PSYC 301 Lecture Notes - Lecture 6: Confidence Interval, Null Hypothesis, Causal Inference
20 Aug 2016
School
Department
Course
Professor
Document Summary
Chapter 6 - independent groups design internal validity. Differences in performance (dv) can be attributed unambiguously to effect of iv. Iv covaries with a different potential iv. Experiment free of confounds has internal validity. Hold conditions constant: iv is only factor that differs systematically across groups. Balancing: random assignment to conditions balances subject characteristics on average; groups are equivalent prior to iv; all subject variables are balanced independent groups design. Different individuals participate in each condition of the experiment (no overlap) Random groups design-individuals are randomly assigned to conditions of iv. Block randomization: random order of all conditions in the experiment. Randomly assign subjects one block at a time. Advantages: creates groups of equal size; controls for time-related events that occur during course of experiment. Natural groups design threats to internal validity. Extraneous variables are not controlled (hold conditions constant) Selective subject loss occurs (mechanical subject loss is not a problem)
