PSYC 1000 Lecture Notes - Lecture 1: Predictive Validity, Publication Bias, Discriminant
Document Summary
Scientific research method: the difference between opinion, objectivity/subjectivity. Reliability: how stable/consistent are my outcomes, uses correlation coefficient= higher correlation=higher reliability. Different types: test-retest- taking the same test twice(e. x testing for depression, alternate forms-taking 2 different versions of a test. Internal consistency- a check to ensure all of the test items are measuring the concept they are supposed to be measuring. Split half- correlation -a test is split in two and the scores for each half of the test is compared with one another: tells us how measurable we are. Generalizability: how broadly can we apply our research to the population. Ideal: large randomly selected people: but not always feasible, demographics can also play a part, convenience samples- (e. x choosing 50 of your own students) Reducing bias: researcher-they might treat subjects differently depending on the measures, observation- (hawthorne effect- behaviour change b/c u find out your being watched), social.