Ch. 7: Randomized Experiments and Causal Inference
Randomized Experiments Experiments where participants are randomly assigned to the experimental groups or conditions. Often referred to as “true experiments.”
Reasons for Using Random Assignment Provides a safeguard against biased assignment of sampling units to the different treatment groups. Distributes the characteristics of the sampling units over the different conditions to prevent biased outcomes. Permits the use of statistical analyses that require certain data characteristics.
Ways of Achieving Random Assignment Presorting booklets or questionnaires Blindly drawing names Flipping a coin Consulting a table of random numbers
Between-Subjects Designs Subjects are exposed to one condition each. Also called nested designs. Condition A Condition B Subject 1 Subject 2 Subject 3 Subject 4 Subject 5 Subject 6 Subject 7 Subject 8 Subject 9 Subject 10
Within-Subjects Design Subjects are exposed to each condition. Also called: Repeated-measures design Crossed design Importance of counter-balancing Condition A Condition B Subject 1 Subject 2 Subject 3 Subject 4 Subject 5 Subject 6 Subject 7 Subject 8 Subject 9 Subject 10
Order of administration Latin Squares Design Order of administration 1 2 3 4 Sequence 1 A B C D Sequence 2 Sequence 3 Sequence 4
Factorial Designs Design has more than one factor and two or more levels of each factor Manipulated conditions Gender Drug Placebo Women A B Men C D
Aristotle’s Four Kinds of Causation Material Formal Final Efficient
Three Criteria of Efficient Causation Covariation Temporal precedence Internal validity
Mill’s Methods Method of agreement Method of difference If X, then Y. X is a sufficient condition of Y. Method of difference If not-X, then not-Y. X is a necessary condition of Y.
Mill’s Methods and the Simple Randomized Design Experimental Group Control Group If X, then Y. (Method of Agreement) If not-X, then not-Y (Method of Difference)
Solomon Four-Group Design Pretested? Experimentally treated? Yes No Group I Group II Group III Group IV
Plausible Causal Events in the Solomon Design Plausible effects Causal events Group I Group II Group III Group IV Pretest Yes No Treatment Sensitization Extraneous effects
Diagramming the Solomon Design Group I R O X Group II Group III Group IV R = Randomization O = Observation X = Treatment exposure
Preexperimental Designs One-shot case study: Symbolized as X-O where X = treatment exposure and O = observation One-group pre-post design Symbolized as O-X-O
Examples of Potential Threats to Internal Validity History Maturation Instrumentation Selection
The Social Psychology of the Experiment Artifact: A finding resulting from conditions other than those intended by the experimenter. Demand characteristics & the good subject Use of quasi-control subjects Experimenter Expectancy Effect Use of blind experimenters and double-blind procedures
Basic Expectancy Control Design Expectancy conditions Treatment conditions Experimental treatment Control treatment Experimental Group A Group B Control Group C Group D
Burnham’s (1966) Use of the Expectancy Control Design Expectancy conditions Treatment conditions Lesioning of brain No lesioning of brain Row means 46.5 49.0 47.75 No lesioning of brain 48.2 58.3 53.25 Column means 47.35 53.65