The Study On English Learning Aptitude Test For Grades 3-6 Schoolchildren

Objective:To construct the English Learning Aptitude Test for Grades 3-6 schoolchildren,to know the current situation of English learning ability of primary schoolchildren,and to provide an empirical basis for English education evaluation,consultation,counselling and intervention.Methods:Based on literature and expert interview,by adopting the combination of qualitative and quantitative methods,built the theory structure of primary school students English learning aptitude,and establish the item bank,then formed the tests of two levels(middle and high),which consisted of speech coding ability,memory,associative ability,grammatical sensitivity and inductive language ability sub-tests,110 items of each initial test.By convenient sampling method,two groups of urban schools and urban and rural schools were selected in four administrative districts in Changsha.In the five primary schools,the students were given 1965 test papers,and the number of valid questionnaires was 1898,with an effective rate of 96.6%.Among them,there were 1,033 male and 865 female students respectively.The middle and high grade was 941 and 957 respectively.CTT and three parameter models of IRT were used to analyze the item quality.Then the final versions were formed.Results:(1)The structure of test:English Learning Aptitude Test For Grades 3-6 were divided into two levels,namely-middle and high grade volume,which consisted of four sub-tests,34 and 40 items respectively.(2)The quality of items:the difficulty parameters of the middle grade volume were[-1.803,2.490],with an average of-0.667;the discriminant parameters were[0.320,4.879],with an average of 1.145.The guessing parameters were[0.031,0.404],with an average of 0.118.The difficulty parameters of the high high volume were[-1.787,1.517],with an average of-0.066;the discrimination parameters were[0.370,1.316],with an average of 0.777,the guessing parameters were[0.049,0.374],with an average of 0.201.(3)The reliability test:the maximum amount of information of middle grade volume was 27,Cronbach's ? was 0.842,and the split-half reliability was 0.692.The maximum amount of information of high volume was 10,Cronbach,s a was 0.868,and the split-half reliability was 0.759.(4)The validity of test:in the second-order CFA model,RMSEA of the middle and high grade volume was 0.049 and 0.037 respectively,and the CFI respectively was 0.917 and 0.910 respectively,and TLI was 0.911 and 0.905 respectively;the correlation between the score and final grade in English was 0.45 and 0.30 respectively.(5)The current situation analysis results:there were significant differences in gender,grade and school types in the total score of the two volumes.In gender,there was no significant difference in the grammatical sensitivity of the two volumes;there was no significant difference in the ability of associative ability and inductive language ability of middle volume;there were significant differences in speech coding ability of the two volumes.In grade,there was no significant difference of the middle volume;there were significant differences in the ability of memory connection and inductive language of the high volume.In school type,there was a significant difference in the memory connection ability of the middle volume,and there were significant differences of the high volume.Conclusion:(1)The item parameters reached the requirements of Psychometrics.(2)The reliability and validity of the test were good.(3)(3)Grades 3-6 students' English learning aptitude was different in gender,grade and school type.