# How to relate raw scores on five-point Likert to three conceptual categories?

My thesis is on employee productivity and I had a questionnaire which consists of a five-point Likert scale (5-strongly agree, 4-agree, 3-agree, 2-disagree, and 1-strongly disagree). On the other hand, I had a conceptual Framework of a 3 point scale (Highly Productive, Productive and Not Productive).

Is there any standard way to identify whether my respondents are highly productive, productive or not?

-
I think you have an error in your description of the Likert scale, as you have 4 and 3 being identical. –  Peter Flom Oct 21 '12 at 10:21
I don't quite understand the question. Do you have data on the productivity of the respondents? If so, you can use it. If not, how could you possibly do this? –  Peter Flom Oct 21 '12 at 10:23

Many psychological tests convert numeric raw scores into categories. For example, Wikipedia mentions cut-offs for the Beck Depression Inventory:

• 0–9: indicates minimal depression
• 10–18: indicates mild depression
• 19–29: indicates moderate depression
• 30–63: indicates severe depression.

Or for example the BMI define various cut-offs (e.g., Cole et al, 2007).

In general, you lose information by collapsing categories or using cut-offs. Psychological reality tends to be more continuous. That said, categories do have heuristic value as decision aides.

A few options for converting scores to a collapsed set of categories

• Use the logical definition of the scale points: For example, you might use "strongly agree" as highly productive, "agree" as productive, and the other categories as "not productive". This is a simple approach that uses the scale anchor points to define the meaning of the categories.
• Use expert judgements: You can ask a set of experts to evaluate where they thinlk the cut-offs between categories should be. These can then be synthesised. This approach is often used to define acceptable standards for various tests.
• Use normative information: You could use information about the normative spread of the variable and an assumption about the prevalence of the phenomena to define cut-offs.
• Use prediction of external criterion: If the thing has objective existence, or if there are things related to it, you could use predictive models of this external criterion to define the categories.

### References

• Cole, T. J., Flegal, K. M., Nicholls, D., & Jackson, A. A. (2007). Body mass index cut offs to define thinness in children and adolescents: international survey. Bmj, 335(7612), 194. FULL TEXT
-