PSYCO212 Lecture Notes - Lecture 10: Confounding, Latin Square, Demand Characteristics
Document Summary
True experimental design experiment where you can manipulate the iv : basic experiment have one iv and one dv . Steps: choose participants and assign them to the levels of the iv, operationally de ne the iv *at least 2 levels, operationally de ne the dv *for comparison. Methods of assigning participants: independent groups design one to one condition aka between-subjects. Solomon four-group design *using latin square: assess impact of pretest by treating pretest as a second iv. Half participants only receive post-test: *no impact: post-test scores would be similar, *an impact of pretest: difference in post-test scores, solutions: Deception: embed test with irrelevant measures, repeated measures design one to all conditions within-subjects. (-) order effects: practice effects, fatigue effects, contrast effects, solutions: Counterbalancing *complete or partial: complete all possible orders are testes , partial only necessary orders are tested based one two criteria *see text book .