Test developers use the item analysis data to select items for the final test form and to
get rid of those that fail to display the required statistical properties. According to Nitko
(1983) item analysis data for teacher-made tests will be used for the number of
purposes that tally with its definition as a process of cataloging, recapitulating and
using information about individual test items, especially information about examinees
responses to items. Some of the importance have been highlighted below.
To begin with is that item analysis helps to figure out whether an item functions as
planned to. There is high possibility that an item cannot be expected to be written
perfectly in one attempt. For an item to be close to perfect, it should assess the
intended objectives, be of an appropriate level of difficulty, differentiate between those
who have mastered the learning objective from those who have not, have a correct
answer and at least have working distracters. Kubiszyn (1993) emphasises that
observation on how students respond to various items, can indicate clear picture that the
item is too easy or very difficult and if the students answer correctly or troubled in
other items can give clear determinant on what learned or not. This helps the item
developers to plan accordingly otherwise the intended purpose of the item might not be
achieved. For example, the pre-tests that offered at the beginning of the course or unit
determine the readiness of learning to aid in instructional planning and to make advance
placements. In case of the physics class, the physics class teacher may plan and select
items in Nuclear physics based on objectives of the topic so that the teaching and
learning procedures of the topic are clearly planned in advance based on fact that the
teacher is at least aware of students’ indigenous knowledge.
Item analysis also helps to give feedback to the teacher on learner’s difficulties.
According to Kubiszyn (2000), difficulty is a ratio or proportionality of students who
give correct responses. Any learning difficulties pupils have become clearly displayed
in item analysis. The analysis displays the percentage of pupils answering the item
rightly or wrongly and helps the teacher to identify the pupils’ problem areas more
easily. This offers differentiation between students who have greater knowledge and
lesser knowledge of the material used. In other words, analysing based on difficulty
index the exam is taken at the perception as: is the item or exam question too easy or
too hard? When the item is one that every student either gets wrong or correct, it
decreases an exam reliability. If everyone gets a particular answer correct, there is less
of a way to tell who really understands the material with deep knowledge. Conversely,
if everyone gets a particular answer incorrect, then there is no way to differentiate those
who have learned the material deeply. Another situation than that might require some
adjustment in difficulty level is when the item developer grades on a point or mastery
system. For example, suppose the grade point is taken as 90% being an A, 80% being
B, and so on. In this scenario, the item developer may not use items with an average
difficulty of say 55%, because the instructor may not obtain the desired distribution of
grades. This implies that in a point or a mastery system, items must be written so that
the distribution of scores result in the score desired distribution (Phatiki, 2011).
It provides clear criticism information to students about their performance reflection. In
a classroom discussions a teacher can use item analysis to categorise the level of
specifications and discrimination expected of students and reinforce good or correct
responses. Does an item discriminate between students who understand the material
and those who do not? Item discrimination emphasises that an item should evaluate the
varying degrees of knowledge students have on the material, reflect by the percentage
correct on the items. It can also be shown by comparing the correct responses to the
total test score of the students (Wiersma, 1990)). That is to say students who scored
high overall have a higher rate of correct responses on the item than those who scored
low overall. If the top scores are separated from the bottom scores, then which group is
getting which response correct?
Now, should all items discriminate? We can simply say yes if and only if the purpose of
the test is to rank students like assigning grades or placing them to the various sections
of the course because item that do not discriminate may contribute nothing to the test
except additional reading time. For example, if a course is conducted at form four level
in secondary school, the most crucial consideration might be: does the item measure the
prominent concept or can demonstrate some diplomatic or intellectual skills? In
attempt to answer this provoking question, the item developers may off course wish to
make sure that the item is well written and covers, an important educational objectives
(Nitko,1983).
Lastly is that item analysis helps to articulate very important areas of curriculum
improvement. Specific problems become clear when item analysis data is used. When
item developers observe that particular kind of items are repeatedly difficult for pupils,
or when certain kind of errors occur often, they opt or wish to review the curriculum.
The revision is usually based on data about pupil responses to and perceptions of other
important aspects to revise it. For example, if the item developers observed a great
number of students were doubtful failures to a certain item like item number six of the
Physics exam paper of final test at form four, the item developers may opt to either
removing or modifying it paraphrasing. This helps to improve the quality of the tests
and has to be done continuously. The improved items can be put in a file, called an item
bank, for future use (Harrow, 1972). Item distractor analysis is also helpful in that it can
help identify misunderstandings students have about the material. If the high percentage
of students selected the same incorrect multiple-choice answer, then that provides
insight into student learning needs and opportunities. Also it can be given a credit in the
sense that the destructors highlights student learning gaps and discriminates student
knowledge (Ott, 1992).
Based on the above discussed points someone can greatly conclude that item analysis
provide information about student’s learning that determine which and have not yet
learned the material either individually or the whole class which eventually promote
better plan for future instruction so as to facilitate and improve learning. This works
hand in hand with the part of item developer including the teacher who improves much
better in item writing skills and test construction which might be accomplished by
acquiring evidence that indicate which item is strong good positive and which one is
weak. Therefore, all item developers are kindly insisted to take the concept of item
analysis into account with prior and great consideration for it to be accurate, precise and
consistent otherwise the whole assessment based on its objectives might be messed up.
REFERENCE
Brookhart,, S. M., & Nitko, A. J. (2018). Education Assessment of Students (8th edition
ed.). New Jearsey: Merril Prentice Hall.
Harrow, A.J. (1972). A Taxonomy of the Psychomotor Domain: A Guide for Developing
Behavioural Objectives. New York: David McKay Company.
Hopkins, K.D., Stanley, J.C. & Hopkins, B.R. (1990). Education and Psychological
Measurement and Evaluation. Needham Heights, Massachusetts: Allyn and Bacon.
Kubiszyn, T. & Borich, G. (1993). Educational Testing and Measurement: Classroom
Application and Practice. New York: Harper Collins College Publishers.
Nitko, A.J. (1983). Educational Tests and Measurement: An Introduction. New York:
Harcourt Brace Jovanovich, Inc.
Nitko, A.J. & Mulgrave, N.W. (1983). Study Guide for Educational Test and Measurement:
An Introduction. New York: Harcourt Brace Jovanovich, Inc.
Ott, R.L., Rexroat, C., Larson, R. & Mendenhall, W. (1992). Statistics: A Tool for the Social
Sciences. Boston: PWS-Kent Publishing Company.
Phatiki, A., & Roever, C. (2011). Current issues and trends in language Assessment
Quarterly, 103-107.
Popham, W.J. (2008). Transformative Assessment. Alexandria: VA: ASCD.
Tavakol , M., & Dennick, R. (2011). Post examination analysis of objective tests. Med
Teach,, 33(6), 447-458.
Thompson , B., & Leviton, J. E. (1985). Using Microcomputers to score and evaluate test
items. Collegiate Microcomputers, 3, 163-168
Wiersma, W. & Jurs, S.G. (1990). Educational Measurement and Testing. Needham Heights,
Massachusetts: Allyn and Bacon
