Primary Data and Its Methods of Collection

Primary data is data collected for the first time, directly from the units of the investigation (people, households, firms), specifically for the purpose of the study at hand. Because the investigator gathers it firsthand, primary data is generally more reliable, more suited to the exact question being asked, and more original — but it is also the more time-consuming and expensive kind of data to collect.

The Gujarat Std 11 Statistics syllabus lists several standard methods of collecting primary data, each with its own trade-offs between accuracy, cost, and coverage:

• Direct Personal Investigation (or Interview Method). The investigator personally contacts each respondent and collects information through face-to-face interviews. This gives the highest accuracy — the investigator can clarify doubtful answers and probe further — but it is slow, expensive, and impractical over a very large or geographically spread area. It suits small, localised, intensive studies.
• Indirect Oral Investigation. Instead of approaching the respondents themselves, the investigator collects information from persons who are believed to be well acquainted with the facts (witnesses, local officials, neighbours). This is used when the actual respondents cannot be reached directly, or where direct questioning may not yield honest answers (e.g., certain socially sensitive enquiries) — but its accuracy depends heavily on how reliable and unbiased the third-party informants are.
• Mailed Questionnaire Method. A questionnaire is prepared and sent by post (or, today, electronically) to the respondents, who fill it in themselves and return it. This is inexpensive and can cover a very wide, scattered area, but suffers from a low and often biased response rate — those who bother to reply may not represent the whole population fairly — and there is no scope to clarify a misunderstood question.
• Schedule Method (through enumerators). Similar to a questionnaire, but here trained enumerators carry the schedule to each respondent and fill it in on the spot, based on the respondent's answers. This combines a wider reach than personal investigation with a much higher and more reliable response rate than mailed questionnaires, since the enumerator is present to explain and clarify — but it needs trained staff and is costlier than a mailed questionnaire. …