Abstract
Discrimination power of category rating scales increases with increases in the number of categories. This increasing fineness of category rating scales was measured directly by having subjects give magnitude estimates of distances between adjacent categories. It was found that the increasing fineness could be well described by a power function. However, distances between categories were not estimated as being equal except on a 5-category scale, which was rated as very nearly equal-interval. The implications of this finding for accepting category ratings at face value and for the construction of interval scales from such data are discussed.
Get full access to this article
View all access options for this article.
