Part 2. Marketing Research
Published Online: 15 DEC 2010
Copyright © 2011 John Wiley & Sons, Ltd. All rights reserved.
Wiley International Encyclopedia of Marketing
How to Cite
Fox, R. J. 2010. Nonprobability Sampling. Wiley International Encyclopedia of Marketing.
- Published Online: 15 DEC 2010
Sampling involves selecting a subset from a population to analyze for the purpose of gaining information about the population. When the subset is chosen in a manner that does not involve randomness or chance, the method used to select the sample is called a nonprobability sampling method. Estimates derived from such samples may not be representative of the population as a whole. Also, the typical statistical calculations for evaluating the precision of estimates based on the sample are not appropriate when nonprobability sampling is used. In other words, there is no valid margin of error that can be associated with the estimates derived from nonprobability samples. Information derived from nonprobability samples may be useful, but they must be analyzed with caution and a clear understanding of its deficiencies.
The most common nonprobability sampling methods are convenience and self-selecting samples. As the name suggests, a convenience sample refers to a selection process in which the items selected are convenient to the person conducting the survey, such as university professors conducting research using students in their classes. Self-selecting samples involve situations in which individuals decide to be included in the sample on their own. Telephone or text polls conducted in conjunction with television shows are good examples of self-selected sampling. Also, the common “suggestion box” survey is a form of self-selected sampling.
A general misconception is that making the sample extremely large compensates for any concerns over the fact that nonprobability sampling was used. No matter how large the sample, if it is selected using nonprobability sampling, concerns regarding how well the sample represents the population are valid. Thousands of people may call in to respond in a telephone poll conducted in conjunction with a television show, but the results are still not necessarily representative of the general public.
- judgment sample;
- convenience sample;
- self-selecting sample;
- snowball sample;
- quota sample