10 Tips for Writing Inclusive Demographic Survey Questions
By Sam Frampton
In 2021, there’s no excuse for writing surveys that aren’t inclusive. This means thinking empathetically about ways your surveys might be making respondents feel uncomfortable (even unintentionally), and ensuring that you’re asking conscious and sensitive demographic survey questions on gender, race, marital status, education, and more.
Writing inclusivity-conscious surveys and survey subject lines is not only the right thing to do, but it’s actually better for your business as well. By allowing for open-ended feedback, you’re ensuring more accurate, impactful data that will help you retain customers and improve CX. Below are 10 tips you should definitely keep in mind.
1. Consider the Necessity of the Information
When writing your survey questions, think about the information you’re actually trying to uncover and if there is a more direct way to ask that doesn’t assume certain traits about the respondent. This article entitled “Respectful Collection of Demographic Data” presents a great example of how to do this.
If you want to know whether a respondent is free on Saturdays, you don’t need to ask them whether they are religious or how often they attend worship services. Instead, ask them directly what their schedule is like — you don’t need the extra context.
2. Make Sure Free-From or Open-Ended Responses are Always an Option
If you’re asking a yes/no or multiple-choice question, make sure a fill-in box is always available as an alternative. Only allowing respondents to choose from a set of answers assumes that these options will apply to anyone, which may not always be the case. This can also skew your data if respondents are forced to select an option that doesn’t apply to them. As mentioned earlier, open-ended survey responses are a more accurate way to survey in general as you’re able to gain more nuanced feedback and insights.
3. Be Upfront About What the Data is Used For
We're all aware of how little we know about how our data is being used. Showing that you care about their data nowadays is a sure way to convey empathy directly, which not only inspires respect from your customers but also engagement with your survey.
At the beginning of the survey, you should state what you will be using this data for, who will have access to it and what information, if any, will remain anonymous or kept on record. This will help respondents feel more confident in providing their data; it will also give them the option to decline taking the survey if they don’t think it will be in their best interest.
4. Allow Users to Skip Questions
In the same vein, make sure you provide an option to skip each question. If someone feels uncomfortable answering a certain question, they shouldn’t feel obligated to provide that information. Allowing for skips will also improve your survey completion rate as respondents are less likely to bounce altogether.
5. Avoid Giving Too Many Multiple Choices
This one may seem a bit counterintuitive. Isn’t providing as many diverse options as possible the most inclusive way to write a survey question? Well, not really. Though the intentions might be good, throwing out a plethora of options can actually be offensive as it implies respondents don’t belong in the broader categories when that may be how they identify. It can also come across as inauthentic, i.e. just trying to use the right lingo but not actually caring about the results. As always, the better option is to provide a free-form response section and let respondents self-identify.
6. Ensure Your Terminology is Up to Date
This one should be obvious, but a reminder can’t hurt. Inclusive terminology frequently changes, so make sure you are continuously researching and referencing authoritative sources when writing your survey questions. What might seem like an innocuous mistake to you could be offensive to someone else. For example, it’s important to understand the difference between race and ethnicity, or the meaning of terms like the singular they, womxn, and Latinx.
7. Randomize List Orders to Avoid Bias
When structuring your multiple-choice lists, make sure you turn on the randomize option. This will help counteract any unconscious bias you had when writing the question, and ensure respondents don’t feel put off by the order of options.
For example, if your question asks respondents to identify their race and the options are in order of largest population in that area to smallest, those selecting the bottom options may feel othered as the bigger populations are presented as the norm. Randomizing the order will help avoid this issue.
8. Design the Survey with Accessibility in Mind
Not only should the copy of your survey be inclusive, but the actual design should be as well. As you would with any online asset, follow guidelines for making sure the piece is accessible to those with low vision or who use screen readers. This includes adding relevant alt text to all images and making colors and fonts highly contrasted and easy to read. There are plenty of resources out there on other considerations for making surveys accessible.
9. Avoid Making Assumptions About Your Audience
As alluded to in tip #1, always read your questions back carefully to make sure that anyone would be able to answer them. This helps avoid making assumptions about the respondent that can be offensive and skew your data.
For example, could someone of any gender identity easily answer the question? Could someone with any education level or income? This doesn’t mean you can’t target surveys to certain segments of your audience, just that you should ensure you have the correct demographic information beforehand.
10. Structure Surveys Using Skip Logic
Most survey tools and other CX technology will allow you to implement logic-based rules for collecting and sorting data. When creating your surveys, be sure to employ skip logic where needed to make sure you aren’t asking respondents irrelevant or insensitive questions.
SurveyMonkey offers a good example of this. If a respondent answers that they are Hindu, and the next question asks them about their favorite Christmas traditions, there is an obvious disconnect that can make them feel undervalued. With skip logic, the initial response would skip the respondent to the next part of the survey, perhaps to a section asking about their favorite Diwali traditions instead.
Below we summarize these points in an infographic:
We hope that these tips help you create a more inclusive survey program for your business in 2021 as part of a unified customer experience. Contact us today to learn how Chattermill can help you gather and analyze open-ended customer feedback for more accurate decision making, and check out our easy-to-use NPS calculator to see how this data can help improve your score.