Error Analysis of Subject-Verb Agreement Made by Students in Meurandeh Langsa

This research aimed at identifying the subject-verb agreement errors in students’ writing. This research applied a descriptive research. The data were collected through test and non-test instruments. Test instruments were conducted through writing test and non-test instruments were through questionnaire and interview. The data obtained from both test and non-test instruments were conducted by employing descriptive analysis. This research analyzed students’ narrative writing based on Surface Strategy Taxonomy proposed by Dulay, Burt and Krashen (1982). This result revealed the types of error in subject-verb agreement in the students’ narrative writing covering omission, addition and misinformation. In conclusion, the students involved in the research made a number of errors. It was found that the students’ ability to use subject-verb agreement in English was still low.

Subject–verb agreement in English

The paper rejects the standard view according to which every tensed verb in English agrees with its subject in person and number. It argues that person is irrelevant to all verbs except BE, and that past-tense verbs and modals (other than BE) have no number agreement features. It discusses agreement mismatches which reflect the subject's meaning, but rejects the idea that subject–verb agreement may be a semantic rule; it proposes instead a new feature ‘agreement-number’. This extra number feature applies only to the subject of a tensed verb and by default has the same value as the subject's ordinary number, while also allowing various kinds of mismatch (for I and you, and for cases of ‘semantic’ agreement). It also offers analyses of agreement with non-nominal subjects and dummy there, and shows how the analysis for Standard English generalizes easily to a range of variations found in nonstandard dialects. The theoretical basis for the analysis is Word Grammar, whose main advantage is that features are free to be assigned by rule because they are not used in classification.