Suppose you develop a new self-report measure of the broad personality trait of extraversion. Distinguish between reliability and validity. Describe how you would establish the reliability and validity of your new test.

Question

What will be an ideal response?

Answer 1

Reliability refers to the consistency with which a test measures a construct such as intelligence. In other words, the same test taken at different times should produce the same results. Validity refers to the degree that a test actually measures what it intends to measure. One way to establish reliability might be to give a sample the test twice, weeks or months apart. If the test is reliable, both of an individual's
scores should be nearly identical. This is test-retest reliability. Another way to establish reliability would to be to compute split-half reliability. Scores on one half of the tests items--the even-numbered items, for instance--should correlate highly with scores on the other half of the items--the odd-numbered items in this example. One could establish a new test's validity by showing that scores on the test correlates at least moderately with scores on measures of personality traits related to extraversion, such as sociability, impulsivity, sensation-seeking, and so on. The validity of a new test might also be established by showing that scores on the new test correlate well with scores on already well-established tests of extraversion.