Research Methods

Experimental design, sampling & ethics — A-Level Psychology

Research Methods Compared

Laboratory Experiment

Controlled environment. IV is manipulated, DV is measured. Extraneous variables controlled.

+ High control + Replicable + Cause & effect

− Low ecological validity − Demand characteristics

Field Experiment

Natural setting. IV manipulated by researcher. Participants may be unaware they're being studied.

+ High ecological validity + Less demand characteristics

− Less control − Hard to replicate − Ethical issues

Natural Experiment

IV occurs naturally (not manipulated). Researcher measures the DV. Quasi-experiment.

+ Studies variables that can't be manipulated + High ecological validity

− No control over IV − Can't establish causation

Observation

Watching and recording behaviour. Can be naturalistic (in natural setting) or controlled (structured setting).

+ High ecological validity (naturalistic) + Rich qualitative data

− Observer bias − No cause & effect − Ethical concerns

Questionnaire / Interview

Self-report methods. Questionnaires (written); interviews (verbal, structured/unstructured).

+ Large samples (questionnaire) + Rich data (interviews)

− Social desirability bias − Leading questions

Correlation

Measures the relationship between two co-variables. Produces a correlation coefficient (−1 to +1).

+ Shows strength & direction + Uses existing data

− Cannot establish cause & effect − Third variable problem

Experimental Designs

Design	Description	Strengths	Weaknesses
Independent Groups	Different participants in each condition	No order effects; quick	Individual differences (participant variables); needs more participants
Repeated Measures	Same participants in all conditions	Controls participant variables; fewer participants needed	Order effects (fatigue, practice); demand characteristics
Matched Pairs	Participants matched on key variables, then split into conditions	Reduces participant variables; no order effects	Time-consuming; impossible to match on all variables

Exam tip: If asked to improve a study's design: repeated measures → counterbalancing reduces order effects. Independent groups → random allocation reduces participant variables. Always explain WHY the improvement works.

Sampling Methods

Click a method to see how participants would be selected from a population of 40.

BPS Ethical Guidelines

The British Psychological Society sets ethical standards that all psychological research must follow.

Informed Consent

Participants must be told the true purpose and give agreement before taking part. Presumptive or prior general consent can be used when deception is necessary.

Right to Withdraw

Participants can leave at any time without penalty and can withdraw their data retrospectively. This must be made clear at the start.

Protection from Harm

Participants should not experience greater risk than in everyday life. Physical and psychological harm must be minimised. Includes distress, embarrassment, and loss of self-esteem.

Deception

Participants should not be deliberately misled unless absolutely necessary and the research couldn't be done otherwise. Must be followed by a thorough debriefing.

Confidentiality

Participants' data must be kept anonymous and secure. Personal information should not be shared without consent. Results should be reported anonymously.

Debriefing

After the study, participants are told the true aims, any deception is revealed, and they're offered the chance to withdraw data. Researcher checks for any negative effects.

Exam tip: When evaluating a study's ethics, don't just list which guidelines were broken — explain the IMPACT on participants and discuss whether the breach was justified by the scientific value of the findings.