UnambiguousWhen it comes to survey design there are many common metaphors one can apply.  Here are a few of my favorites: “junk in = junk out” and “just because you can (ask) doesn’t mean you should (ask).”

Despite the truths behind these metaphors, I’ve met countless “surveyors” who are average Joes tasked with designing a survey.  Why are these employees asked to do something for which they are not skilled?  My guess is the gentle blend of art and science that goes into survey design is lost on them (and their bosses).  If you count yourself among those who think compound questions are more efficient or if you don’t get all the fuss over scale size, then please continue reading.

As the title of this blog implies, the most important part of survey design is making the survey items and their scales unambiguous.  You want as little room for interpretation as possible.  If the items are ambiguous and subjective, the data received from the survey is useless (hence the junk in = junk out metaphor.)  To demonstrate this, let’s look at an example of an unclear item: The training I receive is effective and valuable.

What makes this item ambiguous?  It’s a compound question, which means it addresses more than one construct: value and effectiveness.  How should a survey participant answer if she believes her training is effective, but not valuable or vice versa?  To fix this problem, simply separate the constructs into two items: 1. The training I receive is effective; and 2. The training I receive is valuable.    

Besides keeping the phrasing of your survey items clear, you should also keep your survey clear from unnecessary items.  The items I see most often included unnecessarily in an employee survey are those related to compensation, such as “I believe I am fairly compensated based on my performance.”  This sounds like a good item and compensation is always an appropriate topic for an employee survey, right?  Wrong!  Nine times out of 10, surveyors who include compensation items have very little, if any, intention of doing something with the results.  What they don’t realize is including the item creates an expectation on the part of the participant that action will be taken based on the results.

There are many more examples of ambiguity in survey design, but rather than focus on what to avoid let’s close by concentrating on the goal for which you should strive when designing a survey – unambiguity!  With that word locked in your mind design your survey.  Then re-read every item and ask yourself the following questions:

  • How could this item be misinterpreted? 
  • Are the response options I’m providing necessary?  (This question is of particular importance if your scale has a neutral response option such as “NA.”)  
  • Have I overlooked any obvious responses that should be included in my scale? 
  • Am I going to do anything with this information? 

Revise the item and scale appropriately based on how you answer these questions.  After you finish revising your survey, ask at least one other person to review it as well.  Tell them to play devil’s advocate and try to find ways the survey could be misinterpreted or ask them to provide their opinion on what each survey item is asking.  Then, compare their opinion on what is being asked to what you intended to learn from that survey item.  If you follow these simple steps, even an “average Joe” can create an unambiguous survey that can yield actionable insights!

Leave us a message and a best time to contact you.

* Fields are required