Where does the data come from? How much control does a user have over data collection? This will depend on whether or not they are involved in the design and collection of the data. Often, however, the data has been captured by other persons or agencies for different purposes and applications. The distinction is made as follows:

Primary source - first hand collection of data by the user. In such a situation, the user has maximum control over the data and the sampling decisions to be made. Data can be collected via:

• survey questionnaires, or
• measurement and recording of phenomena in the field
  (using instruments, experimentation , observation, etc.)

Secondary source - data collection is undertaken by another (usually previously). The user usually receives the data "as is" and must manipulate it to the desired form.

The data can be obtained from:

Okay...let's go sampling

If you have the "luxury" (or if you have no choice!) of primary data collection, you may have to go through the sampling process.

Let us consider some specific sampling methods in the following pages...

Which is preferred?

Which data source is preferred: primary or secondary? You must weigh the costs (time and economics) of capturing new data vs manipulating and massaging the data to suit ones needs.

The former usually is expensive and the latter is often time-consuming. Much depends on the quality and suitability of the existing data for the intended application.