2 Methods in Psychology · 2.1 Methods in psychology
2.1.2 Observation, survey, case study, experiment and qualitative methods
Psychologists gather data by watching behaviour (observation), asking people (survey, questionnaire, interview), studying one case in depth (case study), manipulating a cause under control (experiment) or measuring how two things vary together (correlation), using numbers (quantitative) or meanings and words (qualitative), always within ethical rules such as informed consent and confidentiality.
- APPSC Paper I
- Standard level
- 16 min read
- Free
Core facts
- Naturalistic observation: behaviour watched where it normally happens, without interference.
- Controlled (laboratory) observation: behaviour watched in a controlled and structured setting chosen by the researcher.
- Participant observer joins the group; non-participant observer stays outside.
- Hawthorne effect: people change their behaviour because they know they are being studied (Western Electric Hawthorne Works studies, 1924–1932, interpreted by Elton Mayo and colleagues; the term was coined by H.A. Landsberger in 1958).
- Survey: collects opinions, attitudes, facts from many people through questionnaires, interviews, telephone or online forms; good for how often something occurs.
- Interview: structured (fixed questions in fixed order) or unstructured (free, flexible).
- Case study: an in-depth study of one person, group or event, using history, interviews, tests and observation; a non-experimental method.
- Clinical method: case study aimed at diagnosis and treatment; Lightner Witmer opened the first psychological clinic, University of Pennsylvania, 1896.
- Experimental method: studies cause and effect; the independent variable (IV) is changed, the dependent variable (DV) is measured.
- Experimental group gets the treatment; control group does not.
- Correlational method: finds whether two variables vary together; the coefficient runs from −1 to +1; correlation is not causation.
- Cross-sectional design: different age groups at one time; longitudinal: the same people over years.
- Quantitative research collects numbers; qualitative research collects non-numerical data (words, narratives, images). The main difference is the type of data collected and analysed.
- Research ethics: voluntary participation, informed consent, debriefing, sharing results, confidentiality (the NCERT list).
Explained
This note covers the methods as a psychology student meets them. The classroom version (introspection, observation, experiment, case study for teachers) is in EDU-PS-02; read that for introspection, the oldest method. The goals and steps of research, kinds of data, variables, hypotheses and sampling are in PSY-ME-01.
1. Observation
- What it involves: selecting what to watch, recording it (notes, checklists, rating scales, video), and analysing the record. Good observation is planned, not casual watching.
- Naturalistic: children on a playground, shoppers in a market, workers on a shop floor. Behaviour is genuine, but the researcher cannot control what happens or when.
- Controlled: the researcher sets up a room (for example a playroom with chosen toys) and watches through a one-way mirror. More precise and repeatable, less natural.
- Participant: the researcher becomes a member (lives in a village, joins a gang study). Gains trust and inside view, but may lose objectivity.
- Non-participant: watches from outside, without joining.
- Problems: observer bias (seeing what one expects), reactivity / Hawthorne effect (behaviour changes because people know they are watched), slow and costly, and it shows what people do, not why.
2. Survey method
- Purpose: to learn the attitudes, opinions, beliefs or behaviour of a large population, usually from a sample. Psychologists use questionnaire surveys to find how common a problem is and its likely causes (for example stress among students, alcohol use among youth).
- Forms: personal interview, questionnaire (mailed, handed out or online), telephone survey, and controlled observation (NCERT's list).
- Questionnaire: a written set of questions answered by the respondent. Closed (fixed-response) items offer choices (yes / no, multiple choice, rating); open-ended items allow free answers. Cheap, reaches many, anonymous; but needs literate respondents, return rates can be low, and people may give socially desirable answers.
- Interview: a face-to-face (or phone) conversation with a purpose. Structured (standardised): same questions, same order, for everyone; easy to compare. Unstructured (non-standardised): the interviewer follows the answers; rich but hard to compare. Interviews can be individual-to-individual, individual-to-group, group-to-individual or group-to-group. They suit illiterate people and children, but cost time and invite interviewer bias.
- Sampling: a survey is only as good as its sample; a random, representative sample lets the results be generalised. Sampling follows the scientific approach: results from a particular set of units are extended to the population.
3. Case study and clinical method
- Case study: everything about one case: family and developmental history, school or work record, medical history, interviews with the person and others, tests, observation. It is used for rare or unusual cases and to generate hypotheses. Famous examples in textbooks: Phineas Gage (personality change after frontal brain injury, 1848), patient H.M. (memory loss after brain surgery, 1953), and Freud's clinical cases.
- Clinical method: the case study used by counsellors and clinical psychologists to diagnose and treat a person's problem, then follow up. Piaget's clinical interview with children is a related technique.
- Merits: depth; a whole-person picture; useful where experiments are impossible or unethical.
- Limits: cannot be generalised to others; depends on the honesty and memory of informants; the researcher's interpretation can be subjective; time-consuming.
4. Experimental method
- Logic: change one factor (IV), keep everything else the same, and see whether another factor (DV) changes. If it does, the IV caused the change.
- Example: does background noise lower memory scores? IV = noise (present / absent); DV = number of words recalled; controlled: the word list, time, room, instructions.
- Extraneous (confounding) variables: anything other than the IV that could change the DV (age, fatigue, practice, time of day). Three standard controls (NCERT): holding conditions constant (or eliminating them), counterbalancing (varying the order of tasks so practice and fatigue balance out) and random assignment (every participant has an equal chance of being in either group).
- Experimental and control groups: the experimental group gets the IV; the control group is identical but does not; the difference between them is the effect.
- Placebo: a dummy treatment (sugar pill) given to the control group so that expectations are the same in both groups.
- Single-blind: participants do not know their group. Double-blind: neither participants nor the experimenter who measures them know. It controls demand characteristics (participants guessing what is wanted) and experimenter bias (the researcher's expectations shaping results).
- Laboratory experiment: maximum control, artificial setting.
- Field experiment: the IV is manipulated in a real setting (a school, a factory); more natural, less control.
- Quasi-experiment: the researcher cannot randomly assign people to groups because the groups already exist (for example, children who lived through an earthquake compared with those who did not, or men compared with women). Cause can be suggested but not proved as firmly.
5. Correlational method
- Measures two variables as they occur and calculates a correlation coefficient (r): positive (both rise together, e.g. study time and marks), negative (one rises as the other falls, e.g. anxiety and test performance), or zero (no relation).
- Strength depends on how close r is to ±1; the sign gives only the direction. How to read r and how to compute it are in PSY-MS-01 and STA-CR-01.
- Correlation does not prove causation: a third variable may drive both (ice-cream sales and drowning both rise in summer), or the direction may run either way.
- Used when an experiment is impossible or unethical (smoking and illness, poverty and school results), and for prediction.
6. Designs for change with age
| Design | Who is studied | Strength | Weakness |
|---|---|---|---|
| Cross-sectional | different age groups, once | quick, cheap | cohort effect: age groups grew up in different times |
| Longitudinal | the same people, repeatedly over years | shows real individual change | slow, costly, dropouts (attrition), practice effects |
| Cross-sequential | several age groups, each followed for a few years | separates age from cohort effects | complex and costly |
7. Quantitative and qualitative methods
| Point | Quantitative | Qualitative |
|---|---|---|
| Data | numbers, scores, counts | words, narratives, observations, images |
| Aim | test hypotheses, measure, generalise | understand meanings and experience |
| Tools | experiments, tests, structured surveys | in-depth interviews, focus groups, participant observation |
| Sample | large, often random | small, chosen on purpose |
| Analysis | statistics | themes, categories, interpretation |
- Qualitative approaches: in-depth interview; focus group (6–10 people discuss a topic led by a moderator); ethnography (living in a community and describing its culture); narrative / life history; phenomenology (how people experience something); grounded theory (theory built up from the data, not tested on it; Glaser and Strauss, 1967); thematic analysis.
- Content analysis turns text, speeches or media into categories; it can be counted (quantitative) or interpreted (qualitative).
- Mixed methods / triangulation: using both kinds of data on one question, so that each checks the other.
8. Research ethics
- Voluntary participation: people take part only if they choose, and may withdraw at any time without penalty.
- Informed consent: before taking part they are told what the study involves, its risks and benefits, in words they understand. For children, consent from parents and agreement (assent) from the child.
- Debriefing: after the study, participants are told its true purpose and any deception is explained; they should leave in the same state as they came.
- Sharing results with participants where possible.
- Confidentiality: data is kept private and reported without names; anonymity where possible.
- Protection from harm: physical or psychological risk must be minimal and justified by the value of the study. Studies are approved by an ethics committee (in India, institutional ethics committees following ICMR guidelines; in the USA, Institutional Review Boards).
- Studies that changed the rules: Watson and Rayner's "Little Albert" fear conditioning (1920), Milgram's obedience experiments (early 1960s, deception and distress), Zimbardo's Stanford prison study (1971, stopped early).
9. Merits and limits at a glance
| Method | Best for | Main merit | Main limit |
|---|---|---|---|
| Naturalistic observation | real-life behaviour, children, animals | genuine behaviour | no control; observer bias |
| Controlled observation | precise records of behaviour | repeatable | artificial; reactivity |
| Survey / questionnaire | attitudes of large groups | many people, cheap | shallow; false answers; no cause |
| Interview | detail, illiterate respondents | flexible, can probe | slow; interviewer bias |
| Case study | rare or complex single cases | depth | cannot generalise |
| Experiment | cause and effect | control, replicable | artificial; ethical limits |
| Correlational | relations, prediction | studies what cannot be manipulated | no cause |
| Qualitative | meanings, new areas | rich understanding | small samples; subjective |
Common traps
- ✗ "Controlled observation happens in a natural setting." ✓ It happens in a controlled and structured setting; naturalistic observation uses the natural setting.
- ✗ "The DV is what the experimenter changes." ✓ The experimenter changes the IV and measures the DV.
- ✗ "Random assignment and random sampling are the same." ✓ Sampling chooses who takes part (generalisation); assignment decides which group each person joins (cause and effect).
- ✗ "A strong correlation proves cause." ✓ Only an experiment can show cause; correlation shows variables vary together.
- ✗ "Case study is an experimental method." ✓ It is non-experimental: nothing is manipulated.
- ✗ "Longitudinal design compares different age groups at one time." ✓ That is cross-sectional; longitudinal follows the same people over time.
- ✗ "Qualitative and quantitative research differ only in sample size." ✓ They differ mainly in the type of data collected and analysed: words versus numbers.
- ✗ "Debriefing is done before the study." ✓ Informed consent comes before; debriefing comes after.
- ✗ "The Hawthorne effect is observer bias." ✓ Hawthorne is a change in the participants' behaviour because they are watched; observer bias is in the researcher.
Quick revision
- Observation: select, record, analyse; Hawthorne effect = people change when watched.
- Naturalistic = natural setting; controlled = structured setting; participant vs non-participant.
- Survey: many people, questionnaire / interview / phone; how common a problem is.
- Structured interview = fixed questions; unstructured = flexible.
- Case study: one case in depth, non-experimental; clinical method adds treatment (Witmer 1896).
- Experiment: IV changed, DV measured; experimental vs control group; cause and effect.
- Controls: hold constant, counterbalance, random assignment; double-blind stops bias.
- Correlation −1 to +1; never proves cause.
- Cross-sectional = many ages once; longitudinal = same people over years.
- Ethics: voluntary, informed consent, debriefing, sharing results, confidentiality.
Go deeper: Deeper (beyond what APPSC Paper I usually asks)
Deeper Advanced
1. Internal and external validity. An experiment has internal validity when the change in the DV really comes from the IV, not from a confound. It has external (ecological) validity when its result holds outside the laboratory. Tight control raises internal validity but often lowers external validity; field experiments trade the other way.
2. Designs within experiments. Between-subjects (independent groups): each person gets one condition; needs random assignment. Within-subjects (repeated measures): each person gets every condition; needs fewer people but suffers order effects, controlled by counterbalancing (ABBA, or half the group in each order). Matched pairs: pairs matched on a relevant variable (intelligence, age), one to each group.
3. Factorial experiments vary two or more IVs at once (e.g. noise × time of day) and can show an interaction, where the effect of one IV depends on the level of the other.
4. Ex post facto ("after the fact") research studies causes that have already happened (child abuse, brain injury). The researcher selects people by their past, so it resembles a quasi-experiment and cannot rule out other causes.
5. Limits of psychological enquiry (NCERT). Psychological measures have no true zero (zero on an intelligence test does not mean zero intelligence); tests are relative to a culture and need adaptation; and qualitative data are open to subjective interpretation. This is why results are checked with more than one method.
6. Deception and its controls. Deception is acceptable only when the question cannot be studied otherwise, the risk is low, and full debriefing follows. Milgram's work is the standard example of why ethics committees now review designs before a study begins.
Practise this topic. Questions on Observation, survey, interview and case study, with answers explained.
All Methods in psychology notes · APPSC Paper I syllabus · APPSC Paper I Mock 1 with answers · Start free
Medha AI writes its own notes and checks them against the sources listed. Medha AI is an independent preparation service. It is not affiliated with, endorsed by or connected to APPSC, TSPSC, MPSC, UPSC or any other commission or government body.