6: Quasi-Experimental Designs

quasi-experiment has a ______ setting and ____ control

natural, low

things that make a true experiment

• manipulation of independent variable (creation of groups/ conditions

• random assignment to experimental groups (counterbalancing for conditions)

quasi-experiments

manipulation of (sometimes naturally occurring) IV (existence of groups/conditions) but lack random assignment

• NO random assignments to groups (between-subjects)

  • groups are pre-exiting or participants self-select into groups

• NO random assignment to counterbalancing conditions (within-subjects)

• less control compared to experiments

  • more threats to research validity (particularly internal validity)

types of quasi-experimental designs

nonequivalent control groups designs (between-subjects)

• nonequivalent groups pretest-posttest

• delayed control group designs

designs without control groups (within-subjects)

• one-group pretest-posttest

• interrupted time-series designs

• repeated treatment designs

nonequivalent control group designs (mixed)

• (some) mixed factorial designs

• multiple time-series designs

nonequivalent groups pretest-posttest designs

• allocation to group 1 and 2: any nonrandom method

• ceiling effects and floor effects may render results uninterpretable

delayed control group designs

testing of one group is deferred

• two groups are tested sequentially with an appreciable time interval between them

one-group pretest-posttest designs

• pretest, treatment, and posttest applied to group 1

interrupted time-series design

same group is compared over time

• consider the trend of the data before and after the treatment

repeated treatment designs

compare participant responses before and after repeated treatments

mixed factorial designs

at least one between-subjects variable and one within-subjects variable

multiple time-series designs

interrupted time-series design with a nonequivalent control group