You are viewing a free preview of this lesson.
Subscribe to unlock all 10 lessons in this course and every other course on LearningBro.
Not all research in psychology uses the experimental method. Many important questions cannot be answered through experiments — either because the variables cannot be manipulated, because the behaviour of interest occurs in natural settings, or because ethical constraints prevent experimental manipulation. Non-experimental methods include observations, self-report methods (questionnaires and interviews), correlations, and case studies. Each has distinctive strengths and limitations that psychologists must weigh when choosing a research design.
Key Definition: Non-experimental methods are research approaches that do not involve the direct manipulation of an independent variable. They are used to describe, measure, and explore relationships between variables without establishing causation.
Observational research involves watching and recording behaviour as it occurs, either in natural or controlled settings.
| Type | Description | Strengths | Limitations |
|---|---|---|---|
| Naturalistic | Behaviour observed in its natural setting with no intervention | High ecological validity; behaviour is genuine | Low control; extraneous variables; difficult to replicate |
| Controlled | Behaviour observed in a structured environment set up by the researcher | Greater control of variables; replicable | Lower ecological validity; behaviour may be artificial |
| Covert | Participants are unaware they are being observed | Reduces demand characteristics; more natural behaviour | Ethical concerns — lack of informed consent; privacy issues |
| Overt | Participants know they are being observed | Ethically preferable — informed consent can be obtained | Participants may alter behaviour (Hawthorne effect) |
| Participant | The researcher becomes part of the group being observed | Rich, in-depth data; access to insider perspectives | Loss of objectivity; risk of "going native"; ethical issues around deception |
| Non-participant | The researcher observes from outside the group | Greater objectivity; less risk of influencing the group | May miss subtle behaviours or context that a participant observer would notice |
| Structured | Uses pre-determined categories (a behavioural checklist) and coding systems | Systematic; quantitative data; high inter-observer reliability if well-designed | May miss unexpected behaviours; categories may be subjective |
| Unstructured | The observer records all relevant behaviour without a predetermined framework | Captures richness and complexity; flexible | Difficult to analyse; low inter-observer reliability; observer bias |
When observations are conducted by more than one observer, it is essential to ensure that they agree on what they are observing and how to code it. Inter-observer reliability is calculated by comparing the records of two or more observers:
Inter-observer reliability = (Number of agreements ÷ Total observations) × 100
A reliability score of 80% or above is generally considered acceptable. If reliability is low, the behavioural categories need to be redefined and observers retrained.
Exam Tip: When evaluating any observation, always discuss inter-observer reliability. Even a well-designed observation is weakened if different observers record different things. Suggest how reliability could be improved (e.g., training observers, refining behavioural categories, using video recordings for cross-checking).
Self-report methods involve asking participants to report their own thoughts, feelings, attitudes, or behaviours. The two main types are questionnaires and interviews.
A questionnaire is a pre-set list of written questions or statements to which participants respond. Questions may be:
| Strengths | Limitations |
|---|---|
| Can be distributed to large samples quickly and cheaply | Social desirability bias — respondents may give socially acceptable answers rather than truthful ones |
| Easy to replicate due to standardised format | Acquiescence bias — tendency to agree with statements regardless of content |
| Quantitative data from closed questions allows statistical analysis | Response bias — participants may not read questions carefully or may rush through |
| Anonymity may encourage honest responses | Lack of depth — closed questions restrict the richness of responses |
| No investigator effects during completion | Questions may be misinterpreted without a researcher present to clarify |
An interview involves a face-to-face (or sometimes telephone/video) conversation between the researcher and participant.
| Type | Description | Strengths | Limitations |
|---|---|---|---|
| Structured | Pre-determined, fixed questions asked in a set order | Standardised and replicable; easy to compare responses | Inflexible; may miss important information not covered by questions |
| Unstructured | No fixed questions; the conversation develops naturally guided by the researcher | Rich, detailed data; flexible; can explore unexpected topics | Difficult to replicate; time-consuming; hard to analyse; interviewer bias |
| Semi-structured | A set of pre-determined questions with the flexibility to ask follow-up questions | Balances standardisation with flexibility; rich data | Some loss of comparability; interviewer skill-dependent |
Key Definition: Social desirability bias occurs when participants give responses that they believe are socially acceptable or that present them in a favourable light, rather than providing truthful answers.
Exam Tip: When evaluating self-report methods, always consider: (1) social desirability bias, (2) the type of data produced (quantitative vs qualitative), (3) whether the questions allow depth or restrict it, and (4) the role of the researcher (present vs absent).
A correlation is a statistical technique used to measure the strength and direction of the relationship between two co-variables. Note: a correlation is not a research method but a method of data analysis — the data can be collected through any method (e.g., questionnaires, archival records).
Key Definition: A correlation is a measure of the relationship between two co-variables. A positive correlation means both variables increase together; a negative correlation means one increases as the other decreases.
Subscribe to continue reading
Get full access to this lesson and all 10 lessons in this course.