by three factorial design, the test scores of 184 grade seven pupils were compared. The independent levelling variable used to determine cognitive style was based upon performance on the Children's Embedded Figures Test (C.E.F.T.). Years of E.S.S. instruction (one year, two years and three years) comprised the independent blocking variable. Groups were compared on fourteen dependent variables ( nine process variables and five attitudinal variables). Hotellings T² statistic was computed prior to analysis of variance in order to determine if the global group (C.E.F.T. score 0-15) would achieve significantly lower scores than the analytical group (C.E.F.T. score 20-25) on the sets of elementary science competencies (processes and attitudes). The predicted inferior performance of the global group was confirmed on the set of attitudinal dependent variables and on the set of dependent variables concerning processes. The predicted effect of superior performance of students who had received more E.S.S. experience than other students was not generally supported by the statistical tests. The predicted interaction effect was not generally significant either, although their appeared to be a trend which might indicate that the global group did less well when this group had more and more E.S.S. experience. Limitations of the cross-sectional design, however, made it difficult to come to any firm conclusions regarding the interaction effect and the effect of years of instruction. Analyses of variance confirmed the findings that the children with a more global cognitive style achieved significantly lower scores on elementary science competencies than children with more analytical cognitive styles. Based on these findings, the implications of the construct of cognitive style on elementary science education was discussed in terms of methodological reform and curricular reform. Finally, a plan for further research was proposed.Education, Faculty ofGraduat