Evaluation of success lab in terms of learning environment, attitudes toward mathematics and academic efficacy among high school algebra students

Abstract

I investigated the effectiveness of an Algebra 1 intervention program, Success Lab, in terms of learning environment and students’ attitudes toward mathematics and academic efficacy with 20 ninth grade classes (N = 313) in three central California high schools. A group of 56 students concurrently enrolled in Algebra 1 and Success Lab were matched and compared with 56 non-intervention students. Criteria of effectiveness consisted of learning environment scales from the What Is Happening In this Class? (WIHIC), attitude scales from the Attitudes Toward Mathematics Inventory (ATMI), and an academic-efficacy scale based on the Patterns of Adaptive Learning Scales (PALS).Factor analysis (principal axis factoring with varimax rotation and Kaiser normalization) showed that the three learning environment scales accounted for 57.61% of the total variation for the pretest data and 63.17% of the total variation for the posttest data. The Cronbach alpha reliability coefficient exceeded 0.80 for every learning environment scale for both the pretest and posttest. Factor analysis of data for the attitude and efficacy scales of the ATMI/PALS revealed that the Value scale accounted for 47.99% of the total variation for pretest data and 52.49% of the total variation for posttest data. Cronbach alpha reliability coefficients all exceeded 0.85 for the attitude and efficacy scales for both pretest and posttest data.A paired-samples t-test was used with each learning environment and attitude and efficacy scale to determine the statistical significance of differences between the intervention group and the non-intervention group. Posttest differences between the intervention group and the non-intervention group for the three learning environment scales were statistically significant even after a modified Bonferroni correction was applied. Also the between-group differences were sizeable in magnitude for all three learning environment scales (with effect sizes of over half a standard deviation indicating a medium effect).However, posttest differences between the intervention group and the nonintervention group were statistically nonsignificant for the three attitude and efficacy scales (after application of a Bonferonni correction) and were associated with relatively small effect sizes ranging from 0.18 to 0.34 standard deviations. Nevertheless, students participating in the intervention class, Success Lab, had somewhat more positive attitudes towards mathematics and academic efficacy than the non-intervention group.For both pretest and posttest data, simple correlation analysis revealed positive and statistically significant correlations between each student attitude and efficacy scale and each of the three learning environment scales. Multiple regression analysis showed that the multiple correlation between the set of three learning environment scales and each attitude and efficacy scale separately was statistically significant for both the pretest and posttest data. Standardized regression coefficients revealed that Teacher Support and Task Orientation were significant independent predictors of Value for the pretest data and Involvement and Task Orientation were significant independent predictors of Value for the posttest data. All three learning environment scales were significant independent predictors of Enjoyment for both the pretest and posttest data, but only for pretest data for Academic Efficacy. Posttest analysis showed that Involvement and Task Orientation were significant independent predictors of Academic Efficacy.