Kim here, writing again with my favorite collaborator, Sheila B Robinson.
Response rates are everything in survey design. You spend countless hours designing your survey. You work through a design thinking process, getting to know and understand your respondents. You clearly articulate a purpose for the survey and ensure that a survey is indeed the right tool for the job. You carefully craft a set of questions aligned with your research or evaluation questions. Then you launch the survey, only to be disappointed weeks later because out of hundreds of potential respondents, only 22 actually completed the survey. Ugh! The pain is real! It’s like planning a party and no one shows up.
This has happened to us. We don’t want this to happen to you!
Here are 10 of our favorite tips to encourage the best survey response rate for your survey:
- Make sure the survey is engaging to respondents. Questions should be well composed and relevant to them; what do they want to tell us? What is on their minds regarding this topic?
One strategy to ensure your survey is relevant is to conduct a focus group (or series of groups) to determine what is important to respondents about the topic. If it is a program, what do they want to talk about regarding the program? What would they like to be asked or give input on?
- Keep the survey to a reasonable length. There is no magic number of questions or pages we can tell you to use. The length of a survey depends on its purpose, context, and the desired respondents. Keep the survey focused and as concise as possible given information needs, and with consideration for what respondents are likely willing to do.
- Ask easily understandable questions. Be careful not to squeeze too much into one question to avoid asking two. An overly complex question will take longer and require more mental effort to answer and will be more burdensome than two or three short, easy questions. In the example below, Sheila was asked this very confusing question.
This organization could have simply asked two straightforward questions:
1.) Is this your first visit? And 2.) Do you plan to schedule future visits?
AND, take a closer look. This was question NINETEEN! Not a good way to guard against survey fatigue, that’s for sure.
- Have a clearly outlined administration plan long before the survey launches. Will your survey be available online only? Mailed? A telephone survey? Paper and pen? A combination of multiple methods? How long will the survey be open?
Ensure that everyone involved knows their roles, including communicating about the survey and its purpose to encourage response. Create a reasonable timeline that factors in reminders and a potential deadline extension if needed.
- Create a marketing and communication plan. Even if the survey is internal to the organization it is vital to plan ahead for communicating about it and encouraging folks to respond. Will survey invitations be emailed? Communicated on social media channels?
- Consider the messenger(s). Sometimes, it works for the survey researcher to be the main messenger inviting respondents to participate in a survey – especially when distance or anonymity are needed. Often, it is better to have someone the respondents are more familiar with invite them to participate, or to remind or reinforce the value of the survey and their responses. That person, whether a colleague or someone else with influence, may also be able to “speak” to the respondent in a way that is more direct, meaningful or relevant. Of course, power dynamics must also be considered and messengers who could inappropriately influence how respondents answer survey questions should be avoided.
- Consider the message. Be crystal clear about why this data matters to you or your organization as well as why it might be important to those you invite to participate in the survey. In other words, ensure respondents know what is in it for them. Try an appeal to their emotions.
In the example invitation below, the organization explains that by completing the survey, respondents can “help us know what we are doing well on and what we can improve to better care for you” and “help make our care better for future patients and their families.”
- Administer the survey where, when and how people are. Make sure the survey is administered in a format and at a time when respondents are able to complete it.
Administration success story: A school district’s family engagement survey was available online and on paper and translated into several languages. The survey was also made available for completion during parent nights when hundreds of parents came to the schools. Teachers were stationed in school hallways with laptops or paper and pens to encourage parents to stop and complete the brief surveys right there. The result? A great response rate!
Administration fail story: A job satisfaction survey that needed to go through several organization layers to get to our intended respondents wasn’t “advertised” well. There was no opportunity for the evaluator/researcher to speak to the managers/supervisors who were expected to communicate about the survey, and ultimately, they did not receive the information they needed to pass on to potential respondents. Another challenge was that respondents were truck drivers who didn’t have time to sit down in an office at a computer or even a paper survey, and who weren’t likely to complete a survey outside of their workday. We didn’t find the right way to access them. The result? A terrible response rate.
- Offer incentives. Whenever feasible, we encourage incentivizing survey response with a small but meaningful cash or equivalent exchange for participation. Not only does this provide additional motivation, it also communicates that the respondent’s time and insights are valuable. Of course, this must be factored into the project budget. One common way that incentive budgets are kept more manageable is via drawings. A survey response would therefore “enter” a respondent into a drawing for said incentive (e.g., a gift card) but not guarantee it. This comes with an additional layer of logistical work for the survey researcher, and requires capture of respondents’ contact information, which may also not be viable or ideal.
- Send reminders. Plan for 2-3 rounds of reminder messages. Reminders can include thank you messages to everyone who has already completed the survey and other comments about responses that communicate the value of the responses or the need for more responses – E.g., “We are learning so much” or “This data is really going to help us do…”
BONUS:
Because we just KNOW you’re going to ask: What is a “good” survey response rate? The answer is (drum roll please…), it depends (sorry, not sorry!).
Are you in a room full of people who are required to attend the meeting and you’ve asked them to complete a survey during that time? We’d hope for pretty close to 100% if your organizational culture is healthy and morale is good.
Are you sending a survey to hundreds or even thousands of former customers? Anything over 5% might be considered good.
For the many of us who are researchers, program evaluators, trainers, and the like, anything between 5% and 100% is what we’re hoping for, and it’s pretty much up to us to determine what survey response rate will satisfy our information needs.
Want to know a statistically valid survey response rate? That takes a bit more doing. Check out The Complete Guide to Acceptable Survey Response Rates for that.
Need help designing a survey or want your team to grow their own skills in survey design? Contact Sheila or I for a consultation or training.