Lecture 10: Quasi-Experiments & Applied Research
11/14/04, 11/16/04
- Quasi-Experimental Designs
- What is a quasi-experiment?
- Nonequivalent Groups Design
- Control group is different from the experimental group
- Only the experimental group gets the treatment
- Why is this not optimal?
- Solution
- Possible Outcomes
- Problems with matching & regression
- Other Nonequivalent Groups Designs
- Nonequivalent dependent variables design
- Reversed-treatment nonequivalent control groups design
- Interrupted Time Series Design
- How is it done?
- Possible outcomes
- Other Interrupted Time Series Designs
- Combined with nonequivalent control group design
- With Switching Replications
- Archival Data
- What is it?
- Pre-collected data
- Stuff that’s "out there"
- How to use it
- Experimenter bias
- Downsides
- Upsides
- Applied Research
- What is it?
- Designed to solve a specific real-world problem
- Usually conducted outside of laboratory/ university environment
- Often easier to get grant funding for applied research
- Problems & Advantages
- similar to observational research
- ethics
- less internal validity, greater external validity
- Program Evaluation
- What is it?
- Needs Analysis
- Formative Evaluation
- Summative Evaluation
- Cost-Benefit analysis
- Politics
Research
Methods Page
Last updated: 11/11/04