Chapter 9 Slides Flashcards
(21 cards)
Experimental and Quasi-Experimental Designs
Require that the researcher actively intervenes to bring about the desired effect
Test cause-and-effect relationships
Provide level II (Experimental) and III (Quasi-experimental) evidence
Cause and Effect Relationship
Three criteria are needed to infer causality:
The causal (independent) and effect (dependent) variables must be associated with each other.
The cause must precede the effect.
The relationship must not be explainable by another variable.
True Experimental Design
Three properties:
Randomization
Control
Manipulation
A research study using a true experimental design is commonly called a randomized controlled trial (RCT).
Provides Level II Evidence
Randomization
Each subject in the study has an equal chance of being assigned to the control group or the experimental group.
Randomization assumes that any important extraneous variable will be equally distributed between the groups, minimizing variance and decreasing selection bias.
Control
Acquired By:
Manipulating the independent variable
Random assignment
Using a control group
- The control group receives the usual care or a placebo.
Manipulation
The independent variable is manipulated when some subjects (experimental group) receive the intervention and others (control group) do not.
“Doing something” to at least some of the subjects
Treatment Effect
Treatment effect: effect attributed to the intervention
Effect Size
Effect size: a statistical measure of the strength of the relationship between two variables
Power Analysis
Power analysis: informs the researcher of the sample size needed determine the effect of the intervention
Types of experimental designs
Randomized Controlled Trial
Solomon Four Group Design
After Only Design
Randomized controlled trial
Use experimental and control groups.
Intervention fidelity ensures that every subject receiving the intervention receives the identical intervention.
Use statistical comparisons to determine any differences between groups.
Sample size is important—
too large wastes time, resources, and money; too small may lead to inaccurate results.
Solomon Four Group design
Subjects randomly assigned to four groups: experimental, control, experimental after group, and control after group
Experimental after group and control after group receive no pretest, only posttests.
Controls for testing effects that are a threat to internal validity
More expensive because it requires a larger sample
After only designs
Also known as the Post-test only control group design
Two randomly assigned groups, neither group is pre-tested
Used to minimize testing effects or when pretest not possible (e.g. measuring post operative pain)
Strengths and weakness of experimental designs
Strengths:
Most powerful for testing cause-and-effect relationships owing to the use of control, manipulation, and randomization
Weakness:
Complicated to design
Costly to implement
Difficult to implement
Quasi-Experimental Design
Also tests cause-and-effect relationships
Randomization may not be possible, or there may not be a control group or both are absent:
- Less control makes evidence provided by quasi-experimental designs less convincing.
- Provides level III evidence
Common types of quasi-experimental designs
- Nonequivalent Control Group Design
- After-Only Nonequivalent Control Group Design
- One-Group (Pretest-Posttest) Design
- Time series Design
Nonequivalent control group design
Experimental and control group but no randomization.
Researcher can compare the two groups on variables of interest before the intervention.
Threats to internal validity include selection, maturation, testing, and mortality.
After-Only Nonequivalent Control Group Design
Assigned experimental and control groups, but neither is pretested or measured
Particularly useful when testing effects are suspected to be a potential major threat to internal validity
No randomization (Confidence in the findings depends on the integrity of assumption that the two groups are comparable)
Also known as Posttest only control group design
One group only (Pretest-Posttest Design)
Used when only one group is available for study
Only an experimental group; data are collected before and after an experimental treatment
No control group, no randomization
Time series Design
Design is useful for determining trends over time.
Data are collected multiple times before the intervention to establish a baseline.
Data are then collected multiple times afterward to determine a change from baseline.
Testing, maturation, and selection are all potential threats to internal validity.
Strengths and weakness of Quasi-Experimental Designs
Strength:
- Practical, less expensive, generalizable, and sometimes the only feasible alternative because these designs are more adaptable to real-world settings.
- Replication of a study can strengthen evidence.
Weakness:
- Less control Lower on evidence pyramid (unable to make confident cause-and-effect statements)
