An employee survey is as good as the questions it poses; you can act only
on what you ask.
Here are 11 tips for how to craft an effective staff survey:
1. Have one key decision maker on question design. This person understands
the survey strategy and purpose and is responsible for collecting feedback
on the questions from relevant people within your company. Having a point
person keeps the process concise.
2. Start with objectives. Each question must have a purpose defined by the
objectives. Look at the systems and initiatives you have in place, and
consider what you have in mind for the future.
[RELATED: Join us at Disney World to soak up some sun and a wealth of
social media secrets.]
3. Consider your company culture. Ask, “What has been crucial to our
success?” If you’re using a survey template and find some questions aren’t
relevant, rewrite them. Your company is unique, and your survey questions
should reflect that. It might be changing “manager” to “coach” if that’s
how your company functions.
4. Understand the cognitive model of question response. In other words,
understand how someone answering your question processes it in their mind.
Consider the example: “I receive appropriate recognition for my work.”
Step One:
Understand intent of the question—“I understand what recognition is.”
Step Two:
Search memory for information—“When was the last time I felt
recognized? Just last week I got props from a co-worker.”
Step Three:
Integrate information into judgment—“I think, yes, I do receive
appropriate recognition for my work.”
Step Four:
Translate judgment onto response options—“I would rate this
affirmatively.”
5. Familiarize yourself with “bi-directional” questions. Each one sends a
signal. If the statement to be affirmed or rejected is, “My manager gives
me feedback once a week,” it signals that weekly feedback is the norm.
Understanding this idea ensures that you are sending the right signals
about company norms.
6. Have the right scale. Include a midpoint in your Likert scale, because
it accounts for people who feel neutral on a question. If you exclude a
midpoint, your “favorable” sentiment becomes inflated as people tend to
acquiesce to the “yes” or “positive” side of the scale rather than
answering how they truly feel. (Some oppose a precise midpoint, contending
that it affords an easy “out” to noncommittal respondents.)
7. Include benchmark questions. If the platform you’re using includes
benchmarks, retain those questions so you can compare your survey results
against others in your industry.
8. Don’t ask questions you aren’t ready to discuss. Survey respondents will
expect that if you ask questions about compensation, for example, you’re
willing to discuss or change something about that policy.
9. Avoid “mandatory question” fields. These can be frustrating and annoying
for people who honestly can’t answer a given question. Survey participation
should be voluntary, and mandatory questions go against that ideal.
10. Each question should serve a purpose. You should understand the “why”
behind each question, but not every question has to be immediately
actionable. You’re looking to diagnose prevailing problems, not solve
individual quibbles.
11. Put yourself in the mindset of the survey taker. Survey fatigue can
affect results. If you’re asking about every aspect of your company in a
200-question survey, participants will get tired or zone out. Keep them
interested and focused by limiting the number of questions. Take the survey
yourself to estimate completion time.
Alexis Croswell
works in content marketing at Culture Amp. A version of this post first
appeared on the
Culture Amp blog
.