Apr
26

Did you know…

Filed Under (Data Analysis, Methodology, program evaluation) by Molly on 26-04-2011 and tagged ,

…that there is a difference between a Likert item and a Likert scale?**

Did you know that a Likert item was developed by Rensis Likert, a psychometrician and an educator? 

And that the item was developed to have the individual respond to the level of agreement or disagreement with a specific phenomenon?

And did you know that most of the studies on Likert items use a five- or seven-points on the item? (Although sometimes a four- or six-point scale is used and that is called a forced-choice approach–because you really want an opinion, not a middle ground, also called a neutral ground.)

And that the choices in an odd-number choice usually include some variation on the following theme, “Strongly disagree”, “Disagree”, “Neither agree or disagree”, “Agree”, “Strongly Agree”?

And if you did, why do you still write scales, and call them Likert, asking for information using a scale that goes from “Not at all” to “A little extent” to “Some extent” to “Great extent?  Responses that are not even remotely equidistant (that is, have equal intervals with respect to the response options) from each other–a key property of a Likert item.

And why aren’t you using a visual analog scale to get at the degree of whatever the phenomenon is being measured instead of an item for which the points on the scale are NOT equidistant? (For more information on a visual analog scale see a brief description here or Dillman’s book.)

I sure hope Rensis Likert isn’t rolling over in his grave (he died in 1981 at the age of 78).

Extension professionals use survey as the primary method for data gathering.  The choice of survey is a defensible one.  However, the format of the survey, the question content, and the question construction must also be defensible.  Even though psychometric properties (including internal consistency, validity, and other statistics) may have been computed, if the basic underlying assumptions are violated, no psychometric properties will compensate for a poorly designed instrument, an instrument that is not defensible.

All Extension professionals who choose to use survey to evaluate their target audiences need to have scale development as a personal competency.  So take it upon yourself to learn about guidelines for scale development (yes, there are books written on the subject!).

<><><><><>

**Likert scale is the SUM of of responses on several Likert items.  A Likert item is just one 4 -, 5-, 6, or 7-point single statement asking for an opinion.

Reference:  Devellis, R. F. (1991).  Scale development:  Theory and applications. Newbury Park: Sage Publications. Note:  there is a newer edition.

Dillman, D. A, Smyth, J. D., & Christian, L. M. (2009).  Internet, mail, and mixed-mode surveys:  The tailored design method. (3rd ed.). Hoboken, NJ: John Wiley& Sons, Inc.

Print Friendly, PDF & Email
Be Sociable, Share!


Post a Comment
Name:
Email:
Website:
Comments: