Jimmy, this thread seems to indicate that the fault lay in a misprinted version of the ballot paper. If that is the case, no amount of testing the design would have made any difference but the quality control would need to be questioned.

As for the design of forms, I think we agree. I worked as systems co-ordinator for the production control department of a manufacturing company for several years and think I am pretty well aware of the importance of good form design (both paper and computer). IMO, if a form leaves room for interpretation (which in this case seems to be what aligns with what) and mistakes are made, the fault lies with the designer not the user.

Jon