1. We must think about individual Likert items and Likert scales (made up of multiple items)
in different ways.
2. Likert items represent an item format not a scale.
3. Whether Likert items are interval or ordinal is irrelevant in using Likert scale data, which
can be taken to be interval.
4. If a researcher presents the means and standard deviations (interval scale statistics) for
individual Likert items, he/she should also present the percent or frequency of people who
selected each option (a nominal scale statistic) and let the reader decide how to interpret the
results at the Likert-item level.
5. In any case, we should not rely too heavily on interpreting single items because single items
are relatively unreliable.