Abstract

• The present study provided an examination of the effect of personal characteristics of students (Academic Self-efficacy, Academic Motivation) and their home environment (Parental Involvement, Socio-Economic Status) on academic achievement of students at Basic Education High Schools in Myanmar. A total of 1309 Grade-11 students from 15 government schools and 3 private schools in 6 Regions and States. Academic self-efficacy scale and academic motivation scale which had Cronbach’s alpha of 0.929 and 0.839 were used to assess student personal factors and parental involvement questionnaire which had Cronbach’s alpha of 0.889 and socio-economic questionnaire were used to assess home environment. Before executing discriminant analysis, data were firstly screened for outliers and then, assumptions for discriminant analysis were checked. Group statistics showed that there were mean differences among all the student and parent factors under the categories of high, average, and low achieving groups. ANOVA table revealed all the student and parent factors were reliable discriminators of the high, average, and low achieving groups. The conducted discriminant analysis was a three-group analysis, and therefore, two discriminant functions were obtained. Discriminant function 1 was statistically significant with Wilks’ Lambda = .42, chi-square = 398.86 at p < .001 while Function 2 was statistically significant with Wilks’ Lambda = .91, chi-square = 13.64 at p < .001. Standardized canonical discriminant function coefficients and structure matrix revealed that students’ academic achievement was mainly determined by socio-economic status and parental involvement and contributed 39.69 %. Academic self-efficacy and academic motivation contributed additional 8.41% to students’ academic achievement. The discriminant model classified correctly 77.3 % of students as high achievers, 10.8% of students as average achievers, and 74.7% of students as low achievers respectively. A total of 60.9 % of teachers were correctly classified into three groups with 60.2% of cross-validated grouped cases were correctly classified.