In contrast to category proliferation, which seems usually to arise somewhat naturally, scale point proliferation takes some thought and effort. An example is:

1) Never 2) Rarely 3) Occasionally 4) Fairly often 5) Often 6) Almost always 7) Always

Such stimuli run the risk of annoying or confusing the responder with hairsplitting differences between the response levels. In any case, psychometric research has shown that most subjects cannot reliably distinguish more than six or seven levels of response, and that in most cases a very large proportion of item-score variance is due to direction of choice rather than intensity of choice. Offering three, four or five scale points is usually quite sufficient to stimulate a reasonably reliable indication of response direction.

Questionnaire items that ask the responder to indicate strength of reaction on scales labeled only at the end points are not so likely to cause responder antipathy if the scale has six or seven points. However, even for semantic differential items, four or five scale points should be sufficient.