Chapter 8 Experience carefully exploited: Quasi-experimental research designs
Why Quasi-Experiments? Maximize internal and external validity When random assignment can’t be done When random assignment shouldn’t be done
Kinds of Quasi-Experiments Person-by-treatment quasi- experiments Natural experiments Nature and treatment designs –Natural groups with experimental treatment –Nature-by-treatment studies
Comparability and Patching Comparable and non-overlapping groups Patched-up designs: –One group, pretest-posttest design –Posttest-only with nonequivalent groups –Pretest-posttest with nonequivalent groups Time-series designs Internal analyses
When True Experiments and Quasi-Experiments Collide Consistencies and inconsistencies in the results of true experiments compared to quasi-experiments
Key Terms from Chapter 8 Person-by-treatment quasi-experiment Prescreening Extreme groups Median split Natural experiments Natural groups with experimental treatment Comparable and non- overlapping groups Patched-up design Patching One-group design One-group pretest- posttest design Posttest-only design with non-equivalent groups
Key Terms from Chapter 8 Pretest-posttest with non-equivalent groups Time-series design Internal analysis Name-letter effect