In a questionnaire, a respondent may select neutral choice, i.e., 'Neither agree nor disagree'. However, there are chances that a respondent may conveniently choose '4' on a 7-Point Likert scale.
If many questions are rated on neutral point, the study may suffer from reliable insights.
My question is, if a respondent chooses too many neutral choices in a questionnaire, what are the remedial actions for analysis?
Can the researcher ignore the case?
If so, what percentage of neutral choices is considered an acceptable threshold?
or
Are there any other guidelines for treating neutral choices?
Regards.
Hi
In order to assure the accuracy of your questionnaire and the scales of measurement, you needs to pilot your questionnaire and to collect the data for calculating the measures of consistency/reliability:
1. The reliability measures could be developed for the whole questionnaire or for each dimension of the questionnaire .
2. Reliability consists of several measures:
Item alpha reliability and split-half reliability assess the internal consistency of the items in a questionnaire – that is, do the items tend to be measuring much the same thing?
3.Split-half reliability on SPSS refers to the correlation between scores based on the first half of items you list for inclusion and the second half of the items. This correlation can be adjusted statistically to maintain the original questionnaire length.
4.Coefficient alpha is merely the average of all possible split-half reliabilities for the questionnaire and so may be preferred, as it is not dependent on how the items are ordered. Coefficient alpha can be used as a means of shortening a questionnaire while maintaining or improving its internal reliability.
Regards,
Zuhair
Please refer to:
1. Landau, S. and Everitt, B. S.(2004). A Handbook of Statistical analyses using SPSS.
2.Howitt, D. and Cramer, D.(2008). Introduction to SPSS.
Regards,
Dear researcher,
The attached documents may be useful for your research.
Regards,
Zuhair
Al-Hemyari,
Thank you for your inputs and SPSS resources.
With too many neutrals, it is possible to get a good Cronbach Aplha, Composite Reliability and Validity of a construct. But it could impact the AVE (average variance extracted). And in case of a dependent variable the neutral choice might lead lower levels of predictability.
When too many neutrals present in a case, does it imply that respondent lack the experience to answer a set of questions?
So, in this regard, is there any threshold per-centage beyond which a respondent to be dropped in the analysis?
Regards... Saradhi.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Probability
- Probability Theory
- Reliability