Intelligence Tests
Intelligence is often defined as a measure of general mental ability. Of the
standardized intelligence tests, those developed by David Wechsler are among
those most widely used. Wechsler defined intelligence as “the global capacity
to act purposefully, to think rationally, and to deal effectively with the
environment.” While psychologists generally agree with this definition, they
don't agree on the operational definition of intelligence
(that is, a statement of the procedures to be used to precisely define the
variable to be measured) or how to accomplish its measurement.
Test construction. To be useful, tests, including intelligence
tests, must be constructed using the established criteria of standardization,
reliability, and validity.
- Standardization is the process of making uniform and
objective both testing procedures and scoring procedures in order to
obtain meaningful scores. Scores on standardized tests are interpreted in
reference to scores obtained from a standardization sample,
that is, scores from a comparable group of subjects tested under
appropriate conditions.
- The term reliability refers
to the consistency of results. Reliability of a test is determined by one
of the following methods.
- test and retest reliability: comparison of original test scores with
retest scores
- alternate form reliability: comparison of scores obtained on alternate
forms of a test
- split‐half reliability: comparison of scores obtained on two halves
of tests (such as scores on odd‐ versus even‐numbered questions)
- The
term validity refers to the extent that a test measures
what it is supposed to measure. Types of validity include
- content validity: the extent to which a test reflects a sample
of the behavior to be measured
- predictive validity: the extent to which a test can predict a
person's behavior in another situation
- face validity: how appropriate a test “appears” to be, just
from the way the items read
- construct validity: how well a test assesses the construct (for
example, intelligence) for which it was designed
- concurrent validity: how well the results of a test agree with
those of a new test or a different form of the test measuring for the same
construct (for example, intelligence)
