Tan , Yoke Ping (2006) Effects of contrastive analysis and error analysis on English articles performance of east Malaysian second language learners in written production. Masters thesis, Universiti Malaysia Sabah.
It is well known that the Malay language does not have functional equivalents of the English definite and indefinite articles. Correspondingly, there is plenty of observational evidence that second language learners have difficulty with the English article system. This study Investigated the error patterns that occur in the usage of English articles produced by the East Malaysian second language learners of English in their written tasks. The hierarchical order of each kind of error in terms of the frequency of its use was identified as well Contrastive Analysis and Error Analysis were applied in this study to investigate the effects of both elements on the performance of students on article system learning. Contrastive Analysis (CA) is the systematic comparison between the native language and the target language. Error Analysis (EA) is a way to assess the learner's competence in his use and knowledge of a second language. Two tasks, namely 24 multiple-choice article insertion questions and a multiple-choice doze test in which a, an and the were given as options to complete a text, were given to 85 Form 4 students, i.e. 41 and 44 students in treatment group and control group respectively. A pre-test and a post test with the same set of tests in each stage were conducted on these students to investigate the effectiveness of the EA and C4 elements in the article system learning. The results ofthe study Indicated that the most difficult article to be mastered was the ø (zero article), whereas ‘a' and ‘an' were major problems, and 'the' was the least difficult one. The types of error most frequently committed by the subjects were 'omission of the' and the used instead of ø In multiple-choice article insertion questions, it was found that overextension (overproduction of articles) errors were prominent, whereas in multiple-choice doze test, underextension (omission of articles) errors were prominent. Another worth noticing finding was that, the subjects were prone to overspecify the first-mention object In the multiple-choice cloze test. The subjects were also weak in the usage of 'a' and 'an' as they seemed to form false hypotheses that an' Is placed before words that begin with a, e, i, o, u and ‘a’ placed before words that begin with other letters or numbers. It was also found that the more the adjectives that precede the nouns, the higher possibilities the subjects would omit the article. Among the problems faced by the subjects In both tasks were the usage of articles before proper nouns and before singular unspecified nouns. Native Language has no strong direct Influence on the learning of article and the majority of the efforts could be accounted as the mastering of the language In nature, since the English article system is a grammatical item that is comprised of multiple elements. It is believed that there are some aspects of grammar that are not teachable and they must be learned through exposure.
|Item Type:||Thesis (Masters)|
|Uncontrolled Keywords:||English Article System, error pattern, multiple-choice cloze test, English, writing|
|Subjects:||L Education > LB Theory and practice of education|
|Divisions:||SCHOOL > School of Education and Social Development|
|Deposited By:||IR Admin|
|Deposited On:||12 Dec 2014 14:45|
|Last Modified:||12 Dec 2014 14:45|
Repository Staff Only: item control page