Essay Test Types of Test Item

In addition, “item analysis as a whole will be defined here as the systematic statistical evaluation of the effectiveness of individual test items. Item analysis is usually done for purposes of selecting which items will remain on future revised and improved versions of the test. Sometimes, however, item analysis is performed simply to investigate how well the items on a test are working with a particular group of students, or to study which items match the language domain of interest.” 83 Moreover, Arthur Hughes proposes the purpose of item analysis which is “to examine the contribution that each item is making to the test. Items that are identified as faulty or inefficient can be modified or rejected.” 84 Although item analysis is done primarily for response-choice item, it is available for teacher to use several of the techniques described with any items that are scored dichotomously simply as correct or incorrect. 85 In the writer’s opinion, item analysis is statistical evaluation to know the quality of a test by identifying whether every item on a test works appropriately or not. It is done by collecting students’ responses to each item so that it can also be known which items are included as a good one and which items that weaken the test. It is very useful for teacher to performs item analysis since it can be a device for test improvement.

2. Kinds of Item Analysis

Item analysis usually concentrates three vital features: level of difficulty, discriminating power, and the effectiveness of each alternative. “Thus, item analysis can tell us if an item was too difficult or too easy, how well it discriminated between high and law scores on the test, and whether all the alternatives functioned as intended.” 86 82 Anthony J. Nitko, Educational Test ..., p. 284. 83 James Dean Brown and Thom Hudson, Criterion-Referenced ..., p. 113. 84 Arthur Hughes, Testing for Language ..., p. 225. 85 Anthony J. Nitko, Educational Test ..., p. 286. 86 Wilmar Tinambunan, Evaluation of Students ..., p. 137.

a. Level of Difficulty

The first area in item analysis is level of difficulty which concerns on how easy or difficult each item is. According to Kathleen M. Bailey, difficulty level is “an index of how easy an individual item was for the people who took it. It is typically printed as a decimal, ranging from 0.0 to 1.0. It represents the proportion of people who got the item right.” 87 Furthermore, in the book, Language Tests at School , “difficulty level or item facility has to do with how easy or difficult an item is from the viewpoint of the group of students or examiness taking the test of which that item is a part.” 88 In writer’s opinion, level of difficulty deals with how many percentage of students who response an item correctly and those who response incorrectly. By analyzing the difficulty level of each item, it can be inferred whether an item is included as easy, moderate or difficult item. Level of difficulty is interpreted in the form of percentage. The larger the percentage of the correct answer, the easier the item is. Then, the fewer the students who answer correctly, the more difficult the item is. Henceforth, a good test item should have the level of difficulty, which includes easy, moderate, and difficult level. The effective and good test should have the items that belong to moderate level. The item that is too easy or difficult potentially weaken the quality of the test and the valid data of information about student’s achievement will not be acquired. In addition, level of difficulty analysis can be applied for either large group of students or the small one. 87 Kathleen M. Bailey, Learning about Language Assessment: Dilemmas, Decisions, and Directions, New York: Heinle Heinle Publishers, 1998, p. 132. 88 John W. Oller, Language Tests at School, London: Longman Group, 1979, p. 246. As a quote, from Lyle F. Ba chman, states that, “to conduct an item analysis, we first arrange the scored test papers or answer sheets in order from the highest score to the lowest score. Next, we separate the papers into upper and lower groups, according to their total test scores. For large groups, we would choose the upper and lower 27 percent, while for smaller groups, we would typically choose the upper and lower one-third. ” 89 The formula used for analyzing the difficulty level of each item in large group is stated below: In which: TK : Index of difficulty U : The number of students in the upper group who answer the item correctly L : The number of students in the lower group who answer the item correctly T : The number of students in upper and lower group 90 Next, for the small group, teacher or test maker can easily evaluate an item by using all the students’ answer sheets. Then, the formula is: 89 Lyle F. Bachman, Statistical Analyses for Language Assessment, Cambridge: Cambridge University Press, 2004, p. 123. 90 M. Ngalim Purwanto, Prinsip-Prinsip dan Teknik Evaluasi Pengajaran, Bandung: Remadja Karya, 1986, p. 153. B P = JS U + L TK = T