Doing Sociology: Research Methods

Why this chapter matters for UPSC: UPSC loves data. GS1 Mains answers that cite Census 2011 sex ratios, NFHS-5 child nutrition figures, PLFS employment data, or NCRB crime statistics score higher than answers without data. This chapter teaches you where that data comes from, how it is collected, what its limitations are — and how to use it critically. Understanding research methods also helps you evaluate policies and government schemes (are they evidence-based? what data are they using?).

PART 1 — Quick Reference Tables

Major Research Methods in Sociology

Key Indian Data Sources

Sampling Methods

PART 2 — Detailed Notes

Quantitative vs Qualitative Research

Sociological research divides broadly into two traditions:

Quantitative research follows the model of natural science:

• Uses numerical data
• Aims to measure, count, and quantify social phenomena
• Seeks to establish correlations and causal relationships
• Methods: surveys, structured observation, secondary data analysis
• Goal: generalise from sample to population

Qualitative research follows the interpretive tradition:

• Uses textual, visual, and narrative data
• Aims to understand meanings, processes, and contexts
• Seeks to describe and theorise rather than measure
• Methods: participant observation, ethnography, in-depth interviews, case studies
• Goal: depth and complexity, not breadth and representativeness

Most good research uses mixed methods — combining quantitative baseline data with qualitative depth analysis.

The Survey Method

The survey is sociology's most widely used method. It involves administering a standardised questionnaire to a sample of a population. Key steps:

1. Define research question and population (who are you studying?)
2. Choose sampling method (how do you select cases?)
3. Design questionnaire (closed questions for quantitative data; open questions for qualitative)
4. Pilot test questionnaire with a small sub-sample
5. Field work — data collection
6. Data analysis — statistical (SPSS, R) or content analysis
7. Interpretation and reporting

Limitations:

• Social desirability bias: People give answers they think are expected, not true answers (e.g., on domestic violence, caste discrimination, sexual behaviour)
• Literacy and language barriers in India
• Interviewer effects: The social characteristics of the interviewer (caste, gender, age) affect responses
• Snapshot problem: Surveys capture a moment in time; social reality is processual

💡 Explainer: The Census of India

The Census of India is the world's largest administrative exercise in data collection — conducted by the Office of the Registrar General of India under the Census Act, 1948.

Key features:

• Decennial: Every 10 years. Last completed Census: 2011. Census 2021 has been delayed (first due to COVID-19, then administrative delays).
• De facto enumeration: People counted where they are found on census night (not where they usually live)
• Comprehensive: Every person in the country is enumerated — households, individuals, buildings
• Two phases: House-listing and Housing Census; Population Enumeration
• Key data collected: Population size, density, sex ratio, literacy (by age, gender, caste category), religion, language, occupation, migration, disability, housing conditions (pucca/kutcha, water supply, sanitation)
• Schedule Caste/Tribe data: SC and ST populations enumerated separately for reservation purposes

What Census does NOT collect:

• Income or expenditure (that is NSS/PLFS)
• Health data (that is NFHS/SRS)
• Caste data other than SC/ST (the last comprehensive caste census was 1931; the SECC 2011 collected some caste data)

Sample Registration System (SRS): Continuous survey running alongside Census to provide annual vital statistics (birth rates, death rates, infant mortality, maternal mortality) between census years. Based on a nationally representative sample.

📌 Key Fact: Census 2011 — Essential Data Points

These are the figures you need to cite with confidence in UPSC answers (note: Census 2021 is pending as of 2026):

• Total population: 1.21 billion (121 crore)
• Decadal growth rate: 17.64% (2001–2011)
• Sex ratio: 940 females per 1,000 males (overall); child sex ratio (0–6): 919 (alarming)
• Literacy rate: 74.04% (males: 82.14%; females: 65.46%)
• Urban population: 31.16% (377 million)
• Rural population: 68.84%
• SC population: 16.6% of total
• ST population: 8.6% of total

The NFHS: Health and Social Indicators

The National Family Health Survey is India's primary source of data on health, nutrition, and family planning. Conducted by the International Institute for Population Sciences (IIPS), Mumbai, with state nodal agencies. Funded by the Ministry of Health and Family Welfare.

NFHS-5 (2019–21) key findings:

• Total Fertility Rate (TFR): 2.0 (below replacement level of 2.1 for the first time)
• Under-5 mortality rate: 42 per 1,000 live births
• Stunting (under-5): 35.5%
• Wasting (under-5): 19.3%
• Anaemia in women (15–49): 57%
• Child marriage (women 20–24 married before 18): 23.3%
• Women owning mobile phone: 53.9%

🎯 UPSC Connect: Using Data Critically

Merely citing data is not enough for a good UPSC answer. You must:

1. Cite the source and year: "According to NFHS-5 (2019–21)..."
2. Contextualise the trend: Is it improving? At what pace? Compared to what baseline?
3. Note regional variation: National averages mask state-level divergences (Kerala vs UP; Tamil Nadu vs Bihar)
4. Acknowledge data limitations: NCRB undercounts crimes against women; PLFS may undercount women's work in informal agriculture

For example, on the question "Discuss the status of women in India":

• Do NOT just say "women face discrimination"
• DO say: "NFHS-5 data shows 57% of women aged 15–49 suffer from anaemia, reflecting nutritional neglect; the child sex ratio of 919 (Census 2011) indicates son preference and possible sex-selective practices; however, women's workforce participation, at 32.8% per PLFS 2022–23, has increased from a nadir of 23% in 2017–18..."

Observation Methods

Participant observation involves the researcher joining the group being studied and participating in its activities, while simultaneously observing and recording data. It is the hallmark method of social/cultural anthropology and qualitative sociology.

M.N. Srinivas's fieldwork in Rampura village (Karnataka) is India's most famous example. He lived in the village, participated in daily life, attended ceremonies, and recorded detailed field notes — producing his classic The Remembered Village (1976).

Strengths:

• Access to naturally occurring behaviour (not just what people say they do)
• Ability to observe what people take for granted and do not mention
• Understanding context, meaning, and process

Challenges:

• Going native: Risk of over-identification with the group, losing analytical distance
• Observer effect: People modify behaviour when observed
• Ethical issues: In covert observation, the group does not know they are being studied

Non-participant observation is more structured — the researcher watches but does not participate. Used for studying behaviour in public spaces (traffic patterns, queuing behaviour, crowd dynamics).

Case Study Method

A case study is an in-depth examination of a single case — a person, family, village, organisation, community, or event. Famous sociological case studies include:

• Whyte's Street Corner Society (1943): Deep study of an Italian-American gang in Boston
• Srinivas's The Remembered Village (1976): Rampura village, Karnataka
• Ambedkar's The Problem of the Rupee (1923): A single case (Indian currency policy) used to argue for monetary reform

Case studies sacrifice generalisability for depth. They are best used to:

• Generate hypotheses for later large-scale testing
• Understand process and mechanism (how does X cause Y?)
• Study unique or extreme cases

Ethical Issues in Sociological Research

Research ethics govern the relationship between researcher and research subjects.

Core ethical principles:

1. Informed consent: Participants must be told what the research is about and agree to participate freely (without coercion or deception)
2. Confidentiality/Anonymity: Personal information must be protected; participants must not be identifiable from research reports (unless they consent)
3. Do no harm: Research must not damage the physical, psychological, social, or economic interests of participants
4. Voluntary participation: Participants can withdraw at any time
5. Accuracy: Data must be reported honestly; results must not be fabricated or selectively reported

Ethical dilemmas in Indian research context:

• Researching marginalised communities (Dalits, tribal women, sex workers) — power imbalance between researcher and researched
• Covert research in sensitive political contexts (studying Naxal-affected communities)
• Government data that is collected for one purpose being repurposed for surveillance
• Community consent vs individual consent in tribal communities

Limitations of Official Statistics

Official statistics (Census, NCRB, PLFS) are produced by the state for administrative purposes. Sociologists treat them as social constructs, not objective facts, because:

1. Definition effects: "Literacy" defined as ability to read and write one's name — a very low bar; actual functional literacy is far lower
2. Undercounting of stigmatised activities: Rape, domestic violence, caste violence — actual incidence far exceeds reported figures in NCRB
3. Overcounting of political priorities: Some crimes may be over-reported when there is political pressure to show "action"
4. Classification effects: The categories used (SC, ST, OBC, urban/rural) shape what can be measured and what is invisible
5. Gender blindness: Women's unpaid work (care, domestic, subsistence agriculture) is not counted in GDP or employment statistics

PART 3 — Frameworks & Analysis

Evaluating a Research Study: Five Questions

When evaluating any sociological study (or policy evaluation) in UPSC answers:

1. Sample: Who was studied? Is the sample representative? How large?
2. Method: How was data collected? Are there method-specific biases?
3. Operationalisation: How were key concepts defined and measured? (e.g., how is "poverty" measured?)
4. Ethics: Were participants' rights protected?
5. Generalisability: Can conclusions be applied beyond the study sample?

The Data Ecosystem for Indian Society Questions

Exam Strategy

Prelims: Census key statistics (sex ratio, literacy, population), NFHS (fertility rate, child nutrition), SRS (IMR, MMR), difference between PLFS/NSS, what NCRB publishes — all tested directly in Prelims data questions.

Mains GS1: Every answer on Indian society demographics should have at least two precise data citations. Use NFHS-5 for health/women/nutrition. Use Census 2011 (acknowledge 2021 delay) for population/literacy/sex ratio. Use PLFS for employment.

Mains GS3: Research methodology is relevant to questions on data governance, evidence-based policymaking, NSSO controversies (2016–17 consumer expenditure survey not released), and statistical institutions.

Practice Questions

1. UPSC Mains GS1 2020: "Discuss the main features of the National Population Policy 2000. How has it impacted India's demographic transition?" (Use SRS and NFHS data on TFR, MMR, IMR to assess impact.)

2. UPSC Mains GS1 2017: "Critically examine the data regarding child sex ratio in India. What are the socio-cultural factors responsible for this problem?" (Census 2011 child sex ratio 919; use NFHS; apply ethnocentrism vs cultural relativism debate.)

3. UPSC Mains GS2 2021: "Can the National Commission for Women be strengthened as an effective institution to safeguard the rights of women?" (Use NCRB data on crimes against women; NFHS data on domestic violence; critique of official statistics.)

4. UPSC Mains GS3 2018: "How do you think that the role of statistics in social science research has changed over time? Discuss with reference to India." (Apply: evolution from Census to big data; limitations of official statistics; ethics of data collection.)