Meta-Analysis of Experimental Research Based on the Dunn and Dunn Model 6

ABSTRACT : The author performed a quantitative synthesis of experimental research conducted between 1980 and 2000, in which the Dunn and Dunn Learning-Style Model (R. Dunn & K. Dunn, 1993, 1999) was used. Of the 695 citations elicited by the database and reference section searches, 76 original research investigations met the established inclusion criteria. The 7,196 participants provided 168 individual effect sizes. The mean effect-size values calculated and interpreted through this meta-analysis provided evidence for increased achievement and improved attitudes when responsive instruction was available for diagnosed learning-style preferences. Three indicators rejected homogeneity for achievement and attitude effect sizes. Mean effect sizes for achievement and attitude were variable enough to be described as heterogeneous. Therefore, the author searched for variables that moderated the effect sizes; 6 were found. The author compared this investigation and a previous meta-analysis conducted by M. Sullivan (1993) and reported in The Journal of Educational Research (R. Dunn, S. A. Griggs, J. Olson, B. Gorman, & M. Beasley, 1995) and in the National Forum of Applied Educational Research Journal (M. Sullivan, 1996-1997). Mean effect-size results for achievement from the present and previous meta-analyses were consistent. The author suggested that, on average, learning-styles responsive instruction increased the achievement or improved the attitudes toward learning, or both, of all students. Although several moderating variables influenced the outcome, results overwhelmingly supported the position that matching students' learning-style preferences with complementary instruction improved academic achievement and student attitudes toward learning. The Dunn and Dunn model had a robust moderate to large effect that was practically and educationally significant. 56

5

5