This depends on the margin of error. The margin of errorindicates the degree to which you can be certain that the sample from your survey accurately represents the attitudes and conduct of your target audience. It is a statistical method of validating your survey experiment, often known as a confidence interval. You can have less confidence in your conclusions the bigger your margin of error.
The margin of error should ideally be as little as possible. The margin of error for market research surveys is typically around 5%.
What does this indicate for your sample size, though?
400 answers is a reasonable starting point, I suppose.
This implies that 400 people will fully complete my survey; after data cleaning, there are still 400 individuals in your dataset.
A marginof error of 400 responses is equivalent to a littleunder 5% in most surveys with a general demographic or broad consumer base readership.
Why Go Bigger Than That?
Hang on! Why are some surveys then conducted with panels in the tens of thousands? Two main reasons:
Confidence – Sometimes, you want to be 99 percent certain rather than the average 95 percent. As a result, the necessary sample size will rise right away.
Getting accuratedata from questions only applies to a portion of the responders. This becomes crucial if you separate responders based on their responses.
A case of the latter is: In a study of Samsung users, the question “Have you used in-app purchases?” is followed by five more questions about the practice. If you had 400 responses at first, maybe half of them would have responded to the additional questions.
Factors that Impact Response Rate
Survey Purpose
Knowing your objectives is necessary before creating any form of survey. Having a certain objective in mind is crucial. Onceyou have this calculated, persuading people to participate in your survey will be much simpler.
When compared to a more targeted survey introduction like “To improve our customer service, we would like to ask you 5 quick questions,” starting with “Can I ask you a few quick questions?” will lower your survey response rate by 11%.
Survey Audience
A common blunder is expecting the same response rate while delivering the same survey to several audiences.
A response rate 12 percent lower than that of a B2B audience is anticipated when you target a B2C audience. Creating an internal survey will have a 20% greater response rate than an external survey.
Survey Length
It would help if you only asked for necessary inquiries because people are easily distracted.
As was previously indicated, a survey’s response rate drops by 15% if it has more than 12 questions or takes more than 5 minutes to complete. When a survey takes more than 10 minutes to complete, the decline is significantly greater (up to 40%).
Before they begin the survey, let your respondents know how long it will take to complete. And be careful not to go over that window of time.
Bottom Line
The margin of error for a 400-person sample size (n=400) is just under 5%, which is a standard goal for market research studies. Any greater than that, though, you run the danger of having investors or other stakeholders doubt your study’s validity or statistical significance.
For this reason, if a client asks me to recommend a sample size, I usually start with 400. At least 400 people should be included in more than 80% of the survey samples.