EXPERIENCES OF RANDOM (PROBABILITY) SAMPLING.

This article focuses on probability sampling. Probability sampling has been undertaken in every country, but whether it is appropriate to use quota sampling or probability sampling depends upon the nature of the problem being investigated. In any survey based on probability sampling a failure rate o...

Descripción completa

Detalles Bibliográficos
Publicado en:Public Opinion Quarterly Vol. 15; no. 4; pp. 765 - 767
Autor principal: Durant, Henry
Formato: Artículo
Publicado: Oxford University Press / USA Winter51/52
Materias:
Acceso en línea:Ver este registro en EBSCOhost
Descripción
Sumario:This article focuses on probability sampling. Probability sampling has been undertaken in every country, but whether it is appropriate to use quota sampling or probability sampling depends upon the nature of the problem being investigated. In any survey based on probability sampling a failure rate of 15 to 20 percent should be expected. This failure rate depends to some extent upon the nature of the list being used, whether it is compiled in the field or in the office and so on. Considering the circumstances under which most surveys are undertaken, the failure rate mentioned above can be anticipated even after three calls. Only in very exceptional cases, when, for example, a list of policy holders is supplied by an insurance company, the failure rate be as low as two per cent. The speed of interviewing and ultimately the failure rate itself depends very much upon the time of day at which the calls are made.