When conducting a survey, your results can get skewed very easily if you do not include the right people. For instance, if your company mostly caters to middle-aged women, but you survey a group of college students, the results aren’t likely to be very helpful. Or worse, you might think they’re helpful and you end up making changes that alienate your intended audience.
Likewise, if you select a very small group to survey, the results are more likely to be skewed or at least not representative of your audience. In a small group, it just takes a couple of people with unlikely answers to change the whole look of the survey data.
So selecting respondents and figuring out who to survey should not be a process that is taken lightly. You must carefully consider a few different aspects of your survey panel to ensure the best results possible. Below are a few tips:
Determine Your Target Market
In general, the people who take your survey should be a sample of your overall group of customers or target audience. If your company targets middle-aged women, that’s who you should survey. If you target residents of a particular city or state, then that’s who you should survey. Get as close to your target audience as you can.
In the same vein, if you are considering launching a new product aimed at a different demographic group than your overall customer base, then that’s who you should focus your research efforts on. So if your company is one that caters to women of all ages, but you’re working on a new feature specifically for young women, you should focus on getting a group of young women for your survey.
Consider Your Sample Size
In general, the number of people you survey depends on the size of your customer base or the population you want to study. The bigger the population, the more people you should survey in order to get an accurate picture of their opinions.
So if you are developing a product aimed at kids at a particular college campus, your sample size should be representative of the overall college population. If you’re targeting residents of a city, your survey should reflect that city’s population. There are many online sample size calculators that you can use to help determine how many people you need responses from for your survey based on your population size.
You also have to keep in mind that not everyone you send your survey to will actually take it. There are things you can do that can help you increase your response rate, like adding a personalized greeting or offering incentives. But some people will still not respond. So take into account your response rate and calculate the size of the group you’ll need to send your survey to in order for you to get the responses you need.
Be as Accurate as Possible
The larger your sample size in comparison to your population, the more accurate your findings should be. But you also have to take some other factors into account. You have to think about how confident or sure you need to be with your results. In general, you should try to be at least 95 percent confident with your results, but you may want to increase that percentage depending on the subject matter of your survey.
You’ll also have to consider the variability of the population. If you have included a lot of different demographic groups in your survey, their answers are more likely to be varied. So the results might not be as accurate, and you might need to increase the sample size to get the same amount of accuracy.
All of the above factors can contribute to your survey’s margin of error. No survey is going to be absolutely accurate. But your goal should be to increase that accuracy in any way possible. Your survey tool should be able to help you calculate the exact number of respondents you might need based on the above factors.
Survey Photo via Shutterstock
It’s easy to grasp at straws asking any willing participant to answer a survey but the getting your target audience to do so can be challenging! My best advice is to resist asking group members who are not your target to help you out. You won’t get the data you need to serve your ideal client.
Do you think I can use a survey for my forthcoming book on tea?