Options
Social correlates of English language proficiency of Primary Three pupils in Singapore
Loading...
Type
Thesis
Author
Teo, Eng Kee
Supervisor
Yeo, Robert
Abstract
This study set out to examine the relationship between learner variables and English Language proficiency.
The sample was made up of two classes of primary three pupils from each of ten schools. Categorisation of schools was based on the Primary School Leaving Examination results from 1982 to 1984. A questionnaire, an English test (made up of listening comprehension, dictation, sound discrimination, reading comprehension, composition and oral test), and the Raven's Standard Progressive Matrices, were group administered on two days. A pilot-test was conducted earlier in three primary schools.
Six hypotheses were drawn up and a correlational analysis carried out on the data.
Findings from the survey suggest that parental educational levels, language use at home and in school, extent of reading interest and general intellectual ability were significantly and positively correlated with English proficiency. A notable exception was private tuition which was found to be significantly but negatively correlated with English proficiency.
Additionally, the results of the correlational analysis also indicated that the father's educational level was significantly correlated with the mother's as well. Next, it was also revealed that there was a significant relationship between parental schooling and language use at home. Furthermore, parental schooling also correlated significantly with the children's reading interests.
For the final hypothesis, path analysis was used to test a hypothesised theoretical model to explain the causal relationships between the selected variables and English proficiency.
A multiple regression analysis was first employed to determine the contribution of the selected variables to the variance in English proficiency. The six selected variables, namely the father's and mother's educational levels, language use at home and in school, sex and general intellectual ability, collectively accounted for about 40% of the variance in English proficiency. General ability accounted for most of the variance, about 24%.
The relative strength of the influence of one variable on another is reflected by the beta coefficient obtained from the multiple regression analysis. The path coefficients for all variables were calculated. The most important contributing variable affecting English proficiency, according to the hypothesised model, was general intellectual ability. It had a direct effect of 0.486 on English proficiency. Next, mother's schooling had a direct effect of 0.310 and an indirect effect of 0.020. The third significant predictor variable, English language use at home, contributed a direct effect of 0.202. Father's schooling had a direct effect of 0.142 and an indirect effect of 0.086 on English proficiency. The English Language use in school and the sex differences were found to have no significant effects on English proficiency and were therefore deleted from the final theoretical path model.
In conclusion, these results suggest that there are significant learner variables affecting English proficiency. Teachers should be aware of the differences in the children's learning abilities and could incorporate suitable teaching strategies or programme to assist the students who are less proficient in English.
The sample was made up of two classes of primary three pupils from each of ten schools. Categorisation of schools was based on the Primary School Leaving Examination results from 1982 to 1984. A questionnaire, an English test (made up of listening comprehension, dictation, sound discrimination, reading comprehension, composition and oral test), and the Raven's Standard Progressive Matrices, were group administered on two days. A pilot-test was conducted earlier in three primary schools.
Six hypotheses were drawn up and a correlational analysis carried out on the data.
Findings from the survey suggest that parental educational levels, language use at home and in school, extent of reading interest and general intellectual ability were significantly and positively correlated with English proficiency. A notable exception was private tuition which was found to be significantly but negatively correlated with English proficiency.
Additionally, the results of the correlational analysis also indicated that the father's educational level was significantly correlated with the mother's as well. Next, it was also revealed that there was a significant relationship between parental schooling and language use at home. Furthermore, parental schooling also correlated significantly with the children's reading interests.
For the final hypothesis, path analysis was used to test a hypothesised theoretical model to explain the causal relationships between the selected variables and English proficiency.
A multiple regression analysis was first employed to determine the contribution of the selected variables to the variance in English proficiency. The six selected variables, namely the father's and mother's educational levels, language use at home and in school, sex and general intellectual ability, collectively accounted for about 40% of the variance in English proficiency. General ability accounted for most of the variance, about 24%.
The relative strength of the influence of one variable on another is reflected by the beta coefficient obtained from the multiple regression analysis. The path coefficients for all variables were calculated. The most important contributing variable affecting English proficiency, according to the hypothesised model, was general intellectual ability. It had a direct effect of 0.486 on English proficiency. Next, mother's schooling had a direct effect of 0.310 and an indirect effect of 0.020. The third significant predictor variable, English language use at home, contributed a direct effect of 0.202. Father's schooling had a direct effect of 0.142 and an indirect effect of 0.086 on English proficiency. The English Language use in school and the sex differences were found to have no significant effects on English proficiency and were therefore deleted from the final theoretical path model.
In conclusion, these results suggest that there are significant learner variables affecting English proficiency. Teachers should be aware of the differences in the children's learning abilities and could incorporate suitable teaching strategies or programme to assist the students who are less proficient in English.
Date Issued
1987
Call Number
PE1068.S5 Teo
Date Submitted
1987