Quota Sampling

Quota sampling is a type of survey sampling in which interviewers are directed to gather information from a specified number of members of the population belonging to certain subgroups. The subgroups are sometimes called strata or cells. In contrast to stratified random sampling, in which interviewers are given specific instructions by the survey planners concerning which individuals to interview, interviewers in quota sampling are given some latitude in selecting the population members to interview. Quota sampling, therefore, is similar to convenience sampling in that the survey designers and planners do not strictly control the selection of the sample. Some control, however, is maintained in the distribution of some sample characteristics.

Suppose an insurance company wants to gather information on a sample of its policyholders. The company's database of clients contains records on the sex, age, and number of policies of all its policyholders. Interviews are to be conducted by telephone. The interviewers are directed to interview 75 individuals in each of 12 strata. The strata are defined by three factors: sex, age, and number of policies. Specifically, the strata are formed by grouping policyholders by sex (female or male), age group (18–35, 36–65, and over 65 years old), and number of policies (one policy or more than one policy). These instructions will produce a quota sample because the interviewers may call as many customers as needed in order to quickly find people to complete the required interviews. In a stratified random sample, in contrast, the study planners would random select 75 individuals and instruct interviewers to gather data from the selected individuals.

Quota samples can be implemented in contexts other than telephone or in-person interviewing. Suppose a researcher wants to know the amount of herbicide per acre applied to agricultural land planted with corn and soybeans in the state of Iowa. The researcher picks 20 counties from around the state. He or she directs data gatherers to collect information on herbicide application for five farms growing corn and five others growing soybeans in the selected counties. If the data gatherers are allowed to choose the most convenient farms or the first farms that will participate, then it is a quota sample. The cells are defined by county and crop. If the researcher randomly selects the farms that the interviewers should visit, then it is not a quota sample.

Quota sampling is an example of nonprobability sampling. Convenience sampling is another nonprobability sampling scheme. The goal of a survey is to gather data in order to describe the characteristics of a population. A population consists of units, or elements, such as individuals, plots of land, hospitals, or invoices of purchases or sales for a company. A survey collects information on a sample, or subset, of the population. In nonprobability sampling designs, it is not possible to compute the probabilities of selection for the samples overall and, usually, for individuals. The probabilities are unknown, because typically data gatherers are allowed some freedom in selecting convenient units for data collection.

Estimates of population characteristics based on nonprobability samples can be affected by selection bias. Since the interviewers choose respondents that they want to interview, there is a potential for selection bias. If the respondents in the survey are systematically different on the variables being measured from the general population, then estimates of characteristics will be different on average from what they would have been with a controlled probability-sampling [Page 813]scheme. In probability sampling, the survey planner or researcher controls which units are in the sample and selects the sample using known probabilities of selection. The probabilities of selection can be used to produce estimates of population characteristics without the problem of selection bias. Examples of probability sampling include simple random sampling, stratified random sampling, and cluster sampling.

...