Abstract:

The aim of this study was to validate an instrument for assessing motivation to learn science among Namibian senior secondary science students. A five-point Likert scale students’ motivation towards science learning (SMTSL) questionnaire developed by Tuan, Chin & Shieh (2005) was adapted for this study. Five out of the original six factors of motivation to learn science namely, self-efficacy, active learning strategies, science learning value, achievement goal and learning environment stimulation were chosen for this study. The five factors of motivation to learn science, viewed in terms of social cognitive theory of human learning served as the theoretical framework for this study. The paper-and-pencil instruments were used and a total sample of 755 students (male = 403; 53% and female = 352; 47%) from four senior secondary schools in two regions (Omusati and Ohangwena) of Namibia participated in the study. The mean age of students was 18.3 and standard deviation of 1.32. The original SMTSL was administered to junior high school students in Taiwan. The reliability (Cronbach’s alpha) of the whole instrument was 0.79 (n=19). The reliability of the factors ranged from 0.66 to 0.77. Exploratory factor analysis (n= 403) using principal components extraction with varimax rotation revealed an interpretable factor structure and the factor solution accounted for 56.1% of the total variance. The measurement model was validated by means of confirmatory factor analysis (n= 352). The results showed that the model had adequate statistical fit for the data, with the following fit indices: X2/df = 1.54, RMSEA = 0.039, SRMR = 0.047, TLI = 0.94 and CFI = 0.95. Convergent validity was evaluated using average variance extracted (AVE) and composite reliability (CR) which was found to be adequate. The AVE values for the five factors ranged from 0.32 to 0.47. The composite reliability values ranged from 0.63 to 0.78. Discriminant validity was assessed by comparing the square root of the AVE with the correlation of latent factors augmented with the comparison of the maximum shared variance (MSV) with the AVE values and both met the minimum recommended thresholds. The measurement model showed acceptable fit to the data with good fit statistics such as X2/df, RMSEA, SRMR, TLI and CFI. These findings indicate that the adapted instrument has adequate construct validity and internal consistency (reliability). Moreover, the findings suggest that the adapted instrument is valid for assessing Namibian senior secondary science students’ motivation to learn science.

Keywords:

Motivation to learn science, validitation, science students, EFA, CFA.