There are many types of data collection methods and even different types of data that can be gathered. Depending on your goals, you’ll use different approaches. Quota sampling is one such approach.

In this guide, we’ll explore what quota sampling is, how to do it, the types, advantages and disadvantages, and provide examples. After you read to the end of the guide, you’ll know exactly what quota sampling is and whether or not you should use it.

Definition of quota sampling

Quota sampling can be defined as the process of researchers creating a sample group (subgroup) to represent a larger population. It does away with the inherent probability associated with gathering data because samples are chosen based on certain criteria determined by the researchers.

An important thing to note is that the proportion of the prevalence of the characteristic you’re using to create the subgroups should be representative of the population you seek to study. For example, if you’re trying to mimic the prevalence of humans with freckles in your sampling, it wouldn’t make sense for everyone in the sample to have freckles. That would no longer be proportional to the population.

An example of how a quota sampling method could be used is as follows. A clothing brand wants to determine the percentage of genders and age groups that prefer running shoes. They take a sample in equal proportion of men aged 21 – 26, women aged 21 – 26, men aged 27 – 33, women aged 27 – 33, men aged 34 – 39, and women aged 34 – 39.

As you can see, there’s no way to get a parameter of all the men and women in those age groups. So, a quota sample is taken.

How to do quota sampling

  1. Choose the relevant subgroups

There are many potential subgroups no matter the kind of data that you’re trying to collect. It’s important to settle on the right ones before you start sending out surveys or otherwise attempt to collect data.

Why?

If you choose the wrong subgroups, your data will be skewed in the wrong direction and the conclusions you draw won’t be effective. When creating the subgroups, you’ll split the population so that each subgroup – usually two – has unique characteristics. For example, the two groups may be men and women.

  • Determine the right proportions for the subgroups

Since a subgroup is representative of the population from which it was created, quota sampling demands that the proportions of your subgroups are consistent with the overall population.

For example, if you know that 81% of your customers are between 21 and 40, the subgroups you create should have 81% of people between the ages of 21 and 40. This is a broad subgroup and yours will likely be narrower to uncover more useful data.  

  • Make sure the sample is the correct size

After deciding on what the segments should be and the correct proportion, it’s time to focus on the absolute size of the sample. Keep in mind that your sample size should be representative of the entire population.

A general rule of thumb you could follow is to get a sample size that’s at least 10% of your total population. In some cases, this may be unfeasible so try to get as close to the 10% number as possible.

If you have a total population of 1,000 then the sample would be 100. If you have a total population of one million then the sample would be one hundred thousand. Unfortunately, that’s not the most feasible number so you may go with 50,000 or even 10,000. Statistical relevance matters too.

Types of quota sampling

There are two main types of quota sampling depending on how much control over the selection the researcher has.

Controlled quota sampling – the researcher has specific restrictions imposed on their ability to sample populations. This may be due to multiple reasons such as lack of expertise of the researcher or the nature of the sampling.

Uncontrolled – There are no explicit restrictions or limitations that the researcher must adhere to. They’re free to choose the sampling criteria and the samples themselves. Like with controlled sampling, there may be many factors that contribute to choosing uncontrolled sampling.

Neither method is inherently better than the other. The one you choose will be dependent on the knowledge that is being brought to bear before the data collection process starts. For example, if the researcher is a domain expert then they should be given free rein to create the sample populations. If, on the other hand, they’re new to the topic, other data sources may need to be used and the sampling controlled to better test the hypothesis.

Note: Sampling can be done based on multiple criteria at once. Age, gender, location, income, employment, etc. may all be taken into consideration when sampling. You’ll get a few examples of this in a later section.

Quota sampling use cases

  • When time is short. If the person conducting research needs to get a representative sample of the population quickly then quota sampling can be a distinct advantage. Creating criteria and only focusing on the people that meet these criteria will save considerable time for the researcher.
  • When only the subgroup is of interest. For example, you may have the entire population of the United States but only millennials are relevant to your study. Within the group of millennials, you’re only focused on those that live in New York and are currently employed as white-collar workers. With this kind of specificity, creating subgroups is essential.
  • When there aren’t enough resources. It’s no secret that most research has a limited budget. Whether that’s academic research or business research, only so much money can be allocated to unearthing the data. Administering surveys and other data collection methods can be expensive. Phone surveys alone cost about $40 a piece. Quota sampling can drive down the costs while improving results.

I’m sure if you’re hard-pressed, you’ll be able to find even more use cases for this type of sampling. The important thing to keep in mind is that it’s a tool to help you achieve an outcome or a goal. Don’t force it if it won’t benefit you directly.

Examples of quota sampling

Example 1:

A firm is hired to determine the effect of new legislation on underserved communities across France. There are ten cities that are identified for the research and there are dozens of communities spread out across those cities.

The researcher needs to understand how it impacts the family unit and how it affects individuals in those families. The total population is estimated at 100,000 people. They create criteria such as:

  • A total of 10,000 people need to be surveyed
  • There will be 200 men and 200 women from each community to make up those 10,000 people.
  • Of those men and women, 20% will be 18 – 25, 40% will be 26 – 33, 20% will be 34 -41, 10% will be 41 – 48, and the last 10% will be 49 or above.
  • 75% of them will be employed and 25% will be unemployed

Note that these numbers are representative of the 100,000 people that make up the entire population. Specifically, the employment status and the age ranges.

Example 2:

A researcher wants to discover the accessibility of clean drinking water in southeast Asia. For this study, they need 5,000 participants and they’ll be doing the research across ten countries in the region. There are many ways the researchers can create subgroups.

  • First, they can divide by location with 500 people being represented in each country.
  • Then, they can divide it by gender with 2,500 hundred men and women (or close to those proportions) total. Per country, it’ll be 250 men and 250 women.
  • The groups can be further divided based on age with multiple ranges. If there are five age ranges then each one will need to be selected based on the prevalence of that age group in the overall population. This may vary by country.
  • The participants can be further subdivided based on the environment they live in. For example, rural homes vs those living in urban homes.
  • Economic status can also be used for division. It can be income level or whether or not they’re employed.

As you can see, there are many ways to go about using quota sampling to

Advantages of quota sampling

  • Saving time – If you needed to gather data from an entire population, it would usually take a prohibitively long amount of time. With quota sampling, you can use a fraction of the people to still arrive at an accurate answer. This will save a considerable amount of time and, by extension, save money.
  • Curtails costs – This post has highlighted the cost of gathering survey data. That’s not from picking up the phone or your survey software. It’s because of the humans that need to conduct the research and the possible incentives you’ll give respondents.  Quota sampling allows you to get more accurate data with fewer people involved.
  • Easier analysis – The data comes from a specific group of people and it’s standardized in many cases. In other words, close ended questions are often used more than open ended questions which makes it easier to study the results.
  • Usually accurate data – Since the research has a lot of control over whom they’re studying, the data is often more reliable.

Disadvantages of quota sampling

  • If there are errors in the research, it can be easily be projected to the entire population. To prevent this, it’s important to do quota sampling more than once and consider the deviation the results produce before drawing conclusions from the data.
  • It requires an experienced researcher to properly create the right sample sizes and criteria to ensure the data is accurate.

Conclusion

Quota sampling can be an effective way to understand trends and insights from a large population. It comes with many distinct advantages but it’s not useful everywhere.

Use this guide to understand if you should be using quota sampling and the different ways you can make the most of it. IF you have any additional questions, let me know in the comments.