Test Item Analysis for its Effectiveness
In this blog I introduce the following key terms used for
finding the level of questions set by an examiner in an examination and the
perception of the students about it. These terms are the facilitation value and
the discrimination value.
1.
Introduction
A.
Facilitation Value: The
facilitation value of a question indicates how easy the question is. The
facilitation value is calculated by using the following formula.
FV= (RU + RL)(2xN*marks of test item)
Where N= no. of students in a group. The
value is chosen as 30% of the total students from top scorer and 30% of
students from low scorers.
RU = sum of marks obtained by upper group
RL = sum of marks obtained by lower group
When FV=1 means that all students
in lower group and all students in upper group answered the test item. So it
can be so assumed that the test item was very easy.
When FV=0, means nobody answered
the question, it means that either the question was so tough that no one was
able to answer the question or else there was a problem in understanding the
topic by the students which further mean that there might be problem in
teaching learning methodology adopted by the faculty.
A.
Discrimination Index:
Discrimination value is used in used to discriminate among good and the bad
students. This value also indicate how good the test item was in deciding among
the good and the bad students.
The formula for finding the decimation value is:
DI = (RU – RL) / (N*marks of the test item)
DI=1 when only the upper group answered the question
DI=0 when equal number of good and equal number of weak student answered
the test ited and scored the same marks. So the test item is so ambiguous that
it is not able to distinguish among the good and weak students.
DI=-1;
This is when only the weak students answered the test item but all the good
students failed to answer the test item.
It may be that weaker students answered by cheating or by some other
means.
When FV=1, DI is definitely ‘0’
but when DI=1, FV is 0.5.
2. Analysis of the Test Item
For the purpose of calculations, a class size of twenty
student and their performance in some course is taken. The score of a class
obtained by totawas taken
Test Item-->
|
1A
|
1B
|
1C
|
1D
|
1E
|
2A
|
2B
|
Total
|
Student Roll No.
|
2
|
2
|
2
|
2
|
2
|
5
|
5
|
Total
|
1
|
2
|
2
|
2
|
2
|
2
|
5
|
4
|
19
|
2
|
2
|
2
|
2
|
2
|
2
|
4
|
4
|
18
|
3
|
2
|
1.5
|
1.5
|
2
|
2
|
5
|
3
|
17
|
4
|
2
|
2
|
1.5
|
2
|
1.5
|
4
|
4
|
17
|
5
|
2
|
2
|
2
|
2
|
2
|
3
|
3
|
16
|
6
|
2
|
2
|
0.5
|
1.5
|
2
|
4
|
4
|
16
|
7
|
2
|
2
|
|
2
|
2
|
5
|
3
|
16
|
8
|
2
|
1
|
2
|
2
|
2
|
2.5
|
4
|
15.5
|
9
|
2
|
2
|
|
1.5
|
0.5
|
5
|
4
|
15
|
10
|
2
|
2
|
1.5
|
1.5
|
2
|
3
|
3
|
15
|
11
|
1.5
|
2
|
2
|
1.5
|
1.5
|
3
|
3.5
|
15
|
12
|
2
|
2
|
2
|
2
|
1
|
2.5
|
3.5
|
15
|
13
|
2
|
1
|
0
|
1.5
|
1.5
|
5
|
3.5
|
14.5
|
14
|
1.5
|
2
|
0.5
|
2
|
1.5
|
2.5
|
4
|
14
|
15
|
1.5
|
1.5
|
0
|
2
|
2
|
2.5
|
3.5
|
13
|
16
|
0
|
2
|
2
|
2
|
2
|
3
|
2
|
13
|
17
|
1
|
2
|
0.5
|
1
|
2
|
2.5
|
4
|
13
|
18
|
|
2
|
|
1.5
|
2
|
2
|
5
|
12.5
|
19
|
|
2
|
1.5
|
2
|
2
|
4
|
|
11.5
|
20
|
2
|
2
|
2
|
2
|
2
|
|
|
10
|
Index
|
1A
|
1B
|
1C
|
1D
|
1E
|
2A
|
2B
|
|
FV
|
0.69
|
0.96
|
0.65
|
0.92
|
0.98
|
0.65
|
0.61
|
|
DI
|
0.63
|
0
|
0.29
|
0.08
|
-0.04
|
0.37
|
0.25
|
|
Conclusion:
As per the observation from the above representation, it can
be concluded that some test items were very easy; some test items were very
ambiguous as they were not able to discriminate among the good and weak
students. The table below summarizes these facts.
Test Item-->
|
1A
|
1B
|
1C
|
1D
|
1E
|
2A
|
2B
|
How easy is the
question
|
Not very Easy
|
Quite easy
|
Not very Easy
|
Quite easy
|
Very Easy
|
Not very easy
|
Not very easy
|
How Good the Question is in
differentianing Good and weak students
|
Almost equal number of students in both upper and lower group
answered
|
Very Tough Question
|
Quite tough question
|
Quite Tough question
|
Ambiguous question No able to discriminate among good and weak
students
|
Not Very Good as it does not clealy discrimate good and weak
students
|
Not Very Good as it does not clealy discrimate good and weak
students
|
Comments
Post a Comment