A type of validity that is more from the
PSYCHOMETRIC PROPPERTIES: VALIDITY
perspective of the test taker as opposed to
THE CONCEPT OF VALID ITY the test user
Validity: A judgment or estimate of how Example: Personality tests
well a test measures what it purports to
Introversion-Extroversion test will be perceived as a
measure in a particular context.
highly (face) valid measure of personality functioning
Validation: The process of gathering and
evaluating evidence about validity. The inkblot test may not be perceived as a (face)
Both test developers and test users may valid method of personality functioning.
play a role in the validation of a test.
Local validation: Test users may validate a CONTENT VALIDITY
test with their own group of testtaker.
Content validity: A judgment of how
VALIDITY IS OFTEN CONCEPTUALIZED adequately a test samples behavior
ACCORDING TO 3 CATEG ORIES: representative of the universe of behavior
that the test was designed to sample.
Content Validity Do the test items adequately represent the
content that should be included in the test?
- This is a measure of validity based on an
Test blueprint: A plan regarding the types
evaluation of the subjects, topics, or
of information to be covered by the items,
content covered by the items in the test.
the number of items tapping each area of
Criterion Validity coverage, and the organization of the items
in the test.
- This is a measure of validity obtained by Culture and the relativity of content
evaluating the relationship of scores validity: The content validity of a test varies
obtained on the test to scores on other across cultures and time.
tests or measures.
CRITERION - RELATED VALIDITY
Construct Validity
A criterion is the standard against which a
- - This is a measure of validity that is arrived test or a test score is evaluated.
at by executing a comprehensive analysis Characteristics of a criterion: An adequate
of: criterion should be relevant to the matter at
a. How scores on the test relate to other test hand, valid for the purpose for which it is
scores and measures. being used, and uncontaminated, meaning
b. How scores on the test can be understood it is not based on predictor measures.
within some theoretical framework for
understanding the construct that the test
was designed to measure.
FACE VALIDITY
A judgment about the relevance of test
items.
Validity coefficient: A correlation Discriminant evidence: Validity coefficient
coefficient that provides a measure of the showing little relationship between test
relationship between test scores and scores scores and/or other variables with which
on the criterion measure. scores on the test should not theoretically
Validity coefficients are affected by be correlated.
restriction or inflation of range. Factor analysis: Class of mathematical
Incremental validity: The degree to which procedures designed to identify specific
an additional predictor explains something variables on which people may differ.
about the criterion measure that is not
explained by predictors already in use. RELATIONSHIP BETWEEN RELIABILITY AND
To what extent does a test predict the VALIDITY
criterion over and above other variables?
Reliability and validity are partially related
and partially independent.
CONSTRUCT VALIDITY
Reliability is a prerequisite for validity,
Judgment about the appropriateness of meaning a measurement cannot be valid
inferences drawn from test scores regarding unless it is reliable.
individual standings on a construct. It is not necessary for a measurement to be
If a test is a valid measure of a construct, valid for it to be considered reliable.
then high scorers and low scorers should
behave as theorized. VALIDITY AND TEST BI AS
All types of validity evidence, including
Bias: A factor inherent in a test that
evidence from the content and criterion-
systematically prevents accurate, impartial
related varieties of validity, come under the
measurement.
umbrella of construct validity.
- Bias implies systematic variation in test
scores.
EVIDENCE OF CONSTRUCT VALIDITY
- Prevention during test development is the
Evidence of homogeneity - How uniform a best cure for test bias.
test is in measuring a single concept. Rating error: A judgment resulting from the
Evidence of changes with age – Some intentional or unintentional misuse of a
constructs are expected to change over rating scale.
time (e.g., reading rate). - Raters may be either too lenient, too
Evidence of pretest-posttest changes - Test severe, or reluctant to give ratings at the
scores change as a result of some extremes (central tendency error).
experience between a pretest and a Halo effect: A tendency to give a particular
posttest (e.g., therapy). person a higher rating than he or she
Evidence from distinct groups - Scores on a objectively deserves because of a favorable
test vary in a predictable way as a function overall impression.
of membership in some groups. Fairness: The extent to which a test is used
Convergent evidence: Scores on the test in an impartial, just, and equitable way.
undergoing construct validation tend to
correlate highly in the predicted direction
with scores on older, more established tests
designed to measure the same (or a similar)
construct.