PSCH 242 Lecture Notes - Lecture 2: Random Assignment, Confounding
Document Summary
Random sampling: every participant in target population has equal chance of being sampled. Stratified random sampling: random sampling within subgroups of target population. Ad hoc: random sample from accessible population, must generalize with care, should describe sample to help define limits of generalization. Free random assignment: random assignment of participants to conditions. Matched random assignment: random assignment of participants in matched sets to conditions, where confound may happen: broad classes of control procedures. Maximise experimental variance: real differences between the groups. Controlling extraneous variance: make sure the groups are as similar as possible, only difference is level of manipulation(iv) Minimizing error variance: control with careful measurement/special designs. Single-group, post test only design(sgpt): draw a sample, apply treatment, measurement. No control, change in response to treatment at all is in question. Single group, pretest-post test design(sgpptd): pretraining measures, relaxation training, post training measures, compare. Pretest helps to document change but there is no control condition.
