- Quota sampling is a sampling strategy in which the researcher sets numeric targets for subgroups, then recruits until each target is filled.
- Selection within each subgroup is non-random, and that makes quota sampling different from stratified sampling.
- There are two types of quota sampling: proportional, which mirrors the population, and non-proportional, which sets minimum numbers instead.
- Using quota sampling allows you to control the sample’s composition on the variables you chose.
Definition of quota sampling
Quota sampling is a non-probability sampling technique in which a researcher divides the population into subgroups, decides in advance how many participants each subgroup should contribute, then recruits people who fit until every target is filled.
The subgroups are called quota categories or quota cells. The variables used to define them, such as age or sex, are called control characteristics.
Once a category is full, recruitment for it stops. If you have already interviewed 30 women over 50, you have to turn away the next woman over 50 who offers, even though she is willing and available.
The technique is deliberately structured but not random. Within each category, the researcher or fieldworker chooses whoever is convenient, which is why quota sampling is not considered probability sampling.
Advantages of quota sampling
- Fast and inexpensive: fieldwork can be completed in days, with no list-building phase.
- No sampling frame needed: you need population percentages, not a list of names.
- Guaranteed subgroup representation: no category can be squeezed out by chance.
- Enables subgroup comparison: non-proportional quotas ensure each group has enough members to analyze.
- Easy to explain: a sample matching known population figures is persuasive to non-technical audiences.
- Practical under deadlines: it remains the standard technique in commercial market research for exactly this reason.
Types of quota sampling
Proportional quota sampling
Each category contributes in proportion to its share of the population, as in the table above. The finished sample mirrors the population on the control characteristics, which is the usual aim in market research and opinion polling.
Non-proportional quota sampling
The researcher sets minimum numbers for each category rather than matching population shares. A subgroup holding 5 percent of the population might be given 25 percent of the sample, so that it contains enough people to analyze on its own.
This is the better choice when your aim is to compare subgroups rather than to describe the population. Equal-sized categories give every comparison the same statistical footing.
| Feature | Proportional | Non-proportional |
| Category sizes | Match population shares | Set by the researcher |
| Sample mirrors population | Yes, on the control variables | No |
| Best aim | Describing the population | Comparing subgroups |
| Small subgroups | May be too small to analyze | Boosted deliberately |
Interlocking and non-interlocking quotas
Non-interlocking quotas set separate targets for each variable, so a study might require 50 men, 50 women, 40 people under 40, and 60 people aged 40 or over, with no rule about how those overlap.
Although non-interlocking quotas are quick to complete, you could end up with an oddly composed sample. For example, you could hit both the sex target and the age target while every younger participant happened to be a woman.
Interlocking quotas rule that out, but recruitment is longer and more effort-intensive. Interlocking quotas set targets for combinations, such as 20 women under 40 and 30 women aged 40 or over. They give you a much more “balanced” sample, and they are harder to fill, because the last few cells can be difficult to find.
How to conduct quota sampling
- Step 1. Choose control characteristics. Pick variables likely to affect the outcome you are measuring.
- Step 2. Find the population breakdown. Use census data, institutional records, or professional registers to establish each subgroup’s share.
- Step 3. Fix the total sample size. Decide this before allocating anything.
- Step 4. Calculate each quota. Multiply the total sample size by each subgroup’s share of the population.
- Step 5. Recruit until every category is full. Stop recruiting from categories that have reached their target.
Step 1 is where most projects go wrong. Quotas only correct for the variables you chose, so a sample balanced on age can still be badly skewed on income, education, or health status.
Quota sampling vs stratified sampling
Stratified sampling selects members at random. Quota sampling lets the researcher or fieldworker choose whoever fits and is willing.
| Feature | Quota sampling | Stratified sampling |
| Selection within subgroup | Researcher’s choice | Random |
| Technique type | Non-probability | Probability |
| Sampling frame | Not required | Required |
| Supports confidence intervals | No | Yes |
| Cost and speed | Low cost, fast | Higher cost, slower |
Think of quota sampling as the practical substitute used when no sampling frame exists, and stratified sampling as the rigorous version used when one does.
Both quota sampling and stratified sampling produce samples that look identical on paper. Both divide the population into subgroups and both fill numeric targets. The difference is what happens inside each subgroup.
Quota sampling vs convenience sampling
Using convenience sampling means that you take whoever turns up, and its composition is whatever it happens to be. A quota sample takes whoever turns up until each category is full, so the composition is fixed in advance for specific variables.
| Feature | Quota sampling | Convenience sampling |
| Targets set in advance | Yes | No |
| Composition controlled | On the control variables | Not at all |
| Population data needed | Yes, for proportional quotas | No |
| Effort required | Moderate | Minimal |
Both techniques select non-randomly, and both recruit whoever is available. The difference is that quota sampling imposes a structure on the result. Quota sampling is best understood as convenience sampling with a structure bolted on. That removes the grossest imbalances that could arise when the sample is purely convenience-based.
Quota sampling examples
Example 1: mental health
A researcher wants to compare attitudes toward seeking therapy across age groups and between men and women. Comparison is the goal, so equal-sized cells make more sense than population shares.
The design uses interlocking, non-proportional quotas totaling 120 participants.
| Age band | Women | Men |
| Under 30 | 20 | 20 |
| 30-49 | 20 | 20 |
| 50 and over | 20 | 20 |
Every cell holds 20 people, so each of the 6 comparisons rests on the same number of responses. A proportional design would have produced uneven cells and weaker comparisons.
But a problem could arise in how those 20 people are found. For example, a researcher who is based at a university could easily fill up the under-30 cells with students. In contrast, the researcher may choose to go to a mall and fill the over-50 cells with whoever happens to walk past. The categories are apparently balanced but the actual participants are very different from each other beyond just age and sex.
Example 2: medical education
A professional body surveys 300 pediatricians about continuing education needs. The national register shows the workforce split across 3 sectors, so proportional quotas will produce a sample that reflects the profession.
| Sector | Share of workforce | Quota |
| Private clinics | 70 percent | 210 |
| Hospital | 20 percent | 60 |
| Community services | 10 percent | 30 |
| Total | 100 percent | 300 |
The sector split will match the profession exactly, which matters when results are presented to members.
What the quotas cannot control is who responds within each sector. Pediatricians with a strong interest in continuing education are the most likely to volunteer, so the survey findings about “interest in training” or “willingness to listen to medical podcasts” will be much higher than what actually occurs in reality.
Disadvantages of quota sampling
- Non-random selection: generalizing to the population is often difficult.
- Fieldworker discretion: interviewers may approach people who look approachable, leading to bias.
- A sample balanced on age and sex can still be skewed on income, education, or health.
- People who decline are simply replaced, so no response rate can be calculated.
- False confidence: a sample that matches the population on age and sex looks representative even when it is not.
When should you use quota sampling?
Quota sampling is a sensible choice in these situations.
- No sampling frame exists, but reliable population percentages do.
- Subgroup comparison is central to the research question.
- Time or budget rules out random sampling.
- You’re expected to have a sample that visibly matches the population.
Choose a different technique when any of the following apply.
- A sampling frame is available, which makes stratified sampling feasible.
- The findings are meant to support clinical, regulatory, or policy decisions.
- The population is hidden, which means that you should opt for snowball sampling instead.
Frequently asked questions
Is quota sampling probability or non-probability sampling?
It is non-probability sampling. Although the categories and targets are decided systematically, the people who fill them are not selected at random, so the probability that any individual joins the sample is unknown.
This holds no matter how carefully the quotas were calculated.
How do you calculate quotas for a quota sample?
Fix the total sample size first, then split it between categories.
- Proportional quotas: quota = total sample size × the subgroup’s share of the population.
- Non-proportional quotas: assign a minimum number to each category, often an equal number for comparison studies.
Round each quota to a whole number, then check the parts still add up to your intended total. For interlocking quotas, multiply the shares of each variable together to get the target for each combination.
What is the difference between quota sampling and purposive sampling?
Quota sampling fills numeric targets and stops. Purposive sampling selects participants for what they can contribute, with no fixed count per category. You’ll see this difference most clearly when a category is full. A quota sampler turns away an ideal participant because the target is met. A purposive sampler would recruit that participant because they might still get relevant information from the participant.
Is quota sampling qualitative or quantitative?
It is used mainly in quantitative research, particularly surveys, market research, and opinion polling, where there’s a lot of emphasis on matching the population’s demographic breakdown. Quota sampling is used occasionally in qualitative research, usually to ensure a set of interviews covers a spread of ages or backgrounds. In that setting it functions much like purposive sampling with counts attached.
Can you use statistical tests on quota sampling data?
You can run them, and researchers routinely do, but the results should be read with caution. Many tests have an assumption of random selection, which quota sampling does not provide. The practical convention is to report descriptive statistics confidently, treat inferential results as indicative rather than conclusive, and state in the limitations section that the sample was not randomly selected.
Comment