Validity and Reliability Study on Teacher-Made Assessment for English Mid-Term Examination

Abstract

This study is primarily aimed to study the validity and reliability of teacher-made assessment for English mid-term examination used in evaluating students' learning experience: to what extent the test quality is in term of content validity, reliability, index of difficulty, index of discrimination, and effectiveness of distractors. Qualitative and quantitative data were collected, analyzed, and interpreted by employing a descriptive and exploratory study on teacher-made English midterm test as well as 20 students' answer sheets as the source of data. The source of data was collected from participants in one private Islamic Senior High School in Kuningan, West Java. Document analysis was conducted to pull out the findings of the study. The test items were observed through its representation in syllabus to reveal its content validity. Then, the students' answers were analyzed using Anates software version 4 to determine the reliability, index of difficulty, index of discrimination, and effectiveness of distractors of the test. The findings of the study show that content validity and reliability were considered good. However, some items were found to be either too easy or too difficult in terms of its index of difficulty. The test also had 50% poor discrimination index and 40% ineffective distractors. Therefore, it is suggested for the teacher to consider a further revision on the English mid-term examination they made for improving of its quality.

A shortcut in language testing: Predicting the score for paper-based TOEFL based on one sub-score

“…In addition, Nurhalimah et al (2019) found that teachers made tests have low discrimination index. In addition, a large percentage of the distractors used in multiple-choice tests were ineffective (Rohmah, 2019). A research study by Ing et al (2015) found that teachers are not proficient in following the test construct.…”