Errors and judgments : a sociolinguistic study of freshman composition

No Thumbnail Available
Authors
Horvath, Veronika
Advisor
Houck, Charles L.
Issue Date
1996
Keyword
Degree
Thesis (Ph. D.)
Department
Department of English
Other Identifiers
Abstract

This study attempts to discover and describe patterns of variation in college students' overt attitudes toward a limited set of grammatical and lexical variables, the shibboleths of edited written American English usage. The basic instrument used in the study is a 115 item multiple choice questionnaire prepared by the researcher. Fifteen questions were designed to assess the respondents' social, economic, and demographic backgrounds, whereas the major part of the questionnaire elicited judgments about one hundred English sentences offering the choice between the attributes "good," "bad," and "I can't decide." This questionnaire was administered to 172 students in nine freshman composition classes during the spring semester of 1994 at Ball State University. The study sought to discover and describe systematic relationships between the answers to the first set of questions (extralinguistic data) and the second set of questions (linguistic data) by using various analytical methods and statistical techniques, such as correlation coefficients, chi-square tests, and multidimensional scaling.It was hypothesized that variation in subjects' overt judgments about linguistic variables would parallel the findings of numerous sociolinguistic studies about variation in linguistic production, and hence would pattern along the social and demographic characteristics of the subjects. However, although this study found considerable variation in the freshman students' judgments about the usage shibboleths, it did not find social or demographic correlates to the respondents' judgments.By investigating the nature of the variation in freshman composition students' notions about linguistic correctness, this study attempted to answer questions which have not been asked by traditional usage studies, sociolinguistics, or composition research. Moreover, this study has added support for linguists' claims that the traditional "mistakes" in usage handbooks have almost no empirical basis, even if they remain the favorites of most handbook authors and English teachers.