Item analysis evaluates the effectiveness of test items, helping to identify the difficulty level and discriminating power of questions based on student performance. An effective item should have a difficulty index between 0.20 and 0.80, and a discrimination index closer to +1 indicates better differentiability between high and low-performing students. Items that do not meet these criteria, or for which distractors are ineffective, should be revised or eliminated to enhance test quality.

1. Item
analysis

5. Item Analysis can help
you evaluate how well
your objective items are
actually working.

6. These problems can be
corrected, resulting in
a better test, and
better measurement.

7. refers to a statistical
technique that helps
instructors identify the
effectiveness of their
test items.

8. indicates the
proportion of
students who got
the item right.

9. A high percentage
indicates an easy
item/question and a low
percentage indicates a
difficult item.

10. In general,
items should
have values of
difficulty no
less than 20%
correct and
no greater
than 80%.

11. Very difficult
or very easy
items
contribute
little to the
discriminating
power of a
test.

12. each of these groups
consists of
twenty-seven
percent of the
total group of
students who took
the test and is based
on the students’
total score for
the test

13. It is the difference
between the proportion
of the top scorers who
got an item correct
and the proportion of the
bottom scorers who got
the item right

14. each of these groups
consists of twenty-seven
percent of the total
group of students who
took the test and is based
on the students’ total
score for the test

15. The
discrimination
index range is
between -1 and +1

16. The closer the index is
to +1, the more
effectively the item
distinguishes between
the two groups of
students

17. Sometimes
an item will
discriminate
negatively.

18. Such an item should be
revised and eliminated
from scoring as it
indicates that the
lower performing
students actually
selected the key or
correct response more
frequently than the top
performers.

19. It is another useful step in
reviewing the effectiveness
of a test item. All of the
incorrect options, or
distractors, should
actually be distracting.

20. Preferably, each
distracter should be
selected by a greater
proportion of the lower
scorers than of the top
group.

21. In order for a
distractor to
be acceptable
it should
attract at
least one
candidate.

22. If no one selects
a distractor it is
important to
revise the option
and attempt to
make the
distractor a
more plausible
choice.

23. Item
analysis

24. WritingequationsinExcel
•
•
•

25. WritingequationsinExcel
•
•
•
•

26. 1. Arrange the test scores
from highest to lowest.
2. Select the Criterion Group
- identify the high group
(top 27%)
and low group
(lowest 27%)

27. Note:
Top 50% and
bottom 50% for a
class with less
than 30 students
Top 27% and
bottom 27% for a
class with 30 or
more students

28. 3. For each item, count the
number of examinees in the
upper group who have
correct responses. Do a
separate, similar procedure
for the lower group.

29. 4. Solve for the difficulty index
of each item:
(for test items scored with 0 and 1 like matching
type, multiple choice and true or false )
n
cLcH
p
2


Index of difficulty of each item
No. of students in the high group who got correct responses
on the particular item
No. of students in the low group who got correct responses
on the particular item
n
cL
cH
p
No. of students in the criterion group

30. 5. Difficulty index (p) scale:
below 0.25 item is very difficult
0.25 - 0.75 item is of average
difficulty
Above 0.75 item is easy

31. 6. Solve for the discrimination
index (D) of each item:
(for test items scored with 0 and 1 like matching
type, multiple choice and true or false )
n
cLcH
D


n
cL
cH
D
No. of examinees in the high group who have correct responses
to the particular item
Is the discrimination index of a particular item
No. of examinees in the criterion group
No. of examinees in the low group who have correct responses
to the particular item

32. 7. Discrimination index (D) scale:
0.40 and above very good item
0.30 - 0.39 reasonably good item
but possibly subject
for improvement
0.20 – 0.29 marginal item, subject
to improvement
Below 0.20 poor item, to be
rejected or revised