Collection of Data

Once you know what you are studying and whom you are studying, you must decide how to obtain the information — and then actually collect it. The objective of the survey is again the deciding factor: it tells you whether the data should be gathered by the primary method, the secondary method, or both. Getting this choice right determines how reliable and how relevant your data will be.

Primary data — collected first-hand. As you have already learnt in the chapter on collection of data, primary data are collected for the first time by the investigator, directly from the source. They are gathered using a questionnaire or an interview schedule, which can be administered through several channels:

• personal interviews — meeting respondents face to face;
• mailing or postal surveys — sending the questionnaire by post;
• telephone; and
• email.

A postal questionnaire must carry a covering letter explaining the purpose of the inquiry, so that respondents understand why their cooperation is being sought and are more willing to reply honestly and fully.

Fixing the size and characteristics of the target group. A key purpose at this stage is to determine how large your target group is and what its characteristics are. Sometimes the objective requires information from every unit. For example, to study female literacy at the primary and secondary levels, or the consumption of a particular brand of soap in an area, you may have to go to each and every family or household — that is, you collect primary data covering the whole population of interest (a census-type approach).

When sampling is used. If instead of covering everyone you study only a representative subset, you are using sampling. In that case, great care must be taken about the suitability of the method of sampling, because a poorly chosen sample can give a misleading picture of the whole group. The sample must genuinely represent the target group if the results are to be trusted.

Secondary data — already available. Secondary data are data that have already been collected and published by someone else. They can be used provided they suit your requirement. They are especially useful when there is a shortage of time, money and manpower and the needed information is easily available from existing sources. The trade-off is that secondary data were gathered for someone else's purpose, so you must check that they actually fit your objective before relying on them. …