Every NCERT book that matters for UPSC — open a book, then a chapter. Learn each chapter's concepts, then how UPSC tests them — with diagrams, flowcharts and the past questions to attempt.
Methods of Enquiry in Psychology
Start here: why this chapter matters
Chapter 1 defined psychology as a science of mental processes, experience and behaviour, and said a discipline is defined ‘more’ by how it studies. This chapter is that ‘how’. It sets out the goals and steps of research, two paradigms, the nature of psychological data, six methods (observation, experiment, correlational research, survey, psychological testing, case study), quantitative and qualitative analysis, the limits of psychological measurement, and research ethics. It has Figure 2.1, two boxes, five activities, twelve review questions and two project ideas.
- Builds on What is Psychology? (the hypothetico-deductive model; common sense versus tested knowledge; Dweck's experiment). Leads into every later chapter, each of which reports experiments, tests or case studies — and the Class 12 book's intelligence and attitude chapters, which depend on tests.
- Pairs with the live sociology chapter Doing Sociology: Research Methods (participant observation, field work, surveys and sampling, interviews, objectivity) — read that for sampling and the margin of error, which this chapter does not cover.
- UPSC: the chapter's direct use is GS4 Ethics. Its ethical principles — respect for choice, beneficence, justice, informed consent, confidentiality — answered two GS4 questions in 2026: one on using an untested supplement on an endangered tribe, one on beneficence and a reluctant consent to surgery. Its methods help with evidence-based policy answers (surveys, field experiments, correlation versus cause) in GS2 and GS3, with reading surveys critically in Essay, and with understanding aptitude and personality tests (GS4 ‘aptitude’). CSAT data-interpretation questions reward the same habit of asking what a number does and does not show.
| Word | What it means here |
|---|---|
| Hypothesis | A tentative answer to a research problem, to be tested — e.g. ‘more TV violence, more aggression’ |
| Variable | Any stimulus or event that takes different values and can be measured; an object is not a variable, its attributes are |
| Independent variable (IV) | The variable the researcher manipulates — the presumed cause |
| Dependent variable (DV) | The variable on which the IV's effect is observed — the effect |
| Extraneous variable | Any other variable that could affect the DV and must be controlled |
| Experimental / control group | The group exposed to the IV / an otherwise identical comparison group without it |
| Confederate | A person instructed by the researcher to act in a set way (Box 2.1) |
| Correlation coefficient | A number from +1.0 through 0 to −1.0 showing the strength and direction of a relationship |
| Psychological test | A standardised, objective instrument to assess a person's standing relative to others on a mental or behavioural characteristic |
| Reliability | Consistency of scores — over time (test-retest) or within the test (split-half) |
| Validity | Whether a test measures what it claims to measure |
| Norms | The normal or average performance of a defined group, used to interpret a score |
| Informed consent | Agreement to take part given after learning what the study will involve |
| Debriefing | Explaining the study to participants afterwards and removing any ill effects |
At a glance — the six ideas to carry away
- Goals: description, prediction, explanation, control, application; steps: conceptualise (hypothesis) → collect data → draw conclusions → revise.
- Two paradigms: scientific (predict, measure, control) and interpretive (understand the meanings people give to events); plus reflection on one's own experience.
- Experiment: manipulate the independent variable, measure the dependent variable, control extraneous ones (elimination, constancy, matching, counterbalancing, random assignment).
- Correlation runs from +1.0 through 0 to −1.0: it measures association for prediction, not cause; surveys use interviews, questionnaires and telephone.
- A psychological test is standardised and objective; it needs reliability (test-retest, split-half), validity and norms; tests are verbal, non-verbal or performance; individual or group; speed or power.
- Limits: no true zero, context-bound tools, subjective qualitative data. Ethics: voluntary participation, informed consent, debriefing, sharing results, confidentiality.
1. Goals and steps of psychological enquiry
NCERT names five goals, all to be pursued objectively. Description records a behaviour as accurately as possible so it can be distinguished from others; the researcher must define her terms — ‘study habits’ might mean attending classes, submitting work on time, planning and revising. Prediction forecasts that, under certain conditions, a behaviour will occur within a margin of error — if study time and achievement are positively related, a child who studies more is likely to score well; prediction improves as more people are observed. Explanation seeks the causes or antecedent conditions of a behaviour, to establish cause–effect relations — why are some children more attentive? Control means making a behaviour happen, reducing it or enhancing it by changing its antecedents; change brought about by therapy is NCERT's example. Application aims at positive change in people's lives — yoga and meditation to reduce stress — and at new theories that lead to further research.
Science is defined less by what it studies than by how: its method is objective, systematic and testable. Objectivity means that two or more people studying the same event independently should largely reach the same conclusion — like two friends measuring a table with the same tape. Research is systematic because it follows steps.
- Conceptualising a problem — choose a topic, narrow it into research questions using past research, observation and experience, then frame a hypothesis, a tentative answer. Problems can concern one's own behaviour (why do we forget?), other individuals (why do some chronic patients not take their medicines?), group influence on individuals (why does a cyclist do better before a crowd?), group behaviour (why does risk-taking increase in groups?) and organisations (how can employers raise motivation?).
- Collecting data — draw up a research design, a blueprint covering (a) participants, (b) methods, (c) tools such as interview schedules, observation schedules and questionnaires, and (d) the procedure (individual or group administration); then collect the data.
- Drawing conclusions — analyse the data with graphs (pie charts, bar diagrams, cumulative frequencies) and statistics to verify the hypothesis.
- Revising research conclusions — if the data support the hypothesis, it is confirmed; if not, the researcher revises it or states an alternative and tests it on new data. Research is a continuous process.
- The cyclist question echoes Norman Triplett's 1898 study, usually cited as the first social-psychology experiment (social facilitation — see Chapter 1).
- ‘Why does risk-taking increase in groups?’ refers to the risky shift, reported by James Stoner in 1961 and later generalised as group polarisation — groups move to a more extreme version of their members' initial leaning.
- The TV-violence hypothesis is linked to Albert Bandura's Bobo doll studies (from 1961) on learning aggression by watching models (Chapter 5, Learning).
- NCERT lists five goals; ‘application’ is the fifth, often forgotten.
- ‘Control’ means making a behaviour occur, reducing it or enhancing it — not only suppressing it.
- The text names four steps as conceptualising, collecting, drawing conclusions and revising; the chapter summary words them slightly differently (‘analysing data, drawing and revising research conclusions’). Use the text's four.
- A hypothesis is a tentative answer to be tested — not a conclusion.
2. Two paradigms, and the nature of psychological data
One view holds that behaviour can and should be studied like physics, chemistry and biology: it is predictable, caused by internal and external forces, and can be observed, measured and controlled. For much of the 20th century psychology accordingly restricted itself to overt, measurable behaviour and neglected feelings and meanings. More recently an interpretive method has emerged that values understanding over explanation and prediction: because human experience is complex and variable, it should be studied differently from the physical world, attending to how people give meaning to events in context. For people suffering from a tsunami, earthquake, cyclone or prolonged illness, objective measurement is, NCERT says, ‘neither possible nor desirable’; the researcher, like an explorer mapping unknown terrain, records detailed descriptions without disturbing the natural flow. Both traditions study others; psychology should also help us reflect on our own experience — why I feel sad, why I break my own resolutions.
| Scientific paradigm | Interpretive paradigm | |
|---|---|---|
| Aim | explain and predict; cause and effect | understand meanings |
| View of behaviour | predictable; caused by internal and external forces | complex and variable; interpreted by each person in context |
| Typical data | measurements of overt behaviour | descriptions, narratives, field notes |
| Researcher's stance | controls conditions | explores without disturbing the natural flow — like an explorer |
| NCERT's example | study time and achievement; TV violence and aggression | experiences of survivors of tsunami, earthquake or cyclone |
Data (singular datum) concern overt or covert behaviour, subjective experience and mental processes. They approximate reality and let us verify or falsify ideas, but they are not independent: they are tied to the context (we behave differently alone and in a group, at home and in the office), to the method and theory used, to the characteristics of respondents (individual or group, young or old, male or female, rural or urban), and to the time. A student may report one thing in an interview and do the opposite when observed. Data do not speak for themselves — the researcher must infer their meaning in context. NCERT lists four kinds: demographic (name, age, gender, birth order, siblings, education, occupation, marital status, locality, caste, religion, parental education and income), physical (ecological conditions, mode of economy, housing, facilities, transport), physiological (height, weight, heart rate, fatigue, galvanic skin response, EEG, blood oxygen, reaction time, sleep, blood pressure, dreams, salivation, and running or jumping rates in animal studies) and psychological (intelligence, personality, interests, values, creativity, emotions, motivation, disorders, illusions, delusions, hallucinations, perceptual judgement, thought, consciousness and subjective experience).
- The interpretive paradigm stresses understanding over explanation and prediction — it does not reject evidence.
- NCERT's text lists four kinds of data (demographic, physical, physiological, psychological); the summary lists five, adding ‘environmental’. The text's ‘physical information’ already covers ecology and housing.
- ‘Data’ is plural; ‘datum’ is singular.
- NCERT writes ‘Galvanic Skin Resistance’; the more common term is galvanic skin response — the same measure of sweat-gland activity.
3. The observational method
Observation is a powerful method of describing behaviour. Scientific observation differs from everyday seeing in three ways: selection (the researcher chooses which behaviour to observe, either with a prepared list of activities or to discover what happens), recording (tallies, notes in shorthand or symbols, photographs, video) and analysis (deriving meaning from the record). A good observer knows what, whom, when and where to observe, how to record and how to analyse.
- Naturalistic vs controlled observation — naturalistic observation happens in real-life settings (homes, schools, hospitals, day-care centres) without controlling the situation; when some factors must be controlled, observation is done in the laboratory — controlled laboratory observation, as in Box 2.1, where smoke could be introduced only under controlled conditions.
- Non-participant vs participant observation — the non-participant observer watches from a distance (a video camera, or a seat in the corner of a classroom); the danger is that an outsider's presence may change behaviour. The participant observer becomes part of the group, first building rapport so that members accept her; the degree of involvement depends on the study.
Strengths: people are studied in natural situations, as behaviour occurs. Weaknesses: labour-intensive, time-consuming and open to observer bias — ‘we see things as we are and not as things are’. Hence the rule: record behaviour as it happens; do not interpret it at the time of observation.
- The change in behaviour caused by an observer's presence is called reactivity. A well-known label is the Hawthorne effect, after studies at the Western Electric Company's Hawthorne Works near Chicago in the 1920s and 1930s; later scholars have debated how far the original data really show it.
- Researchers reduce reactivity by staying long enough for people to get used to them (habituation), by unobtrusive recording, and by having two observers code the same behaviour and checking their inter-observer agreement.
- The sociology chapter Doing Sociology treats participant observation and field work in depth (Malinowski; M.N. Srinivas's village studies).
- Naturalistic observation does not control the situation; controlled observation is done in the laboratory.
- In non-participant observation the person may not know she is observed; in participant observation the observer joins the group.
- Observation is mainly a method of description, not of establishing cause.
- The observer should not interpret at the time of observing.
4. The experimental method
Experiments establish cause–effect relationships in a controlled setting: one factor is changed and its effect on another is studied while other factors are held constant. The cause is what is manipulated; the effect is the behaviour that changes. A variable is any stimulus or event that varies and can be measured. An object is not a variable, but its attributes are — a pen is not a variable, but pens of different shapes, sizes and colours are; a room is not, its size is. Variation may be in quality or quantity — hair colour, height, intelligence, or the presence or absence of other people in a room.
The independent variable is the one the researcher manipulates or varies; the dependent variable is the one on which its effect is observed — the phenomenon to be explained. The IV is the cause, the DV the effect. The two are interdependent: neither can be defined without the other. The IV is not the only influence on the DV; other relevant or extraneous variables must be controlled so that the IV–DV relationship can be pin-pointed.
Experiments compare one or more experimental groups, exposed to the IV, with one or more control groups, treated identically except that the IV is absent. In Box 2.1, NCERT counts the room with no one present as the control group and the two rooms with two others present as experimental groups — one with confederates told to do nothing, one with uninstructed participants. The control group reported most, the uninstructed group next, and the confederate group least. Conditions — the speed and amount of smoke, the rooms — were kept the same, and participants were assigned randomly, so that each person had an equal chance of being in any group; had one group been all men and another all women, gender rather than the IV could explain the result.
| Extraneous variable (NCERT) | Examples | Main control technique |
|---|---|---|
| Organismic | anxiety, intelligence, personality, fear, motivation | matching (matched pairs across conditions); random assignment |
| Situational / environmental | noise, temperature, humidity | elimination (a sound-proof, air-conditioned room) or constancy (hold it the same throughout) |
| Sequential | fatigue or practice effects when the same person is tested in several conditions | counterbalancing — order A then B for half, B then A for the other half, or A-B-B-A for each person |
| Background | rural/urban, caste, socio-economic status | matching |
| Any systematic group difference | — | random assignment |
A well-designed experiment gives relatively convincing evidence of cause. Its limits: laboratory settings only simulate the outside world, so results may not generalise — low external validity; some problems cannot ethically be studied experimentally (no one may be starved to study nutrition and intelligence); and it is hard to know and control all relevant variables.
Field experiments are run in natural settings for greater generalisability — e.g. comparing lecture and demonstration teaching in a school — with less control and more time and cost. Quasi experiments (Latin quasi, ‘as if’) are used when the IV cannot be manipulated: the researcher selects naturally occurring groups instead — e.g. children who lost parents in an earthquake (experimental group) and children who lived through it without losing parents (control group).
| Laboratory experiment | Field experiment | Quasi experiment | |
|---|---|---|---|
| Setting | laboratory | natural setting (e.g. a school) | natural setting |
| Independent variable | manipulated | manipulated | selected, not manipulated — naturally occurring groups |
| Control of extraneous variables | highest | less | least |
| Generalisability | lowest (low external validity) | higher | higher |
| NCERT's example | Box 2.1, smoke-filled room | lecture vs demonstration teaching | children orphaned by an earthquake vs not |
- Date: NCERT says the study was conducted ‘in 1970’. Beyond NCERT: Latané and Darley's smoke-filled-room study was published in 1968 (Journal of Personality and Social Psychology); 1970 is the year of their book The Unresponsive Bystander. Safe line: ‘Latané and Darley's smoke-filled-room experiment (late 1960s)’.
- Which group is ‘control’: NCERT calls the alone condition the control group. That is NCERT's labelling; many textbooks simply call it the ‘alone condition’. Either way, the IV is the presence of others and the DV is reporting.
- Latané and Darley's work was prompted by the 1964 murder of Kitty Genovese in New York, where newspaper reports claimed that many neighbours heard and did nothing; later investigations found those reports exaggerated. Their finding, the bystander effect, is explained by diffusion of responsibility and by people reading others' calm as a sign that nothing is wrong.
- Randomised field experiments are now used to test public policy: Abhijit Banerjee, Esther Duflo and Michael Kremer received the 2019 economics Nobel for an experimental approach to alleviating global poverty, much of it based on field trials in India and Kenya.
- IV = cause (manipulated); DV = effect (measured). In Box 2.1, IV = presence of others; DV = reporting of smoke.
- An object is not a variable; its attributes are.
- In a quasi experiment the IV is selected, not manipulated.
- Counterbalancing controls sequence effects; matching controls organismic and background variables.
- The main criticism of laboratory experiments is low external validity.
5. Correlational research
Correlational research asks whether two variables are associated — whether they covary — for prediction, without manipulating either. Is ‘amount of study time’ related to ‘academic achievement’? The strength and direction are expressed as a correlation coefficient from +1.0 through 0.0 to −1.0. Positive: as X rises, Y rises (and as X falls, Y falls) — more study, higher scores; +.85 would be a strong positive association. Negative: as X rises, Y falls — more hours of study, fewer hours for other activities (between 0 and −1.0). Zero: no relationship; exact zeros are rare, but values close to zero (−.02, +.03) show the variables are unrelated.
- A correlation between X and Y may arise because X causes Y, Y causes X (reverse causation) or a third variable causes both (a confounding variable — hot weather in the ice-cream example). Only an experiment, or careful design, can separate these.
- The sign gives the direction; the size gives the strength — −.80 is a stronger relationship than +.30.
- The commonest coefficient is Pearson's r, developed by Karl Pearson in the 1890s; whether a small correlation is ‘significant’ also depends on the sample size.
- Correlational research does not manipulate variables; its purpose is prediction.
- A negative correlation is still a relationship — only zero means none.
- The coefficient cannot exceed +1.0 or fall below −1.0.
6. Survey research: interviews, questionnaires, telephone
Surveys — familiar from election polls — arose to study opinions, attitudes and social facts, at first to establish a baseline of existing reality (literacy, religious affiliation, income; attitudes to family planning or to giving panchayati raj institutions powers over health, education and sanitation). NCERT says they have since evolved into a sophisticated technique that can also help infer some causal relationships. Their techniques are personal interviews, questionnaires, telephone surveys and controlled observation.
An interview is a purposeful, usually face-to-face conversation in which an interviewer asks and a respondent answers, to obtain facts, opinions, attitudes and reasons. Its list of questions is the interview schedule. In a structured (standardised) interview the wording and order are fixed and responses are often specified in advance (close-ended questions); in an unstructured (non-standardised) interview the interviewer decides the questions, wording and order as she goes, and the respondent answers freely (open-ended — ‘How happy are you?’). Combinations are individual-to-individual, individual-to-group (one variant is the focus group discussion), group-to-individual (a job interview panel) and group-to-group. Interviewing is a trained skill: putting respondents at ease, probing vague answers. Interviews give in-depth, flexible information, work with children and non-literate people, and let the interviewer check understanding — but take time and may not be cost-effective.
The questionnaire is the most common, simple, versatile and low-cost self-report method: a predetermined set of questions answered in writing, distributed to groups or by mail — like a highly structured interview. Questions may be open-ended or closed-ended (yes/no, true/false, multiple choice, rating scales of 3, 5, 7, 9, 11 or 13 points, or ranking). Questionnaires collect background and demographic information, past behaviour, attitudes and opinions, knowledge, and expectations. Mailed questionnaires suffer poor response. Telephone surveys (and SMS polls) save time but meet uncooperativeness and superficial answers, and those who respond may differ from those who do not in age, gender, income, education and psychological traits — producing biased results.
| Interview | Questionnaire | |
|---|---|---|
| Mode | oral, usually face-to-face (sometimes by phone) | written self-report, in person or by mail |
| Flexibility | high, especially when unstructured; can probe and rephrase | low — fixed questions, like a highly structured interview |
| Depth | in-depth | usually shallower |
| Who can respond | also children and non-literate people | requires literacy |
| Cost and time | time-consuming; an hour or more per person | cheap and quick for large groups |
| Main risk | interviewer influence; cost | poor response (mailed); misreading questions |
Advantages: information from thousands of people can be gathered quickly and efficiently, and opinion on new issues obtained almost as soon as they arise. Limitations: people may give inaccurate information through memory lapses or reluctance to reveal what they really believe, and may give the answers they think the researcher wants.
- Answering what the researcher seems to want is social desirability bias; differences between responders and non-responders produce non-response bias; a sample of volunteers or Internet users suffers self-selection or coverage bias.
- The 5-point agree–disagree format is the Likert scale, after Rensis Likert (1932).
- How a sample should be drawn, and how large it must be, is covered in the live chapter Doing Sociology: Research Methods (random and stratified sampling, margin of error). India's official surveys — the National Sample Survey and the National Family Health Survey — are large-scale examples.
- Structured interview → close-ended questions; unstructured interview → open-ended.
- A focus group discussion is a variant of the individual-to-group interview.
- The questionnaire is a self-report method; its main problem when mailed is poor response.
- NCERT's survey techniques include controlled observation as well as interviews, questionnaires and telephone.
7. Psychological testing
Assessing individual differences has been a concern of psychology from the start. Tests exist for intelligence, aptitude, personality, interests, attitudes, values and educational achievement, and are used for selection, placement, training, guidance and diagnosis in schools, clinics, industry and defence. A test consists of items (questions with possible responses) related to one clearly defined characteristic; it is usually meant for a given age group and may or may not have a time limit. Technically, a psychological test is a standardised and objective instrument used to assess an individual's standing relative to others on some mental or behavioural characteristic.
- Objectivity — two or more researchers testing the same group should obtain more or less the same scores. Items must mean the same to every reader; instructions, conditions, time limit, mode of administration and scoring must be specified in advance.
- Reliability — consistency of scores. Test-retest reliability (temporal stability) is the correlation between scores on two occasions, say 20 days apart. Split-half reliability (internal consistency) divides the test into odd and even items and correlates the two halves — items from the same domain should correlate; ‘apples and oranges’ will not.
- Validity — does the test measure what it claims to? A ‘mathematics achievement’ test may actually be measuring language proficiency.
- Norms — the average performance of a large group, set by age, sex, place of residence and so on. A test becomes standardised when norms are developed; norms let an individual's score be interpreted against others like her.
| Basis of classification | Types | Notes (NCERT) |
|---|---|---|
| Language | verbal · non-verbal · performance | verbal needs literacy; non-verbal items use symbols or pictures; performance tests require moving objects in a set order |
| Mode of administration | individual · group | individual: one person, face to face, time-consuming but suits children and those without the language; group: quick and easy, but respondents may lack motivation or fake answers |
| Difficulty and time | speed · power | speed: time limit, items of equal difficulty; power: no time limit, items in increasing difficulty. Most tests mix both |
Tests must be selected and used carefully. No decision should rest on a single test; test data should be combined with information about the person's background, interests and past performance.
- Because each half of a split test is shorter than the whole, the half-test correlation is usually adjusted upwards with the Spearman–Brown formula.
- Common kinds of validity: content (do items cover the domain?), criterion (do scores predict an outside measure, such as job performance?) and construct (does the test behave as the theory of the trait predicts?).
- A test can be reliable without being valid, but not valid without being reasonably reliable.
- The armed forces' Services Selection Boards use psychological tests in officer selection, developed with the Defence Institute of Psychological Research (DRDO); the Civil Services Examination's interview is formally a Personality Test.
- Test-retest = temporal stability; split-half = internal consistency.
- A test becomes standardised when norms are developed.
- Speed tests: items of equal difficulty; power tests: items in increasing difficulty, no time limit.
- Performance tests need movement of objects, not language.
8. Case study, combining methods, and analysing data
A case study is an in-depth study of a case that can give critical information on a little-understood phenomenon: an individual (a patient with a psychological disorder), a small group with something in common (creative writers such as Rabindranath Tagore and Mahadevi Verma), an institution (a poorly or successfully run school or office) or an event (children exposed to tsunami, war or vehicular pollution). Cases are chosen because they are unique and information-rich. Case studies use multiple methods — interviews, observation, tests — and many informants, and produce detailed narratives of feelings, fantasies, hopes, fears, trauma and upbringing. They are valuable in clinical and developmental psychology: Freud's psychoanalytic theory grew from meticulous case records, and Piaget developed his theory of cognitive development from observing his three children. Minturn and Hitchcock studied the socialisation of children among the Rajputs of Khalapur, and S. Anandalakshmy studied childhood in a weavers' community in Varanasi.
But generalising from one case needs great caution; validity is a real problem. NCERT recommends collecting information by multiple strategies, from different sources, by several investigators, with careful planning and a chain of evidence linking each source to the research questions. More generally, every method has strengths and limits, so researchers should combine two or more; if the methods converge, confidence rises.
| Method | Main purpose | Key strength | Key limitation (NCERT) |
|---|---|---|---|
| Observation | describe behaviour | natural behaviour as it occurs | labour-intensive; observer bias; observer may change behaviour |
| Experiment | establish cause and effect | relatively convincing evidence of cause | low external validity; some problems unethical to test; hard to control all variables |
| Correlational | find association; predict | studies variables that cannot be manipulated | does not manipulate — so cause is not established |
| Survey | opinions, attitudes, social facts | fast, large numbers, new issues quickly | memory lapses; concealment; answers the researcher wants |
| Psychological test | individual differences | standardised, objective comparison with norms | context-bound; group tests invite faking; never rely on one test |
| Case study | in-depth understanding of one case | rich narrative detail | hard to generalise; validity problem |
Quantitative analysis: tests, questionnaires and structured interviews use close-ended, often scaled responses (1 to 5, 7 or 11), or right/wrong answers scored 1 and 0, summed into an aggregate score — the attribute is converted into a number. Scores are compared between individuals and groups using statistics (mean, median, mode; range, quartile deviation, standard deviation; correlation). Qualitative analysis: some experience — how a mother feels about losing her child — loses its organisation if quantified; one needs to hear her story (narrative analysis). Data from participant observation or unstructured interviews (people's own words, field notes, photographs, recordings, informal talk) are analysed by content analysis, building thematic categories with examples. The two approaches are complementary, not contradictory.
- Leigh Minturn and John T. Hitchcock's study of the Rajputs of Khalapur (in present-day Uttar Pradesh) was part of the Six Cultures project on child-rearing, published in the early 1960s.
- Famous single cases in psychology include Phineas Gage (1848), whose personality changed after an iron rod passed through his frontal lobe, and the patient H.M. (Henry Molaison), whose 1953 surgery left him unable to form new long-term memories (Chapter 6). They show both the power of a case and why it needs confirmation.
- Using several methods so that they check one another is called triangulation.
- A ‘case’ can be a person, a small group, an institution or an event.
- Piaget observed his three children.
- Narrative analysis and content analysis are qualitative methods.
- Quantitative and qualitative methods are complementary, NCERT says — not rivals.
9. Limitations of psychological enquiry, and research ethics
NCERT names three general limitations. (1) No true zero point: physical measurement starts from zero, but no one has zero intelligence; psychologists fix a zero arbitrarily, so scores are relative. Ranks show the problem: out of 50, the first rank holder may score 48, the second 47 and the third 40 — the gap between first and second is not the gap between second and third. (2) Relative nature of tools: a test made for urban students (multistoreyed buildings, aeroplanes, the metro) does not suit tribal children more familiar with local flora and fauna; western tests may not suit India without adaptation. (3) Subjective interpretation of qualitative data: interpretations vary, so field work should be done by more than one investigator who agree on meanings — ideally with respondents involved in the meaning-making.
Research on people must follow ethics. NCERT names three principles — respect for persons' privacy and choice to participate, beneficence (protecting participants from harm) and justice (sharing the benefits of research with all participants) — and five practices:
- Voluntary participation — people choose whether to take part, without coercion or excessive inducement, and may withdraw without penalty once the study has begun.
- Informed consent — participants learn what will happen before data are collected. This matters because some experiments use electric shock or unpleasant stimuli, ask for private information, or use deception (false feedback such as ‘you are very intelligent’ or ‘you are incompetent’).
- Debriefing — after the study, participants are told what they need to complete their understanding, especially if deceived; they should leave in the same physical and mental state as they came, reassured, with any anxiety removed.
- Sharing the results — it is the researcher's obligation to return and share findings; this meets participants' expectations and their comments may yield new insight.
- Confidentiality — information is kept in strict confidence, used only for research and never passed on. Best of all, do not record identities; where that is impossible, use code numbers, keep names separately, and destroy the identification list when the research is over.
| Classic study (Beyond NCERT) | What was done | Critics argued | Defenders argued |
|---|---|---|---|
| Solomon Asch, conformity (published 1951) | participants judged line lengths after confederates gave wrong answers | deception, and mild embarrassment | harm was minimal; the deception was needed and explained afterwards |
| Stanley Milgram, obedience (published 1963) | participants were told to give what they believed were increasingly strong electric shocks to a ‘learner’ (a confederate; no shocks were real) | Diana Baumrind (1964): severe stress, deception and loss of trust were not justified | Milgram (1964): participants were debriefed, a follow-up found most glad to have taken part, and the findings on obedience were important |
| Philip Zimbardo, Stanford Prison Experiment (1971) | students role-played guards and prisoners; planned for two weeks, stopped after six days | real distress; the researcher was also ‘superintendent’; later critics (e.g. Thibault Le Texier, 2019) argued guards were coached, questioning the findings | it dramatised the power of situations and roles; Zimbardo himself later acknowledged its ethical failings |
| Latané and Darley, smoke-filled room (1968; Box 2.1) | participants were not told the ‘emergency’ was staged | deception and brief alarm | short, low-risk, debriefed; impossible to study real emergencies otherwise |
- Strict view (associated with Diana Baumrind and others): deception violates informed consent and erodes public trust in research, and should be avoided.
- Conditional view (the American Psychological Association's ethics code): deception is acceptable only when the study's value justifies it, no non-deceptive method is feasible, it is not expected to cause pain or severe distress, and participants are debriefed as early as possible.
- NCERT's three principles match the Belmont Report (United States, 1979): respect for persons, beneficence, justice. Earlier milestones: the Nuremberg Code (1947) and the World Medical Association's Declaration of Helsinki (1964).
- In India, research on human participants follows the ICMR's National Ethical Guidelines for Biomedical and Health Research Involving Human Participants (2017), reviewed by institutional ethics committees; personal data are also protected by the Digital Personal Data Protection Act, 2023.
- NCERT's three principles: respect, beneficence, justice. Justice here means sharing the benefits of research with participants.
- The text lists five practices; the summary mentions only three (voluntary participation, informed consent, sharing results). Learn all five, including debriefing and confidentiality.
- Debriefing comes after the study; informed consent before.
- ‘No true zero’ is why psychological scores are relative, not absolute.
- Asch 1951, Milgram 1963, Zimbardo 1971 are Beyond NCERT — do not attribute them to the textbook.
NCERT's activities and in-text questions — short answers
The chapter has five activities and two project ideas. Short outlines follow.
Activity 2.1 (page 26): Observe one psychology period — what the teacher does, what students do, and their interaction. Compare observations.
Activity 2.2 (page 29): Identify the independent and dependent variables.
- IV: teachers' classroom behaviour; DV: students' performance.
- IV: parent–child relationship; DV: children's emotional adjustment.
- IV: level of peer pressure; DV: level of anxiety.
- IV: enrichment of the environment (special books and puzzles); DV: children's performance.
Activity 2.3 (page 32): Will an Internet questionnaire on welfare programmes reflect the general population? Why or why not?
Activity 2.4 (page 34): Examine a test and its manual.
Activity 2.5 (page 35): Choose the most appropriate method for each problem.
- Noise and problem solving — experiment (noise as IV).
- Dress code for college students — survey of opinion.
- Attitudes of students, teachers and parents to homework — survey (questionnaire or interviews), perhaps an attitude scale.
- A student's behaviour in a playgroup and in class — naturalistic observation.
- Major life events of a favourite leader — case study (biographies, speeches, interviews).
- Anxiety level of Class XI students — psychological test (a standardised anxiety scale).
Project 1 (page 39): Survey after-school activities of Class V and Class IX students (10 each).
Project 2 (page 39): Effect of reading aloud on learning a poem (10 six-year-olds, two groups).
NCERT exercise — answer outlines
The twelve review questions at the end of the chapter (page 39). Try each first, then open the outline.
1. What are the goals of scientific enquiry?
2. Describe the various steps involved in conducting a scientific enquiry.
3. Explain the nature of psychological data.
4. How do experimental and control groups differ? Explain with the help of an example.
5. A researcher is studying the relationship between speed of cycling and the presence of people. Formulate a relevant hypothesis and identify the independent and dependent variables.
6. Discuss the strengths and weaknesses of experimental method as a method of enquiry.
7. Dr. Krishnan is going to observe and record children's play behaviour at a nursery school without attempting to influence or control the behaviour. Which method of research is involved? Explain the process and discuss its merits and demerits.
8. Give two examples of the situations where survey method can be used. What are the limitations of this method?
9. Differentiate between an interview and a questionnaire.
10. Explain the characteristics of a standardised test.
11. Describe the limitations of psychological enquiry.
12. What are the ethical guidelines that a psychologist needs to follow while conducting a psychological enquiry?
What to remember
Numbers
One-liners
- Methodology is what makes psychology a scientific endeavour.
- Science is defined less by what it investigates than by how — objectively, systematically, testably.
- A hypothesis is a tentative answer to a research problem.
- Data are tied to context, method and theory; they must be interpreted.
- The interpretive paradigm values understanding over explanation and prediction.
- Scientific observation means selection, recording and analysis — record first, interpret later.
- Independent variable = cause, manipulated; dependent variable = effect, measured.
- Quasi experiments select the independent variable from naturally occurring groups.
- Correlation predicts; it does not by itself prove cause.
- A test is standardised once norms are developed; it must be reliable and valid.
- Never rely on a single test or a single method; convergent methods raise confidence.
- Psychological measurement has no true zero — scores are relative.
Traps to avoid
- Five goals of enquiry, not four — application is the fifth.
- In Box 2.1 the IV is the presence of others; the DV is reporting the smoke.
- NCERT dates Box 2.1 to 1970; the published paper is from 1968.
- An object is not a variable; its attributes are.
- In a quasi experiment the IV is selected, not manipulated.
- Counterbalancing controls sequence effects; matching controls organismic and background variables.
- A negative correlation is still a relationship; only zero means none.
- Test-retest reliability = temporal stability; split-half = internal consistency.
- Speed tests: equal difficulty with a time limit; power tests: increasing difficulty, no time limit.
- Structured interview → close-ended questions; unstructured → open-ended.
- Box 2.2's multiple-answer percentages add to more than 100 — that is not an error.
- Debriefing is after the study; informed consent is before.
- Justice, in NCERT's ethics, means sharing research benefits with participants.
How to strategise
Prelims — what UPSC asks
- No Prelims question in the archive tests this chapter directly. What it gives Prelims is precision with words that recur in science, economy and governance statements: ‘randomised’, ‘control group’, ‘sample survey’, ‘correlation’, ‘reliability’. Read statements about surveys (the National Sample Survey, the National Family Health Survey) or trials with the chapter's distinctions in mind.
Mains — where it appears and how to answer
- GS4 Ethics: research and medical ethics questions — the 2026 case of an untested supplement for an endangered tribe and the 2026 question on paternalism and beneficence in a reluctant consent to surgery. Use the chapter's frame: respect (voluntary, informed consent), beneficence (protection from harm), justice (vulnerable groups not used as test populations; benefits shared), confidentiality; add the deception debate (strict vs conditional) when relevant. The 2026 case of a surgery performed on a sister-in-law's consent (Lata) is also about informed consent, but there the issue is clinical surrogate consent rather than research.
- GS4 aptitude and the civil service: psychological tests show how aptitude and personality are assessed; their reliability, validity and norms — and the warning never to decide on one test — make a good critical paragraph on selection and assessment.
- GS2 / GS3 evidence-based policy: surveys establish baselines; field experiments and randomised trials test schemes (the 2019 economics Nobel); correlation is not causation; tools must be adapted to context. A sentence of each gives policy answers analytical depth.
- Essay: topics on truth, data, perception or judgement can borrow the chapter's point that data are tied to context, method and theory and must be interpreted — and the interpretive paradigm's insistence on people's own meanings.
Self-check
Tap an option. NCERT's own exercise questions come first; the rest test the figures and common confusions.
1. Which of the following is NOT one of the five goals of psychological enquiry named by NCERT?
2. In Latané and Darley's smoke-filled-room study (Box 2.1), the dependent variable was:
3. A researcher compares children who lost parents in an earthquake with children who lived through it without such loss. This is a:
4. Giving half the participants task A then B, and the other half B then A, controls:
5. A correlation coefficient of −0.80 between two variables indicates:
6. Correlating scores on the odd-numbered and even-numbered items of a test gives an estimate of its:
7. A test becomes ‘standardised’, according to NCERT, when:
8. Explaining the purpose of a study to participants after it ends, and removing any anxiety caused by deception, is called:
UPSC questions from this chapter
Open any question to attempt it in the Question bank — your attempt is saved to your progress.