Home Statistical Dictionary Quota Sampling
Quota Sampling
A non-probability sampling method that sets target quotas for how many units to include from each of several subgroups, matching the overall sample's subgroup proportions to the population, but selects units within each quota by convenience rather than randomization.
In Plain English
Quota sampling looks a lot like stratified sampling at first glance, split the population into groups (say, by age and gender), decide how many people you need from each group to match the population's makeup, then go fill those quotas. The critical difference is how you fill them: stratified sampling randomly samples within each group, while quota sampling just grabs whoever's convenient until each quota is met, an interviewer might simply approach people on the street until they've got their target number of, say, women aged 30-45. That convenience-based selection within quotas is exactly what keeps quota sampling a non-probability method, even though the overall sample's demographic proportions can look reassuringly close to the population's.
Definition
Quota sampling divides the population into subgroups (quota cells, typically defined by demographic variables like age, gender, or region) and specifies a target sample size, or quota, for each cell, usually set to match known population proportions, but fills each quota through non-random, convenience-based selection rather than probability sampling within cells. The result superficially resembles stratified sampling in that both control the subgroup composition of the final sample, but quota sampling provides no guarantee that units are selected representatively within each quota cell, an interviewer filling a quota might disproportionately approach easier-to-reach or more willing individuals within the target demographic, introducing selection bias that matching the marginal demographic totals cannot detect or correct, which is why quota sampling remains classified as a non-probability method despite its resemblance to a probability design.
Properties
- Matching a quota sample's demographic proportions to the population, by age, gender, region, and so on, does not by itself guarantee representativeness on the actual variable of interest, an interviewer could fill an age-30-45 quota entirely with unusually opinionated or unusually available people within that age band, and the demographic match would look perfect while the underlying attitudes remain badly skewed.
- Quota sampling is historically significant as the method behind several well-known polling failures, most notably the 1948 U.S. presidential election polls, which used quota sampling and badly mispredicted the outcome, an episode frequently cited in survey methodology as a cautionary example of quota sampling's hidden within-cell bias.
- Despite its known limitations for rigorous inference, quota sampling remains common in commercial market research and rapid-turnaround polling, where its speed and lower cost relative to true probability sampling are often judged an acceptable trade-off for less methodologically demanding applications.
At a Glance