import Orange
import numpy as np

data = Orange.data.Table("voting.tab")
for x in data.domain.attributes:
    n_miss = sum(1 for d in data if np.isnan(d[x]))
    print("%4.1f%% %s" % (100.0 * n_miss / len(data), x.name))