My reference of survey instrument used 5 likert scale, but I used 4 likert scale to avoid having too many "3 or neutral" responses. Is this okay? or does this affect the results in any way?
Hello Anom,
Yes, it can affect your results. About the only way to be completely certain of what degree, if any, of practical import the difference might make would be to either administer both versions to a common group of respondents, or to randomly assign respondents to "version A" or "version B" and then compare the results.
There's been a lot of discussion on inclusion/exclusion of neutral categories, both here on Rgate and elsewhere. You shouldn't have any trouble in identifying the usual arguments each way.
Good luck with your work.
Certainly, the results can be affected. Examining the proportion of neutreal responses, may give you the hint.
Dear Anom Klmo
I suggest 5 extra tips on how to use Likert scales
Kind Regards
Qamar Ul Islam
- Badges
- Science topic
- Similar topics
- Instrumentation Engineering
- Instruments