The purpose of this study was to determine the frequency of errors in English usage, punctuation, and spelling made by grade 12 students in the two paragraphs that each student wrote on the June, 1953, English 40 (Language) University Entrance examination in British Columbia. The errors were classified within each of fourteen major categories. These categories were further divided to give a total of seventy-four classes. In order to record specific errors some of the seventy-four classes were further subdivided to increase the number of classes to 104, excluding spelling errors. Furthermore an attempt was made to discover a relationship between the incidence of errors in English and certain factors that possibly may have been associated with such errors. These factors were: the student's (a) intelligence (scholastic ability); (b) sex; (c) socioeconomic status as determined by the father's occupation; (d) interest in English as determined by the student's choice of major subjects; (e) choice of topics on which the student wrote his paragraphs, and (f) choice of high school program: University or General. Furthermore, in order to determine the extent to which the number of words in the paragraphs might have influenced the number of errors, this writer found a relationship between the number of errors students made and (a) the number of words written on the two paragraphs on the examination, and (b) the number of words written on (i) the expository paragraph and (ii) the descriptive or narrative paragraph. By discovering the extent of the relationship between errors made in the paragraphs and the marks that teachers gave to the paragraphs, this investigator attempted to find out the degree to which markers took into consideration mechanical errors in English.
On examining 599 paragraphs written by 300 grade 12 students, this writer found the number of words written and errors in usage, punctuation, and spelling as summarized in the table below.
(Tables omitted)
Students wrote the mean number of words and made the mean number of errors as shown in the following table.
(Tables omitted)
When one considers the fourteen main categories of errors, he finds that spelling and punctuation account for slightly more than two-thirds of the errors. If four other categories (capitalization, the apostrophe, omissions, misuse of quotation marks) are added to the punctuation and spelling, one finds that non-usage errors account for nearly 80 per cent of the total number of errors. Those errors ranking 1-7 account for nearly 93 per cent of all errors. Ten kinds of errors in punctuation accounted for 89.9 per cent of all such errors. By applying appropriate statistical analyses, this investigator attempted to determine the relationship between errors and the elements mentioned in the first paragraph. The writer found that the coefficient of correlation between errors and scholastic ability was .304. On both paragraphs boys made a mean of 16.73 errors and girls 13.41; t was found to be 3.12. For 293 degrees of freedom t is 2.59 at the 1 per cent level or less. Consequently, for t = 3.12, the hypothesis of no difference in the means can be rejected. The writer found that students whose fathers were in the professional, semi-professional, and managerial vocations made a mean of 11.17 errors, and students whose fathers were in the skilled, semi-skilled and unskilled vocations made 14.8. For 98 degrees of freedom t is 1.984 at th 5 per cent level. But for the means just given t is 2.062. Therefore the hypothesis of no difference can be rejected.
If choice of majors is used as a criterion of interest, students who are primarily interested in English make fewer errors than those who are not. The former made a mean of 12.61 errors on both paragraphs; the latter, 14.00. For 267 degrees of freedom t is 1.969 at the 5 per cent level of significance; therefore the hypothesis of no difference in the means can be rejected. Turning to a consideration of errors made by University Programme students and those made by General Programme students, one finds that the former made a mean of 12.35 errors; the latter, 17.55. For 287 degrees of freedom t is 2.592 at the 1 per cent level of significance. One can therefore reject with considerable confidence the hypothesis of no difference in the means. That the number of errors on a paragraph does not increase directly as the number of words written is shown by the fact that the coefficient of correlation between the number of errors and the number of words written is .574. Consequently the use of the paragraph as a unit on which to base the numbers of errors need not invalidate the statistical analyses and inferences previously made. Finally, examiners probably took errors into consideration when they marked the paragraphs as the coefficient of correlation between errors and the marks the examiners gave the paragraphs was - .202, which is significant at the 1 per cent level. / Education, Faculty of / Graduate
Identifer | oai:union.ndltd.org:UBC/oai:circle.library.ubc.ca:2429/39727 |
Date | January 1960 |
Creators | Matheson, Hugh Naismith |
Publisher | University of British Columbia |
Source Sets | University of British Columbia |
Language | English |
Detected Language | English |
Type | Text, Thesis/Dissertation |
Rights | For non-commercial purposes only, such as research, private study and education. Additional conditions apply, see Terms of Use https://open.library.ubc.ca/terms_of_use. |
Page generated in 0.0019 seconds