Elidio Armando Miguel
Ferraz Fai Sufo Júnior
Guisado Jaime Duarte
Laila António Laila
Micail José André Caribo
Objective vs Subjective and Norm- Referenced vs. Criterion-Referenced Tests
(Licenciatura Degree in English Language Teaching )
Universidade Rovuma
Nampula
2021
1
Elidio Armando Miguel
Ferraz Fai Sufo Júnior
Guisado Jaime Duarte
Laila António
Micail José André Caribo
Objective VS. Subjective and Norm- Referenced VS. Criterion-Referenced Tests
(Licenciatura Degree in English Language Teaching)
Work to be submitted in the department
of Language science, communication
and arts, for the evaluation purpose of
Didactics of English 3
Lecturer: Samuel Canda
Universidade Rovuma
Nampula
2021
2
CONTENTS
INTRODUCTION.........................................................................................................................................4
OBJECTIVE vs. SUBJECTIVE AND NORM-REFERENCEDvs. CRITERION-REFERENCED . .5
TESTS.............................................................................................................................................................5
NORM-REFERENCEDvs. CRITERION-REFERENCED TESTS........................................................5
CHARACTERISTICS OF NORM-REFERENCED ANDCRITERION-REFERENCED TESTS.....6
CONSTRUCTIONOF CRITERION-REFERENCED TEST...................................................................8
CONCLUSION............................................................................................................................................10
REFERENCES.............................................................................................................................................11
3
INTRODUCTION
Many teachers develop classroom tests without spending enough time planning what they want
to do. With experience, they soon learn that planning is critical to developing high-quality tests.
Teachers need to spell out the purpose of the test, the material that will be covered, the types of
item that will be used, the difficult level of the items, and the time that is available for the test.
Testing, whether by tradition, conviction or practice is an indispensable part of teaching
methodology in the learning process. In general, it allows teachers to measure the knowledge
acquired by each individual student in the classroom, and also test the validity of didactic
strategies being used and to ascertain whether the proposed aims and objectives are being
achieved. By the same token, the effective use of this testing tool in the classroom setting
requires more experience, additional training, extra effort and a fresh commitment by teachers as
the main engine in the learning process. In this context, this paper aims at focusing on Objective
vs. Subjective Tests; and Norm-Referenced Test vs. Criterion-Referenced Test.
4
OBJECTIVE VS. SUBJECTIVE AND NORM-REFERENCEDvs. CRITERION-
REFERENCED TESTS
An educational test is not just a test that measures achievement in subjects of study, but it is also
a psychological test that leads to an assessment of the overall development of a student. KUMAR
and ROUT (2016:54).
Testing- in an educational system, is considered as an attempt to identify and explain the
effectiveness of a teaching program and an attempt to find out how much or little the student has
achieved in a course or learning.
OBJECTIVEVS. SUBJECTIVE TESTS
Objective test- a test or assessment that is composed of items for which the correct answers are
known in advance of test administration and are unaffected by scoring opinion or judgment.
Objective tests typically have “keyed” responses, such that the items can be scored mechanically
or by computer. Examples of objective tests are those that contain item formats such as multiple-
choice, selected-response, matching, true-false, or other forced choice type of item for which
there is an indisputable correct answer. HMH ASSESSMENTS (2014:18).
Subjective test- a test or assessment for which some subjectivity on the part of the examiner or
rater (as in a panel of score judges) is inherent in the determination of an examinee’s scored dara.
Projective devices, free response examinations, holistic evaluations, observation data, all require
some opinion or judgment on the part of the rater in the assessment of scores, even if “objective”
scoring guidelines are included as part of the process. HMH ASSESSMENTS (2014:27).
NORM-REFERENCED VS. CRITERION-REFERENCED TESTS
To understand non-referenced type of evaluation, we have to first learn about the term “norm”,
which has two meanings. One is the established customs of the society which most of the people
follow without question. The other meaning of the term, which is meaningful, is the average
performance of the group. KUMAR and ROUT (2016:49).
5
Norm-Referenced Test (NRT) is a test design to provide a measure of performance that is
interpretable in term of an individual‘s relative standing in some known group. BLERKOM
(2009:19).
Norm group may be made up of examinees at the local level, district level, state level or national
level. Since the development of norm-referenced tests is expensive and time consuming.
Bormuth
(1970) writes that Norms is to measure the growth in a student‘s attainment and to compare his
level of attainment with the levels reached by other students and norm group.
Criterion-Referenced Test (CRT) -a test designed to measure a student’s performance as
compared to an expected level of mastery, educational objective, or standard. The type of scores
resulting from this type of test provide information on what a student knows or can do in with
respect to a given content area as opposed to a score indicating how that student ranks among his
age or grade peers (norm group). HMH ASSESSMENTS (2014:8).
For VALKENBURGH (1974:7) currently, it is necessary to use norm-referenced tests to obtain
information for decision relating to placement, diagnosis, assessment, prediction, and evaluation.
Criterion-referenced tests are necessary, however, and are currently being used to obtain specific
information needed in decisions involving placement (of the student materials and methods
appropriate to his own skills and development in a particular subject), diagnosis, assessment,
and, to a limited extent, evaluation.
CHARACTERISTICS OF NORM-REFERENCED ANDCRITERION-REFERENCED
TESTS
We should not design norm-referenced and criterion-referenced tests in the same way.
For BLERKOM (2009:23) when designing a test, there are several characteristics that you should
keep in mind:
6
Item Difficulty
When using a norm-referenced approach you essentially want to rank students from those who
know the material well to those who are less prepared. You want to differentiate among the
students, to separate the best prepared and the least prepared students. In order to do this reliably
you need to see a large standard deviation, a large spread among the scores. If all of the scores
bunch up together, then it is difficult to tell whether differences in student performance on the
test are because of differences in the various students’ skills or as a result of simple measurement
error. However, if the scores are spread out further, you can feel more confident that the test is
truly differentiating among the students based on their skills. The amount of spread that you
obtain in the test scores is often related to the average item difficulty. For example, if all of the
items are relatively easy, most students will do quite well on the test and there will be little
spread in the scores. The same will happen if the items are all very difficult. Most students will
do poorly on the test and, again, there will be little spread in the scores. You can frequently
obtain the greatest spread of scores (the largest standard deviation) when most of the items are of
moderate difficulty. Therefore, when using a norm-referenced interpretation, you frequently want
the majority of your test items to be moderately difficult. In that case you will obtain a large
standard deviation and will be able to differentiate reliably among students. However, if you plan
to use a criterion-referenced interpretation, then you do not need to differentiate among the
students and item difficulty is somewhat less important.
Number of Items on the Test
In the perspective of BLERKOM (2009:24) In general it is a good idea to use as many test items
as time allows. Longer tests (those with more items) are almost always more reliable and have
greater content-related evidence of validity. The number of items needed on a criterion
referenced test is dependent primarily on the number of objectives covered by the test. You need
at least one item for each objective. With a fairly large number of objectives, the test will
typically have only one item for each objective. Time restrictions will simply not allow for more
items. However, having only one item per objective is not the ideal situation as greater reliability
and accuracy can be achieved by having more than one item for each objective. If the number of
7
objectives is relatively small, then it might be feasible to have two or more items for each
objective. With norm-referenced interpretations there are no hard and fast rules for the number of
items required. In this case you will be sampling objectives to cover on the test and will not need
to measure each and every objective. You could, theoretically get away with fewer items.
However, it is best to include as many items as time will allow.
Although many test items can be used appropriately on either norm-referenced or criterion
referenced tests, there are times when teachers might want to design items specifically based on
the type of interpretation they plan to use. When planning on the number of items to use on the
test they also may want to keep in mind how the test scores will be interpreted.
CONSTRUCTIONOF CRITERION-REFERENCED TEST
According to KUMAR and ROUT (2016:50) during the preparation of a criterion-referenced test,
the test constructor is required to take the following steps:
Step 1: Identifying the purpose of the test- the test developer must know the purpose of the test
and he should know the aspects, like:
• Content areas of the test from where the items will be developed;
• Level of students or examinees for whom test is being prepared;
• Difficulty level of the test items;
• Types of the test objective type or subjective type or mixed type of test;
• Criteria for qualifying the test
Step 2: Planning the test-
(i) Content analysis
(ii) Types of items- what is planned at this stage is the proportion of objective and
subjective type items in terms of marks. In case of subjective type, it may be essay
type, short-answers type and very short-answer type. In case of objective type, it may
be multiple-choice question, fill in the blanks, true or false type, sentence completion
type, one word answer type, etc. If the test is of mixed type, then questions are
developed accordingly.
8
(iii) No of items- total number of questions of each type which is to be included in the test.
(iv) Weightage
(v) Duration of the test
(vi) Mechanical aspects- the quality of paper, ink, diagrams, type setting, font size and
printing of the test papers.
(vii) Development of key objective scoring
(viii) Instructions for the test
Step 3: Preparing blueprint of the test
Blueprint is a specification chart which shows the details of the test items, all the content areas
and the number and type of questions from those areas. It reflects the objectives to be tested,
describes the weightage.
Step 4: Construction of test items- questions may be objective or subjective type as mentioned
in the blueprint. As objective-based and objective-type questions, multiple choice questions, true
false, matching, fill in the blanks, one word-order questions.
Step 5: Selecting the items for the test- the selection of the right item for the test is done
through a process known as “try-out”, and it involves the following steps:
(i) Sampling of subjects
(ii) Pre-try-out
(iii) Proper try-out
(iv) Final try-out
Step 6: Evaluating the test and preparing final draft of the paper- for establishing quality, is
prepared an index test manual about the test’s norm, scoring key, reliability and validity.
9
CONCLUSION
Teachers must make every possible effort to make students realize that a test is a worthwhile goal
for them. With this regard, when it is to testing, teachers should show that they really know what
they are doing, designing tests of objective type or subjective type or even more of a mixed type
where the question are developed accordingly. Thus, the norm-referenced and criterion
referenced tests should be administered by teachers wisely in the classroom with the intention of
measuring a student’s performance as compared to an expected level of mastery, educational
objective, or standard.
10
REFERENCES
BLERKOM, Malcolm L. Van. 2009. Measurement and statistics for teachers. University of
Pittsburgh at Johnstown.
VALKENBURGH, Marilyn W. Van. 1974. A Study of the Relationship between Norm
Referenced Tests and Criterion Referenced Tests. Western Michigan University
HMH ASSESSMENTS. 2014. Glossary of Testing, Measurement, and Statistical Terms.
Washington, DC: American Educational Research Association.
KUMAR, Dr. Harish and Santos Kumar Rout. 2016. Measurement and Evaluation in Education.
MA [Education], Second Semester, EDCN 804E, [English Edition], Directorate of Distance
Education, TRIPURA UNIVERSITY.
11
