An experiment is said to have external validity when its results can be generalized to other settings (ecological validity), other people (population validity), and over time. The external validity can be affected by multiple factors.
- Between Testing and Treatment: When a pre-test exposes the individual to the purpose of the test and, in turn, causes the change in the dependent variable.
- Between Selection and Treatment: Specific characteristics of the subjects that alters their response to the independent variable (e.g. some individuals having better metabolism of vitamin D than others).
- Reactivity: when individuals change their behaviors after they realize they are being watched.
- Demand Characteristics: when subjects figure out what is being studied and respond accordingly.
- Experimenter Expectancy: when the experimenter unintentionally gives out hints or cues to the subjects, which alters their response.
- Evaluation Apprehension: subjects begin to act in a way that helps them avoid negative evaluation.
Can be controlled by using deception, a nonreactive measure, or single/double-blind study design. In a Single-Blind Study, subjects do not know whether they are in the experimental or control group. In a Double-Blind Study, neither the subject, nor the experimenter know which subject belongs to what group.
- Multiple-Treatment Interference: When multiple treatments are being administered to subjects and effects of one treatment carry over to the other.