A study was conducted to determine whether different overall evaluation rat
ing scales would produce different results. Six different rating scales wer
e used to measure three stimuli (the respondents' college, the Governor of
New York State, and the quality of the current season's television programm
es). The six overall evaluation rating scales were: an 'overall how would y
ou rate' scale, a 'compared to what you expected' scale, a 'compared to the
ideal' scale, a 'likely to recommend' scale, an 'amount of improvement nee
ded' scale, and an 'if you had to do it all over again, would you?' scale.
The results indicated that the six scales, although producing different mea
ns, were all measuring the same underlying factor. In addition, the 'improv
ement' scale produced more negative evaluations than the other scales and t
he overall performance rating scale tended to produce more positive evaluat
ions.