Rating Scale Research. Aoife Kilduff. Rating Scale Usage – 6 point Scale.
Both the development and frequency scales differ significantly from the normal distribution.
There is only a slight difference in the average skewness of the rating scales and they are both biased towards positive ratings with 84% of ratings being 3 or more.
However the frequency scale received a negative kurtosis value which suggests that the distribution is flatter than normal. On the other hand the development scale displayed a positive kurtosis value which suggests a distribution which is more peaked than normal. This suggests that in the frequency scale the scores are more spread out.
6 point frequency rating scales with only end labels are more negatively skewed than scales with all labels present.
End label scales display even more of a bias for positive ratings with 87% of ratings being 3 or over compared to 84% on normal 6 point frequency scales.
Research also suggests that fully labelled scales are more effective as they ensure that respondents understand the meaning of each point on the scale.