제품

SurveyMonkey는 모든 사용 사례와 요구를 다루도록 구축되었습니다. 제품을 둘러보고 SurveyMonkey로 어떤 효과를 누릴 수 있는지 알아보세요.

온라인 설문조사의 글로벌 리더로부터 데이터 기반 인사이트를 얻으세요.

100개 이상의 앱 및 플러그인과 연동하여 업무 효율성 향상

정보 수집과 결제를 위한 온라인 양식을 만들고 맞춤화하세요.

빌트인 AI을 통한 더 나은 설문조사 작성과 빠른 인사이트 발견

시장 조사에 필요한 모든 것을 갖춘 솔루션

기본서식

비즈니스에 대한 고객 만족도와 충성도를 측정

고객을 만족시켜 지지자로 만드는 것이 무엇인지 파악

실행 가능한 인사이트를 얻어 사용자 경험을 개선

잠재 고객, 참석자 등으로부터 연락처 정보를 수집

다음 이벤트를 위해 쉽게 RSVP를 받고 확인

다음 이벤트 개선을 위해 참석자가 무엇을 원하는지 파악

참여도를 높이고 더 나은 결과를 이끌어낼 인사이트를 발견

참석자들의 피드백을 받아 회의 운영 방법을 개선

동료 피드백을 통한 직원 성과 향상

더 나은 코스를 만들고 교수법을 개선

학생들이 코스 자료 및 프레젠테이션을 어떻게 평가하는지 파악

신제품 아이디어에 대한 고객의 생각을 파악

리소스

설문조사 및 설문조사 데이터 사용에 대한 모범 사례

설문조사, 비즈니스를 위한 팁 등에 관한 블로그

SurveyMonkey 이용에 대한 튜토리얼 및 사용법 가이드

일류 브랜드들이 SurveyMonkey로 성장을 견인하는 방법

영업팀에 문의로그인
영업팀에 문의로그인

What is snowball sampling?

Snowball sampling can help you find survey participants with hard-to-find characteristics.

Finding a sample audience with relevant variables can be challenging. This is particularly common when you're looking for personal information about health, criminal records or sexual orientation. When this issue arises, researchers can rely on the snowball sampling method to get the best population sample.  

In this article, we'll define snowball sampling and explain the advantages and limitations of using it. You’ll also learn about different types of snowball sampling and how to make it work for your business, followed by examples of how it can apply to certain situations.

Snowball sampling is a research technique used to build a population sample of traits that are difficult to find. This method of research, also known as chain-referral sampling, avoids non-probability sampling

When trying to get information from a niche audience, referrals are the best way to obtain a good sample. For example, if you're researching the effects of smoking or vaping, you want a sample to involve people who smoke or vape. Otherwise, you’ll have skewed data. One way to ensure you get relevant information is through snowball sampling. 

TIP: Get the ultimate guide to market research and learn how to build a survey design, and use your audience sample to conduct a quick and efficient analysis. 

There are different ways to go about obtaining a snowball sample. You can use the linear approach, the exponential non-discriminative approach, or the exponential discriminative sampling approach. These options depend on your research strategy. Let’s take a closer look at how these three methodologies work for your study. 

The process of linear snowball sampling involves obtaining a sample population one individual at a time through a referral. Once that person has been added to the sample, they refer another person, and this process continues until you reach the desired number of people to investigate. The advantage of using this technique is that it saves time searching for the right people. The challenge is that it can take time to build a sample audience one referral at a time. 

Exponential non-discriminative sampling provides a solution to linear snowball sampling limitations. This type of sampling method speeds up the process by selecting the first person from the sample population and having them refer multiple subjects to be included in your study instead of just one. And all those subjects refer multiple people until you reach your desired number of subjects for your sample audience. The challenge here is ensuring all these referrals are relevant to your research.

The process of exponential discriminative sampling can alleviate the non-discriminative challenges of snowball sampling by being more selective with the type of subjects chosen. This method borrows from the linear and non-discriminative approach. The process involves repeating the systematic approach of recruiting one subject, but this person refers multiple subjects. Exponential discriminative sampling involves the extra step of the researcher selecting the people from the group of people each subject has referred. While this process takes time, you'll get a concentrated sample of traits relevant to your study. 

TIP: Learn different types of sampling methods for market research. 

How can snowball sampling help your marketing research? There are several benefits to using the snowball sampling approach. It's cost-effective, calling for minimal sourcing planning. It's a budget-friendly process that makes it quicker to find samples of hesitant subjects who might not otherwise want to participate in your study. Furthermore, referrals avoid the risk of including non-relevant subjects in your research, and you'll most likely get educated on certain population characteristics.

Here’s a quick checklist that outlines these advantages of snowball sampling:

  • Quicker to find samples: The subjects do the work for you by finding the people they think are relevant to your study. They provide you with an audience sample you might otherwise not find. 
  • Cost effective: Chain referrals are the most budget-friendly way to source your sample population. So, you spend less time and resources populating your sample.
  • Sample hesitant subjects: Having a subject referring someone who shares the same traits you're looking to study can be easier than approaching them yourself.
  • Avoids risk: Chain referrals help you avoid selecting subjects who don't have any of the traits you want to study, which can result in inconclusive insights.
  • Minimal sourcing planning: You spend less time planning how to populate your sample audience because subjects are coming to you through referrals. 
  • Education on population characteristics: Through chain referrals, you might learn about certain attributes within your sample population that can help your study.

The overall limitation of snowball sampling is that it's biased and may not represent the entire population you want to study. And even through chain referrals, there's still a possibility the subjects you want to study might not wish to participate.

Here’s a quick checklist of some of the overall limitations of snowball sampling:

  • Sampling bias/margin of error: Chain referrals can make it challenging to catch a mistake, resulting in skewed data results.
  • Lack of cooperation: Because of the possible subject of sensitivity, the people you’re looking for may not feel comfortable talking about their experience or situation. Learn more about asking sensitive questions in a survey
  • Not always representative of greater population: Chain referrals can funnel into a pipeline of familiar traits that may leave other important characteristics out and fully represent the population you want to study.

Your research will ultimately dictate your snowball sampling approach, but there are general steps to take. First, you'll need to identify the possible subjects, contact them and encourage them to refer people with similar traits to participate in your study. Once you reach that milestone, the next step is evaluating the referrals to make sure they fit the criteria of your population sample. Then, repeat the process until you feel you've reached the desired number of people to represent your audience sample. Let’s take a closer look at these five steps.

Identifying possible subjects is the most important step because this is the base of your chain referrals. Every referral will represent your beginning subjects, so taking time to identify the traits related to your sample population is recommended. Find out how to create the ideal population sample for optimal feedback and insights.

Once you've identified the type of subjects you want to examine, the next step is to reach out to them and ask if they're interested in participating in your study. This approach can be made in a few ways: in person, by phone, through email, and through social media. Email and social media will be the most cost-efficient ways to reach subjects. Be sure to inform potential subjects about your research and what you hope to achieve. 

TIP: Learn how to ask respondents the right questions with these survey sample questions. 

Once you've established enough trust with your subjects, you can encourage them to send other people who share the same traits your way. This approach is better than having your subjects name them and give you their information, because you'll have to start the process from the beginning. Having your subjects refer others establishes more trust and snowballs into gaining a relevant sample audience.

Verify your referrals fit the criterion of subjects you’ve identified. While chain referrals can guide you towards building a relevant sample audience, taking the extra step to evaluate those referrals will ensure all your subjects have the characteristics you want to study. This approach is especially necessary if you plan to use exponential discriminatory sampling.  

If the snowballing sampling is done right, the only step you'll need to repeat is evaluating referrals. Once you get enough referrals, this method "snowballs" into subjects coming to you rather than approaching them. You'll only need to repeat the entire process if you find that the caliber of referrals doesn't represent the population you're trying to study, even when using the discriminatory sampling approach. 

Whichever type of snowball sampling method you choose to use, it’s best practice to follow these steps:

  • Analyze responses as they are received: Evaluating responses as they come in can help quickly assess if you have the right population sample for your research and adjust if necessary.
  • Establish a margin of error: It's best practice to calculate a reasonable margin of error so you can incorporate that into your assessment when it's time to analyze the data.
  • Encourage participation: Make respondents comfortable with participating in your study by informing them about it. If possible, make it anonymous and communicate that to them. 

What type of industries might call for the application of snowball sampling? Medical research can aid in preventative care and treatment of a disease, and social research can use snowball sampling to study marginalized populations. Often, subjects related to the healthcare industry require anonymity; this is where snowball sampling can work for medical and social researchers. 

Cases of discord are another example of where snowball sampling can be useful. Subjects who have been involved or witnessed serious incidents might be easily swayed into participating through chain referrals than a cold introduction of a basic case study. 

So, what is an example of snowball sampling? Finding people who had been infected during the height of the Covid-19 pandemic. At that time, there wasn’t a vaccine and the only way to create one was to study patients who survived the disease. With quarantine and social distancing, it was likely difficult to locate those people, so snowball sampling may have been useful. 

Another example where snowball sampling can work is for mental illness research. More people are talking about mental health these days. Still, many people might not be comfortable disclosing that kind of personal information. Overall, snowball sampling allows researchers to reach people they wouldn’t otherwise have been able to do.  

Identify a relevant sample population with just a few clicks and see how snowball sampling can work for your research. Then send out your survey using an experienced management platform where you can easily analyze feedback and insights.

Collect market research data by sending your survey to a representative sample

Get help with your market research project by working with our expert research team

Test creative or product concepts using an automated approach to analysis and reporting

To read more market research resources, visit our Sitemap.