Many researchers believe you should only report the results for individual likert items using the proportion of responses for each scale point. For example, 17% strongly agreed, 32% agreed, 10% neither agreed or disagreed and so on. The reason they say this is that likert data is not "equal interval" - the difference between strongly agree and agree is not the same as the difference between neutral and agree, for example. The data is said to be ordinal, not metric. However, this is actually not so much of an issue. Several research studies show have calculated the numerical difference between Likert-type scale points and showed they are very, very close to "equal interval". References for this are given in this paper: Dawes, John. "Do Data Characteristics Change According to the Number of Scale Points Used - an Experiment using 5-point, 7-point and 10-point scales", International Journal of Market Research Vol 50, no 1, 2008. In fact the data used for this experiment is available on the web, go to www.johndawes.com.au and click on "free data".
Chat with our AI personalities
6
See the related link for more information. The Likert scale with 5 points typically has a middle neutral (neither agree or disagree) choice. In the 4-point scale, the neutral choice is removed, so that person who is uncertain is forced to lean (even if slightly) one way or the other.
6
On the "category axis", the scale may be nominal, ordinal, interval or ratio scale. On the frequency axis the scale must be numerical.On the "category axis", the scale may be nominal, ordinal, interval or ratio scale. On the frequency axis the scale must be numerical.On the "category axis", the scale may be nominal, ordinal, interval or ratio scale. On the frequency axis the scale must be numerical.On the "category axis", the scale may be nominal, ordinal, interval or ratio scale. On the frequency axis the scale must be numerical.
The scale factor.The scale factor.The scale factor.The scale factor.