When Proctor & Gamble started to survey their employees, they used anonymous surveys, under the common assumption that people are more honest when protected by anonymity. Today, Proctor & Gamble has moved completely away from anonymity in their surveys. If you’re a Proctor & Gamble employee, your full name is attached to every employee survey you fill-out. And it’s working. Proctor & Gamble reports better quality of responses, higher response rates, and an overall positive culture around surveys1.
Does anonymity matter? Scrapping anonymity worked for Proctor & Gamble, but would it work for your organization? What does the scientific evidence say about surveys with names and surveys without names?
There are basically two approaches to employee surveys...
- Anonymous Survey: You do not collect any names or identifying information (e.g., age, race). Somebody hates the coffee machine.
- Identified Survey: Employees respond to surveys under their real first and last name. Bob Jenkins in Accounting dislikes the coffee machine.
Gray Area Note: There are also partially-identified surveys, but we’ll get into that later.
The Great Debate: Anonymous or Identified?
There are convincing arguments for and against anonymous surveys. Here are a few examples...
Arguments FOR anonymous surveys...
- Why Your Employee Surveys Should be Anonymous12
- Keep Your Employee Surveys Anonymous3
- Beware employees! Make sure that survey is truly anonymous8
Arguments AGAINST anonymous surveys...
- 4 Reasons to Abandon Anonymity in Employee Surveys6
- Is Your Anonymous Employee Survey **Doing More Harm Than Good?2
Spoiler: Scientific research argues against anonymous surveys for most situations. Non-anonymous (identified) surveys typically provide more useful insights. But organizations keep sending anonymous surveys.
It really comes down to this question...
“But I want people to be honest. If the survey is anonymous, people will feel safe to share how they really feel, right?”
The answer is “kind of.”
First, remember your goal: Why are you doing a survey?
- A true sense of “where everybody is at” on different things.
- Constructive, meaningful suggestions for how to improve your organization and keep employees happier (and more productive, ideally).
- To be alerted to any major issues in your organizational culture.
There are several problems with anonymous surveys that prevent attaining these goals.
Under the anonymity cloak, people don't always behave well...
Studies show that with anonymity, people will...
- Not respond at all. Who's going to know if you don't take this survey? It's anonymous! Can totally skip it1,4.
- Do the minimum. You're anonymous. Nobody is going to know if you slack on the survey, so why put the effort in? Just lazily check through all those "strongly agrees" and type a few lines in the comment boxes, submit, and get back to your work4,7.
- Voice more narrow complaints. Anonymous complainers tend to fixate on one really vexing thing, and rant longer about that one thing, instead of giving a more complete picture of a struggle9.
- Be rude more easily. There's a reason why YouTube tried forcing you to use your real name. Studies show that when people are "masked" we are much less civil11.
Short version: Anonymous users tend to do the minimum, and when they complain they rant rudely about one complaint.
On the other hand, there are lots of advantages to using Identified surveys.
When the survey is personalized/identified, you get all kinds of benefits...
- Higher response rates. Studies show higher response rates for Identified surveys. If your name's on it, then whoever manages the survey is going to know if you didn't reply1.
- Broader, more thoughtful responses. When your name is on it, you have a tendency to be more well-rounded and detailed in your comments (both positive and negative)9.
- Fully honest responses. Remember the common assumption that anonymous surveys will be more honest? Studies show that for most questions asked on employee surveys, there is no difference in how people answer when they're anonymous or when they're identified1.
- Preserves organizational structure. An anonymous survey will tell you that somebody somewhere is very unhappy with their manager, but that doesn't help you figure out where to take action. When your survey is fully-identified, it's easy to track who's unhappy where.
- Easier to follow up. One study found that when people take a survey and put their name on it, they kind of expect that at some point somebody will follow up with them about their comments. From the manager's perspective, this makes it really easy to get more detail about a point raised in the survey (of course, you could always just walk in to your big Monday meeting and be like "Which one of you brought up this thing?")1.
After reading those, you might think "OK, I'm sold. Identified surveys for life." Not so fast. There are still some advantages to anonymous surveys.
Anonymous surveys still have advantages...
- Calm privacy fears. When employees have privacy fears, anonymity is one way to alleviate them5.
- Examine sensitive issues. The one place anonymity is better is for sensitive issues. For example, comments like "I'm going to quit soon" and "My immediate manager Dave is the root cause of all human suffering."10.
It's about fear. When responding to a survey, people can have a fear of being exposed, a fear of something being used against them, or a fear of looking bad.
The only research-supported benefit of making a survey anonymous is to soften fears by establishing a sense of safety. Specifically, psychological safety.
But wait - is your survey the best way to create psychological safety? If you can lessen fear by providing a sense of safety some other way, then you don’t need anonymity to create safety, and you can reap all the benefits of identified surveys.
So, we have this situation...
The big question is: How else can you give people psychological safety, besides making a survey anonymous?
Things you can do to make Identified surveys work, ranging from easy-to-do to hard...
- Create a culture of feedback, input, and surveys. At Proctor & Gamble, the big employee survey is a cultural tradition, and everybody expects it. On every survey, they communicate how data will be used, who will see it, and give employees the opportunity to opt in/out of any part they're uncomfortable with (but most people opt-in). Their surveys aren't anonymous, but they take pains to ensure that responses are highly confidential1.
- Try implementing performance chats to improve communication between employees and managers, and boost psychological safety. Performance chats are like performance reviews, but less stressful and more focused on two-way communication. Watch for our Performance Chats ToolKit.
- Ask questions that matter, and that you're prepared to act on. Make it easy for people to assume their response matters to you and will make a difference in the organization. Response rates are stronger and open-ended comments are more meaningful where there is a history of action.
- Share the results of your survey. You shared meaningful questions. Next share the responses and plans for action. Make it easy for people to see that they're contributing to something that's moving forward.
If you suspect that trust is fragile enough that people are afraid to share real opinions, use anonymous surveys. But plan to move towards identified surveys as you build a track record of productive action. If you've already got the track record and trust, use identified surveys. Then enjoy the insights!
- Biga, A., Silke McCance, A., & Massman, A. J. (2011). Identified Employee Surveys: Lessons Learned. Industrial and Organizational Psychology, 4(4), 449–451. https://doi.org/10.1111/j.1754-9434.2011.01370.x
- Cancialosi, C. (2015). Is Your Anonymous Employee Survey Doing More Harm Than Good? Forbes, January 12. https://www.forbes.com/sites/chriscancialosi/2015/01/12/is-your-anonymous-employee-survey-doing-more-harm-than-good/
- DeFranzo, S. E. (2013). Keep Your Employee Surveys Anonymous. Snap Surveys Blog, March 25. https://www.snapsurveys.com/blog/employee-surveys-anonymous/
- Gordon, R. A., & Stuecher, U. (1992). The Effect of Anonymity and Increased Accountability on the Linguistic Complexity of Teaching Evaluations. The Journal of Psychology, 126(6), 639–649. https://doi.org/10.1080/00223980.1992.10543394
- Jako, R. A. (2011). Respondent Privacy Versus Accountability and Some Situational Considerations. Industrial and Organizational Psychology, 4(4), 452–454. https://doi.org/10.1111/j.1754-9434.2011.01371.x
- Laurtisen, J. (2016). 4 Reasons to Abandon Anonymity in Employee Surveys. Fistful of Talent, August 22. https://fistfuloftalent.com/2016/08/4-reasons-abandon-anonymity-employee-surveys.html
- Lelkes, Y., Krosnick, J. A., Marx, D. M., Judd, C. M., & Park, B. (2012). Complete anonymity compromises the accuracy of self-reports. Journal of Experimental Social Psychology, 48(6), 1291–1299. https://doi.org/10.1016/j.jesp.2012.07.002
- Lucas, S. (2012). Should you fill out a worker satisfaction survey? CBS News, September 21. https://www.cbsnews.com/news/should-you-fill-out-a-worker-satisfaction-survey/
- Mueller, K., Straatmann, T., Hattrup, K., & Jochum, M. (2014). Effects of Personalized Versus Generic Implementation of an Intra-Organizational Online Survey on Psychological Anonymity and Response Behavior: A Field Experiment. Journal of Business and Psychology, 29(2), 169–181. https://doi.org/10.1007/s10869-012-9262-9
- Saari, L. M., & Scherbaum, C. A. (2011). Identified Employee Surveys: Potential Promise, Perils, and Professional Practice Guidelines. Industrial and Organizational Psychology, 4(4), 435–448. https://doi.org/10.1111/j.1754-9434.2011.01369.x
- Santana, A. D. (2014). Virtuous or Vitriolic: The effect of anonymity on civility in online newspaper reader comment boards. Journalism Practice, 8(1), 18–33. https://doi.org/10.1080/17512786.2013.813194
- TINYpulse. (2020). Is TINYpulse Really Anonymous? TINYpulse. https://www.tinypulse.com/anonymity