A comparative analysis of sentences written by eight-grade students instructed in transformational-generative grammar and traditional grammar

Abstract

PURPOSE: To investigate the effects of direct instruction in the kernel sentences of transformational-generative grammar and in paralleled concepts of traditional grammar upon the following variables of sentences written by eighth grade students: 1) noun phrase element, 2) verb phrase element, 3) predicate expansion element, and 4) average length of clauses.

PROCEDURES: The concepts for transformational-generative grammar kernel sentences were selected by means of a survey of tre literature and by expert opinion. A series of lessons was developed by the investigator and evaluated by a jury of English teachers and a text-book author of transformational grammar. Paralleled concepts in traditional grammar were outlined. A sentence analysis instrument was constructed. The fourth and final draft was applicable for written sentences. Eighty-four students in an urban junior high school comprised the groups included in the study. Experimental and control groups were equated on the basis of: 1) chronological age, 2) intelligence test scores, and 3) achievement test scores. The developed lessons were used by the experimental group and the paralleled lessons in traditional grammar were used by the control group. The experiment extended over a period of fourteen weeks. Both groups wrote compositions for pre-test and post-test. From each test two hundred running words were extracted and analyzed for four sentence variables.

RESULTS AND CONCLUSIONS: 1. The developed lessons are effective in promoting sentence growth for three sentence variables. The mean for the noun phrase element increased from 28.93 at pretest to 31.07 at post-test. The mean for the predicate expansion increased from 36.51 to 40.07. The mean for total number of clauses increased from 26.18 to 29.78. The increases for predicate expansions and clauses were statistically significant. 2. The lessons are not effective in promoting growth or expansion of the verb phrase element. The mean for pre-test was 31.44 and 26.93 for post-test; the decrease was significant. 3. The lessons are effective in promoting sentence "packing," a pooling of phrases and clauses into the predicate expansion. In this study, sentence "maturity" or growth was determined by the "packing" process. 4. On the dependent and control variables, positive correlations were significant for the Experimental Group at post-test for I.Q. and the noun phrase element and for achievement test and the noun phrase, the predicate expansion, and clauses. For the Control Group, the only posttest significant positive correlation was between chronological age and the verb phrase element.

IMPLICATIONS: This study supports the belief that instruction in the kernel sentences of transformational-generative grammar promotes growth in sentence writing. It appears that the "packing" of phrases and clauses into the predicate expansion clearly identifies the ability to think and to write beyond simple, elementary levels of structure. Careful consideration should be given to the revamping of the English curriculum at the junior high school level. That the so-called "school" grammar contributes little to the improvement of syntax is clearly shown in this study. On the other hand, the "new" grammar offers promise to the adolescent's understanding and writing "matured" syntactic structures of the language.