What is the Ceiling Effect?
A ceiling effect happens when your questionnaire or test components/problems aren’t hard enough; An artificially low ceiling is created that is easy to achieve.
If it exists in your test or survey, it becomes a problem when you’re trying to compare two groups with a hypothesis like “Group A’s mean will be higher than group B’s.” If the test questions are too easy, both groups will score equally well, reaching the “ceiling”. This results in meaningless test results and pointless retests, because participants have no room for improvement.
Ceiling effects can also happen when rating scales are skewed so that it’s too easy to reach a perfect or near perfect score. If everyone places at the top of the scale, it becomes impossible to rank order the participants, making everyone equal.
Calling it a “ceiling” effect can be a little confusing in practice, because the ceiling can be reached from above or below. For example, you might want participants to score 100 on a math test, and your participants climb to that goal. Alternatively, you might want participants to complete parts in as little time as possible. In that scenario, lower is better, and the “ceiling” might be an easily achievable run time of 2 minutes.
You can avoid the ceiling effect by carefully choosing test questions.
More Technical Definition
More technically, a ceiling effect happens when both of the following are true:
- The control and treatment conditions have equal dependent variables,
- The dependent variable is at a maximum. In other words, the independent variable no longer has an effect on the dependent variable.
See also: What is the Floor Effect?
Howe, Adele. Ceiling and Floor Effects. Retrieved from https://www.cs.colostate.edu/~howe/EMAI/ch3/node7.html on June 14, 2018.
Howe, Adele. How to Detect Ceiling and Floor Effects. Retrieved from
https://www.cs.colostate.edu/~howe/EMAI/ch3/node8.html on June 14, 2018
Orsini, Pezzuti, & Hulbert. Beyond the floor effect on the Wechsler Intelligence Scale for Children–4th Ed. (WISC-IV): calculating IQ and Indexes of subjects presenting a floored pattern of results. Retrieved from J Intellect Disabil Res. 2015 May;59(5):468-73. doi: 10.1111/jir.12150 on June 14, 2018.