5.3 Experimental and Quasi-Experimental Designs Flashcards
Types of Quantitative Research Designs
- Experimental
- Quasi-experimental
- Non-experimental
Experimental
Gold Standard - Randomized control trial (RCT) - Level 2 evidence.
Systematic review of RCT’s are level 1 evidence
CONCEPTS THAT MUST BE PRESENT IN EXPERIMENTAL DESIGN
- Randomization
- Control
- Manipulation (All experimental designs involve the manipulation of the independent variable (cause) and measurement of a dependent variable (effect).
- NOTE casual relationships are difficult to establish so researchers should avoid the word “prove” when discussing hypothesis testing
Types of Experimental Designs
- Classic Randomized Clinical Trials
- Solomon four-group design
- After-Only Experimental Design
Classic Randomized Clinical Trial
STEPS
- Sample is selected from the population
- Baseline is collected
- Subjects are randomized
- 2 Groups (intervention and control group)
- Post intervention data is collected from each group
Solomon Four-Group Design
STEPS
- Sample is selected from the population
- Subjects are randomized into 4 groups that vary based on collected baseline data
- Control and intervention groups are created for both when baseline data is collected and baseline data is not collected
- After this postintervention data is collected
After-Only Experimental Design
STEPS
- Subjects are randomized into an intervention and control group
- Post intervention data is collected
Strengths/Weaknesses of an Experimental Design
Strengths
- Gold standard and best at testing cause-effect relationships.
- Control, manipulation, and randomization is one of the best ways to measure differences between 2 groups
Cons
- Complicated and costly
- Many variables related to patient care outcomes cannot be manipulated due to ethical concerns
Quasi-Experimental Design
- Difference between experimental and quasi-experimental is that quasi does not contain either a control group or random assignment
- Weaker cause-effect relationship
- Level 3 evidence
Types of Quasi-Experimental Designs
- Nonequivalent Control Group
- After-Only Nonequivalent control design
- One-group (Pretest/Posttest)
- Time series design
Nonequivalent Control Group Design
STEPS
- 2 groups are created, control and intervention group
- Baseline data is collected
- Intervention is applied to the experimental group
- Postintervention data is collected for the experimental group and data is collected again for the control group.
After-Only Non-equivalent Control Group Design
STEPS
- 2 Groups created (control and intervention)
- Intervention is applied to experimental group
- Postintervention data collected for experimental group, data collected from control group
One-Group (Pretest, Posttest) Design
STEPS
- Only experimental group is created
- Baseline data is collected
- Intervention is applied
- Postintervention data is collected
Time-Series Design
STEPS
- Experimental design is collected
- Data is collected twice prior to intervention
- Intervention is applied
- Post-Intervention data is collected twice
Strengths/Weaknesses of Quasi-experimental Design
Strengths
- Practical, feasible, cost-effective, potential generalizable findings, more adaptable to real-world settings
Weakness
- Inability to make clear cause-effect determinations. Did manipulation of independent variable actually affect the change in the dependent variable