Types of Data — Primary and Secondary

Once a statistical survey has defined its objective and scope (previous chapter), the very first practical task is to obtain the data. All statistical data falls into two broad types, distinguished not by what the figures are but by who collected them and for whose purpose. The Odisha CHSE +2 Commerce Business Mathematics and Statistics syllabus draws on the same standard treatment of data sources found in any course on descriptive statistics.

Primary data. Primary data is data collected afresh, for the first time, by the investigator (or their agency) specifically for the enquiry in hand. It is 'first-hand' or 'original' data. Because it is gathered directly for the present purpose, it fits the enquiry exactly, its degree of accuracy and method of collection are known, but it is costly, time-consuming and needs more effort to collect. Example: a firm sending its own field staff to ask 500 households about their monthly grocery spending, for a study the firm is itself conducting.

Secondary data. Secondary data is data that has already been collected by someone else, for some other (earlier) purpose, and is now used again by the present investigator. It is 'second-hand' data. Because it already exists, it is cheaper, quicker and easier to obtain, but the investigator must check its suitability and reliability — how, by whom, when and for what purpose it was originally collected — before using it. Example: the same firm instead taking household-expenditure figures from a published government survey report.

The relative nature of the distinction. A crucial point: the same set of figures can be primary data for one person and secondary data for another. When a government agency collects population figures itself, those figures are primary data for the agency; but when a researcher later takes those published figures for their own study, the very same figures are secondary data for the researcher. What decides the type is not the figures but the relationship between the data and the user — original collection makes it primary, re-use of someone else's collection makes it secondary.