The q/m ratio (JJ Thompson's experiment)

We did this one in class, but it seems everyone got a result about 70-150% too high. This would seem to be a rather easy experiment to reproduce, seeing as how we used multimiters to measure V for the parallel plates and I for the coils, and the geometry-to-magnetic field calculation for the coils was known (and verified by the teacher).

No my class is not filled with idiots either.

But 70-150% is ridiculously bad. What gives? has anyone ever done this one and gotten wtihin 10%?