The Measurement of a Multidimensional Learning Strategy in Upper Secondary School Students: Competing Model, and Predicting on Academic Achievement

Abstract

The purposes of this study were 1) to compare the efficiency of learning strategies measurement models consisted of (a) Weinstein and Palmer‘s Model, (b) Stevens and Tallent-Runnels‘s Model, (c) Cano‘s Model, and (d) Model developed by the researcher, and 2) to assess the prediction of measurement efficiency of students‘ learning strategies on academic achievement. The participants were 2,187 upper secondary school students from schools under the jurisdiction of the Office of the Basic Education Commission of Thailand. The instrument for data collection was a multidimensional learning strategies scale for upper secondary school students. In the results of this study, the model developed by the researcher was the most efficient model consisted of three dimensions: cognitive, affective, and skill strategies. This model was identified by 2=24.666 (df =17, p=.102), 2/df=1.451, GFI=.998, AGFI=.993, RMR=.008, RMSEA=.014, CFI=1.000, and AIC=100.666 (Saturated
AIC=110.000). The skill strategy, affective strategy, and cognitive strategy had significant positive effects on academic achievement. The standardized regression coefficients were .274, .241, and .227, respectively. Each
strategy accounted for 7.50, 5.80, and 5.10 percents of variance in the academic achievement. There was low error of prediction for .008, .007, and .010, respectively.