An added point to the posts by Carol and George on student evaluations: They are intended to be a form of assessment, but everybody recognizes how unreliable are the respondents, as well as the frequent mismatch between quality of teaching and the final rating. Paradoxically, though, there are times the invalidity of the practice make it attractive. This happens with hiring and promotion decisions, when faculty members and administrators flounder for reasons to decide one way or the other, or are motivated by unprofessional reasons to support or trash the candidate, and cherry pick from the evaluations to find desirable evidence. Precisely because of the vague, inconsistent, and subjective nature of the responses, they are nicely adaptable to the vagaries of the judges.