defects in constructing the questions, poor quality questions as well as questions that were too difficult for the students. Thus it provides a basis for remedial work as it brings to light general areas of weakness of students requiring more extended attention. DEFINITION
A type of analysis used to assess whether
items on a scale one tapping the same construct and are sufficiently discriminating. PURPOSES OF ITEM ANALYSIS 1. To determine whether the item is of appropriate level of difficulty for the batch of students tested 2. To determine whether the item is capable of discriminating between knowledgeable and the poor students. 3. Item analysis describes the statistical analysis which allow measurement of the effectiveness of individual test items. 4. It helps in selecting the best item for the final test and reject poor items and modify some of the items PARAMETERS FOR ITEM ANALYSIS
Item analysis is a measure of three important
parameters of a multiple choice items after it has been administer to a group of students. They are 1. Difficulty index 2. Discrimination index 3. Functionality or effectiveness of distracters DIFFICULTY INDEX STEPS 1.Score the whole test for all students. 2.Rank the students in order of merit based on their test scores. 3.Take the bottom third on the low achievers and the top third as the high achievers. 4.Prepare a table for each item as follows. For example:- Suppose 12 pupil out the 24 pupils in the upper group and 5 pupils in the lower group selected the right alternative. That makes a total of 17 out of 24, who got the item correct indicating that the item has a low level of difficulty. Since more pupils in the upper group than the lower group got them right, indicating that it is discriminating positively Estimating item difficulty: Item difficulty of a test is indicated by the percentage of pupils who get the item right. The formula for this is, Difficulty = (R/t x 100) Where, R - is the number of pupils who got the item right. T - is the number of pupils who tried in the item. DISCRIMINATION INDEX
An indicator showing how significant a question
discriminate between ‘high’ and ‘low’ students. It varies from -1 to +1. Calculations: The following formula is used (Guilbert) Discrimination index 2 x (H-L / N) DISCRIMINATION INDEX
Guilbert suggests the following index to judge
questions. 0.35 and over – Excellent questions 0.25 to 0.34 – Good Questions 0.15 to 0.24 – Marginal questions, to be revised Under 0.15 – poor questions, to be discarded. EVALUATING THE EFFECTIVENESS OF DISTRACTORS (FUNCTIONALITY)
• When a distractor attracts more pupils from the upper
group & less pupils form the lower group, it is because the distractor is a poor one. • An item ie maximum positive discrimination power would be one where all the pupils in the upper group got the item correct and all the pupil from the lower group got the item wrong. This results is an index of 1.00. • An item & no discrimination power would be one which an equal number of pupils in both the upper & lower groups got the item right & the index will be 00. CAUTION WHEN INTERPRETING ITEM ANALYSIS RESULTS
• Item analysis data are not synonymous item
validity • The discrimination index is not always a measure of item quality • Item analysis data are tentative USES OF ITEM ANALYSIS
• Creation of item banks
• It gives the feedback to the teacher • It can be used to improve teaching methods, introduce audio-visual aids or use them more effectively of determine areas requiring emphasis and reinforcement • It also pinpoint the questions on which good students are confused and which students did not attempt • Teachers consequently can examine the amount of time allocated to those areas/and/or clarify of teaching