An assessment of the effectiveness of an experimental approach for teaching applied English grammar to adults in a public setting

This study assessed the effectiveness of the experimental approach "Unified, Conceptual, Integrated Model (U.C.I.M.) for English Grammar" for increasing the skills of a sample of adult learners in elements of standard written English as defined and measured through the "Usage" (assessing use of pronouns and modifiers) and "Sentence Structure" (assessing complete sentence use, appropriate use of subordination and coordination, appropriate modifier placement) subtests of the Descriptive Tests of Language Skills (DTLS). Two hypotheses guided the development of the study: (a) the mean DTLS "Usage" subtest scores and (b) the mean DTLS "Sentence Structure" subtest scores for the study sample taught by the experimental approach would be significantly higher than those of a comparison group taught by a traditional approach. The study followed a quasi-experimental pretest-posttest design, involving experimental (N = 202) and comparison (N = 238) groups. The homogeneity of the groups was established through a chi-square test of difference on selected demographic characteristics. The subjects participated in seven weeks of instruction (ten hours) in standard applied, English grammar. The Traditional group was treated with a "traditional" model constructed by the writer, based on analyses of 32 texts and other materials, which revealed the model as generally authoritative, rule-based, and characterized by fragmentation and variability in content, presentation, and learning sequence. The Experimental group was treated with the concept-based "Unified, Conceptual, Integrated Model for English Grammar" approach devised by the author. Hypothesis testing utilizing pre- and posttest data entailed the t-test statistic. Significance was set at p $<$.05. The results of hypothesis testing led to the rejection of both hypotheses. The Experimental group did not show a significant gain. The data did show greater gains for the experimental group in the parts of the tests that assess communicative areas. The findings warrant the conclusions that the U.C.I.M., as conducted under the conditions of this study, is not significantly more effective than the traditional model, that no definitive changes in approach should be made as a result of this study, but that, from the findings of the subtest cluster scores, selected aspects of the U.C.I.M. concerned with linguistic connectivity may have significance for teaching models.