changing opinions. The discussions,
while carried on for the same length of time, were on the whole less
animated. There was less desire to convince and more restraint, but
the record, which was taken in shorthand, showed nearly the same
variety of arguments which the men had brought forward. Everything
agreed exactly with the experiments with the men, and the only
difference was in the results. The first vote of all experiments with
the women showed a slightly smaller number of right judgments. The
women had 45 per cent. correct judgments, as against the 52 per cent.
of the men. I should not put any emphasis on this difference. It may
be said that the men had more training in scientific observations and
the task was therefore slightly easier for them than for most of the
women. I should say that, all taken together, men and women showed an
equal ability in immediate judgment, as with both groups about half of
the first judgments were correct. The fact that with the men 2 per
cent. more, with the women 5 per cent. less, than half were right
would not mean much. But the situation is entirely different with the
second figure. We saw that for the men the discussion secured an
increase from 52 per cent. to 78 per cent.; with the women the
increase is not a single per cent. The first votes were 45 per cent.
right, and the last votes were 45 per cent. right. In other words,
they had not learned anything from discussion.

It would not be quite correct if we were to draw from that the
conclusion that the women did not change their minds at all. If we
examine the number of cases in which in the course of the first,
second, and third votes in any of the experiments some change
occurred, we find changes in 40 per cent. of all judgments of the men
and 19 per cent. of all judgments of the women. This does not mean
that a change in a particular case necessarily made the last vote
different from the first; we not seldom had a case where, for
instance, the first vote was larger, the second equal, and the third
again larger. And as a matter of course, where a change between the
first and the last occurred, it was not always a change in the right
direction. Moreover, it must not be forgotten that the votes always
covered three possibilities, and not only two. It was therefore
possible for the first vote to be wrong, and then for a change to
occur to another wrong vote. The 19 per cent. changes in the decisions
of the women contained acc