The Likert scale contains options such as Strongly Disagree, Disagree, Neutral, Agree, and Strongly Agree.
One of the WORC teams worked on answering the following research questions:
- Is the Likert scale an ordinal scale or an interval scale, and why does it matter?
- Would you include a midpoint (e.g., Neutral) or not, and why?
- What wording would you use for the midpoint (e.g., Neutral, Undecided, No opinion), and why?
- What impact does inclusion or exclusion of a midpoint have on the validity and reliability of the survey instrument?
- What evidence-based recommendations does the body of current research provide to the OPWL practitioners regarding the use of a midpoint in response scales ?
The WORC team’s article has been published in Performance Improvement Journal:
Chyung, S. Y., Roberts, K., Swanson, I., & Hankinson, A. (2017). Evidence-based survey design: The use of a midpoint on the Likert scale. Performance Improvement Journal, 56(10), 15 – 23. doi: 10.1002/pfi.21727
WORC researchers who participated in this project are:
The researchers also presented the research findings at the ISPI BABS chapter meeting and 2018 and 2019 ISPI conferences:
Chyung, S. Y., Swanson, I., & Shamsy, J. (2017, August 30). Write a strong survey instrument using evidence-based practices. Presented at ISPI-BABS Chapter meeting.
Chyung, S. Y., Swanson, I., Roberts, K., & Kennedy, M. (2018, April 8). Evidence-based survey design with Likert scales. Presented at ISPI’s Performance Improvement Conference, Seattle, WA.
Swanson, I., Roberts, K., Shamsy, J., & Chyung, S. Y. (2019, April 14). Bring me data! Evidence-based survey design. Presented at ISPI’s Performance Improvement Conference, New Orleans, LA.