PSY 130 Lecture Notes - Lecture 20: Discriminant Validity, Convergent Validity, Test Validity
Spring Semester
2018
Psych 130
Psychology
Personality Development and Effective Behavior
● Congruent Validity:
o Assume you are interested in using a new intelligence test that is apparently less
time-consuming to perform and more affordable than the most widely used test
o You would probably want to see how scores from this new test compare on an
existing intelligence test with scores from that
o Yet assume you gave a group of people both tests, and found a connection
between the test scores of just20
o Since scores on the two tests aren't strongly correlated, a individual may achieve
a high score on one intelligence test and a low score on the other, leaving you
wondering what the true measure of intelligence is
o This doesn't mean that the old scale is intellect, and the new scale isn't, just that
they can't both measure the same concept
o The congruent test validity, also referred to as convergent validity, is the degree to
which the test scores agree with other tests of the same type
o When two tests calculate the same thing, the scores from the two tests will be
very strongly linked
o Congruent validity evidence is not, however, limited to correlations with other
personality assessments
o For instance, if you wanted to assess the construct validity of a new anxiety
measure, you might equate test scores with rates of anxiety as assessed by a team
of qualified psychologists
o Discriminant Validity:
▪ Unlike congruent validity, discriminant validity applies to the fact that a
test score does not correspond with the scores of the potentially unrelated
tests
▪ Let's go back to developing a creativity test
▪ It is necessary to note that the test only tests creativity and not anything
like imagination, like intelligence
▪ You could give a group of people both the imagination test and a standard
intelligence test to determine discriminant validity
▪ If scores from the two tests are strongly correlated, someone might argue
that the creativity test does not at all assess creativity, but only intellect
▪ Unless the correlation between the two tests is weak, there is proof that
the structures of the two tests measure diff erent
▪ Notice that a low correlation does not tell you the measures of the study,
but rather what it does not calculate
▪ Nonetheless, this is an important step in the validity of the test model
o Compliance Validation
Document Summary
Congruent validity: assume you are interested in using a new intelligence test that is apparently less time-consuming to perform and more affordable than the most widely used test. You would probably want to see how scores from this new test compare on an existing intelligence test with scores from that. Yet assume you gave a group of people both tests, and found a connection between the test scores of just20. Since scores on the two tests aren"t strongly correlated, a individual may achieve a high score on one intelligence test and a low score on the other, leaving you wondering what the true measure of intelligence is. When two tests calculate the same thing, the scores from the two tests will be very strongly linked. Congruent validity evidence is not, however, limited to correlations with other personality assessments. Let"s go back to developing a creativity test.
