SlideShare utilise les cookies pour améliorer les fonctionnalités et les performances, et également pour vous montrer des publicités pertinentes. Si vous continuez à naviguer sur ce site, vous acceptez l’utilisation de cookies. Consultez nos Conditions d’utilisation et notre Politique de confidentialité.
SlideShare utilise les cookies pour améliorer les fonctionnalités et les performances, et également pour vous montrer des publicités pertinentes. Si vous continuez à naviguer sur ce site, vous acceptez l’utilisation de cookies. Consultez notre Politique de confidentialité et nos Conditions d’utilisation pour en savoir plus.
item or a combination
A test may consist of a single
of items. Regardless of the number of items in a test,
every single item should possess certain
Having good items , however does not necessarily
lead to a good test, because a test as a whole is more
than a mere combination of individual items.
Therefore, in addition to having good items, a test
should have certain characteristics.
1.Reliability 2.Validity 3.Practicality
There are different theories to explain the
concept of reliability in a scientific way.
Firs and simplest: A test is reliable if we
get the same results repeatedly.
Second: when a test gives consistent
Third: reliability is ratio of true score
variance to observed score variance.
order to explain the
concept of reliability in a
non-technical term is to say:
imagine that feeling that
someone did not as well on
a test as he could have.
Now imagine if he could
take the test again, he
This may be quit true.
Nevertheless, one should
also admit that some
factors, such as a good
chance on guessing the
correct responses, would
raise his score higher than
it should really be. Seldom
ow, if one could take a test
over and over again, he would
probably agree that his average score
over all the tests in an acceptable
estimate of what he really know or
how he really feels about the test.
On a “reliable test, one’s score on its
varies administrations would not
differ greatly. That is, one’s score
would be quit consistent.
On an “unreliable” test, on the other
hand one’s score might fluctuate
from one administration to the
other. That is, one’s score on various
The notion of consistency of one’s
score with respect to one's average
score over repeated administration is
the central concern on the concept of
change in one’s
Some of the changes
steady increase in one’s
score. The increase
would most likely be
due to some sort of
learning. This kind of
change, which would
be predictable, is called
contributes to the
reliability and the
variation, which is
contributes to the
unreliability of a
takes a test. Since all
Let’s assume that someone
measurement devices are subject to error, the
score one gets on a test cannot be true
manifestation of one’s ability in that particular
trait. In other words, the score contains one’s true
ability along with some error. If this error part
could be eliminated, the resulting score would
represent an errorless measure of that ability. By
definition, this errorless score is called a “true
The true score is almost always different from the score
one gets, which is called the “observed score”. Since the
observed score includes the measurement error, i.e., the
error score, it can be grater than, equal to, or smaller
than the true score. If there is absolutely no error of
measurement, the observed score will equal the true
score. However, when there is a measurement error,
which is often the case, it can lead to an overestimation
or an underestimation of true score. Therefore, if the
observed score is represented by X, the true score by T
and the error score by E, the relationship between the
observed and true score can be illustrated as follows:
Standard Error of
It is necessary to find an index of error in
measurement which could be applied to all occasions
of a particular measure. This index of error is called
standard error of measurement, abbreviated as SEM.
By definition, SEM is the standard deviation of all
error score obtained from a given measure in
In this method reliability is obtained through
administrating a given test to a particular group
twice and calculating the correlation between the
two sets of score obtained from the two
Since there has to be a reasonable amount of time
between the two administrations, this kind of
reliability is referred to as the reliability or
consistency over time.
Disadvantages of Test-Retest
It requires two administrations.
Preparing similar conditions under which the
administration take place adds to the complications
of this method.
There should be a short time between to
administration. Although not too short nor too long.
To keep the balance it is recommended to have a
period of two weeks between them.
In the parallel-forms method, two similar, or parallel forms of the
same test are administrated to a group of examinees just once.
The problem here is constructing two parallel forms of a test which is a
difficult job to do.
The two form of the test should be the same. It means all the elements
upon which test items are constructed should be the same in both
forms. For example if we are measuring a particular element of
grammar, the other form should also contain the same number of
items on the same elements of grammar.
Subtests should also be the same, i.e., if one form of the test has tree
subsection of grammar, vocabulary, and reading comprehension, the
other form should also have the same subsections with the same
In split-half method the items comprising a test are
homogeneous. That is, all the items in a test attempt
to measure elements of a particular trait, E.g., tenses,
propositions, other grammatical points, vocabulary,
reading and listening comprehension, which are all
subparts of the trait called language ability.
In this method, when a single test with
homogeneous items is administrated to a group of
examinees, the test is split, or divided, into two equal
halves. The correlation between the two halves is an
estimate of the test score reliability.
In using this method, two main points should be
taken into consideration. Firs, the procedure for
dividing the test into two equal halves, and
second, the computation of total test reliability from
the reliability of one half of the test.
In this method easy and difficult items should be
equally distributed in two halves.
Split-Half Advantages and
Advantages: it is more practical than others. In using the
Split-Half method, there is no need to administer the
same test twice. Nor is it necessary to develop two
parallel form of the same test.
Disadvantages: the main shortcoming of this method is
developing a test with homogeneous items because
assuming the quality between the two halves is not a safe
assumption. Furthermore, different subsections, in a test
,e.g., grammar, vocabulary, reading or listening
comprehension, will jeopardize test homogeneity, and
thus reduce test score reliability.
Which method should we use?
It depends on the function of the test.
Test-retest method is appropriate when the
consistency of scores a particular time interval
(stability of test scores over time) is important
The Parallel-forms method is desirable when the
consistency of scores over different forms is of
When the go-togetherness of the items of a test is of
significance (the internal consistency), Split-Half
and KR-21 will be the most appropriate methods.
Factors Influencing Reliability
To have a reliability estimate, one or two sets of
scores should be obtained from the same group of
testees. Thus, two factors contribute to test reliability:
the testee and the test itself.
The Effect of Testees
Since human beings are dynamic creatures, the attributes
related to human beings are also dynamic. The
implication is that the performance of human beings
will, by their very nature, fluctuate from time to time, or
from place to place. (e.g., students misunderstanding or
misreading test directions, noise level, distractions, and
sickness) can cause test scores to vary.
Heterogeneity of the Group Members.The greater the
heterogeneity of the group members in the
preferences, skills or behaviors being tested, the greater
the chance for high reliability correlation coefficients.
The Effect of Test Factors
Test length. Generally, the longer a test is, the more
reliable it is, however the length is up to a point.
Speed. When a test is a speed test, reliability can be
problematic. It is inappropriate to estimate reliability
using internal consistency, test-retest, or alternate form
methods. This is because not every student is able to
complete all of the items in a speed test. In contrast, a
power test is a test in which every student is able to
complete all the items.
Item difficulty. When there is little variability among test
scores, the reliability will be low. Thus, reliability will be
low if a test is so easy that every student gets most or all of
the items correct or so difficult that every student gets
most or all of the items wrong.
The Effect of Administration
• Poor or unclear directions given during
administration or inaccurate scoring can affect
For Example - say you were told that your scores on
being social determined your promotion. The result
is more likely to be what you think they want than
what your behavior is.
The Influence of Scoring Factors
the likes and dislikes of
In an objectively-scored test,
the scorers will not influence the results.
In a subjectively-scored test, the likes and dislikes of
the scorers will influence the results and as a result
Intra-rater errors (Errors which are due to
fluctuations of the same rater scoring a single test
Inter-rater errors (Errors which are due to the
fluctuations of different scorers-at least two- scoring
a single test.
The second major
characteristic of a good test is
What does validity mean?
A test is valid if it
measures what we want it
to measure and nothing
The extent to which a test
measures what it is
supposed to measure or
can be used for the
purposes for which the test
Validity is a more-testdependant concept but
reliability is a purely
So, validity refers to the
extent to which a test
measures what it is
supposed to measure.
There are four types of
Types Of Validity
Relevance of the test item to the purpose of the test.
Does the test measure the objectives of the course?
It refers to the correspondence (agreement) between
the test content and the content of materials (subject
matter and instructional objectives) taught to be
The extent to which a test measures a representative
sample of the content to be tested at the intended
level of learning.
Content Validity is called appropriateness of the test;
that is appropriateness of the sample and the
Content Validity is the most important type of
validity which can be achieved through a careful
examination of the test content.
It provides the most useful subjective information
about the appropriateness of the test.
Criterion-related Validity investigates the correspondence
between the scores obtained from the newly-developed test
and the scores obtained from some independent outside
The newly-developed test has to be administered along with
the criterion measure to the same group.
The extent to which the test scores correlate with a relevant
Refers to the extent to which different tests intended to measure
the same ability are in agreement.
Depending on the time of administration, two types exist:
Correlation between the test scores (new test)
with a recognized measure taken at the same
Comparison (correlation) of students' scores
with a criterion taken at a later time (date).
Refers to measuring certain traits or theoretical construct
Refers to the extent to which the psychological reality of a
trait or construct can be established.
It is based on the degree to which the items in a test
reflect the essential aspects of the theory on which the test
is based on.
Construct validity also refers to the accuracy with which
the test measures certain psychological/theoretical traits
Reading comprehension – Oral language ability.
This is done through factor analysis
Factors Affecting Validity
a. Directions (clear and simple)
b. Difficulty level of the test (not too easy nor too
c. Structure of the items (poorly constructed and/or
ambiguous items will contribute to invalidity)
d. Arrangement of items and correct responses
(starting with the easiest items and ending with the
difficult ones + arranging item responses randomly
not based on an identifiable pattern)
Validity and Reliability
Reliability is a purely statistical parameter; that is, it
can be determined fairly independently of the test.
But Validity is a test-dependent concept.
We have degrees of validity:
very valid, moderately valid, not very valid
A test must be reliable to be valid, but reliability
does not guarantee validity.
Reliability, Validity and
How reliable and valid should a test be?
The more important the decision to be made, the
more confidence is needed in the scores, an thus, the
more reliable and valid test are required.
Nevertheless, it is a generally accepted tradition that
validity and reliability coefficients below 0.50 ( low )
0.5 to 0.75 ( moderate), 0.75 to 0.90 ( high )
practicality refers to the
ease of administration
and scoring of a test.
Ease of administration
It is = Clarity, simplicity and the ease of reading
Fewer numbers of subtests
The time required for test
Ease of scoring
A test can be scored subjectively or
Since scoring is difficult and time
consuming, the trend is toward
objectivity, simplicity and machine
Ease of Interpretation and
The meaningfulness of scores obtained from that
If the test results are misinterpreted or
misapplied, they will be of little value and may
actually be harmful to some individual or group.