Small n designs
aka single-subject experiments
AB design
ABA (reversal) design
alternating treatments design (ACABCBCB)
multiple baseline
changing criterion design
Problems with small n designs
low generalizability
can’t do if IV is not reversible
Other small n designs
case studies
deviant case analysis
Threats to internal validity
History
Maturation
Testing
Instrumentation
Regression to the mean
Selection
Mortality
Quasi-experiments
No random assignment
Some participants receive "treatment"
Different from true experiments
Different from field experiments
Example
Why do quasi-experiments?
Advantages/disadvantages
higher external validity than exp.
lower internal validity than exp.
Types
Nonequivalent control group