Types of Experiment

The Experimental Method

The experiment is psychology's most powerful research method for establishing cause and effect. By manipulating an independent variable (IV) whilst controlling all other potentially relevant variables, and then measuring any resulting change in a dependent variable (DV), the experimenter can infer that the IV caused the observed change in the DV. This ability to demonstrate causation distinguishes the experiment from all other research methods, which can establish association but not direction of cause.

Laboratory Experiments

A laboratory experiment is conducted in a controlled, artificial environment where the researcher has maximum control over variables. All participants experience identical conditions (except for the IV manipulation), extraneous variables are minimised, and procedures are standardised. This produces high internal validity — confidence that the IV caused the change in the DV — and enables replication.

Strengths: high control over extraneous variables; high internal validity; replicable; cause-and-effect conclusions can be drawn. Weaknesses: artificial environment reduces ecological validity — findings may not generalise to real-world settings. Participants aware of being studied may show demand characteristics — altering their behaviour to match perceived expectations.

Field Experiments

A field experiment manipulates an IV in a natural, real-world setting. The researcher still controls the IV but the environment is not artificially constructed. Because participants are in their natural environment, ecological validity is higher and demand characteristics are reduced (participants may be unaware they are in a study).

Strengths: higher ecological validity than lab experiments; reduced demand characteristics. Weaknesses: less control over extraneous variables; replication is harder; ethical issues arise if participants are unaware they are being studied (lack of informed consent).

Natural Experiments

A natural experiment takes advantage of a naturally occurring change in the IV — the researcher does not manipulate it. For example, studying the psychological effects of a real earthquake on the affected population. Because the IV is not experimentally controlled, strict cause-and-effect conclusions are weaker, and the researcher cannot randomly allocate participants to conditions.

Strengths: allows study of variables that cannot be manipulated for ethical or practical reasons; high ecological validity. Weaknesses: IV not directly controlled by researcher; random allocation is not possible; confounding variables are harder to rule out; cannot be replicated (the natural event is unique).

Quasi-Experiments

A quasi-experiment involves an IV that differs between groups but is not directly manipulated by the researcher and participants are not randomly allocated — for example, comparing anxiety levels between males and females, or between diagnosed clinical groups and healthy controls. The IV (sex, diagnostic status) is a pre-existing characteristic, not an imposed manipulation.

Strengths: allows comparison of groups that differ on variables of theoretical interest. Weaknesses: lack of random allocation means groups may differ on confounding variables; cause-and-effect conclusions are weakened.