This document discusses different types of test validity and reliability. It defines content validity, criterion-related validity including predictive and concurrent validity, construct validity, and face validity. It explains that validity is the degree to which a test measures what it claims to measure. Reliability refers to a test's consistency and includes test-retest reliability, equivalent forms reliability, internal consistency, and factors that can influence reliability such as test length, difficulty, and objectivity of scoring. The document provides examples and methods for establishing different types of validity and reliability.
Characteristics of a good psychological test santos
2. What
is the difference?
A.Criterion-related validity
B.Construct Validity
C.Content Validity
D.Face Validity
What are the factors that influence test
validity and reliability
3. This
is the degree to which the test
actually measures what it purpots to
measure.
It provides a direct check oh how well the
test fulfills its functions
Example:
The PUPCET is check with external criteria
like rating of teachers, success or failure,
grades or training.
4. When
test show consistent scores but it is
not useful then it is not valid or if it is not
consistent in whatever it is measuring then
it cannot be valid for any purpose.
6. CONTENT VALIDITY
The extent to which the content of test
provides an adequate representative of the
conceptual domain it is designed to cover,
test of achievement and ability are of this
type of examples
It is extablished through logical analysis.
There is a careful and critical examination
of the test items based on the objectives of
instruction.
7. CRITERION RELATED VALIDITY
Is established statistically through correlation
between the set of scores revealed by a test
and some other predictor or data collected by
other test or external measured
Evidence tells us just how a test corresponds
with particular criterion. Such evidence is
provided by high correlations between a test
and well defined criterion measures
8. A.Predictive
Validity
Evidence tells us the forecasting functions of
test. The criterion evidence are obtained in
the future generally months or years after
the test scores are known
Example:
SAT serves as predictive evidence as
college admission test it is forecast how
wel high school students do well in their
academic performance
9. B.Concurrent-related evidence
Comes from assessments of simulateous
relationship between the test and school
performance;in industrial setting job
samples to correlate with performance in
the job.
Appropriate for licensure examination,
achievement test and diagnostic test.
10. CONSTRUCT VALIDITY
Is determined by analyzing the
psychological qualities traits, or factors
measured by a test
Psychological constructs include
Intelligence, Mechanical ability, Perceptual
ability and critical thinking.
Construct is defined by showing relationship
between test
11. D.T. Campbell and Fiske (Kaplan, 2001)
Introduced an important set of logical
considerations for constructing evidence of
construct validity
A.Convergent Evidence-when a measure
correlate well with other test believe to
measure the same construct.
B.Discriminant Evidence-indicates that the
measure does not represent other than the
one for which it was devised
12. FACE
VALIDITY
Is the extent to which items on test appear
to be meaningful and relevant. It is merely
the appearance that a measure has
validity
13. A.
Appropriateness of test items
B.Direction
C.Reading vocabulary and sentence
structures
D. Difficulty of items
E.Construction of test items-no ambiguous
items or leading items
G. Length of the test-Sufficient length
H. Arrangement of items-no patterns T-T;FF; or A-B,A-B;C-d,C-D and so on.
14. This
refers to consistency of scores by the
same person when reexamined with the
same test on different occasions
16. Scores
Inconsistency
Scoring should be objective by providing a
scoring key so that corrector’s biases donot
effect examinee’s scores
Limited sampling of behavior
This happens with too much accidental
inclusion of certain items and the inclusion of
other.
Instability of examinee’s performance
Fatigue practice, and examinee’s mood are
among the factors that may affect
performance.
17. A. Measures of stability-often called test-retest
estimate or reliability and is obtained by
administering a test to a group of individuals,
re-administering the same test to the same
individual at a later date, and correlating the
two sets of scores. The correlation must be
high, .80 is a min figure. To be trustworthy two
testing administrations should be separated
by a least a three months gap subjects should
be large
18. B.
Measures of Equivalence
The equivalent forms of reliability is obtained
by giving two forms (with equal content,
means, and variance) of a test to the same
group of individuals on the same day and
correlate the results.
We can generalize a person’s score as to
what to receive taking similar test but
different questions.
19. Measures of Equivalence and Stability
Could he obtained by giving one form of the test and
after a period of time, administering the other form
and correlate the results.
This procedure allows for both changes in scores
due to trait instability and changes in scores due to
item specifity.
Measures of Internal Consistency
To be reliable internal consistency must be high,
coefficient alpha is bet index of internal
consistency a simple approximation to alpha is the
split half reliabilty
20. Split
half
Is a method of estimating reliability that is
theoretically the same as the equivalent forms
method. It is ordinarily considered as a
measure of internal consistency because the
two equivalent forms are contained within a
single test. One test is administered in
estimating reliability, one obtains a sub-score
for each of the two halves and two subscores
are correlated using Speaman-Brown
prophecy formula
22. The
administration, scoring, and interpretation
of scores are objective insofar as they are
independent of the subjective judgement of
the particular examiner
Objectivity of the examiner can be achieved if
a qualified test examiner who is competent
and responsible handles the test.
Major way of objective test is administration
of difficulty level of an item or of a whole test
based on objective, empirical procedues
employed by test developer.