Experimental Methods and Designs
Why Use Experiments?
The experiment is the gold standard of psychological research. When a psychologist wants to establish that one thing causes another — not merely that the two are associated — an experiment is the method that can demonstrate causation. This is because experiments involve directly manipulating one variable while keeping everything else as constant as possible, then measuring the effect on another variable.
The key distinction the spec requires you to understand is between the research method (how the data is gathered) and the experimental design (how participants are allocated to conditions within the experiment). These are two separate decisions a researcher makes. A laboratory experiment can use any of the three experimental designs; so can a field experiment.
Research method = how data is collected (lab, field, natural experiment, observation, questionnaire etc.). Experimental design = how participants are divided between conditions (independent groups, repeated measures or matched pairs). These are distinct decisions — a common exam error is confusing the two.
Experimental Designs
Once a researcher decides to run an experiment, they must choose how to use their participants. There are three main experimental designs.
In an independent groups design, different participants take part in each condition. One group does condition A, a different group does condition B. For example, Group 1 revises with flashcards; Group 2 revises by rereading notes. They are separate, independent groups.
Strength: there are no order effects, because participants only experience one condition. Weakness: individual differences between the groups may affect results (one group might naturally have better memories), and more participants are needed because each person only provides data for one condition.
In a repeated measures design, the same participants take part in all conditions. Every person does condition A and condition B. This eliminates individual differences as a confounding variable, because the same person is compared to themselves.
Strength: individual differences are controlled — each participant acts as their own control. Fewer participants are needed. Weakness: order effects can occur. If participants do condition A first and then condition B, they may perform better on condition B because of practice (‘practice effect’), or worse due to fatigue or boredom. Counterbalancing (half the participants do A then B; the other half do B then A) can reduce order effects.
A matched pairs design pairs participants on a relevant variable (such as age, IQ or previous experience) before assigning one from each pair to each condition. It combines the advantages of independent groups (no order effects) and repeated measures (individual differences partly controlled).
Strength: individual differences are reduced because each pair is similar on the key variable; no order effects because each person only takes part in one condition. Weakness: very time-consuming to match participants carefully; impossible to match on every relevant variable; if one participant drops out, both members of the pair are lost.
Types of Experiment: Laboratory, Field and Natural
As well as choosing a design, a researcher must choose the setting and level of control. The three main types of experiment each involve the manipulation of an independent variable, but they differ in where they take place and how much control the researcher has.
A laboratory experiment takes place in a controlled setting, usually a specially designed room. The researcher can control many variables: lighting, noise, temperature, the exact stimuli presented and so on. Strength: high control means results more clearly reflect the IV; easier to replicate. Weakness: the artificial setting reduces ecological validity — participants may not behave as they would in everyday life.
A field experiment takes place in a real-world setting. The researcher still manipulates the IV but the participants are in their natural environment and may not know they are being studied. Strength: high ecological validity because behaviour occurs in a natural setting. Weakness: less control over extraneous variables; ethical concerns if participants have not consented.
A natural experiment involves a naturally occurring event as the IV — the researcher does not manipulate it but measures the effect of something that has happened independently. Strength: high ecological validity; allows study of variables that would be unethical to manipulate deliberately. Weakness: low control; participants cannot be randomly allocated to conditions because the IV was not chosen by the researcher.
Choosing the Right Design and Method
The choice of experimental design and research method depends on the research question. If a researcher is concerned about order effects and has enough participants, independent groups is the safest design. If individual differences are a major worry and order effects can be counterbalanced, repeated measures is preferable. If the research involves sensitive topics requiring real-world settings — for example, studying helping behaviour in a public space — a field experiment has advantages over a laboratory setting, even though it sacrifices some control.
Three experimental designs: independent groups (different participants per condition — no order effects but individual differences may confound); repeated measures (same participants, controls individual differences but risks order effects); matched pairs (paired on a variable, avoids order effects and reduces individual differences but time-consuming).