PSYB04H3 Lecture Notes - Lecture 11: Observer-Expectancy Effect, The Control Group, Random Effects Model
Document Summary
Chapter 11 more on experiments: confounding and obscuring variables. The really bad experiment (a cautionary tale: one-group, pretest/protest design: an experiment using an independepnt-groups design in which participants are tested on the key dependent variable twice: one before and once after exposure to the independent variable. Scenario 2: dr. yuki recruits 40 depressed women, all of who are interested in receiving psychotherapy to threat their depression, their measure of depression is measured using a standard depression inventory at the start of the therapy. For 12 weeks, all the women participate in cognitive therapy: after the 12 weeks, she measured the women again and finds that their levels of depression have significantly decreased. Six potential internal validity threats in one-group, pretest/protest designs. Twelve internal validity threats in total: design confounds, selection effects, order effects, maturation threats, history threats, observer bias, demand characteristics, and placebo effects.