Ajeng Desy H, 2201405080 (2009) The Analysis of Validity, Reliability, Discrimination Power, and Level of Difficulty of First Mid-Term Test. The case of eighth grade students of SMP 33 Semarang. Under Graduates thesis, Universitas Negeri Semarang.

Preview

PDF (The Analysis of Validity, Reliability, Discrimination Power, and Level of Difficulty of First Mid-Term Test. The case of eighth grade students of SMP 33 Semarang) - Published Version Download (7MB) | Preview

Abstract

A good English test will help students to learn the language by requiring them to study hard, emphasizing the objectives of the course and also showing them in which parts of the course they need improvement. A test, which is intended to measure the students’ achievement, has to fulfill the requirements of good test such as validity and reliability. There are several factors that influenced in building good test. There are relevance, balance, efficiency, specificity, difficulty, discrimination, variability and reliability. In this study, the writer would like to focus her research on the English midterm test which is administered to eighth grade students of SMP 33 Semarang in the academic year of 2008/2009. In this study, the writer would like to find the answer of the following question: “how good is the English mid-term test for eight grade of SMP 33 Semarang in the academic year 2008/2009.” The general objective of the study is obtaining an objective description of the structure of a good test item. The method that the writer used in analyzing the data this study is quantitative approach. In writing this final project, the writer conducts to activities. The first is library activities, the writer select some books which give information, or supporting data for reference. Then the second is field activity, it is used to collect the data. From the result of the analysis the test there are 33 valid items and 17 invalid items. The reliability of the test is 0.39, so this test is still reliable. From the point of view of discrimination power, it can be concluded as poor because the mean of the discrimination power is 0.17. There are 8 good items, 13 marginal items and 29 poor items. In the term of difficulty level this item categorized as moderate item because the mean is 0.41. There are 11 difficult items, 34 moderate items, and 5 easy items. Based on the result above, the writer would like to offer some suggestions. First, the constructor of the test should be aware the characteristic of good test, especially in determine difficulty levels and discrimination power. Second, items that still can be used should be revised and save, while items which have negative value should be discarded, because it means that the students in the lower group performing better than the students in the upper group. Finally, the writer suggest that the test should not be used in the English final test, it can still be used unless it has makes some revisions. The writer hopes that the result of this item analysis could be used as an example in analyzing other test item and encourages teacher to make good English test.

Actions (login required)

View Item