Revision notes

Chapter 2 Collection of Data: Foundation

The key points, rules to remember, tables and diagrams from all 5 lessons, on one page for quick revision. To keep a copy, use your browser's Print and choose "Save as PDF".

Lesson 1

Data, variables and the sources of data

Illustrative data, made up for practice
Year (X)Milk collected, thousand litres (Y)
Year 1240
Year 2255
Year 3250
Year 4270
Year 5290

Primary and secondary data

  • Primary data
    • Collected by the investigator through his or her own enquiry
    • First-hand information
    • Example: asking students which film actor they like
    • Costs more time and money
    • Fits the purpose exactly
  • Secondary data
    • Already collected and processed (scrutinised and tabulated) by another agency
    • Taken from government reports, documents, newspapers, books by economists, websites
    • Saves time and cost
    • Must be checked for suitability

Data are primary to the source that collects and processes them for the first time, and secondary for every source that uses them later.

Situation (illustrative)Type of dataReason
A student interviews 60 shopkeepers about their daily salesPrimaryShe collects it first-hand by her own enquiry
A researcher copies crop output figures from a government reportSecondaryThe figures were collected and tabulated by another agency
A newspaper reports a survey done by its own reportersPrimary for the newspaperThe newspaper collected it first
A teacher uses that newspaper survey in a class projectSecondary for the teacherSomeone else collected it

Key takeaways: data are collected to give evidence for solving a problem. A variable is a quantity that varies; each of its values is an observation. Primary data are first-hand, collected by the investigator; secondary data have been collected and processed by another agency. The same data are primary for the collector and secondary for later users.

Read the full lesson

Lesson 2

Designing a questionnaire

Points to keep in mind while preparing a questionnaire
RuleFaulty question (made up)Better question
Keep it short: as few questions as possibleA 70-question form for a 5-minute street surveyOnly the questions needed for the purpose
Easy words, no difficult or ambiguous termsIs your household's fuel consumption elastic?Did your family buy less cooking gas when its price rose?
Move from general to specificIs the new bus fare justified? (asked first)First: Do you use the city bus daily? Then: Is the new bus fare justified?
Precise and clearWhat share of income do you spend on food to stay healthy?What share of your income do you spend on food?
Not vague; give ranges to tickDo you spend a lot on mobile data?How much do you spend on mobile data in a month? Less than ₹100 / ₹100-200 / ₹200-300 / more than ₹300
No double negatives or 'Don't you...' openingsDon't you think plastic bags should not be allowed?Do you think plastic bags should be banned?
Not leading (no clue to the answer)How do you like this delicious new biscuit?How do you like this biscuit?
Do not suggest only certain alternativesWill you study further or start working after school?What do you plan to do after school?

Types of questions in a questionnaire

  • Closed-ended (structured)
    • Two-way question: only 'Yes' or 'No'
    • Multiple-choice question: more than two options, often with 'Any other (please specify)'
  • Open-ended (unstructured)
    • Respondent answers in own words
    • Example: What is your view about online classes?

Strengths and weaknesses

  • Closed-ended
    • Easy to use, score and codify for analysis
    • Hard to write: options must cover both sides of the issue
    • The true answer may not be among the options
    • Options can restrict or shape answers
  • Open-ended
    • Allows individualised responses
    • Hard to interpret
    • Hard to score because answers vary a lot

Closed-ended questions are easy to score but may miss the respondent's true answer; open-ended questions capture it but are hard to interpret and score. The 'Any other (please specify)' option is a bridge between the two.

Key takeaways: a questionnaire should be short, simple, ordered from general to specific, precise, unambiguous, free of double negatives, not leading and not limited to suggested alternatives. Closed-ended questions (two-way or multiple-choice) are easy to score; open-ended questions give richer but harder-to-score answers.

Read the full lesson

Lesson 3

Modes of data collection and the pilot survey

Comparing the three modes
FeaturePersonal interviewMailed questionnaireTelephone interview
CostMost expensiveLeast expensiveRelatively low
Response rateHighestLow; long response timeRelatively high
Clarifying questionsYes, fullyNoYes
Influence on respondentPossibleNone; anonymity keptRelatively less, but possible
ReachNeeds access to membersReaches remote areasOnly people with telephones
SuitsOpen-ended and all types of questionsSensitive questions; literate respondentsQuick surveys; reluctant respondents
Situation (illustrative)Suitable modeReason
Collect information from a person living in a remote village with no phone networkMailed questionnaireThe only way to reach remote areas cheaply (if the person can read)
Ask parents about teaching quality while the school principal is presentAvoid personal interview here; prefer mailThe principal's presence may stop parents from speaking freely
Quick opinion survey of 200 shopkeepers in a city in two daysTelephone interviewCheaper and faster than visits; questions can be clarified
Detailed study of how farm families take loans, with many open-ended questionsPersonal interviewAllows explanation, follow-up questions and observation

Where the pilot survey fits

  1. Draft the questionnaire Following the rules for good questions
  2. Pilot survey on a small group Try out the questions
  3. Find shortcomings Unclear questions, weak instructions, slow enumerators
  4. Revise; estimate cost and time Improve questions and plan resources
  5. Main survey Collect the data

A pilot survey (pre-test) gives a preliminary idea of the survey and checks the questions' suitability, the clarity of instructions, the performance of enumerators, and the cost and time of the actual survey.

Key takeaways: data are collected by personal interviews (highest response, costliest), mailed questionnaires (cheapest, reaches remote areas, low response) or telephone interviews (cheap and quick, limited to phone owners). The choice depends on cost, reach, the kind of questions and the respondents. A pilot survey pre-tests the questionnaire and helps estimate cost and time.

Read the full lesson

Lesson 4

Census, sample surveys and methods of sampling

Population (or universe): every item that the study is about, so that its findings are meant to hold for all of them. Sample: the part of the population that is actually questioned or measured.

Research problem (illustrative)PopulationSample
Opinion of Class 11 students of a made-up school on the new canteen menuAll Class 11 students of that school40 students selected from them
Average yield of rice in a made-up districtAll rice fields in the districtA selection of 150 fields
Cropping pattern in a village of 400 farms, of which 80 are surveyedAll 400 farmsThe 80 farms surveyed (sample size 80)

Methods of sampling

  • Random sampling Every unit has an equal chance
    • Lottery method
    • Random number tables
    • Computer programmes
  • Non-random sampling Selection depends on the investigator
    • Judgement
    • Purpose
    • Convenience
    • Quota

Key takeaways: a census covers every unit; a sample survey covers a part. The population is the whole group studied; a representative sample reflects it at lower cost and time. Random sampling gives every unit an equal chance (lottery, random number tables, computers); non-random sampling relies on judgement, purpose, convenience or quota and may be biased.

Read the full lesson

Lesson 5

Sampling errors, non-sampling errors and agencies that collect data

Sampling error is the difference between the actual value of a population parameter and its estimate from a sample. It can be reduced by taking a larger sample.

Sample (illustrative)SizeSample mean (₹)Sampling error (₹)
320, 360234060
320, 360, 400, 480439010
All six (census)64000

Kinds of non-sampling error

  • Sampling bias Some members of the target population could not possibly be included
    • Example: a phone survey that leaves out people without phones
  • Non-response error A listed person cannot be contacted or refuses to answer
    • The remaining sample may not be representative
  • Errors in data acquisition Incorrect responses are recorded
    • Different measuring tapes give different lengths
    • Prices differ by shop and quality, so only averages can be used
    • Recording or copying mistakes, such as 31 written as 13

Sampling and non-sampling errors

  • Sampling error
    • Arises because only a part of the population is studied
    • Difference between parameter and estimate
    • Falls as sample size rises
    • Absent in a census
  • Non-sampling error
    • Arises from bias, non-response or faulty data acquisition
    • Not cured by a larger sample
    • Can occur even in a census
    • More serious
Two major national agencies, as described in the textbook
AgencyWhat the textbook says it does
Census of IndiaGives the most complete and continuous demographic record of the population; described as conducted every ten years since 1881, with the first after Independence in 1951; covers size, density, sex ratio, literacy, migration and rural-urban distribution
National Sample Survey (NSS)Set up by the Government of India for nation-wide surveys on socio-economic issues, in successive rounds; reports and the journal Sarvekshana; estimates of literacy, school enrolment, employment, unemployment, enterprises, health, use of the public distribution system; also industrial activity and retail prices, used for planning

Key takeaways: sampling error is the gap between a population parameter and its sample estimate and shrinks as the sample grows. Non-sampling errors (sampling bias, non-response, errors in data acquisition) are more serious because a bigger sample does not remove them. The Census of India and the NSS are major national sources of data on population and socio-economic conditions.

Read the full lesson

Privacy · Terms · Refunds · Delivery