Does the number of the response categories (points of the Likert scale) affects the reliability of the scale. Actually, i think that when we use 5 point Likert scale (strongly agree, agree, neutral, disagree, strongly disagree) on the responses on "neutral" we don't get anything about the aspect in which we are interestead. i am in the process of developing a scale in which i will use likert scale but i am confused in taking the desicison that which type of likert scale (four, five or six) will be best for the tool. please give suggestions and any autentic reference if have.
Yes. - Does the number of the response categories (points of the Likert scale) affects the reliability of the scale.
The larger the better.
odd, because it offers neutrality.
so you may prefer five to 'four, five or six' or Seven
Unfortunately Likert scale is qualitative (Though it delivers quantitative data for experimental research). Well if the width is large, the information tends to be better differentiated.
My experience shows that it is difficult to generalize the answer to your question. In most cases, its the quality of the question and the way that it is being asked rather than the number of points on the scale. Poorly written questions lead to unreliable data.
@ Douglas D. Gransberg
the quality of the question is most vital. And at times the best could be on a scale of two - Yes versus No.
i.e Dichotimic: that are are represented as being opposite or entirely different.
- A Likert scale of 2.
Of course in that case no of respondents need be large.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Probability
- Probability Theory
- Reliability