Interpreting Questionnaire Data

#1
As someone who does not, as a rule, do much work with statistics, I find myself wondering about a report that recently crossed my desk summarizing/analyzing some questionnaire data.

The questionnaire has to do with consumer satisfaction: Each item is answered on a Likert scale ranging from:

Very Satisfied
Satisfied
Neither
Dissatisfied
Very Dissatisfied

The report summarizes responses using a "Satisfaction Index" which is the percentage of respondents who answer a given question "Very Satisfied," "Satisfied," or "Neither." There is also a "Dissatisfaction Index" which is the percentage of respondents who answer a given question "Dissatisfied" or "Very Dissatisfied."

I questioned one of the managers presenting the report--It seemed to me that including "Neither" in the Satisfaction Index skews the results of the survey in a positive light. For my part, I would think that "Neither" should not be included in either the Satisfaction or the Dissatisfaction index.

I was assured that this technique is very common, and is statistically "ok".

I'm wondering if anyone out there can confirm that this is indeed common practice and point me toward any resources/materials that justify (or at least document) that this is "ok." If it's NOT common practice, or is NOT "Ok" pointers toward materials supporting that argument would also be useful.

Eventually, the results of this survey will be used to determine whether corrective action should be taken among health care providers in my area. Thus, I'd really like to understand and feel comfortable about the data that's being presented to me.

Thanks.
 

JohnM

TS Contributor
#2
I agree with you - including "Neither" in the satisfaction index skews the results. Although it may be "common practice," all that means is that "it's the way we've always done it" and it doesn't mean that it's correct.

Off the top of my head I can't think of any references to back up my position, but it just fails the "common sense" test --> people who respond "Neither" are not necessarily Satisfied.
 
#3
Thanks!

Thanks for the reality check!

I (and several of my staff) continue to investigate. I would have thought that "common usage" would have made the practice easier to find among others who do questionnaires and/or materials associated therewith!