Skipping Questions in Employee Surveys
If you run an employee survey that allows individual questions to be skipped (which it should!) then you might wonder if this has an impact on representation or margin of error. This is what to expect from employees skipping questions.
I spent seven years leading methodology and data science for employee engagement surveys at Peakon and Workday. Over that time, I’ve worked with everyone from tiny tech startups to massive retail brands, mining operations, football clubs, and even government entities. Now that I’m no longer directly involved in survey creation, I can share my honest thoughts on the most effective ways to run employee surveys.
See the complete set of tips & tricks for running employee engagement surveys.
If you run an employee survey that allows people to skip individual questions, good. You should. Nobody wants to be forced to answer something irrelevant or awkward. But then the question arises. Does this mess with your data? Does it skew representation, or inflate your margin of error?
Let’s look at what actually happens when employees skip questions.
Scale Questions
On rated scale questions, the skip rate is low. Consistently low. In almost every case, fewer than two percent of respondents skip them. The only time you see a meaningful jump is when the question simply doesn't apply to a chunk of your workforce. For example, asking frontline staff about remote work policies.
I’ve seen this enough times that I built automatic alerts to flag any question with a skip rate above one percent. In my experience, that is almost always a signal that something is off. Either the question is poorly phrased, irrelevant, or risks fatiguing people.
Avoid irrelevant questions. Avoid inappropriate questions. Both erode trust and tank response quality and, eventually, lead to survey fatigue.
On scale questions you should aim for skip rates of around 1%. Any more and you should review the question to check it is relevant to all employees.
Comments
If you're also collecting text comments, and you should be, expect much higher skip rates. This isn’t a fault in the system. It’s human behaviour. People are more selective when typing out feedback. How many skip will depend on the question topic, the survey cadence, company culture, even the mood of the week. Anyone offering a neat rule of thumb here is probably oversimplifying.
Open-ended Questions
Now, for fully open-ended questions, like the classic magic wand prompt, prepare yourself. Three-quarters of your respondents will likely skip it. And that’s normal. People don’t always have something to say, or don’t feel like writing essays during their coffee break.
None of this invalidates your data. It just means you should design your surveys knowing how people naturally behave, not how you wish they would.
Summary
In short, skip rates are not a flaw, they’re a feature of honest data collection. Scale questions barely suffer. Open-ended ones will always have high drop-off. The key is not to panic, but to design surveys that respect respondents' time and relevance. Clean questions, clear purpose.