What are your thoughts on using Likert-type scales with another instrument that must total 100? Specifically, I'm curious about using 5-point or 7-point Likert scales in the same study with an instrument that offers a scale where the participant must total 100 with four dimensions. The Organizational Culture Assessment Instrument (OCAI) is the survey I am referring to. Can you pair that with typical Likert-type scales such as the Minnesota Satisfaction Questionnaire (MSQ) which is a 5-item Likert-type scale?
I am not familiar with the The Organizational Culture Assessment Instrument, but in general, measures that call for a fixed sum are hard to analyze because the separate sections are not independent. For example, imagine a simple version that requires a sum of 100 across only two dimensions, so that if a respondent scores dimension one at 51, then the score for dimension two is automatically forced to be 49. In this case, the two dimensions would be correlated 1.0 because the second one is a linear combination of the first ( Y = 100 - X ). Similar but more complex patterns apply when there are more than two dimensions.
You can read more about this issue under the heading of "ipsative measures."
Using ipsative measures as would be the case when you require the sum to be 100 is fraught with analytic difficulty. If you were to analyze each dimension separately, it should not be a problem. If you try and combine them requiring the total points on each to be 100, using traditional statistics will produce results that could be misleading. It is partially due to the underlying mathematics that would be akin to dividing by zero.
- Badges
- Science topic