Analysis of Test Items used in an Achievement Test in Physics at Secondary Level

Aim of present study was the analysis of test items constructed in an achievement test in the subject of physics for class IX. It was used to study the difficulty level, discrimination power and distracter analysis of each test item. The achievement test was comprised of 40 multiple choice questions. Test was administered by researcher on a sample of 250 students from Islamabad Model Schools in Sector G-9. Findings of the study showed that most of the test items fall in acceptable range of difficulty index and discrimination power. In keeping view, the findings of the study, researcher comes to know that most of the test items prove to be satisfactory regarding difficulty index and discriminating power. However, ten out of forty test items were discarded due to high or very low level of difficulty and poor discrimination power. This work is equally significant for the researchers and for the subject teachers in preparing achievement tests in order to assess performance of students with optimum level of difficulty index and discriminating power.


Significance of the Study
Successes of any education system rely solely on the well-established system of examination. Examination on one hand play the role for promoting students to next grade and on the other hand provides scope for assessment of curriculum, program and performance of teachers. The significance of the study may be understood as · It enables subject teacher to know about the too easy and too difficult test items (Items difficulty). · It enables him to differentiate between high achievers and low achievers (Discrimination power). · It enables him to construct useful distracters (Distracters analysis). · It also provides scope for checking whether the test under consideration is reliable or unreliable (Test reliability).

Instrument of the Study
The instrument of the study was comprised of 40 test items for data collection. Initially test was consisting of 60 test items from unit on motion and rest in physics for class IX by Punjab text book board, Lahore. The test was developed by researcher himself and approved by experts in the subject of physics at secondary level. All experts ensured that various levels of Bloom's Taxonomy have got due weightage. Test items which found ambiguous, providing clues and of language issues were discarded. All necessary instructions were provided to the participants to ensure reliability.

Population of the study
The population of the study comprised of all students of class 9 th in Islamabad Model Schools in G-9 sector, Islamabad.

Sample of the Study
Simple random sampling technique was used to obtain a sample of 250 students at secondary level. Sample comprised of the students of both genders. Sample of the study in detail shown in figure 1.

Data Collection
Achievement test in physics was administered by the researcher himself to the sample of 250 students. Initially students were informed about the particular unit for achievement test and concerning directions for its administration.

Data Analysis
Student's scores on test items were arranged by using MS-Excel in descending order. In order to start the item analysis top 30% scorers and bottom 30% scorers of the entire sample were selected. Upper and lower group of students each comprised of 75 students was used. Nunnally (1972) suggested 25% while SPSS (1999) uses the highest and lowest one third (33%).

Item Difficulty (F)
It deals with how difficult is the test item. It is indicated by the percentage of the pupils who got the item right. It is recommended that test item would be neither too easy nor too difficult. It involves following basic steps: 1) Arrange the papers in order from the highest to the lowest scores (say 250 papers).
3) Select 30% papers with the lowest scores (low achievers) -75 papers. 4) 40% papers in the middle (100 papers) would not be taken in account. 5) Calculate the correct responses of high achievers and low achievers on each item. 6) Apply the formula and calculate F (Facility Index) F= NR/NT Where NR= Number of students who got the item right.  Vol.10, No.10, 2019 92 NT=Total number of students. 7) F is acceptable when it ranges from 0.30 to 0.90. 8) Value more than 0.90 indicates that item is very easy. 9) Value less than 0.30indicates that item is very difficult.

Discrimination Power (D)
It refers to the degree to which test item discriminates between pupils with high and low achievement. One purpose of testing is to discriminate between high and low achievers. It involves following basic steps: 1) Arrange the papers in order from the highest to the lowest scores (say 250 papers).
3) Select 30% papers with the lowest scores (low achievers) -75 papers. 4) 40% papers in the middle (100 papers) would not be taken in account. 5) Calculate the correct responses of high achievers and low achievers on each item. 6) Apply the formula and calculate D D=NH-NL/n Where n= Number of high or low achievers. NH=Number of high achievers who got the item right. NL= Number of low achievers who got the item right. 7) D is acceptable when it ranges from 0.30 to 1.0. 8) Value more than 0.40 indicates 100% discrimination. 9) Value less than 0.30 indicates incapability of the item to discriminate. Poor item to be rejected or improved by revision Source: Ebel (1972) in Ovwigho (2013).

Distracters Analysis
The last but not least stage in item analysis procedure is to assess the quality of distracters for each test items. These are incorrect alternatives alongside correct option in case of multiple choice items. The quality of good distracters is to "distract" examinees which are unaware of the right response.
Distracter analysis provides the information to instructor to observe that how many numbers of testers in the upper group and bottom group tick out each option on a multiple-choice item. Distracter fails to do its job when it is not selected by examinees at all. In other words, the usefulness of distracter relies solely on its selection by some examinees. Otherwise it is ineffective and must be revised or replaced. When distracter attracts more examinees from the bottom group than the upper group it is negative discrimination. Similarly, when distracter attracts more examinees from upper group than the bottom group it is positive discrimination. Distracters should demonstrate negative discrimination.  2,3,4,5,6,7,8,10,11,12,13,14,15,16,17,18,19,21,22,23,24,25,26,27,28,30,31,32,33,34,35,36,37,38,39,40.

Test Reliability
An instrument is considered reliable if the instrument produces same results every time when use to evaluate identical measurement. A Kuder Richardson formula, KR20 and KR21 for analyzing test items, which is based on item difficulty, was used to analyze internal consistency of the test. The value of KR20 and KR21 range between 0 to 1. The closer the value to 1 the better the internal consistency.
Where K= Number of test items Pj=Number of subject in the sample who answered item j correctly qj= Number of subject in the sample who didn't answered item j correctly σ 2= Variance of the total scores of all the subject taking the test Where K= Number of test items μ= Mean score σ 2= Variance of the total scores of all the subject taking the test The value KR20 =0.7966 and KR21 = 0.7515 shows that the test has high reliability.

Conclusions and Recommendations
Results of this study signify the importance of item analysis for constructing a useful, reliable and valid test. Its main focus was to manage item bank for individual discipline during an academic calendar so that a collection of items within acceptable range of difficulty index, discrimination power and suitable distracters obtained. In the present study an achievement test on a unit from subject of physics comprised of 40 test items was prepared and administered to check the level of difficulty, discrimination and standard of distracters. It clearly highlighted the poor, confusing items which need improvement. The findings of the present study clearly highlighted the usefulness of procedure of items analysis for subject teachers to develop quality test items in particular subject area. It is recommended that further studies should be conducted on different topics in physics and in other subjects like biology, chemistry and mathematics.