The research, entitled “Levels of English Skills of Thai Students”, aimed to:
1. study the levels of both unitary and integrative English skills and the knowledge of language components of secondary school students in Matayom Suksa Three and Matayom Suksa Six and those of university students;
2. compare the English skills and knowledge of language components of students in small-sized, medium-sized, and large to extra-large sized schools, in the first, second, and third years;
3. study the relationships among the students’ English language skills and their knowledge of the English sound system, vocabulary, and structure; and to
4. pave the way for developing a university English curriculum appropriate for students’ levels of English skills, abilities, interests and their demands for English use.
The research samples at the secondary level were obtained by using the multi-stage stratified random sampling method on students in schools of small, medium, large and extra-large sizes in Bangkok under the jurisdiction of the Department of Curriculum and Instruction of the Ministry of Education, numbering 697 Matayom Suksa Three students and 525 Matayom Suksa Six students. The 493 tertiary students were first, second and third year Chulalongkorn University students, and were obtained by using the stratified random sampling method. The confidence level of the sampling was 99 per cent.
The research instruments consisted of three sets of constructed English skill tests based on amplified objectives of the English syllabus at each level of education. The tests were standardized through the classical model of test-item analysis to improve the items, and through the Rasch model to select only those items that fit the ICC curve for the research. Statistical analyses used to study and compare the students’ English skills were arithmetic means, coefficients of variation, skewness, t-tests, and one-way ANOVA. In addition, the Pearson product-moment correlations were employed to discover the relationships between the skills in the sound modality and the graphic modality, and between the skills of using English and knowledge of the English language components.
The research findings could be summarized as follows:
1. Matayom Suksa Three students were rather weak in the skills of using English in both the sound modality and the graphic modality, needing improvement, especially in the skills of reading and writing.
2. Matayom Suksa Six students had a medium level of English skills in the sound modality needing improvement, while their skills in the graphic modality were rather weak needing improvement, especially in the writing skill.
3. University students had a medium-level of skills in both the sound and the graphic modalities needing improvement, especially in the integrative skills of reading-writing and the writing skills, which were very weak.
4. Concerning language components, it was discovered that Matayom Suksa Three and Six students had a low level of ability which needed improvement while university students exhibited a medium level of ability.
5. Concerning test score distribution, it was discovered that the scores of Matayom Suksa Three students were most widely spread having 6 levels of dispersion, followed by those of Matayom Suksa Six students with 5 levels and university students with 4 levels.
6. Concerning the skewness of score distribution, it was revealed that while the test scores from the English skill tests and the language component tests of Matayom Suksa Three and Matayom Suksa Six students were generally below the mean scores, those of university students were above.
7. Secondary school students from small and medium sized schools exhibited statistically significantly lower levels of skills in both the sound and the graphic modalities than students from large and extra-large sized schools (p < .01). It was also discovered that the English skills of first, second and third year university students were significantly different with second-year students having significantly higher scores than first-year students (p < .05).
8. Language skills correlated significantly with language components at all three levels of education; however, the correlational patterns differed.
Generally speaking, integrative skills correlated more highly with language components than unitary skills (integrated listening-reading-writing skills of Matayom Suksa Three and Six students correlated with language components having rxy of .397** and .490** while their unitary listening skills had only .163** and .445** respectively; the university students’ unitary skills of speaking (self-assessment) correlated with language component, having rxy .264** while their integrative listening-reading-writing skills correlated with rxy .457**).
9. The productive skill of speaking (self-assessment) correlated more highly with the productive unitary skill of writing than with the receptive skill of reading (rxy for Matayom Suksa Three were .275** and .164**, .264** and .252** for Matayom Suksa Six, and .224** and .112** for university students).