Use of Statistical Tools — Designing a Survey Project
A hands-on capstone chapter that teaches how to apply every statistical tool learned earlier by designing, conducting and reporting a survey-based project on a real economic issue.
Prelims seldom quotes this chapter directly but heavily tests its vocabulary and method — census vs sample, primary vs secondary data, class midpoint, attribute vs variable — and the agencies behind official data (Census, NSSO/MoSPI). For Mains GS-III it underpins the case for reliable statistics and evidence-based policymaking, while its survey logic powers CSAT data-interpretation.
Understand the chapter
What This Chapter Is Really About
This is the capstone chapter of Class 11 Statistics for Economics: instead of teaching a new tool, it shows how all earlier tools fit together in a real survey project. The aim is to take an economic problem and move systematically from question to data to conclusion. It trains you to analyse a product, service or social issue and suggest improvements.
- Tools apply to production, consumption, distribution, banking, insurance, trade, transport, etc.
- Typical study areas: demand for a product, drinking-water/electricity problems, consumer awareness.
- End product = a survey-based project report with actionable suggestions.
The Seven-Step Project Pipeline
A statistical project follows a fixed logical sequence, each step feeding the next. Skipping or reordering steps weakens the analysis. Memorising this pipeline is the single most exam-useful takeaway.
- 1 Identify the problem/area; 2 Choose the target group; 3 Collect data.
- 4 Organise and present (tabulation, diagrams); 5 Analyse and interpret.
- 6 Draw conclusions with predictions/policy suggestions; 7 List the bibliography.
Target Group and Choice of Data
The objective decides whom you study and what kind of data you need. The target group narrows respondents (e.g., middle/high-income groups for cars; all consumers for soap) so questions can be framed sharply. Data may be primary (first-hand), secondary (already collected), or both.
- Primary data: via questionnaire or interview schedule — personal interview, postal/mail, phone, email.
- Secondary data: used when there is paucity of time, money and manpower and information is readily available.
- Census method = study every unit; sampling = study a representative subset (ensure the sampling method is suitable).
- Postal questionnaire must carry a covering letter stating the purpose of inquiry.
Designing the Questionnaire
The questionnaire is the engine of primary data — it must generate exactly the information the study needs, no less. A tested questionnaire from a similar study can be reused after suitable modification; otherwise build one carefully. The toothpaste sample captured income, current brand, monthly expenditure, preferred ingredients and media influence.
- Mixes personal details, closed questions (Yes/No, tick-options) and quantitative questions.
- Targets six key facts: average spend, brands in demand, attitudes, ingredient preference, media influence, and their relation to income.
Organising, Presenting and Analysing Data
Raw data is first organised through classification and tabulation, then presented visually. Bar diagrams, pie diagrams and histograms make patterns readable at a glance. Statistical analysis then extracts meaning from the numbers.
- Central tendency (e.g., mean) gives the average/representative value.
- Dispersion (e.g., standard deviation) gives variability/spread.
- Correlation gives the relationship between variables (e.g., income vs expenditure).
The Sample Project — Toothpaste Survey
Entrepreneur X wants to set up a toothpaste factory and commissions a primary survey of 100 households to study tastes, spending and brand demand. The findings profile the consumer and guide the business decision. These exact numbers are the most quotable part of the chapter.
- Sample = 100 households; Urban 67%, Rural 33%; most users aged 25-50 with 3-6 members.
- Mean monthly income Rs 18,000 (SD Rs 9,000); mean toothpaste spend Rs 104 (SD Rs 35.60).
- Top brands: Pepsodent, Colgate, Close-up; preferred ingredients: gel and antiseptic.
- Biggest media influence: Television, then Newspaper.
Conclusion and Bibliography
The final steps convert analysis into insight: draw meaningful conclusions and, where possible, predict future prospects and suggest growth or policy measures. The bibliography then credits every secondary source used. Together they close the loop from problem to recommendation.
- Conclusion should link findings back to the original objective.
- Bibliography lists only secondary sources — magazines, newspapers, research reports.
Key terms
- Census Method
- Data collection in which observations are taken on every individual in the population (complete enumeration).
- Attribute
- A qualitative characteristic that cannot be measured numerically (e.g., gender, occupation).
- Constant
- A quantity describing an attribute that does not change during the investigation.
- Continuous Variable
- A quantitative variable that can take any numerical value within a range.
- Discrete Variable
- A quantitative variable that takes only certain values, changing by finite jumps.
- Class Midpoint (Class Mark)
- The representative middle value of a class = (upper class limit + lower class limit)/2.
- Assumed Mean
- An approximate value chosen to simplify the calculation of the mean.
- Bivariate Distribution
- A frequency distribution involving two variables together.
- Decile
- A partition value that divides ordered data into ten equal parts.
- Enumerator
- The person who actually collects the data in a survey.
Must-know facts exam-ready
- Sample project = toothpaste by entrepreneur X; sample size 100 households (Urban 67%, Rural 33%).
- Mean monthly family income = Rs 18,000; standard deviation = Rs 9,000.
- Mean monthly toothpaste expenditure = Rs 104 per household; standard deviation = Rs 35.60.
- Top three preferred brands: Pepsodent, Colgate, Close-up.
- Most influential medium = Television (followed by Newspaper).
- Census method = data on ALL individuals in the population; sampling studies only a subset.
- Class midpoint (class mark) = (upper class limit + lower class limit)/2.
- Decile divides data into 10 equal parts; assumed mean is an approximate value to simplify calculation.
- Secondary data is preferred when there is paucity of time, money and manpower.
- Primary data tools = questionnaire and interview schedule; postal questionnaire must carry a covering letter.
- Tied fact: Census of India is conducted under the Census Act, 1948 by the Registrar General and Census Commissioner (Ministry of Home Affairs).
- Tied fact: MRP is regulated under the Legal Metrology Act, 2009; food adulteration under the Food Safety and Standards Act, 2006 (FSSAI); consumer rights under the Consumer Protection Act, 2019.
Memory tricks remember it for good
Traps to avoid
- Census vs Sample: census = ALL units (complete enumeration), sample = a subset — UPSC swaps these definitions.
- Primary vs Secondary data: primary is first-hand (questionnaire/interview); secondary is pre-existing and chosen to save time, money and manpower — not because it is more accurate.
- Class mark/class midpoint = (upper + lower limit)/2 — it is NOT the class frequency or the class interval.
- Discrete vs Continuous: discrete moves in finite jumps and takes only certain values; continuous can take any value — questions flip these.
- Attribute is qualitative and cannot be measured; a variable is quantitative — don't treat occupation or gender as variables.
- Sample size is 100 households, but the age table totals 500 persons (family members) — don't misread the base.
Exam focus
🧠 Prelims angles
- Definition-matching of glossary terms: census method, assumed mean, class midpoint, decile, attribute, discrete vs continuous variable.
- Census vs sampling and primary vs secondary data — sources, merits and when each is used.
- Questionnaire vs interview schedule; the covering-letter requirement for postal questionnaires.
- Matching data to diagram: bar diagram, pie diagram, histogram (and which suits a frequency distribution).
- Official data agencies tied to such surveys: Census of India (Census Act, 1948; Registrar General, MHA) and NSSO under MoSPI.
- Regulatory pegs from the project list: MRP (Legal Metrology Act, 2009), food safety (FSSA, 2006/FSSAI), Pulse Polio (1995; India certified polio-free 2014).
✍️ Mains angles GS-III
- Reliable official statistics and surveys are the backbone of evidence-based policymaking in India.Use the project pipeline (objective to data to analysis to conclusion) to argue for strengthening Census/NSSO and timely data release.
- Primary vs secondary data — trade-offs in conducting large socio-economic surveys.Balance cost, time and manpower against accuracy and sampling error; illustrate with literacy or drinking-water surveys.
- Data-driven consumer protection — surveys to detect overcharging and adulteration.Link survey evidence to enforcement under the Legal Metrology Act 2009, FSSA 2006 and Consumer Protection Act 2019.
Last-minute revision tick as you recall
- 7 steps: Identify, Target, Collect, Organise, Analyse, Conclude, Bibliography.
- Primary = first-hand (questionnaire/interview); Secondary = ready-made, saves time/money/manpower.
- Census = every unit; Sample = subset.
- Analysis trio: Mean (centre), SD (dispersion), Correlation (relationship).
- Sample project = toothpaste, 100 households, Urban 67%.
- Mean income Rs 18,000 (SD 9,000); toothpaste spend Rs 104 (SD 35.60).
- Top brands = Pepsodent, Colgate, Close-up; top media = Television.
- Class midpoint = (upper + lower limit)/2; Decile = 10 equal parts.
- Bibliography credits only secondary sources.
Distilled from NCERT Class 11 · Statistics for Economics for UPSC. Always cross-check facts with the original NCERT.