Experimental Methods

Experiments are the primary method psychologists use to establish cause-and-effect relationships. In an experiment, the researcher manipulates the IV and measures the DV while controlling extraneous variables. However, there are different types of experiment, each with its own strengths and limitations.

flowchart TD
    A[Experiments] --> B[IV manipulated by researcher]
    A --> C[IV NOT manipulated]
    B --> D["Laboratory<br/>controlled setting<br/>high control"]
    B --> E["Field<br/>real-world setting<br/>moderate control"]
    C --> F["Natural<br/>IV occurs naturally<br/>e.g. disaster"]
    C --> G["Quasi<br/>IV is participant trait<br/>e.g. age, gender"]

---

Types of Experiment

1. Laboratory Experiment

A laboratory experiment takes place in a controlled environment (usually a laboratory or specially prepared room). The researcher has high control over variables.

Example: Loftus and Palmer (1974) — participants watched a film of a car accident in a controlled setting and were asked leading questions.

Advantage	Disadvantage
High control over extraneous variables — easier to establish cause and effect	Low ecological validity — the artificial setting may not reflect real-life behaviour
Easy to replicate due to standardised procedures	Demand characteristics — participants may behave differently because they know they are being studied
Can precisely measure the DV	May lack mundane realism — tasks may not be similar to real-world activities

2. Field Experiment

A field experiment takes place in a natural, real-world setting (e.g. a school, workplace, or public place). The researcher still manipulates the IV but has less control over extraneous variables.

Example: Piliavin et al. (1969) — studied helping behaviour on the New York subway.

Advantage	Disadvantage
Higher ecological validity — behaviour is more natural because participants are in their real environment	Less control over extraneous variables — harder to establish cause and effect
Fewer demand characteristics — participants may not know they are being studied	Difficult to replicate due to variations in the natural environment
More mundane realism	Ethical issues — participants may not have given informed consent

3. Natural Experiment

In a natural experiment, the IV is not manipulated by the researcher — it occurs naturally. The researcher takes advantage of a naturally occurring situation.

Example: Studying the effects of a natural disaster on mental health — the researcher cannot manipulate whether a disaster occurs, but can compare people who experienced it with those who did not.

Advantage	Disadvantage
Allows study of variables that cannot be ethically manipulated (e.g. trauma, deprivation)	The IV is not controlled by the researcher — cannot establish cause and effect with certainty
High ecological validity	Confounding variables are more likely because the researcher cannot control the situation
Can provide valuable data on rare or unique events	Participants cannot be randomly allocated to conditions

4. Quasi-Experiment

A quasi-experiment is similar to a natural experiment, but the IV is based on an existing characteristic of the participants (e.g. age, gender, occupation) rather than being manipulated.

Example: Comparing memory performance between older and younger adults — age is the IV, but the researcher cannot assign participants to an age group.

Advantage	Disadvantage
Allows comparison of pre-existing groups	Cannot establish cause and effect — the IV is not manipulated
Useful for studying characteristics that cannot be manipulated	Participant variables between groups may confound results

---

Comparing Experimental Methods

Feature	Laboratory	Field	Natural	Quasi
IV manipulated?	Yes	Yes	No	No
Control	High	Moderate	Low	Low
Ecological validity	Low	High	High	Variable
Cause and effect?	Strong	Moderate	Weak	Weak
Demand characteristics	Risk	Lower risk	Lower risk	Variable
Ethical issues	Moderate	Risk (consent)	Moderate	Moderate

---

Key Points

• Laboratory experiments: high control, low ecological validity.
• Field experiments: natural setting, higher ecological validity, less control.
• Natural experiments: IV occurs naturally, cannot establish cause and effect.
• Quasi-experiments: IV is a pre-existing characteristic (e.g. age, gender).
• Only experiments where the IV is manipulated can establish cause-and-effect relationships.

---

The Spectrum of Experimental Control

A useful way to think about the four experimental types is as a spectrum running from highest control to lowest control:

1. Laboratory experiment — highest control; the researcher manipulates the IV and controls almost every aspect of the setting
2. Field experiment — moderate control; the researcher manipulates the IV but cannot control the natural setting
3. Natural experiment — low control; the IV occurs naturally but the researcher can still plan the measurement of the DV
4. Quasi-experiment — low control; the IV is a pre-existing characteristic, not manipulated at all

As control decreases, ecological validity tends to increase — real-world behaviour is richer and messier than lab behaviour. The ideal study design balances control against realism, which is why many psychologists combine methods across their career: a lab study might establish a basic effect, followed by a field study to check whether the effect holds in real life.

---

Experiments and Cause-and-Effect

The central reason psychologists value experiments is that, when conducted properly, they allow a cause-and-effect conclusion. This is because the researcher manipulates the IV directly and controls extraneous variables, isolating the effect of the IV on the DV. Non-experimental methods — observations, correlations, case studies — can describe relationships and patterns, but they cannot prove that one variable causes another.

However, not all experiments produce equally strong causal conclusions. Laboratory experiments allow the strongest conclusions because of the high control. Field experiments support moderate causal conclusions. Natural experiments and quasi-experiments allow only weak causal conclusions because the IV is not manipulated — participants are not randomly allocated to conditions, and pre-existing differences may explain the results.

Remembering this hierarchy is essential for any exam question asking you to evaluate the conclusions of a study.

---

Classic Experiments: Matching Method to Method

Match each famous study to its experimental type — a useful revision exercise.