The pre-writing experience in composition teaching for grade ten pupils
Date
1979
Authors
Mitchell, Robert Sinclair
Journal Title
Journal ISSN
Volume Title
Publisher
Abstract
The purpose of the study was to determine what significant differences, if any, existed among three groups of Grade X students with respect to the fluency (number of words and number of T-units) and syntactic maturity shown in a composition test written after exposure to differing pre-writing strategies and experiences. One treatment group (Group A) was required to work independently, with no opportunity for verbal exchange with others; a second group (Group B) participated in a teacher-directed lesson which involved some student-teacher interaction; the third group (Group C) engaged in "free discussion" of the topic to permit maximum pupil talk (student-student interaction).
Theoretically, this research investigation operated in the belief that written composition is essentially a communication process and that it must be viewed in that light. Further, it was held that the composing process itself must be the basis for research on writing, since classroom instruction cannot be substantially bettered without further explanation as to how and why students write in the first place. The present study was conceived, more specifically, on the premises that pre-writing is an important and integral part of the composing process; that, to be meaningful, cognitive processing in terms of language must be internalized for the student writer (that is, that others cannot do this for him/her); that such processing occurs more readily in an environment where pupil talk (verbal interaction among students without teacher intervention) predominates.
The subjects involved in this study were taken from six classes of Grade X (English 10) students from three junior secondary schools in School District No. 61 (Greater Victoria), Victoria, British Columbia. All students wrote a pre-measure composition on the same topic and followed the same procedure (the method used by Group A for the composition test: independent work). The composition test itself was written the following week, with the classes assigned to one of the three treatment groups. (In practice, the students in each school followed a different method.) From the total population of 156, subsamples from each of the three treatment groups of thirty students (15 boys and 15 girls) were randomly selected for statistical analysis. As a result, the pre-measure and composition test papers for ninety students were utilized. Subsequently, one (male) student from Group A had to be deleted because of incomplete data provided for the composition test. The final study population, therefore, was eighty-nine students (44 males and 45 females).
For the purposes of this study, the subjects wrote only in one mode of discourse (the expressive mode). Both the pre-measure and composition test papers were written in this mode, and all of the students wrote on the same topics.
Four hypotheses were tested in this study:
1. The hypothesis that there would be no significant differences among the six classes involved in this study on the pre-measure composition with respect to fluency (as measured by the number of words and the number of T-units written) and syntactic maturity (the number of words per T-unit) was not accepted when analysis of variance procedures were applied in the statistical treatment of the data. It was found that significant differences (beyond the .001 level) existed among the classes for all three variables.
(Because significant differences were found for the pre-measure composition, analysis of covariance procedures were applied to the three variables examined in the composition test situation to control for these differences.)
2. The hypothesis that there would be no significant differences among the three treatment groups in terms of fluency (as measured by the mean number of words in the compositions) was accepted. No significant differences were found.
3. The hypothesis that there would be no significant differences among the three treatment groups in terms of fluency (as measured by the mean number of T-units in the compositions) was also accepted, although the results did approach significance (beyond the .07 level).
4. The hypothesis that no significant differences would exist among the three treatment groups with respect to the variable of syntactic maturity was not accepted. Differences significant beyond the .01 level were found.
A review of the literature would suggest that the findings of this study with respect to the variable of syntactic maturity are new. It would be desirable, therefore, that a replication of this study be attempted to confirm the results and, further, that additional research investigations be initiated with different populations, with different modes of discourse, and with a view to analyzing the effects, if any, on writing quality.
The results of this study would seem to indicate that composition instructional strategies ought to include provision for pre-writing and particularly for activities which permit maximum pupil talk, at least as far as the expressive mode of discourse is concerned. In addition, there appears to be a need for further research into the areas touched upon in the present study.