Lecture 4B: Quasi-experimental Studies Flashcards
What are the major characteristics of quasi-experimental design?
- No random assignment to groups
- No control/comparison group
What is the weakest design among all quasi-experimental designs?
One group pretest-posttest design
List the five types of quasi-experimental designs.
- One group pretest-posttest design
- Interrupted time series
- Patient as own control
- Historical controls
- Non-equivalent pretest-posttest control group design
What is a disadvantage of the one group pretest-posttest design?
- Lack of control group
- History effect
- Maturation effect
- Testing/Practice effect
Describe the interrupted time-series design.
One treatment group with multiple pretests and multiple posttests after intervention.
What are the advantages of the interrupted time-series design?
- Lessen the effects of maturation
- More confidence in baseline and outcome data
What are the disadvantages of the interrupted time-series design?
- More time consuming
- No control group
- Difficult to achieve stability of measurements
In the patient as own control design, what is the main characteristic?
The subjects serve as their own control, measured on outcome in presence /absence of a single/various intervention
What is an advantage of the patient as own control design?
Have some control over maturation/practice.
What are the disadvantages of the patient as own control design?
- Period of no treatment or delayed treatment
- Carry-over effect possible
- No control over history
What are historical controls in quasi-experimental design?
Use of a control group who received a different intervention during an earlier time period.
What are the advantages of historical controls?
- Having a comparison group
- Alternative when ethical concerns preclude a true control group
What are the disadvantages of historical controls?
- Different subject characteristics
- Concerns with data collection from historical controls
- Exposure to different factors for the two groups
What distinguishes the non-equivalent pretest-posttest control group design from true experimental design?
The group allocation is NON-RANDOM.
What are the advantages of non-equivalent pretest-posttest control group design?
- Strongest quasi-experimental design
- Some control over history, maturation, testing/practice, and instrumentation effects
What is a disadvantage of the non-equivalent pretest-posttest control group design?
Comparison groups may not be equivalent.
What are some advantages of quasi-experimental design?
- Practical
- No need to deny treatment to one group
- More generalizable to clinical settings
- Reasonable alternative when randomization isn’t feasible
What are some disadvantages of quasi-experimental design?
Less control over factors that influence internal validity.
Fill in the blank: The _______ is a threat to internal validity that involves changes in the subjects over time.
Maturation
True or False: In quasi-experimental designs, random assignment to groups is a requirement.
False