Edexcel GCSE Psychology: Research Methods
Topic 11: Research Methods — How do you carry out psychological research?
Paper 2 | Compulsory | 37 marks
Research Methods is the largest assessed topic on the specification and appears on both papers. On Paper 1 it is tested through application to the compulsory topics. On Paper 2 it has its own dedicated section worth 37 marks, including a 12-mark extended question. A solid understanding of research methods is essential for achieving a good grade across the whole qualification.
Designing psychological research
Variables
An independent variable (IV) is the variable a researcher deliberately changes or manipulates. It is the thing that differs between conditions.
A dependent variable (DV) is the variable that is measured in a study — what the researcher expects to change as a result of the IV.
An extraneous variable is any variable other than the IV that could affect the DV and threaten the validity of the results. Situational variables are environmental factors such as noise, temperature or lighting. Participant variables are individual differences between participants such as age, memory ability or prior knowledge.
Controlling extraneous variables
Standardised procedures involve following exactly the same method with every participant to reduce situational variables. Counterbalancing reduces order effects in repeated measures designs by alternating the order of conditions between participants. Randomisation uses chance to determine the order of stimuli or conditions. A single-blind technique means participants do not know which condition they are in, reducing demand characteristics. A double-blind technique means neither the participant nor the researcher interacting with them knows the condition, reducing both demand characteristics and investigator effects.
Hypotheses
A null hypothesis predicts no difference or relationship between variables and that any difference found will be due to chance. An alternative hypothesis predicts that a difference or relationship will be found. A directional hypothesis specifies the direction of the expected difference. A non-directional hypothesis predicts a difference without specifying its direction.
Sampling methods
A target population is the entire group a study aims to investigate. A sample is selected from this population to take part in the study.
Random sampling: every member of the target population has an equal chance of selection. Highly representative but time-consuming and difficult to achieve in practice.
Stratified sampling: participants are selected from subgroups in proportion to their representation in the population. Very representative but complex to organise.
Volunteer sampling: participants self-select by responding to an advertisement. Easy and ethical but may produce a biased sample if volunteers differ systematically from non-volunteers.
Opportunity sampling: participants are selected because they are available and willing at the time. Quick and convenient but often unrepresentative of the wider population.
Experimental and research designs
Independent measures: different participants in each condition. Avoids order effects but requires more participants and means participant variables may differ between groups.
Repeated measures: the same participants in all conditions. Controls participant variables and requires fewer participants, but introduces order effects and increases the risk of demand characteristics.
Matched pairs: different participants in each condition, matched on key variables before allocation. Balances the advantages of both other designs but is time-consuming and imperfect.
Reliability and validity
Reliability refers to the consistency of a measure — a reliable study produces the same results when repeated.
Validity refers to whether a study measures what it intends to. Internal validity refers to whether results are genuinely due to the IV rather than extraneous variables. External validity refers to whether findings can be generalised to the target population or real-life settings.
Unrepresentative samples reduce external validity. Order effects in repeated measures designs reduce internal validity. Qualitative methods tend to produce rich data but are harder to replicate, reducing reliability. Quantitative methods are more replicable but may oversimplify complex behaviour.
Ethical issues in psychological research
The British Psychological Society (BPS) provides ethical guidelines for all psychological research.
Informed consent: participants should be told what the study involves and agree freely to take part.
Deception: participants should not be misled. Where deception is necessary, participants must be fully debriefed afterwards.
Confidentiality: participants' personal data and responses should be kept private.
Right to withdraw: participants should be free to leave the study at any time without penalty.
Protection of participants: researchers must prevent physical or psychological harm. Studies causing unexpected distress should be stopped.
Research methods
Laboratory experiment
Conducted in a controlled environment where the researcher manipulates the IV and measures the DV while controlling extraneous variables.
Strengths: high control allows cause and effect to be established. Replicable, improving reliability. Weaknesses: the artificial setting reduces ecological validity. Participants may display demand characteristics.
Field experiment
Conducted in a real-world setting with the researcher still manipulating the IV.
Strengths: natural setting improves ecological validity. Demand characteristics are reduced if participants are unaware. Weaknesses: less control over extraneous variables. Ethical concerns if participants have not given informed consent.
Natural experiment
The IV changes naturally rather than being manipulated by the researcher.
Strengths: studies real events that could not ethically be manipulated. High ecological validity. Weaknesses: no random allocation to conditions, meaning participant variables may differ. Extraneous variables are hard to control.
Interview
Involves asking participants questions directly. A structured interview uses fixed questions in a set order. A semi-structured interview has core questions but allows follow-up. An unstructured interview follows the participant's responses more freely.
Strengths: can gather rich, detailed data. Can be tailored to the individual. Weaknesses: unstructured interviews are hard to replicate and may introduce interviewer bias. Social desirability may affect responses.
Questionnaire
A set of written questions. Closed-ended questions offer set responses and produce quantitative data. Open-ended questions allow free responses and produce qualitative data.
Strengths: large amounts of data can be collected quickly. Quantitative data is easy to analyse. Weaknesses: social desirability may distort responses. Closed questions may miss important information.
Correlation
Examines the relationship between two co-variables. A positive correlation means both variables increase together. A negative correlation means as one increases the other decreases. No correlation means no clear relationship exists.
Strengths: useful for investigating relationships that cannot be manipulated. Weaknesses: correlation does not establish cause and effect. A third variable may explain the relationship.
Case study
An in-depth investigation of a single individual or small group, often involving rare or unusual circumstances.
Strengths: provides rich, detailed data impossible to gather through other methods. Useful for studying unique cases. Weaknesses: findings cannot be generalised. Researcher bias may affect interpretation.
Observation
Involves watching and recording behaviour. Observations can be covert (participants unaware) or overt (participants aware), and participant (researcher joins the group) or non-participant (researcher remains separate). Structured observations use predetermined coding schemes.
Strengths: captures natural behaviour. High ecological validity in naturalistic settings. Weaknesses: covert observations raise ethical concerns about consent. Observer bias can affect coding. Inter-rater reliability should be checked using more than one observer.
Data analysis
Descriptive statistics
The mean is calculated by adding all values and dividing by the number of values. It uses all the data but is affected by extreme scores. The median is the middle value when scores are ordered — not affected by extreme scores. The mode is the most frequently occurring value. The range is the difference between the highest and lowest values, showing the spread of data.
Data types
Primary data is collected directly by the researcher. Secondary data already exists, collected by someone else. Qualitative data is descriptive and non-numerical — rich and detailed but harder to analyse systematically. Quantitative data is numerical — easier to analyse and compare but may oversimplify behaviour.
Graphs
Bar charts display frequency or average data for discrete categories. Histograms display continuous data with the area of each bar representing frequency. Scatter diagrams plot two co-variables to show correlation. Frequency tables organise how often each value occurs.
Issues and debates: ethical issues in psychology
Ethical evaluation runs through all topics on the specification. Key examples include Milgram's study (deception, psychological distress, right to withdraw), Piliavin et al. (lack of informed consent from subway passengers) and Haney, Banks and Zimbardo (psychological harm leading to early termination).
The BPS guidelines exist to ensure psychology maintains public trust and does not harm participants. Where deception or lack of consent is unavoidable, a thorough debrief helps to address any distress and restore transparency.
Key terms
Independent variable (IV): the variable deliberately changed by the researcher.
Dependent variable (DV): the variable measured in a study.
Extraneous variable: any variable other than the IV that could affect the DV.
Situational variable: an environmental factor that could affect results.
Participant variable: an individual difference between participants that could affect performance.
Standardised procedure: following exactly the same method with every participant.
Counterbalancing: alternating the order of conditions between participants to reduce order effects.
Null hypothesis: a prediction that there will be no difference or relationship between variables.
Alternative hypothesis: a prediction that there will be a difference or relationship between variables.
Directional hypothesis: specifies the direction of the expected difference.
Non-directional hypothesis: predicts a difference but does not specify its direction.
Target population: the entire group a study aims to investigate and generalise to.
Sample: a group of participants selected from the target population.
Independent measures: different participants in each condition.
Repeated measures: the same participants in all conditions.
Matched pairs: different participants matched on key variables before allocation to conditions.
Reliability: the consistency of a measure or result.
Validity: whether a study measures what it intends to measure.
Internal validity: whether results are genuinely due to the IV rather than extraneous variables.
External validity: whether findings can be generalised beyond the study.
Ecological validity: whether findings reflect real-world behaviour.
Informed consent: participants agreeing to take part having been told what the study involves.
Deception: misleading participants about the purpose or nature of the study.
Confidentiality: keeping participants' personal data private.
Right to withdraw: the freedom to leave a study at any time without penalty.
Protection of participants: the researcher's duty to prevent physical or psychological harm.
Demand characteristics: when participants change behaviour because they have guessed the study's purpose.
Positive correlation: as one variable increases, the other also increases.
Negative correlation: as one variable increases, the other decreases.
Quantitative data: numerical data.
Qualitative data: descriptive, non-numerical data.
Primary data: data collected directly by the researcher.
Secondary data: data that already exists, collected by someone else.
Looking for revision resources for this topic? The Research Methods Knowledge Organiser covers all of the above in a format designed for efficient revision.