You need to conduct survey research on a budget. You need to know more about a specific subset of your target population. You need representative results that will lead your organization in the right data-driven direction.
What you most likely need is quota sampling.
Proper data collection and effective sampling matter—which is why quota sampling can make a big difference in market research. With quota sampling, you can make sure your survey results closely resemble your target population. That way, you’ll get results you can really use, and avoid all kinds of survey research no-nos like researcher bias, population definition errors, sampling frame errors, and nonresponse errors.
We’ll take you through the ins and outs of quota sampling, including what it is, how to use it, and how some real-life examples may apply to your own research.
What is a quota sample? It’s a sample group that represents your project’s target audience or population. Let’s dig into what quota sampling actually involves.
Quota sampling is similar to stratified sampling, in that it involves segmenting a population into mutually exclusive subgroups. However, quota sampling is a non-probability sampling method, meaning it doesn’t use random sampling. Researchers divide the target population into subgroups based on certain known characteristics, traits, or interests, then gather information from a specified number of people in those subgroups.
Let’s say that you’re conducting market research on the popularity of meal kit delivery services in a particular city—and you’ve opted to use quota sampling as your sampling method. You might first split your population into age groups or gender, and take a sample from each group to meet an agreed-upon quota.
Since quota sampling relies on sampling a proportional number of individuals relative to the population, you would likely use census data to help determine your quotas. For example, if you determine that the city is 55% female and 45% male, your quota for those subgroups should be consistent with those percentages so that your sample is nice and representative.
As far as survey research goes, quota sampling can be used to monitor the number of respondents that are allowed to complete a survey based on particular traits like age, gender, race, and location.
Why does this matter? Well, people who belong to one descriptive group (like women from the Pacific Northwest who are ages 18-24) may share common beliefs, behaviors and attitudes. Neglecting to give that group the same representation in a survey that they have in the entire population will almost certainly bias your results.
Keep in mind, quota sampling should only be carried out when a researcher does not have access to the entire population. So if you’re carrying out an email survey in which 100% of your target audience is sent an invitation, quota sampling is not for you. Quotas are most useful when your respondents come to you randomly, like through pop-up surveys, embedded surveys, or in-person kiosk or tablet surveys. In these cases, the researcher isn’t reaching every potential respondent in the population, so it’s important to ensure there is proper representation through quotas.
If you have different target audiences, you may face some challenges when it comes to finding the correct quota numbers. However, regardless of the population specifics, the key to proper quota sampling is to make sure your quota numbers reflect the percentages of traits that are present in your population. Let’s take a look at some of the audiences that you could be using quota sampling on:
1. General Population
Most countries have their demographic information freely available to the public. When doing a general population survey, you can usually find all the percentages you need through a country’s census survey statistics and create your quotas accordingly.
For example, Census Canada shows that Canada has an almost perfect split between men and women in its population. This means that a representative sample of Canadian citizens should be 50% male and 50% female. Once you have these percentages, it’s time to set your quotas. Take your calculated sample size and divide it into groups based on the quota. If your sample was comprised of 300 people, then it should have 150 male responses and 150 female responses.
2. Special Groups
Sometimes your sample group won’t reflect the general population. Take, for example, a survey on federal election voters. We know that older demographics have a higher turnout rate at the voting polls than young adults, so we also know the quota numbers for age groups won’t be the same for our voter-based survey as a general population survey. Instead of using a census survey, try researching the demographic makeup of voters in the past election. This will give you a good base for how to divide your sample group.
3. Customer Feedback
Quotas should also be used when businesses have high or low proportions of certain demographics in their customer base. If we look at the video game industry, we have a business sector with an abundance of consumers that are young males and an extremely low number of senior women. Any surveys in this industry should take into account this customer makeup and plan their sample group accordingly. Usually businesses will gain the statistical percentages they need for proper quotas through big data, adding necessary information from sign-up pages, or demographical research.
Quota sampling is generally seen as more reliable than other non-probability methods like convenience sampling and snowball sampling. But just like any sampling method, quota sampling has its pros and cons. It’s not right for every research project, particularly if, as we mentioned, you already have access to an entire population. However, it is a popular way to survey a representative sample of your population without overspending time, energy, or money.
Here’s a breakdown of the main advantages and disadvantages of quota sampling: