PSYC 3530H Lecture Notes - Lecture 3: Primum Non Nocere, Institutional Review Board, Dependent And Independent Variables
Document Summary
Selection of participants: representativeness through random selection, problems with selection (example: selection bias). Content validity refers to whether the content of a measure corresponds to the content of the attribute of interest. Construct validity refers to whether a measure corresponds to the construct (concept) underlying the attribute of interest. Face validity refers to whether a measure, on its surface, seems appropriate to the attribute of interest. Concurrent validity refers to whether the scores on a measure correlate with scores on another acceptable measure of the attribute of interest. Predictive validity refers to whether the scores on a measure predict later scores on another acceptable measure of the attribute of interest or other outcomes of interest. Extent to which other explanations can be ruled out: external: Research methods no single way to categorize. Useful distinction: descriptive portray phenomenon (ie: naturalistic observation), nonexperimental may involve statistical and other analysis (ie: correlations), experimental come closest to establishing cause and effect.