2. Shy away from open-ended questions
Sumit Aneja has a background in strong leadership, with his experience founding companies and his private equity firm, Pivoton Capital. His great knowledge in research stems from his work at IMF as well as the Bank of America. He also has a background in Engineering and Data Science.
If you’ve ever invested ample resources in building, programming, and designing a survey, only to receive very few responses, you’re not alone!
Researchers across all industries and verticals are experiencing a downturn in their average response rates.
That being said, there are many factors that can contribute to a decreased response rate, for example: inadequate distribution, survey length, asking too many personal questions, and verbiage of your survey questions, to name a few.
For the purpose of this particular blog, we’ll focus in on how to write survey questions that really drive the performance of your survey and resonate with respondents.
If you’ve ever felt the frustration of pouring valuable time and energy into a survey, only to get disappointing results, keep reading:
1. Keep the survey pithy
Let’s start simple: do not ask any questions that are unnecessary to the study. It happens more often then you might think – surveys slip by with redundant or even irrelevant questions. Pay close attention to what you are asking of respondents: if there is more than one survey question that is asking the same thing, consider removing it.
Not only will encountering the same question repeatedly cause an unnerving experience for the respondent, but these multiple questions contribute to the overall length of the survey, which should be kept as short as possible.
2. Shy away from open-ended questions
Beyond the actual length of the survey, the level of effort that the survey asks of the respondent can also be a deterrent for survey completes. In most (but not all) cases, open-ended questions ask more of the respondent, because they need to craft their answer by hand.
There are certainly cases where open-ended questions are required in order to get the right data for the study, or for a specific kind of question even, but please use sparingly!
3. Don’t use double-barreled question types
Writing questions in this manner is actually quite common, as it’s a natural way of speaking/writing. A double-barreled question is an informal fallacy, which is absolutely fine in most contexts – but not for writing survey questions!
A few examples of double-barreled questions:
- Did you find our service helpful and friendly?
- Should the President lower taxes and decrease the unemployment rate?
- Do you like cats and dogs?
You are much better off framing these as two separate questions, which will make your survey question easier to respond to – and your end-data will be cleaner!
4. Give respondents an out
Not in the sense of exiting the survey, but if they encounter questions that don’t necessarily apply to their situation. It’s something we can relate to even as researchers, sometimes you’re asked a question that isn’t relevant to you, or you don’t really know the answer to.
That doesn’t mean the whole survey is a waste – don’t throw the baby out with the bath water! Just make sure to include answer options like “I don’t know” or “Non-applicable”.
Note: there has been a lot of talk about the importance of inclusivity lately. For example, for gender/sex questions, consider not only “Male” and “Female” but also “Non-binary” and “Prefer not to say” so people of all identities feel included in demographic screening questions.
5. Don’t be afraid to get personal
We know that respondents are more likely to engage with a survey if they feel it has a sense of personalization within it. When we talk about “getting personal” in a survey, we’re not referring to asking personal questions necessarily (however, we recognize that is an important facet of certain studies).
However, in this context, we’re referring to adding a personal touch to the way survey questions are written. Ask respondents to share their personal experiences with your client’s products or services. Examples of this may include “How has our services impacted your day-to-day?” or “What kind of uses have you found for product x?”.
6. Test, test, test!
We’ve talked in the past about the importance of soft launching a survey in order to verify the flow and logic of a survey. But it’s also crucial to test the language of the survey: for typos, yes, but also to see if the phrasing of the questions are effective.
If you notice that a certain question is not receiving the desired response, you should take the time to assess the verbiage used and modify if needed.
Crafting survey questions that garner the best response rate is partially a matter of “trial and error” and partially a matter of sticking with best practices in the industry. Use a combination of the methods mentioned above and test your surveys in order to get the best results!