• Foundation
  • About 18 min
  • 5 practice questions

Designing the survey for a project

List the information a project needs, write a questionnaire that collects it, and plan the sample, using the made-up soap project.

If something looks wrong, please tell us.

By the end of this lesson you can

  • Turn a project objective into a list of information needed.
  • Write questionnaire items of different types that follow the rules of good questions.
  • Plan the sample and summarise its composition.

The questionnaire is the bridge between a project's objective and its data. Whatever questionnaire you use must be able to generate exactly the information the study needs. A questionnaire already tried and tested for a similar study can be adapted; otherwise you prepare your own and check that every needed item is asked.

Throughout this chapter we follow one made-up project: Navya, a young entrepreneur (made up), wants to start a small unit making bathing soap and asks you to study the market first. Every figure in the project is illustrative.

From objective to information needed

  • Average monthly household spending on bathing soap.
  • Brands of soap currently in use and how satisfied users are with them.
  • Features customers want (for example herbal, fragrance, moisturising, low price).
  • The media through which people learn about soaps.
  • Background facts: family size, income, occupation, rural or urban residence.
  • How spending on soap relates to income.

Worked example 1: writing the questions

Each information need becomes one or more questions. Background questions come first (general), then questions about soap (specific), following the rule of moving from general to specific.

Information needQuestion (made up for this project)Type
Family sizeHow many members live in your household?Open, numerical
IncomeWhat is your household's monthly income? Below ₹10,000 / ₹10,000-20,000 / ₹20,000-30,000 / ₹30,000-40,000 / above ₹40,000Multiple choice (ranges)
ResidenceDo you live in a village or a town? Village / TownTwo-way
Soap useDoes your household buy bathing soap? Yes / NoTwo-way
SpendingAbout how much does your household spend on bathing soap in a month (₹)?Open, numerical
Brand and satisfactionWhich brand do you mainly use? Are you satisfied with it? Yes / NoOpen and two-way
Features wantedWhich features would you like in a new soap? (tick all that apply) Herbal / Fragrance / Moisturising / Low price / Any other (please specify)Multiple choice, several answers allowed
MediaWhere do you mostly learn about new soaps? Television / Newspaper / Internet / Shop display / Any otherMultiple choice

Checks against the rules of Chapter 2: questions are short and clear; none is leading (we do not ask 'Would you like a superior herbal soap?'); ranges replace vague words like 'a lot'; 'Any other (please specify)' catches unexpected answers; and nothing asks for more personal detail than the study needs. Income is asked in ranges, which respondents find easier to answer.

Planning the sample

Nearly every household is a possible soap buyer, so a census is out of the question. A random sample of households is drawn from a list of households in the chosen area (the sampling frame), for instance by lottery or random numbers, so that every household has an equal chance. A pilot survey with a handful of households tests the questionnaire first.

Worked example 2: describing the sample

Illustrative: 120 households were surveyed: 78 in towns and 42 in villages. Urban share: (78 ÷ 120) × 100 = 65 per cent. Rural share: (42 ÷ 120) × 100 = 35 per cent. Check: 65 + 35 = 100. Observation: most respondents were urban, so conclusions apply more strongly to urban buyers; if rural buyers matter to Navya, a larger rural sample may be needed.

Worked example 3: sample size and coverage

Illustrative: the chosen area has 2,400 households. Sampling fraction: 120 ÷ 2,400 = 0.05, or 5 per cent. If the pilot showed that about 1 in 10 households refuses to answer, to end with 120 completed questionnaires Navya should approach about 120 ÷ 0.9 ≈ 134 households. Planning for non-response this way keeps the final sample at the intended size, though it does not remove the risk that refusers differ from those who answer.

Good and poor survey design for the project

  • Good design
    • Every question tied to an information need
    • General to specific order
    • Clear ranges and options
    • Random sample with a pilot survey
  • Poor design
    • Questions with no use in the analysis
    • Leading or vague wording
    • Only friends and neighbours asked
    • No pre-testing

Write down the information the study needs before writing a single question; then make sure each need is covered by a clear, unbiased question, and each question serves a need.

Key takeaways: list the information needed from the objective, then design questions (two-way, multiple choice, open) that collect it, ordered from general to specific and free of leading or vague wording. Plan a random sample from a suitable frame, pre-test with a pilot, allow for non-response, and describe the sample's composition in percentages.

Check your understanding

  1. 1. A good questionnaire for a project must above all:

    1. be as long as possible
    2. contain only open-ended questions
    3. be able to generate the information the study needs
    4. include leading questions to save time
    Show answer

    Answer: C. The chapter stresses that the questionnaire must be capable of generating the information needed for the study; length and question type follow from that.

  2. 2. Illustrative: of 150 households surveyed, 90 live in towns. The rural share of the sample is:

    1. 60%
    2. 40%
    3. 90%
    4. 66.7%
    Show answer

    Answer: B. Rural households = 150 − 90 = 60. Rural share = (60 ÷ 150) × 100 = 40%. 60% is the urban share.

  3. 3. Which question is best suited to a soap project questionnaire?

    1. Don't you think our new soap is better than others?
    2. Do you spend a lot on soap?
    3. Would you like a superior quality herbal soap?
    4. About how much does your household spend on bathing soap in a month?
    Show answer

    Answer: D. It is clear, precise and neutral. The others use a 'Don't you' opening, vague words ('a lot') or leading words ('superior quality').

  4. 4. Illustrative: a project needs 200 completed questionnaires and the pilot shows that 20 per cent of households refuse. About how many households should be approached?

    1. 250
    2. 220
    3. 240
    4. 160
    Show answer

    Answer: A. Completed = 80 per cent of those approached, so approached = 200 ÷ 0.8 = 250.

  5. 5. Assertion (A): In the soap project, background questions such as family size are asked before questions about soap brands. Reason (R): A questionnaire should move from general to specific questions.

    1. Both A and R are true, and R is the correct explanation of A
    2. Both A and R are true, but R is not the correct explanation of A
    3. A is true, but R is false
    4. A is false, but R is true
    Show answer

    Answer: A. Both are true, and R explains A: background questions are general and put respondents at ease; specific questions about brands follow.

Privacy · Terms · Refunds · Delivery