PSYC 2030 Chapter Notes - Chapter 4: Criterion Validity, Convergent Validity, Discriminant Validity

• School: York University
• Department: Psychology
• Course: PSYC 2030
• Professor: John Grundy

Document Summary

While the constructs cannot be directly observable-each must be inferred from the measures used to investigate it. A measure of behavior is said to be reliable if its results are repeatable when the behaviors are remeasured, without it there is no way of determining wht a score on a particular measure means. A behavior measures reliability is a direct function of the amount of measurement-error present. (great deal of error = low reliability; vice versa). Reliability also allows the researcher to make more meaningful comparisons with other sets of scores (high similarity = strong correlation). Confidence in the reliability of a measure develops over time, a benefit of the replication process. A behavioral measure is said to be valid if it measures what it has been designed to measure. Conceptually, the simplest level of validity is called content validity: whether or not the actual content of the items o(cid:374) a test (cid:862)(cid:373)akes se(cid:374)se(cid:863) i(cid:374) te(cid:396)(cid:373)s of the (cid:272)o(cid:374)st(cid:396)u(cid:272)t (cid:271)ei(cid:374)g (cid:373)easu(cid:396)ed.