Follow

Likert response formats and the Agree Format


So what is the Likert question format?

The Likert question format is commonly used in surveys and questionnaires (See: Wikipedia). The general format is to present a statement and then a scale of response options representing different response levels to that statement. In Culture Amp we most often use a Likert question format based on the "Agree Format". The Agree Format consists of a statement that then asks the participant how strongly they agree. For example, "I am proud to work at ACME" or "My manager gives me constructive feedback". Participants may answer: Strongly Disagree, Disagree, Neither Agree nor Disagree, Agree or Strongly Agree.

 

 

 

The aim in Culture Amp is to provide a strong anchor for each survey question. This avoids biases that might be present in more ambiguous response formats. For example, everyone will tend to interpret a 1 to 5 scale differently. Some people may never give a "5", others may use it freely. More importantly, individuals will tend to add their own bias. Asking an employee to "Rate from 1 to 5 their Managers' feedback" invites the participant to add their own interpretation to the score. The employee might not feel feedback is important to their role and give a 3. This is fine, however the intent of the original question was to see if the employee is getting feedback, not if they believe it is important. Being clear on the intent of the question becomes particularly important when using Driver Analysis. In Culture Amp, the our approach is to present the ideal or target state - for example, employees get enough feedback, are proud to work at the company or rarely look for jobs elsewhere. You can see some more examples by taking a look at our Engagement Questions or a full preview of our 50 Question Engagement Survey.

 

How is this Reported?

Our agree questions are generally reported as a simple percentage. If Culture Amp reports 85% for an agree question, this means that 85% of respondents either agreed or strongly agreed with the statement. If the statement was "I am proud to work at ACME", this means 85% of employees agree they are proud (Congratulations!). When you drill into Culture Amp reports you get a little more detail. This will be the percentage of "Disagree and Strongly Disagree", the percentage of "Agree and Strongly Agree" as well as the "Neither Agree nor Disagree". The latter is quite important as you survey over time. These represent the "swinging voters" that may tend one way or another over time.

 

Why don't we use different scales and formats?

Culture Amp does support a variety of scale types, such as quality and satisfaction scales, as well as other question formats. However, we recommend keeping to a single scale where possible - it creates less cognitive load for respondents. In either case, we stick to a 5-point scale. There is much academic literature that debates using various point scales (here's a great summary). Our general principle, however, is to (1) keep consistent and (2) make it easy to read and quick to answer. A survey with more detailed scales may add some fine element of nuance to your survey, but we err on getting broader participation and being quick to complete. For that, the 5-point scale is ideal. In that spirit, keep your surveys simple with straightforward to-the-point questions. We see it being much more productive to survey often than survey in inordinate levels of detail.



Was this article helpful?
0 out of 0 found this helpful
Have more questions? Submit a request

Comments