Why not run a poll rather than speculate idly? (Of course, the interesting comparison is probably between the percentage that would say that with "better" and w

Why not run a poll rather than speculate idly? (Of course, the interesting comparison is probably between the percentage that would say that with "better" and with "worse" respectively. You'd also want some restrictions on the set X is drawn from to prevent worthless positives from low-status values of X such as sports, social skills or rote memorisation)