Avoiding the pitfalls of low-quality surveys leading to low-quality data collection.
Low quality surveys, or surveys that don't comply with our best practices, can lead to low quality survey responses. This can have a negative impact on the accuracy of the data you collect. In other words, the data you have collected might be biased or wrong.
What are some survey mistakes you can make that can lead to this?
- Your survey can be too long
Surveys that are too long (above 35 questions) will fatigue respondents. This can lead to respondents dropping out without completing the survey or providing random answers for the sake of finishing the survey. - You can have too many grid questions or grid questions with too many statements
Grid questions can be very useful, but you need to use them wisely and not too often. When you do use grid questions, limit them to 3-5 statements per grid question. If you use too many grid questions or too many statements, this can also lead to the survey feeling very long to respondents, fatiguing them and causing the same outcome as a too-long survey. - Your survey could contain too many open-ended questions
Open-ended questions are objectively harder to answer than close-ended questions, which is why they should be used wisely and very sparingly. See which of your open-ended questions can be changed to close-ended question, or run the risk of fatiguing respondents, causing them to leave low-quality responses to your questions. - Your survey could contain leading or biased questions
There are many ways in which a question can be leading or biased. Asking "Which of these brands have you heard of before?" and then not including "None of the above" as an answer option is a simple example. It's biased because the survey crafter is assuming that every single person answering this question has heard of at least one of the brands they're asking about, and we shouldn't be making these assumptions when crafting a question. Refer to our survey crafting best practices to avoid including leading or biased questions. - Your survey could be difficult to answer
Sometimes, a survey is just objectively difficult to answer. Some examples are surveys containing difficult questions that respondents may not have the knowledge to answer, surveys that include referring to an external document or website while answering questions and so on. A survey crafter's focus should be on making respondents' experiences as simple and straightforward as possible. Are the survey questions too difficult? Simplify them. Is referring to an external document or website necessary for answering the survey? Include the document or website as an image, video or gif within the survey itself instead.
Find workarounds that would help make respondents' experiences better.
Our researchers' purpose is to guide you through these pitfalls, and they're there to make sure you don't make these mistakes. Survey reviews are especially important, and listening to their feedback and making the relevant changes to your survey would help you create an easy-to-answer survey that would lead you to collecting accurate data.
If you were to disregard the feedback given, this would make the team unable to determine if a response given by a respondent was genuine, causing us to be unable to clean your data. You would be left with data that may not be as clean or as accurate as you need in order to make the correct business decisions.
Reach out to us via the chat below if you have any questions on the article, and we'd be happy to help!