This MCQ module is based on: Sources of Data, Survey Methods & Questionnaire Design
TOPIC 2 OF 15
Sources of Data, Survey Methods & Questionnaire Design
🎓 Class 11
Social Science
CBSE
Theory
Ch 2 — Collection of Data
⏱ ~25 min
🌐 Language: [gtranslate]
🧠 AI-Powered MCQ Assessment
▲
📝 Worksheet / Assessment
▲
This assessment will be based on: Sources of Data, Survey Methods & Questionnaire Design
Upload images, PDFs, or Word documents to include their content in assessment generation.
Sources and Methods of Data Collection
Every statistical insight begins with a deceptively simple step: collecting the data. Where do numbers come from? How does a researcher walk up to strangers and gather honest answers about price, income, opinion or behaviour? And why does the wording of a single question decide whether the data is gold or garbage? This part introduces primary vs secondary sources, the three modes of survey, and the craft of writing an unbiased questionnaire.
2.1 Why Do We Collect Data?
In Chapter 1 you learned what economics is and why statistics matters to it. The next obvious question is: where do all these numbers come from? The aim of data collection is to assemble enough credible evidence so that an economic problem can be analysed and solved. Without good data, every conclusion is a guess; with good data, even small problems can be tackled with confidence.
Consider this NCERT example of how data make a story visible. Indian food-grain output rose from 108 million tonnes in 1970-71 to 132 million tonnes in 1978-79, briefly slipped back to 108 mn t in 1979-80, and then climbed steadily to 252 million tonnes in 2015-16 and 272 million tonnes in 2016-17. The numbers themselves are the production of food grain — but what they reveal is a country slowly winning its long battle against hunger.
| Year (X) | Production in million tonnes (Y) |
|---|---|
| 1970–71 | 108 |
| 1978–79 | 132 |
| 1990–91 | 176 |
| 1997–98 | 194 |
| 2001–02 | 212 |
| 2015–16 | 252 |
| 2016–17 | 272 |
Fig 2.1 — The same data plotted as a line chart shows a clear long-run upward trend with year-to-year variation.
Each year (X) and each output figure (Y) is a single observation?; together they form the data. Because the values change from year to year, X and Y are called variables?. Variables are usually written using the letters X, Y or Z. Every individual reading of a variable is one observation.
📖 Definition — Data, Variable, Observation
Data are facts (often numerical) collected for analysis. A variable is a quantity that takes different values across cases or time. Each separate value of the variable is an observation. In Table 2.1, X and Y are variables; the entry "1990–91 → 176" is one observation.
2.2 Where Does Data Come From? Primary vs Secondary Sources
Statistical data flows from two distinct sources. You can either go out and collect numbers yourself — by asking, observing, measuring — or you can lift numbers that someone else has already collected, cleaned and published.
Primary Data
Collected directly by the researcher for the specific question at hand. First-hand information, never used before. Costs more time and money but matches the researcher's exact need.
Secondary Data
Already collected, processed and published by another agency — government reports, books, websites, RBI bulletins. Saves time and money but may not exactly match the new study's purpose.
💡 Same Data, Two Labels
Whether a number is "primary" or "secondary" depends on who is using it for what. The same observation is primary for the agency that collects it the first time and secondary for everyone who reuses it later. If you survey 200 classmates about their favourite filmstar and publish the result, those numbers are your primary data — but if a magazine quotes them next month, they become secondary data for the magazine.
Fig 2.2 — Sources of statistical data: a single tree with two branches and many leaves.
2.3 How Do We Collect Primary Data? The Survey Method
How does a manufacturer know whether shoppers like a new shampoo? How does a political party gauge a candidate's popularity? They run a survey — a structured way of gathering information from a group of people. The point of a survey is to describe characteristics of a product (price, quality, usefulness) or of a person (popularity, honesty, loyalty), and the raw fuel of every survey is data.
2.3.1 The Instrument: Questionnaire / Interview Schedule
The most common survey instrument is the questionnaire? (sometimes called an interview schedule). It is either filled by the respondent themselves (self-administered) or read out by a trained researcher / enumerator who records the replies. A badly written questionnaire will collect badly distorted data, no matter how big the sample. So writing one is half art, half science. NCERT lists six rules of thumb.
⚖️ Six Rules for a Good Questionnaire
(1) Keep it short — minimum number of questions.(2) Use simple, unambiguous language.
(3) Order questions so the respondent feels comfortable; move from general to specific.
(4) Make every question precise and clear.
(5) Avoid double negatives ("Wouldn't you...", "Don't you...").
(6) Avoid leading questions and questions that hint at an answer.
2.3.2 Poor Questions vs Good Questions — Five Worked Examples
NCERT pairs a poorly framed question with a corrected version. Each pair illustrates one of the rules above.
(a) Order — general before specific:
Poor Q
(i) Is the increase in electricity charges justified?
(ii) Is the electricity supply in your locality regular?
(i) Is the increase in electricity charges justified?
(ii) Is the electricity supply in your locality regular?
Good Q
(i) Is the electricity supply in your locality regular?
(ii) Is the increase in electricity charges justified?
(i) Is the electricity supply in your locality regular?
(ii) Is the increase in electricity charges justified?
Why? Asking the broad fact-of-life question first relaxes the respondent and produces honest answers to the loaded follow-up.
(b) Precision — drop vague phrases:
Poor Q
What percentage of your income do you spend on clothing in order to look presentable?
What percentage of your income do you spend on clothing in order to look presentable?
Good Q
What percentage of your income do you spend on clothing?
What percentage of your income do you spend on clothing?
Why? "Look presentable" is subjective and biases the answer. Strip such phrases.
(c) Avoid ambiguity — quantify the response options:
Poor Q
Do you spend a lot of money on books in a month?
Do you spend a lot of money on books in a month?
Good Q
(Tick one) How much do you spend on books in a month?
(i) Less than ₹200 (ii) ₹200–300 (iii) ₹300–400 (iv) More than ₹400
(Tick one) How much do you spend on books in a month?
(i) Less than ₹200 (ii) ₹200–300 (iii) ₹300–400 (iv) More than ₹400
Why? "A lot of money" means different things to different people. Range buckets standardise the reply.
(d) Avoid double negatives:
Poor Q
Don't you think smoking should be prohibited?
Don't you think smoking should be prohibited?
Good Q
Do you think smoking should be prohibited?
Do you think smoking should be prohibited?
Why? Phrases beginning with "Wouldn't you" or "Don't you" push the respondent towards agreement and produce biased data.
(e) Avoid leading questions:
Poor Q
How do you like the flavour of this high-quality tea?
How do you like the flavour of this high-quality tea?
Good Q
How do you like the flavour of this tea?
How do you like the flavour of this tea?
Why? Calling the tea "high-quality" tells the respondent what to think before they answer.
(f) Avoid questions that suggest the alternatives:
Poor Q
Would you like to do a job after college or be a housewife?
Would you like to do a job after college or be a housewife?
Good Q
What would you like to do after college?
What would you like to do after college?
Why? Pinning the respondent to two alternatives hides every other career or life choice.
REWRITE — Spot the Loaded Question
Bloom: L4 Analyse
Below are three poorly framed survey questions. Identify which design rule each one violates and rewrite it as a good question.
- "Don't you agree that online classes are worse than offline classes?"
- "Do you spend a reasonable amount on monthly internet recharge?"
- "Will you become a doctor or an engineer after Class 12?"
✅ Sample Rewrites
(1) Violation: double negative + leading. Rewrite: "Compared with offline classes, are online classes better, the same, or worse?"(2) Violation: ambiguous "reasonable". Rewrite: "How much do you spend on internet recharge per month? (a) Less than ₹200 (b) ₹200–500 (c) ₹500–1000 (d) More than ₹1000."
(3) Violation: forced alternatives. Rewrite: "What career do you wish to pursue after Class 12?"
2.3.3 Closed-Ended vs Open-Ended Questions
Once a question is well-worded, the next design choice is the response format. There are two broad families.
Closed-ended (Structured)
Fixed reply choices: either a two-way Yes/No or a multiple choice with an "Any other (please specify)" escape. Easy to score and codify, but options must cover both sides fairly.
Open-ended (Unstructured)
Respondent writes in free language. Captures individuality and surprise responses, but is harder to interpret and score because answers vary widely.
📜 NCERT Multiple-choice Example — "Why did you sell your land?"
(i) To pay off the debts. (ii) To finance children's education. (iii) To invest in another property. (iv) Any other (please specify).
— NCERT Statistics for Economics, Ch. 2
Closed-ended questions are preferred when the researcher already knows the universe of likely answers. Open-ended questions are preferred when the goal is to discover new viewpoints — for example, "What is your view about globalisation?"
⚖️ The Trade-off
Closed-ended questions are easy to use and score but may exclude the respondent's true answer if not all options are listed. Open-ended questions allow individualised responses but are slow to interpret. A well-designed questionnaire usually mixes both.
2.4 The Three Modes of Data Collection
Once the questionnaire is ready, how does it actually reach a respondent? NCERT identifies three classic modes.
2.4.1 Personal Interviews
The investigator meets the respondent face-to-face. This works when the entire population (or sample) is reachable in person. The interviewer can explain difficult words, expand on important answers, and read the respondent's body language for extra clues.
✅ Advantages
- Highest response rate
- All types of questions can be used
- Best for open-ended questions
- Ambiguous wording can be clarified on the spot
❌ Disadvantages
- Most expensive — needs trained interviewers
- Possibility of the interviewer influencing the respondent
- Takes the longest time
2.4.2 Mailing (Mailed) Questionnaires
The questionnaire is posted to each respondent with a request to fill it in and return it by a deadline. Today this also includes online forms and SMS surveys.
✅ Advantages
- Least expensive of the three
- Reaches respondents in remote areas
- No interviewer to influence answers — preserves anonymity, best for sensitive topics
- Respondent has time to think before replying
❌ Disadvantages
- Cannot be used by illiterate respondents
- Long response time
- No one to clarify ambiguous questions
- Reactions of the respondent cannot be observed
- Low response rates — forms get lost, returned blank, or never returned
2.4.3 Telephone Interviews
The investigator asks the questions over a phone call. This is faster and cheaper than personal interview, allows clarification, and works well when respondents are shy about face-to-face questioning on a sensitive topic. The catch: many people may not own a telephone.
✅ Advantages
- Relatively low cost
- Comparatively less interviewer influence
- Reasonably high response rate
- Quicker than personal interview
- Better for questions respondents are reluctant to answer in person
❌ Disadvantages
- Limited use — requires the respondent to own a telephone
- Reactions cannot be watched
- Some interviewer influence still possible
2.4.4 Comparison Table — Mode at a Glance
| Feature | Personal Interview | Mailed Questionnaire | Telephone Interview |
|---|---|---|---|
| Cost | Highest | Lowest | Moderate |
| Time per interview | Long | Long (waiting for return) | Short |
| Response rate | Highest | Lowest | High |
| Reaches remote areas | Difficult | Easy (with postal/online access) | Only if telephone available |
| Suitable for illiterate respondents | Yes | No | Yes |
| Risk of interviewer bias | High | None | Moderate |
| Best use of open-ended questions | Excellent | Limited | Limited |
| Sensitive / personal questions | Difficult | Best (anonymity) | Reasonable |
2.5 Pilot Survey — Test Before You Trust
Before unleashing the final questionnaire on hundreds of respondents, a researcher runs a small trial called a Pilot Survey? (also known as pre-testing). A pilot is a miniature version of the planned survey on a small group, used to expose flaws before scaling up.
📖 Definition — Pilot Survey
A pilot survey is a try-out of the questionnaire on a small group before the actual survey, to test the questions, instructions, enumerator performance and likely cost & time. It gives a preliminary preview of the data the full survey will yield.
A well-run pilot answers four practical questions:
- Are the questions intelligible? Do respondents misread, skip or struggle with any item?
- Are the instructions clear? Can a respondent follow the form without help?
- Are the enumerators well-trained? Are they recording answers consistently?
- What will the full survey cost in money and time? The pilot gives the first realistic estimate.
EXPLORE — Pilot Survey for Your School Canteen
Bloom: L3 Apply
Imagine you have to find out which snack the school canteen should add next term. You have written a 10-question questionnaire for 600 students. Design a 30-student pilot survey: who do you pick, what do you watch for, and what changes would you make to the questionnaire after the pilot?
✅ Sample
Pick 30 students drawn from different classes (5 each from VI to XII) so that the pilot mirrors the full population. Watch for: questions students skip, words they ask the meaning of, options they tick "Other" for. After the pilot, drop questions with high "no answer" rates, simplify wording where students paused, and add new options that "Other" responses revealed. Estimate average completion time × 600 to plan the rollout.
DISCUSS — Choosing the Right Mode
Bloom: L5 Evaluate
- You have to collect information from a person who lives in a remote village in India. Which mode of data collection will you use, and why?
- You have to interview the parents about the quality of teaching in a school. If the school principal is present in the room, what kinds of problems can arise?
✅ Discussion
(1) A mailed questionnaire (or even a personal interview if accessible) is preferable. Telephone surveys fail when phones are scarce; mailed forms reach remote villages cheaply. If the respondent is illiterate, a personal visit by an enumerator is best.(2) The principal's presence creates an authority bias — parents may hide complaints out of fear that their child will be penalised, or they may praise the school excessively. The interviewer cannot guarantee anonymity. Ideally the parents should be interviewed away from the school in a neutral setting.
2.6 Bringing It Together — A Quick Worked Application
📊 Case-Based Question — The Filmstar Popularity Survey
A Class 11 Statistics student wants to find out which Bollywood actor is most popular among the 800 students of her school. She has two weeks, ₹500 of budget, and access to all classrooms. She drafts a questionnaire, picks 40 students from different classes for a pilot, and then surveys 200 students.
Q1. Identify the source of the data the student is collecting.
L1 RememberAnswer: Primary data — the student is collecting first-hand information for the first time.
Q2. Which mode of data collection is most cost-effective for a school survey of 200 students with low budget? Justify.
L3 ApplyAnswer: A self-administered (mailed/handed-out) questionnaire is cheapest and feasible because all students are accessible inside the school. Personal interviews would consume too much time; telephone interviews are unnecessary when respondents are physically reachable.
Q3. The student tested the questionnaire on 40 students before rolling out to 200. What is this step called and why is it useful?
L4 AnalyseAnswer: The step is called a Pilot Survey (pre-testing). It exposes confusing or ambiguous questions, gauges how long the form takes, checks the clarity of instructions and provides a realistic estimate of cost and time before the full rollout.
Q4. The student's draft included the question: "Don't you think Actor X is the best actor of his generation?". Diagnose the flaw and rewrite the question.
L5 EvaluateAnswer: Two flaws — a double negative ("Don't you think") and a leading prompt ("the best actor"). Both bias the respondent toward agreement. Rewrite: "Among the following actors, whom do you consider the most popular today? (i) Actor X (ii) Actor Y (iii) Actor Z (iv) Any other (please specify)."
⚖️ Assertion–Reason Questions (Class 11)
Choose: (A) Both A and R are true and R is the correct explanation of A. (B) Both A and R are true but R is not the correct explanation of A. (C) A is true, R is false. (D) A is false, R is true.
Assertion (A): The same numerical record can be primary for one researcher and secondary for another.
Reason (R): A datum is primary to the agency that collects it for the first time and secondary for everyone who reuses it later.
Correct: (A) — Both statements are true, and R correctly explains A. The label depends entirely on the user, not on the data itself.
Assertion (A): Mailed questionnaires are unsuitable when respondents are illiterate.
Reason (R): Mailed questionnaires require the respondent to read the questions and write the answers without any researcher present to assist.
Correct: (A) — Both statements are true and R explains A precisely. Without literacy or a present helper, the form is useless for the respondent.
Assertion (A): A pilot survey can be skipped if the researcher has previous experience.
Reason (R): A pilot survey detects flaws in the questionnaire, instructions, enumerator performance and cost estimates that even an experienced researcher cannot anticipate for a fresh study.
Correct: (D) — Assertion is false; even experienced researchers run pilots because every new context (population, region, language) introduces fresh flaws. Reason is a true general statement about why pilots are valuable.
Frequently Asked Questions — Sources and Methods of Data Collection
What is the difference between primary and secondary data in NCERT Class 11 Statistics?
Primary data is information collected first-hand by the investigator for a specific study, while secondary data is information already collected and published by another agency that is re-used for a new investigation. NCERT Class 11 Statistics Chapter 2 explains that primary data is more reliable and tailored to the research question but is expensive and time-consuming to collect, whereas secondary data from sources like NSSO, RBI, Census of India and the Economic Survey is faster and cheaper but may be outdated, in different definitions, or not perfectly suited to the new question. The investigator must judge fitness for purpose before using any secondary source.
What are the three main methods of data collection in Class 11 Statistics?
The three main methods of primary data collection in NCERT Class 11 Statistics Chapter 2 are personal interview, mailed questionnaire and telephone interview. A personal interview gives high response rates and clarification but is costly and slow; a mailed questionnaire reaches a wide geographic area cheaply but suffers from low response and unclear answers; a telephone interview is fast and inexpensive but excludes people without phones and limits how complex the questions can be. The investigator chooses the method based on budget, speed, coverage required and the literacy level of respondents.
How do you design a good questionnaire in NCERT Class 11 Statistics?
A good questionnaire in NCERT Class 11 Statistics Chapter 2 is short, simple and uses easy-to-understand language with no leading or ambiguous questions. Questions should follow a logical sequence, move from easy to difficult, and avoid personal or sensitive items unless essential. Use multiple-choice or yes/no formats where possible, define units clearly, and avoid double-barrelled questions that ask two things at once. Always pilot the questionnaire on a small sample first, refine the wording based on confusion or non-response, and then deploy the final version for the main survey.
What is a pilot survey and why is it important in data collection?
A pilot survey is a small trial run of the questionnaire on a few respondents before the main study is conducted, and it is important because it reveals problems that would otherwise spoil the entire investigation. NCERT Class 11 Statistics Chapter 2 explains that a pilot survey checks whether questions are understood correctly, whether the time taken is reasonable, whether response options are exhaustive, and whether sensitive items cause non-response. Issues uncovered in the pilot are fixed before the full survey, saving cost and improving the quality of primary data substantially.
Which agencies publish secondary data in India for Class 11 Statistics?
Major agencies that publish secondary data in India, as listed in NCERT Class 11 Statistics Chapter 2, include the National Statistical Office (NSO) and its successor body MoSPI, the Reserve Bank of India (RBI), the Census of India (decennial population census), the Central Statistics Office (CSO), the Labour Bureau, the Indian Council of Agricultural Research (ICAR) and the National Sample Survey Office (NSSO). International sources include the World Bank, IMF, WTO and UN agencies. These agencies publish official data on prices, employment, output, population and prices that researchers and policymakers re-use widely.
What is the difference between a leading question and a neutral question in survey design?
A leading question is worded in a way that pushes the respondent toward a particular answer, for example 'Don't you agree that prices are too high?', while a neutral question lets the respondent answer without bias, such as 'How do you rate the current price level — high, average or low?'. NCERT Class 11 Statistics Chapter 2 warns that leading questions destroy data quality because the answers reflect the questioner's view rather than the respondent's true opinion. Survey designers must use balanced wording, avoid loaded terms, and offer all reasonable response options to keep results objective.
AI Tutor
Class 11 Economics — Statistics for Economics
Ready