Collection of Data
Data collection is the first and most critical step in any statistical investigation. The quality of conclusions drawn depends entirely on the quality of data collected. In this chapter, we learn the types of data, sources of data, and methods of collecting primary data.
Types of Data
Primary Data: Data collected for the first time, directly from the source, for a specific purpose. It is original and first-hand. Example: A researcher personally interviewing farmers about crop yield.
Secondary Data: Data already collected by someone else for another purpose, which is now being used by the researcher. Example: Using Census of India data for a study on urbanisation.
Sources of Secondary Data
- Published sources: Government publications (Census, RBI reports, Economic Survey), international organisations (IMF, World Bank), research journals, newspapers.
- Unpublished sources: Records of businesses, schools, hospitals that are not officially published.
Methods of Collecting Primary Data
1. Direct Personal Investigation (Interview Method):
The investigator personally collects data from respondents. High accuracy but expensive and time-consuming.
2. Indirect Oral Investigation:
Information is obtained from witnesses or third parties rather than the direct subject. Used when direct contact is not possible.
3. Information from Local Correspondents:
Local agents/correspondents collect and report data. Used by newspapers, government departments. Less expensive but may lack uniformity.
4. Mailed Questionnaire Method:
A questionnaire is sent to respondents by post/email. Covers a wide area at low cost but response rate is often low.
5. Schedules and Questionnaires filled by Enumerators:
Trained field workers personally fill questionnaires through direct interview. Used in Census. More accurate but expensive.
Census vs. Sample Survey
Census: Every unit of the population is studied. Complete and accurate but very costly and time-consuming.
Sample Survey: Only a portion (sample) is studied and results are generalised to the whole population. More practical and economical.
Qualities of a Good Questionnaire
- Questions should be clear, simple, and unambiguous.
- Avoid leading or biased questions.
- Questions should be arranged logically.
- Should be as brief as possible.
- Sensitive questions should be handled carefully.
Worked Examples
A student collects rainfall data from the Meteorological Department website. Is this primary or secondary data for the student?
Secondary data — the student did not collect it originally; it was collected by the Meteorological Department.
The same Meteorological Department collected rainfall readings from its stations daily. For the department, this data is:
Primary data — the department collected it directly from its own instruments for the first time.
A company wants to know customer satisfaction. Identify the best method.
Mailed questionnaire or direct interview — direct interview gives better accuracy; mailed questionnaire covers more respondents at lower cost.
The government wants data on every household in India. Which approach is used?
Census — since every unit must be covered, a complete enumeration is needed (like the Decennial Census of India).
A researcher cannot directly interview accident victims for a road-safety study. Which method is appropriate?
Indirect Oral Investigation — collecting information from witnesses, police, or hospital records.
Distinguish between a questionnaire and a schedule.
A questionnaire is filled by the respondent themselves; a schedule is filled by a trained enumerator who asks questions and records answers.
Why is secondary data less expensive than primary data?
Secondary data already exists; the researcher only needs to locate and access it. No fresh field survey is needed, saving time, money, and effort.
Common mistakes
Common mistakes
Students often confuse questionnaire (self-filled by respondents) with schedules (filled by enumerators). Also, remember — data that is primary for one researcher may be secondary for another. The key question is: who collected it first?
Summary
Data can be primary (original, first-hand) or secondary (already collected). Primary data is collected through interviews, questionnaires, or enumerators. Secondary data comes from published or unpublished sources. Choice of method depends on accuracy required, resources available, and the nature of investigation.