ITEM ANALYSIS OF STUDENT COMPREHENSIVE TEST FOR RESEARCH IN TEACHING BEGINNER STRING ENSEMBLE USING MODEL BASED TEACHING AMONG MUSIC STUDENTS IN PUBLIC UNIVERSITIES

Abstract

Beginner string ensemble model based teaching is a systematic instruction which incorporate improvement and new strategies for teaching string ensemble. The instructional design model by Dick and Carey laid the groundwork for these model based teaching. In the design process, achievement test were constructed to evaluate learner’s progress and instructional quality. Using student comprehensive test as instrument, the study attempts to examine test items of a researcher made test in the research of string ensemble course for university music students. The quality of each particular item was analyzed in terms of item difficulty, item discrimination and distractor analysis. Reliability test were also conducted in addition to the item analysis to observe the quality of the test as a whole. Two statistical tests were used to compute the reliability of the test: The Kuder-Richardson Formula (KR20) and the Kuder-Richardson Formula 21 (KR21). The comprehensive test was piloted in one university in Perak involving 16 music students. The difficulty index of 41 test items was computed using Microsoft Office Excel. The discrimination analysis was conducted by means of Pearson product-moment correlation using SPSS 17.0. Eventually Microsoft Office Excel was used to compute the reliability statistics. The result indicates that forty four percent of the total test items exceed the difficulty index of 0.8 suggesting easy items. Fifty nine percent of items obtained acceptable range of discrimination index. Distractor analysis reveals that some distractors were not effective. The quality of the item as a whole indicates a reliable value Kuder-Richardson 20 (KR20) value of 0.717 and Kuder-Richardson 21(KR21) value of 0.703. The findings suggest that in order to measure students’ performance effectively, necessary improvement need to be done where items with poor discrimination index should be reviewed.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 50 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.