Methods of research

There are two basic types of research:

1. Quantitative research provides data that can be analyzed statistically and results that can be expressed numerically.

2. Qualitative research provides data on subjects that do not lend themselves to simple quantitative analysis, such as attitudes, perception, motivation etc.

The three principal methods of primary data collection are observation, experimentation and surveys.

Observation

Observation can be carried out mechanically, using cameras or by eye. It does not involve the respondent who is normally unaware of the process. It is useful for the analysis of shopping behaviors, store layout design, traffic flow monitoring etc. Observation can also be used to gain a greater understanding of product – in – use behavior, although it may not be possible to achieve this without the subject being aware of the process.

Experimentation

Experimentation is more than just data – gathering; it involves the attempt to measure the effects of controlled change-significant variables. For example, it is used to measure the effects of a change in elements such as pricing policy, pack design, store layout and advertising treatments.

Surveys

Surveys are the most frequently used data collection technique. They rely on collecting data from a sample of the population under investigation and using the results to draw the conclusions about the population as a whole. The survey can be used to gather both quantitative and qualitative data, and may be continuous (collecting data from the same respondent over time, using a panel) or ad hoc (one-off) surveys. Particular applications may include on-street interviews, mystery shopper surveys, business – to – business interviews, customer satisfaction surveys etc.

Traditionally, surveys have been carried out using three communications vehicles – by post, by telephone or face-to-face. These have now been joined by the internet.

Postal

While postal surveys can offer advantages in terms of speed and cost in the correct circumstances, they are limited to the collection of fairly simple Information. It is difficult to control the identity of the respondent and be sure of obtaining sufficient replies to be confident that the response is representative of the population under investigation. Similar problems may occur with the Internet.

Telephone

Telephone surveys offers more flexibility than postal surveys, and can also be cost effective in the appropriate circumstances (i.e. well targeted, business – to – business). It is possible to control the identity of the respondent and, there is an adequate sampling frame in the form of the telephone directory. Care has to be taken to control interviewer bias.

Дата добавления: 2015-11-14