Research Methods: Non-Experimental Research

What you'll learn

Non-experimental research methods allow psychologists to study behaviour without manipulating variables. You'll learn about observational research, correlational studies, self-report techniques (questionnaires and interviews), and case studies. Understanding when and how to use each method is essential for OCR GCSE Psychology Paper 2, where you'll need to identify, design and evaluate research methods.

Key terms and definitions

Naturalistic observation — watching and recording behaviour in the environment where it would normally occur, without interference from the researcher.

Controlled observation — watching and recording behaviour in an environment manipulated by the researcher, often in a laboratory setting.

Correlation — a relationship between two variables where changes in one variable are associated with changes in another variable.

Questionnaire — a set of written questions used to collect self-report data from participants about their attitudes, opinions or behaviours.

Structured interview — a research method using pre-determined questions asked in a fixed order to all participants.

Unstructured interview — a research method where the interviewer has general aims and topics but develops questions freely based on participants' responses.

Case study — an in-depth investigation of a single individual, group or event, typically using multiple research methods.

Coding frame — a systematic set of categories used to classify and record observations of behaviour.

Core concepts

Observational research

Observational research involves watching and recording behaviour systematically. Psychologists use different types depending on their research aims.

Types of observation:

• Naturalistic observation: Behaviour is studied in its natural environment without researcher interference. Example: observing playground aggression during break time.
• Controlled observation: Behaviour is observed in a structured environment created by the researcher. Example: Ainsworth's Strange Situation observing infant attachment in a laboratory playroom.
• Participant observation: The researcher becomes part of the group being studied. Example: a researcher joining a sports team to study motivation.
• Non-participant observation: The researcher remains separate from those being observed.

Recording observations:

Researchers must decide how to record what they observe:

• Event sampling: Recording every time a specific behaviour occurs. Example: tallying each aggressive act during a playground observation.
• Time sampling: Recording behaviour at set intervals. Example: noting what children are doing every 30 seconds during a 10-minute session.

Developing a coding frame:

A coding frame categorises behaviour into observable, measurable units. Good coding frames are:

• Operationalised: Behaviours are clearly defined so observers know exactly what counts. Example: "aggression" becomes "physical contact involving hitting, pushing or kicking."
• Mutually exclusive: Categories don't overlap, so each behaviour fits only one category.
• Exhaustive: All relevant behaviours can be categorised.

Strengths of observational research:

• High ecological validity when conducted naturally, as behaviour occurs in realistic contexts.
• Can study behaviours that cannot ethically be manipulated experimentally.
• Useful for studying children and animals who cannot complete questionnaires.

Limitations of observational research:

• Observer bias: Researchers may interpret behaviour according to their expectations.
• Low inter-observer reliability: Different observers may code the same behaviour differently. This is improved by training observers and correlating their recordings.
• Difficult to determine cause and effect relationships.
• Ethical issues in naturalistic settings where informed consent may be impossible.

Correlational research

Correlation measures the relationship between two co-variables (not independent and dependent variables, as there's no manipulation).

Types of correlation:

• Positive correlation: As one variable increases, the other also increases. Example: hours of revision and exam scores both increase together.
• Negative correlation: As one variable increases, the other decreases. Example: as hours of sleep decrease, reaction time increases (gets slower).
• Zero correlation: No relationship exists between the variables. Example: shoe size and intelligence.

Correlation coefficients:

The strength of a correlation is expressed as a number between -1 and +1:

• +1.0 = perfect positive correlation
• 0 = no correlation
• -1.0 = perfect negative correlation

Values closer to +1 or -1 indicate stronger relationships.

Scattergrams:

Psychologists display correlations visually on scattergrams (scatter graphs):

• Each dot represents one participant's scores on both variables.
• The pattern shows the type and strength of correlation.
• A line of best fit helps identify the trend.

Strengths of correlational research:

• Can investigate variables that cannot be manipulated for ethical or practical reasons.
• Provides a starting point for experimental research by identifying relationships worth investigating.
• Uses quantitative data that can be analysed statistically.

Limitations of correlational research:

• Cannot establish cause and effect — correlation does not equal causation.
• A third, unknown variable may cause both measured variables to change.
• Misinterpretation by media and public who assume correlation proves causation.

Self-report techniques: Questionnaires

Questionnaires collect data through written questions. They're widely used in psychology to gather information about attitudes, behaviours and experiences.

Question types:

• Closed questions: Offer fixed response options. Example: "How many hours did you sleep last night? a) 0-4 b) 5-6 c) 7-8 d) 9+"

  • Strengths: Easy to analyse, produce quantitative data, quick for participants to complete.
  • Limitations: Lack detail, may not include the response a participant wants to give.

• Open questions: Allow participants to answer in their own words. Example: "Describe how you felt when you woke up this morning."

  • Strengths: Rich, detailed qualitative data; participants can express themselves fully.
  • Limitations: Time-consuming to analyse, difficult to compare responses, participants may not want to write extensively.

Designing effective questionnaires:

• Avoid leading questions that suggest a particular answer. Poor: "Don't you agree that violent video games are harmful?" Better: "What is your opinion on violent video games?"
• Avoid double-barrelled questions that ask two things at once. Poor: "Do you enjoy psychology and find it useful?" Better: separate into two questions.
• Use clear, simple language appropriate for your target population.
• Include a mixture of question types for comprehensive data.
• Pilot test the questionnaire with a small sample to identify problems.

Strengths of questionnaires:

• Can reach large samples quickly and cost-effectively.
• Participants may answer honestly about sensitive topics due to anonymity.
• Standardised format allows for replication.

Limitations of questionnaires:

• Social desirability bias: Participants may answer to present themselves positively rather than truthfully.
• Response bias: Some participants always agree (acquiescence bias) or select middle options.
• Low response rates, particularly with postal questionnaires.
• Participants may misunderstand questions with no researcher present to clarify.

Self-report techniques: Interviews

Interviews involve direct verbal questioning, allowing researchers to explore topics in depth.

Structured interviews:

• All participants receive identical questions in the same order.
• Responses are often recorded using fixed categories.
• Similar to a verbal questionnaire.

Strengths: Easy to replicate, straightforward to analyse, reduced interviewer bias.

Limitations: Lacks flexibility, cannot explore interesting responses, may miss important information.

Unstructured interviews:

• The interviewer has general topics but develops questions naturally during conversation.
• Allows exploration of unexpected but relevant areas.
• Responses are recorded in detail, often audio-recorded.

Strengths: Rich, detailed qualitative data; flexibility to follow up interesting points; participants can express themselves fully.

Limitations: Difficult to analyse and compare; interviewer bias may influence questions and interpretation; requires skilled interviewers; time-consuming.

Semi-structured interviews:

• Combine both approaches with some fixed questions and opportunities for follow-up.
• Balance structure with flexibility.

General interview considerations:

• Interviewer effects: The interviewer's characteristics (age, gender, ethnicity, behaviour) may influence how participants respond.
• Rapport: Building trust helps participants answer honestly.
• Recording: Audio or video recording preserves detail but may inhibit participants.

Case studies

A case study is an in-depth investigation of a single individual, small group, institution or event. Multiple research methods are typically used to build a comprehensive picture.

Examples from psychology:

• Phineas Gage: Brain injury case showing the frontal lobe's role in personality.
• Genie: A child raised in isolation, studied to understand language acquisition critical periods.
• Little Hans: Freud's case study of a five-year-old boy's phobia.

Methods used in case studies:

Case studies typically combine several techniques:

• Interviews with the individual and people who know them.
• Observations of behaviour in various contexts.
• Psychological tests and assessments.
• Analysis of documents, records and historical information.
• Brain scans or other physiological measurements where relevant.

Strengths of case studies:

• Provide rich, detailed qualitative data about complex behaviours.
• Can study rare conditions or situations impossible to create experimentally.
• Generate hypotheses for future research.
• High ecological validity as they investigate real-life cases.

Limitations of case studies:

• Cannot generalise findings to the wider population from a single case.
• Researcher bias in selecting and interpreting information.
• Often retrospective, relying on memory which may be inaccurate.
• Difficult to replicate due to unique circumstances.
• Ethical issues around anonymity and consent, particularly with vulnerable individuals.

Worked examples

Example 1: Identify and explain (4 marks)

Question: A psychologist wants to investigate whether there is a relationship between the number of hours students spend on social media and their stress levels. Identify an appropriate research method and explain why it is suitable. [4 marks]

Mark scheme answer:

• The psychologist should use a correlational study/design [1 mark].
• This is appropriate because the psychologist wants to investigate whether a relationship exists between two variables [1 mark].
• The variables (social media use and stress) cannot be manipulated by the researcher for ethical reasons [1 mark].
• A correlation will show whether increased social media use is associated with higher or lower stress levels [1 mark].

Example 2: Design question (6 marks)

Question: Design a questionnaire to investigate students' attitudes towards exam stress. Your answer should include two different types of question with examples. [6 marks]

Mark scheme answer:

One closed question example: "How stressed do you feel about upcoming exams? a) Not at all stressed b) Slightly stressed c) Moderately stressed d) Very stressed e) Extremely stressed" [1 mark for type, 1 mark for appropriate example].

One open question example: "Describe the physical symptoms you experience when you feel exam stress" [1 mark for type, 1 mark for appropriate example].

Explanation: The closed question provides quantitative data that is easy to analyse and compare across participants [1 mark]. The open question provides detailed qualitative data about individual experiences that closed questions might miss [1 mark].

Example 3: Evaluation (4 marks)

Question: Evaluate the use of case studies in psychological research. [4 marks]

Mark scheme answer:

Strengths [2 marks available]:

• Case studies provide rich, detailed data about complex behaviours that other methods cannot capture [1 mark].
• They can investigate rare conditions or situations that cannot be studied experimentally for ethical or practical reasons [1 mark].

Limitations [2 marks available]:

• Findings cannot be generalised to the wider population as they involve only one or a few individuals who may be atypical [1 mark].
• There is potential for researcher bias in selecting and interpreting information, reducing objectivity [1 mark].

Common mistakes and how to avoid them

• Confusing correlation with experiments: Remember, correlations measure relationships between co-variables without manipulation. Experiments manipulate an independent variable. Never say a correlation has an IV and DV.

• Claiming correlation proves causation: Correlation only shows a relationship exists, not that one variable causes the other. Always explain that a third variable might be responsible or the relationship might be coincidental.

• Vague coding frames in observations: Don't just say "record aggressive behaviour." Operationalise it: "Record incidents of physical contact involving hitting, pushing, kicking or throwing objects at another person."

• Confusing interview types: Structured interviews use fixed questions in a set order. Unstructured interviews develop questions flexibly during the conversation. Don't mix up their characteristics.

• Forgetting multiple methods in case studies: Case studies aren't just interviews or just observations — they combine multiple research methods to build a detailed picture.

• Not linking method to research aim: In design questions, always explain why your chosen method is appropriate for the specific research question, not just what the method is.

Exam technique for "Research Methods: Non-Experimental Research"

• Command word awareness: "Identify" requires naming only (1 mark). "Explain" requires reasons or elaboration (2+ marks). "Evaluate" requires both strengths and limitations. "Design" requires specific, detailed examples.

• Structure evaluation answers: Use clear PEE (Point, Evidence, Elaboration) structure. For 4-mark evaluation questions, give two strengths or two limitations (or one of each), with each point fully elaborated for 2 marks.

• Design questions need detail: When asked to design a questionnaire or observation, provide actual examples of questions or coding categories, not just descriptions. "Ask about their sleep" scores 0 marks; "How many hours did you sleep last night?" scores marks.

• Link evaluations to context: When a question gives a specific scenario, tailor your evaluation to that context. Generic strengths and limitations score lower marks than contextualised evaluation.

Quick revision summary

Non-experimental research includes observations (naturalistic or controlled, using coding frames), correlations (measuring relationships between co-variables, shown on scattergrams), self-report techniques (questionnaires with open/closed questions, structured/unstructured interviews), and case studies (in-depth investigations using multiple methods). Remember: correlation doesn't prove causation, observations need clear operationalisation, questionnaires should avoid leading questions, and case studies provide rich detail but cannot be generalised. Always link the research method to the specific aim and context in exam answers.