Statistical Analyses For Quality Of English Test Paper

Language testing emerges with the development of language teaching and they are closely interrelated with each other. Teachers always make qualitative analyses to test paper quality before, but this approach is not very deep and comprehensive. This study combines qualitative analyses and quantitative analyses to analyze the selected test paper. Evaluating test quality with statistical methods is systematic, in-depth, objective and scientific and it will provide more convincing results.Taking a college English test paper for second semester final examination as the object, this study adopts statistical methods to make a comprehensive quality evaluation and conducts an in-depth analysis of the factors leading to poor quality items. A sample consisting of 300 examinees is drawn randomly from the population and analyzed from aspects of test validity, reliability, item difficulty, and item discrimination.The results of data analyses indicate that the test reliability is relatively high. Though the test has good face validity and content validity, its construct validity is not very good. Item difficulty is not distributed adequately and more easy items exist. Item discrimination is affected by item difficulty, not showing very satisfactory results.The factors leading to poor quality items fall into two categories, violation of the basic multiple-choice-item writing principles and the unique factors leading to poor quality items in listening and reading comprehension tests. This study is an effort to apply statistical approaches in language testing, in which limitations exist and hoped to be improved in the future research.