ability; namely, the muscular speed of the right hand. Psychological
laboratories have delicate apparatus for making such a study. But let
us see how we can do it, roughly at least, without any apparatus. Let
each member of the class take a sheet of paper and a pencil, and make as
many strokes as possible in a half-minute, as shown in Figure I. The
instructor can keep the time with a stop watch, or less accurately with
the second hand of an ordinary watch. Before beginning the experiment,
the instructor should have each student taking the test try it for a
second or two. This is to make sure that all understand what they are to
do. When the instructor is sure that all understand, he should have the
students hold their pencils in readiness above the paper, and at the
signal, "Begin," all should start at the same time and make as many
marks as possible in the half-minute. The strokes can then be counted
and the individual scores recorded. The experiment should be repeated
several times, say six or eight, and the average score for each
individual recorded.

[Illustration: FIGURE I.--STROKES MADE IN THIRTY SECONDS
A test of muscular speed]

Whether the result in such a performance as this varies from day to day,
and is accidental, or whether it is constant and fundamental, can be
determined by repeating the experiment from day to day. This repetition
will also show whether improvement comes from practice.

If it is decided to repeat the experiment in order to study these
factors, constancy and the effects of practice, some method of studying
and interpreting the results must be found. Elaborate methods of doing
this are known to psychologists, but the beginner must use a simpler
method. When the experiment is performed for the first time, the
students can be ranked with reference to their abilities, the fastest
one being called "first," the second highest, "second," and so on down
to the slowest performer. Then after the experiment has been performed
the second time, the students can be again ranked.

A rough comparison can then be made as follows: Determine how many who
were in the best half in the first experiment are among the best half in
the second experiment. If most who were among the best half the first
time are among the best half in the second experiment, constancy in this
performance is indicated. Or we might determine how many change their
ranks and how much they change. Suppose there are thirty in the clas