Gathering accurate, useful employee benefits feedback can be tricky. Good data can help you shape your benefits strategy with confidence; bad data can send you spinning off in the wrong direction.
To help you collect the most useful intel you can, we put together these six survey-building tips with the help of the good folks at Harris Insights, the data science wizards who helped us conduct our most-recent financial wellness survey and benefits communication survey.
Let’s jump right in:
1. Embrace the five-point Likert scale
A yes/no question works great for assessing certain concrete behaviors (Do you participate in our wellness program? Do you receive XYZ benefits?). But when you want to measure employee attitudes or opinions, which offer a range of possibilities, make sure to use what’s called a 5-point Likert scale.
Here’s an example:
Overall, how satisfied are you with our online enrollment system?
a) Not at all satisfied
b) Somewhat satisfied
d) Very satisfied
e) Extremely satisfied
Bonus tip: To reduce bias, always present the ‘negative’ end of the scale first, as in the example above.
2. Offer a valid way to respond to every question
If I were taking a survey about candy bars and saw the question below, I’d feel annoyed.
What is your favorite type of candy bar?
d) Three Musketeers
Why is this so irritating? Because my favorite candy bar isn’t listed among the choices, and I don’t have the option to say, “it’s actually something else.”
Given these bad choices, I’m either just going to pick one just to move on (giving you bad information) or I’m going to get frustrated and quit the silly survey altogether. Either way, you’re not getting the data you want.
The takeaway here? Be careful with how you word your questions and your possible answers. If you’re not sure you’re accounting for every conceivable response, throw in an ‘other’ option just to be safe.
3. Only ask about one aspect of a topic per question
If you ask about more than one thing in a question that only allows one answer, you’re asking for trouble. And by trouble I mean inaccurate, not-helpful feedback.
Here’s an example of what NOT to do:
How easy was our online benefits enrollment system to find and use?
a) Not at all easy to find and use
b) Somewhat easy to find and use
c) Easy to find and use
d) Very easy to find and use
e) Extremely easy to find and use
The problem here is that if an employee thought your system was easy to find but confusing to use, she’d be incapable of giving you a truthful response. Break this up into two questions–one about finding and one about using–to avoid confusion.
4. Be careful with benefits jargon and terms
If you’re going to ask questions that include health care terms like ‘premiums’ or ‘deductibles,’ which employees often don’t understand, make sure to define those terms the first time you use them in the survey. (For example: ‘1. How affordable are your health insurance premiums (the amount for health insurance that comes out of your paycheck)?’)
If you don’t take this extra step, you run the risk of employees not really understanding what they’re being asked. This is a recipe for really unhelpful answers.
5. Try to get the biggest possible rate of response
Obviously, a survey that a lot of your employees respond to is more representative of your workforce and more useful than one that only a few people respond to.
So after you’ve carefully crafted your survey, be strategic about motivating as many people as possible to participate. Some ideas:
- Ask company leaders to vouch for the importance of the survey and/or ask managers to ask their teams to participate during group meetings.
- Explain the goals of the survey and how you’ll be using employee feedback: we all like to know our .02 matters.
- Allow employees time in the workplace/coffee break rooms to complete the survey. This is critical for non-office-based or shift-working/operations-based employees.
6. Keep your survey in the field until you get a representative response
If you’re planning to make business decisions based on your survey results, make sure the group of people providing the feedback reflects the makeup of your organization as much as possible.
Let’s say your company’s employee population breaks down like this:
- 5% senior leaders
- 20% middle managers
- 75% associates
If finish your survey and the respondents end up being 15% senior leaders, 40% middle managers, and 45% associates, your sample will be skewed–extremely–in favor of senior leaders and middle managers.
To fix this, consider leaving the survey in the field a little longer and send reminder emails to associates so you can get more people from that group to better represent themselves.
You might also enjoy:
5 Simple Tricks for Your Last-Minute Open Enrollment Communications
3 Reasons to Add an Interactive Benefits Decision Support Tool to Your 2018 Budget NOW
The ALEX Guide to Epic HDHP & HSA Communications: Helping Your Employees Master Their Plans