EXPERIMENTS

1
EXPERIMENTS

• Experiments involve a) taking action b)
  Observing the consequences of that action.
• Designed to determine causation.

2

• CLASSICAL EXPERIMENT
• Has the following components
• Formulate a hypothesis
• Randomly assign participants to the intervention
  group or to the control.
• Measure the dependent variable.
• Introduce the treatment or intervention.
• Measure the dependent variable(s) again.

3
Important Concepts

• a)  Independent and Dependent variable
• b)  Pretesting and Posttesting
• c)  Experimental and control group
• d) Double-Blind Experiments

4
SELECTING SUBJECTS

• Most experiments done are done on campus
  students.
• The experimental and control groups should be as
  similar as possible.
• RANDOMIZATION This is random assignment and is
  different from random sampling.

5
INTERNAL AND EXTERNAL VALIDITY

• Experiments have a virtue of high internal
  validity (causal validity), but they have the
  liability of having low external validity (cannot
  can be generalized across subgroups).
• What makes experiments have low external
  validity?

6

• No random selection of subjects.
• Experiments are usually conducted in limited
  settings.

7
THREAT TO INTERNAL VALIDITY

1. History
2. Maturation
3. Testing and Instrumentation
4. Regression to the mean
5. Selection of participants
6. Mortality
7. Diffusion of treatment