One pole or two?

When we create a scale to measure consumer attitudes, be it for consumer or B2B marketing research, we have a choice in how we create the scale. One of the first questions to ask ourselves is how many poles should be used? By poles I mean how we specify the end points for the scale. The two common derivatives are unipolar and bipolar.

Unipolar scales seek to focus respondent thinking on the presence or absence of an attribute or quality. For example when assessing customer satisfaction we might ask the respondent to select from a five-point scale such as the following:

How satisfied are you with the service you received from Acme, Inc. during your last service call?

  • Not at all satisfied
  • Slightly satisfied
  • Moderately satisfied
  • Very satisfied
  • Completely satisfied

Typically unipolar scales are assigned a point value from 0 – 4 or 1 – 5, with the higher numbers corresponding to greater levels of satisfaction or agreement.

An alternative to the unipolar scale is the use of two extreme endpoints to form a bipolar scale. This task requires the respondent to balance two opposite attributes. The graphic below is an example of a bipolar scale applied to political viewpoints (liberal vs. conservative). For survey data analysis these scales are often assigned values ranging from -3 to +3 with a zero midpoint. Bipolar scales require an odd number of points, typically seven.

bipolar

Which to choose? That depends upon the research question you are attempting to answer. However there are considerations. Unipolar scales tend to be easier to administer (five vs. seven-points), they are less mentally taxing (you are focused on one attribute instead of two), and they require less effort on the part of the survey creator (you don’t have to find two polar opposites.) There is a strong case for using a bipolar scale instead of the Net Promoter question for assessing a customer’s willingness to recommend the organization. This scale use endpoints of extremely likely to recommend against and extremely likely to recommend.

Choose your scales wisely!
 

blog comments powered by Disqus
Awkward Moments in Employee Feedback