Take the 2-minute tour ×
User Experience Stack Exchange is a question and answer site for user experience researchers and experts. It's 100% free, no registration required.

I am wondering what are the factors involved in deciding the cutoff value / cut score for 5-point likert scale from:

1 = not ready at all,
2 = not ready,
3 = moderately ready,
4 = ready,
5 = highly ready.

Is it proper for the cut score (in mean value) to be as such if i want it in 3 categories: below 2=not ready, 2-4=ready, above 4=highly ready - and if yes, how do I justify my decision? Are there any references or empirical ways to help decide? I was told that I could justify that the inherent meaning of the scale helps to make such decision - especially using the information embodied in the numerical value - which expert said that?

share|improve this question

2 Answers 2

Here are some resources about likert scales:

In 1999 an interesting sequence of discussions regarding the use of the Likert scale appeared on the AERA Division D LIST SERV. Consider a scale such as: Agree Tend to agree Undecided Tend to disagree Disagree Dr. Dennis Roberts of Penn State, on August 30 1999, regarding the existence of the middle option, made the following statements regarding use of this scale: “There is no assurance whatsoever that a subject choosing the middle scale position harbors a neutral position. A subject’s choice of the scale midpoint may result from ignorance, uncooperativeness, reading difficult, reluctance to answer, or inapplicability (Note that Dr. Roberts included definitions in his email for each of these possible interpretations). “In all the cases above, the investigator’s best hope is that the subject will not respond at all (?). Unfortunately, the seemingly innocuous middle point counts, and, when a number of subjects choose it for invalid reasons, the average response level is raised or lowered erroneously (unless, of course, the mean of the valid responses is exactly at the scale midpoint).

[...]

Think twice before using summed or scaled Likert scores, especially if there is more than a single construct in the instrument. Consider using percentages of individual ordinal categories or, if the sample size is small, collapsing categories and using percentages.

AN EMPIRICAL EXAMINATION OF THE LIKERT SCALE: SOME ASSUMPTIONS, DEVELOPMENT, AND CAUTIONS, Chester H. McCall

"One of the issue about analyzing data from likert scale is assuming that you can use things like means.

Likert scales fall within the ordinal level of measurement.2–4 That is, the response categories have a rank order, but the intervals between values cannot be presumed equal, although, as Blaikie3 points out, ...researchers frequently assume that they are. However, Cohen et al.1 contend that it is illegitimate to infer that the intensity of feeling between strongly disagree and disagree is equivalent to the intensity of feeling between other consecutive categories on the Likert scale. The legitimacy of assuming an interval scale for Likerttype categories is an important issue, because the appropriate descriptive and inferential statistics differ for ordinal and interval variables 1,5 and if the wrong statistical technique is used, the researcher increases the chance of coming to the wrong conclusion about the significance."

Likert scales: how to (ab)use them, Susan Jamieson

share|improve this answer

From what I can remember from classes so dry I'd rather have forgotten, the informative power of a likert/5-7 point scale is closely tied to the symmetry of possible respondent choices.

Applying a 3-group 'cut-off' to a five-point scale can only result in a skew toward an extreme. I can't say with absolute certainty that such a design would have facilitative value, but I can't image how you'd go about crunching the numbers in any comparatively meaningful way...

I'm also a little unclear on what you mean by 'cutoff value'? Clear standards exist for rejecting or failing to reject a null hypothesis (which in the case of a 5 or 7 point scale fall beyond all but one of the representative 'ranges' of the response options).

Are you simply trying to organize groups of respondents for later research? And if so, perhaps there's a more appropriate scale for promoting a more balanced and representative sample?

Or I've completely misinterpreted your question and I should stop before I waste any more of your time...

share|improve this answer

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.