This chapter sets out to explain the concepts behind the use of sampling in market research. It outlines some of the main options available to the survey designer and discusses the way sampling is carried out in practice. More complex topics are only introduced, with recommendations for more detailed reading.
The quota method is a statistical annoyance. In order to obtain a representative cross-section of the population using this procedure, interviewers are, of course, given very precise instructions as to how many interviews, they are to make in various social groups of the population, but within this exactly prescribed limits of "quotas" they can select people at .their own discretion. This freedom is the jumping off point for all criticism of the quota method. Nobody could foresee what yardstick the interviewers would apply when making their selection. All that could be said with any certainty is that they wouldn't succeed in making a random selection. In one sort of hedonistic theory it is assumed that the interviewers cater as much as possible to their own convenience in quota sampling - that they seek out only particularly pleasant respondents, cover as little ground as possible, and so on. Errors of this nature, it is argued, can be avoided solely by genuine random sampling in which, for example, the addresses of the respondents are prescribed and binding on the interviewers. In practice, examples have actually been found, which show that quota surveys can result in extreme distortions of cross-sections.
The quota method is a statistical annoyance. In order to obtain a representative cross-section of the population using this procedure, interviewers are, of course, given very precise instructions as to how many interviews, they are to make in various social groups of the population, but within this exactly prescribed limits of "quotas" they can select people at .their own discretion. This freedom is the jumping off point for all criticism of the quota method. Nobody could foresee what yardstick the interviewers would apply when making their selection. All that could be said with any certainty is that they wouldn't succeed in making a random selection. In one sort of hedonistic theory it is assumed that the interviewers cater as much as possible to their own convenience in quota sampling - that they seek out only particularly pleasant respondents, cover as little ground as possible, and so on. Errors of this nature, it is argued, can be avoided solely by genuine random sampling in which, for example, the addresses of the respondents are prescribed and binding on the interviewers. In practice, examples have actually been found, which show that quota surveys can result in extreme distortions of cross-sections.
For years now, something like a cold war has been waged between the protagonists of random sampling and those of quota sampling. In a way, it has been a war between mathematicians and practitioners of survey research.And as things go in such situations, the heat of battle may have blinded both sides to a certain extent, so that they became more and more dogmatic believers in one or the other principle of sampling, thus losing both the will and the ability to try and look at the problem from more than the one and only point of view they had chosen.
The purpose of the first pilot study was to establish that an enquiry into attention levels of the type envisaged would be practicable and that meaningful results could be obtained. 200 interviews were completed with housewives in I. T. V. homes in this study and the interviews took place in the latter part of June and early July 1960 in London, Birmingham, Leeds and Glasgow. Quota sampling methods were used with a social class control. During the interviews, housewives were questioned / about their activities during the preceding 30 minutes from the time at which contact was first established. 100 interviews began at some time between 5.30 and 6.00 p.m., and 100 between 9.00 and 9.30 p. m. All the interviews took place on weekday evenings.