I needed to research a question this week for a client project: “On an attendee evaluation, is it better to list scale points from high/satisfied to low/unsatisfied or the other way around?” My research so far shows that this question has not drawn as much attention as other questions about clear labeling of each point along the scale, how many options to present and whether to have a unipolar or bipolar scale.
I did find one reference about the starting point of the labels:
Survey Gizmo’s advice points to a low-neutral-high progression from left to right or top to bottom:
This mini research project reminded me that, like it or not, evaluation is equally art and science. With a dash of trial and error. In that spirit, here’s some background on “Likert” scales and some best practices to get you started on a solid footing:
A Bit of Background on Likert Scales
“Likert (1932) developed the principle of measuring attitudes by asking people to respond to a series of statements about a topic, in terms of the extent to which they agree with them, and so tapping into the cognitive and affective components of attitudes.
Likert-type or frequency scales use fixed choice response formats and are designed to measure attitudes or opinions (Bowling, 1997; Burns, & Grove, 1997). These ordinal scales measure levels of agreement/disagreement.” From Simply Psychology
Additional Best Practices for Likert Scales
- Label clearly
“The most accurate surveys will have a clear and specific label that indicates exactly what each point means. …we want all respondents to easily interpret the meaning of each scale point and for there to be no room for different interpretations between respondents. Labels are key to avoiding ambiguity and respondent confusion.” From Qualtronics blog
- Keep it odd
“Scales with an odd number of values will have a midpoint. How many options should you give people? Respondents have difficulty defining their point of view on a scale greater than seven. If you provide more than seven response choices, people are likely to start picking an answer randomly, which can make your data meaningless. Our methodologists recommend five scale points for a unipolar scale, and seven scale points if you need to use a bipolar scale.” From Survey Monkey
- Stick with unipolar scales
“Do you want a question where attitudes can fall on two sides of neutrality–“love” vs. “hate”– or one where the range of possible answers goes from “none” to the maximum? The latter, a unipolar scale, is preferable in most cases. For example, it’s better to use a scale that ranges from “extremely brave” to “not at all brave,” rather than a scale that ranges from “extremely brave” to “extremely shy.” Unipolar scales are just easier for people to think about, and you can be sure that one end is the exact opposite of the other, which makes it methodologically more sound as well.” From Survey Monkey
What do you think?
What guidelines do you swear by for Likert scales?
You are welcome to add your thoughts and input by commenting on this post. I also welcome recommendations on resources for inclusion in the forthcoming Resources section of the Learn-o-rama blog.
Also, please consider following Learn-o-rama to receive a summary of each article via email. You can find the Follow link in the bottom right corner of every page.