STANDARDS FOR MENTAL TESTS
To test or measure mental capacity or any of the dimensions and powers
of the human mind, two preliminary steps are necessary.
First, it must be determined what particular powers or qualities of the
mind it is desired to measure.
Second, there must be prepared a standard or scale that is, primarily at
least, adapted to the measurement of those particular qualities.
While it is, in practice, as has been heretofore pointed out, impossible
entirely to segregate a particular mental quality or power from all the
other abilities and capacities possessed by a particular individual, it
is possible to select certain characteristics or abilities which, by the
degree of their presence or absence, give a fair index of certain mental
dimensions or capacities, and to devise tests that, when taken together,
will measure these “key-abilities” and so reflect the general ability
and capacity of the subject. The standards by which the results of such
tests are gauged must necessarily, therefore, be such as have been
shown, by experiment and experience, to give the closest possible
measurement of the individual’s ability in these particular directions,
by enabling the examiner to compare each subject’s performance under the
test, or series of tests, with the records made under precisely similar
tests by individuals and groups of known ability.
Mental capacity tests may be devised that will measure certain mental
qualities of an infant who has not yet learned to talk, and by thus
providing a comparison between this particular child’s capacities and
the average of children of the same age, enable parents and physicians
to determine in what direction efforts looking toward its mental
development may most helpfully be undertaken. Thus we may test the
infant’s power of observation and perception of shapes, of colours, of
sounds and familiar objects before it is able to talk, measuring these
by standards derived from experience with similar tests applied to a
large number of healthy, normal infants, and by this means determining
whether the subject is above or below the normal average for its age and
if so in what respects.
At the other end of the scale of mental development, let us assume, is
the possessor of the degree of Doctor of Philosophy from any of the
great universities, since this is the principal degree the possession of
which tends to show the possession of unusual mental powers, if not
necessarily of wisdom. By applying to a large number of Ph.D.’s tests
which are designed to require for their successful performance the
utmost use of all their inherent mental abilities, and arriving at an
average of performance by tabulating and comparing the degrees or
percentages of perfection achieved by all of the individuals so tested,
a standard is set up by which to measure the mental capacity of any
individual or group of individuals of superior, or presumably superior,
intelligence. By such a standard there may be measured also the mental
capacity of men and women who have never seen the inside of a
university, but whose education has been acquired in the course of their
business and professional activities. This is so because _what is
measured is not acquired knowledge, but the ability to acquire
knowledge_, which is quite a different thing.
The simplest way to measure the capacity of a circular tank is to pump
it full of water and then measure the water as it is drawn off. But it
would be absurd to contend that because there has never been any water
pumped into the tank it is therefore impossible to determine how much
water it would hold. And what the Doctor of Philosophy has got out of
his university course is comparable to the water in the tank. The
university may have assisted, and if its faculty were competent
undoubtedly did assist him, in discovering earlier in life than he
otherwise would have discovered the actual capacity of his mental tank.
But there are probably as many men of equal mental capacity whose mental
tanks have never been filled with the particular kind of intellectual
fluid that the Ph.D. carries about with him, whose capacity there is no
other means of measuring than by the application of mental tests based
upon the known capacities of Doctors of Philosophy.
The process of measuring the human mind is, indeed, precisely like the
process of measuring an automobile by an engineer, as was pointed out in
the preceding chapter. Back of the tests that are applied to the
automobile to determine its abilities and capacities there must lie a
mass of very definite, exact knowledge of all automobiles or all types
of automobiles already in existence and whose capacities and limitations
are already definitely known. It is of no service to ascertain that the
engine cylinders are of four-inch bore and that the piston has a
six-inch stroke, unless it is well known what the possession of a given
number of cylinders of that particular bore and stroke signifies as to
the ability or capacity of an automobile engine. That knowledge has been
acquired by the observation and measurement over a period of years of
the performance of many automobiles of varying cylinder sizes and number
of cylinders, and the comparison of each size and type with all the
others.
Similarly, it is of no service to apply a test of any kind to a human
being unless we have, in the first place, determined just what
particular abilities or capacities we want to measure, and, in the
second place, possessed ourselves of knowledge as to the significance of
these capacities, after they have been measured.
Here, again, the reader should keep constantly in mind the warnings set
forth in the preceding chapter and try to think of mental abilities and
qualities not as detached, separate, sharply defined parts of a mental
whole (as the engine, transmission and bearings of the wheels of an
automobile are detachable, separate entities) but rather as qualities so
intermingled and connected by an infinite number of attachments to all
the other mental qualities and abilities that no one particular ability
can be measured separately or even positively delimited by any sort of
test. Even if this could be done in the case of one individual, the
process would have to be repeated in each separate, individual case, as
in no two human beings is there found exactly the same combination and
correlation of the manifold manifestations of conscious sensation and
thought that together make up the human intelligence.
But having determined just what qualities and abilities it is desired to
measure, we must set up a standard of measurement by which to compare
the indicated ability of each individual examined, or we shall have
nothing as a result of our test but a mass of information, of the
significance of which we cannot possibly be aware. This standard, for
some purposes, may be merely a composite record of the performances of a
particular group or class examined simultaneously and under the same
conditions. That is to say, if all that is required is to determine
which individual of a group has the greatest ability in certain
directions (and by inference the greatest capacity for further
development along similar lines) then all that is necessary is to apply
a test that will give a comparative measurement of the intelligence of
this particular group. But if the purpose is to ascertain how a
particular individual, or the average of a group of individuals,
compares in particular kinds of capacity with the average or the most
highly developed persons of the same status, education, occupation, or
age, then the standard by which the subject must be measured must be one
derived from the observation and measurement of the mental capacities of
as large a number as possible of individuals engaged in all sorts of
occupations and of all degrees and grades of educational attainment. And
even where the purpose is merely to determine the relative
qualifications and capacities of a particular limited group, it is as a
matter of practice desirable, it might almost be said necessary, to
compare the performance of each individual of the group with a standard
previously fixed and determined as a result of a much broader series of
observations and experiments than can be made within the limits of any
group to which it is practicable to apply any given set of tests as a
whole.
This is true for two reasons. First, without such an outside standard of
comparison all that is determined by the application of even the most
carefully devised tests to any group is that certain individuals are
more and certain others are less able in particular ways than the
average of the group. The net result is of service, but of nowhere near
the service of a record of the same individuals’ performances graded in
accordance with their approach to conformity with a universal standard.
For example, one might take two, three, or a dozen automobiles on a
speedway and quite readily determine which was the fastest and which the
slowest, but unless one were possessed of certain standards of
measurements that in themselves have no relation whatever to
automobiles, the net result would be of little consequence and of no
value whatever in comparing any one of these cars with another
automobile that had not taken part in the particular test. In this case,
two standards are requisite, namely, distance and time. The length of
the course must be definitely ascertained. The time required for each
automobile under test to cover the course must be accurately recorded.
Now we have a record of performance that compares at all times with
universal standards. If we add another automobile to the group we do not
need again to run all the cars, including the new one, along the
speedway to determine where the added member of the group ranks with
reference to the others; we can apply to it alone a test based upon the
universal standards of time and distance with which we have already
compared the others, and the new one falls instantly into its proper
rank among its fellows. So, too, we are enabled by this means to compare
any member of the group with any automobile anywhere in the world, the
performance of which has been gauged by these same universal standards
of time and space, and we are thus able to tell, not only how each
particular car ranks with reference to the limited group of cars, but
how it ranks with reference to all cars of all kinds or of a particular
type so far as these have been tested by the universal standard.
So in testing groups of individuals as to their intelligence or mental
capacity, the use of universal standards of comparison makes the
relative grading of the members of the group with reference to each
other just as easy and simple as though the only standard were that of
the group’s collective performance, and at the same time furnishes a
record of the performance of each individual member of the group by
which he or she may be readily compared with the members of any new
group to which he or she may be at some subsequent time attached, and at
all times with the general run of men or women of the same or differing
social, economic, vocational, or educational status.
It is in the determination of these universal standards and the
preparation of tests, the results of which indicate the individual’s
relative approximation to these standards, that the scientific training
of the psychologist comes principally into play. Rough standards for
testing the more obvious mental capacities might be set up by any
intelligent person who would take the pains to collect the essential
data. These standards would not, however, be universal unless they were
based upon research and experimentation covering as broad a field as
that in which the psychologists have been working for many years. Nor
would they, except by accident, be as simple and as accurate as the
universal standards compiled by scientifically trained persons. For just
as the average untrained individual cannot form an accurate or even an
approximately accurate estimate of another person’s character and
abilities by observation alone, so persons untrained in the study of the
human mind are prone to be misled by the obvious and to lay undue
emphasis upon external indications which do not, as a matter of
scientific fact, actually signify what they are popularly believed to
indicate. The scientific psychologist’s training enables him to
eliminate to a large extent the non-essentials and to include, in the
establishment of standards of mental measurement and the preparation of
tests or methods of applying these standards, many facts which, to the
untrained mind, do not at once present themselves as important elements.
Even in the simplest of mechanical operations every workman knows that
it is not safe to trust to the accuracy of homemade measuring
implements. In the absence of a try-square made by a responsible
manufacturer in conformity with the universal standard right angle, even
the most expert carpenter will refuse to run the risk of error until he
has either obtained a new standard from the hardware store or by the
application of geometrical science and the exercise of careful and
painstaking technical skill constructed for himself a new try-square
that conforms, without the variation of a hair’s-breadth, to the
universal standard to which he must work. Still less would a good
machinist undertake to gauge the close tolerances of an automobile
bearing with a homemade micrometer. He knows it is not sufficient merely
to have a perfect fit of this particular bearing, which might be worked
out by rule of thumb, but that it is essential that the dimensions of
the bearing, down to within a thousandth of an inch, must conform to the
universal standards for automobile bearings, and that the best implement
with which to test the degree of conformity to the universal standard is
the standardized micrometer, prepared by specialized methods and
produced only by the exercise of highly trained technical skill. Once
given such implements of precision, any good workman can readily apply
all the scientific intelligence that went into the devising of the
standards and the preparation of the methods of applying them.
So, once there are at hand scientifically devised standards with which
the mental qualities of any individual may be gauged and compared, and
tests have been prepared for the scientific measurement of these
qualities with reference to the established standards, the application
of these tests to individuals may be made by anybody sufficiently
intelligent to grasp their purport and follow directions exactly. It is
not necessary, in other words, even for the testing of the most complex
and highly developed mental powers, that the actual application of the
test be made by the scientific psychologist. It is possible, and it has
been the purpose in the preparation of the tests which are presented in
this book, to devise mental tests which, if applied precisely as
indicated in the instructions accompanying them, will yield the same
results in the hands of the wholly untrained examiner as though the
actual administration of the tests had been made by the scientist who
devised them.
It must not be thought that the result of any test is always 100 per
cent. accurate. Even good workmen sometimes make errors in the use of
the most precise scientific instruments. Even though constructed with
the most painstaking care, according to the truest scientific formulas
and by men of the highest technical training and skill, the mechanical
instruments of precision are occasionally found to be inaccurate. If
this is the case with material implements and dimensions which are
finite, concrete, and tangible, how much greater is the liability to
error in dealing with the intangible, infinite, and more or less
abstract qualities of the human mind. The scientific psychologist is,
after all, merely another human being, and as such equally liable with
all other human beings to human error. Of no man or woman can it be said
that he or she is infallible, and as every one who applies a
psychological test is human, and so liable to error either in its
application or the reading of its result, conclusions drawn from the
results of any particular test should be accepted as accurate only when
they have been checked by the results of other tests applied to the same
subject, and substantial conformity of the results of one to those of
the others has been obtained. For this reason, among others, no single
test can be expected to yield definite and complete information as to
any particular individual’s mental capacity or ability, whether gauged
by the universal standard or by group comparison. It has, therefore,
been necessary to establish, as preliminary to the preparation of the
Mentimeter tests, a variety of standards, and to prepare a considerable
number of tests under each of these standards, all or most of which must
be used in each instance if anything approaching scientific accuracy is
to be reflected in the resulting scores.
As has previously been pointed out, however, the scientific method is
incomparably freer from the liability to error than any method of
determining human ability and capacity that depends upon unaided
personal observation. How completely this has been demonstrated in
practice in a wide range of fields is set forth in subsequent chapters.
To yield results of maximum accuracy, however, scientific mental tests
must be used only with reference to the standards on which they are
based.
Lest it has not been made clear already to the reader how the method of
establishing mental standards of comparison operates, let us again
briefly try to point out just what is meant by a universal standard of
mental capacity.
It is a comparatively simple matter, involving merely a considerable
amount of painstaking search and the expenditure of a good deal of time
to find, let us say, a thousand engineers, each of whom has demonstrated
in the course of his professional practice that he possesses unusual
ability to project and design bridges and viaducts. Let us suppose that
we wish to take the average capacity of these thousand engineers as the
standard by which to measure every budding engineer in the technical
schools with reference to the capacity of each to become a planner and
designer of bridges and viaducts.
The scientific psychologist must first familiarize himself with the
essentials of that combination of artistic, technical, and mathematical
skill which makes a great engineer. This is not a simple or easy task to
begin with, and to accomplish it calls for the exercise of highly
trained mental powers on the part of the investigator as well as a
thorough understanding of the operation of the various processes of the
human mind. Then there must be devised methods by which, as simply and
yet as precisely as possible, each of these thousand engineers of known
capacity may be tested as to the degree in which he possesses the
various abilities, the sum total of which is the measure of his capacity
as an engineer. It may be necessary to make these tests over a period of
years, and the tests themselves may and probably will require frequent
revision and amendment as it is found in the course of their application
that some of them are unnecessary and others inadequate. If it is found
that any of the tests so applied is readily fulfilled by every subject
examined, the effort is made to increase the difficulty of the test,
until it has reached a stage where the perfect performance of all its
requirements is barely within the reach of the ablest and most competent
of all the engineers under examination. Indeed, some of the tests may be
so difficult that none of those examined may conform precisely to the
set requirements. In respect of some classes of tests this is, in fact,
desirable, as what is being sought is an average of group capacity, and
if any considerable percentage exhibit a capacity greater than can be
measured by the tests set there arises an element of doubt as to the
accuracy of the average combined score, since some of those contributing
to it have obviously greater mental ability than can be measured by the
particular scale used.
Once, however, tests have been applied to the supposititious thousand
expert engineers, and the performance of each of them in each test has
been given its proper place in the scale, and an average struck, there
has come into existence a preliminary standard; which, however, before
being offered for general use in the testing of engineering students and
others, must first be tried out by experimental application on as many
individuals and groups as are available, and their performance with
reference to the standard checked up by all other means available. It
may be, and quite frequently is, the case that this preliminary try-out
of a standard results in the elimination of some of its elements, the
modification of others, and the necessary preparation of a new series of
tests based upon the altered standards. But in this fashion, in the
course of time and as the result of the combined effort of many trained
minds, there is at last set up a standard which is substantially
universal in its application, and by which it may readily be determined
whether or not any particular individual possesses the mental capacity
and particular abilities that have been found to be necessary if he is
to develop into a competent engineer.
As psychological tests are more and more widely applied and there is
consequently accumulated an increasing volume of data which can be
collected, classified, and compared, standards become either more firmly
established as a result of experience or subject to modification in the
light of the wider range of knowledge. In science nothing is final. What
psychology offers to-day is a method of mental tests, the soundness of
which in principle is unchallenged, though the application in detail of
these principles is subject to constant improvement and refinement.