Daniel Starch.

Educational psychology online

. (page 5 of 41)
Online LibraryDaniel StarchEducational psychology → online text (page 5 of 41)
Font size
QR-code for this ebook

correlation is that it is possible to estimate a person's general ability in
the first year (H. S.) class from the marks he has received in the last four


years of elementary school with accuracy represented by a coefficient of
correlation of .789, and that individual idiosyncrasies may be estimated,
in the case of mathematics and English, with an accuracy represented
by a coefficient of correlation of .515. . . . Indeed, it seems that an
estimate of a pupil's ability to carry high school work when the pupil is
in the fourth grade may be nearly as accurate as a judgment given when
the pupil is in the seventh grade."

A study of the permanency of interests was made by Thorndike
('12) by comparing the relative strength of interests and abilities
during each of three periods of a person's school career, during the
elementary school, high school, and college. These comparisons
were made by asking one hundred individuals to estimate in ret-
rospect, their relative interests and abilities in mathematics,
history, literature, science, music, drawing, and manual work.
Such data are necessarily subject to the errors of memory and
judgment, but they are practically the only results available so far
as strength of interests is concerned. Thorndike inferred from these
estimates that early interests are not passing whims, but rather
prophetic, with a fair degree of certainty, of later interests anS
abilities. He concludes that "A correlation of .60 or .70 seems to
be approximately the true degree of resemblance between the
relative degree of an interest in a child of from ten to fourteen and
the same person at twenty-one." "Interests are shown to be not
only permanent but also symptomatic to a very great extent, of
present and future capacity or ability. Either because one likes
what he can do well, or because one gives zeal and effort to what
he likes, or because interest and ability are both symptoms of
some fundamental feature of the individual's original nature, or
because of the combined action of all three of these factors, interest
and ability are bound very closely together. The bond is so close
that either may be used as a symptom for the other almost as well
as for itself. The importance of these facts for the whole field of
practice with respect to early diagnosis, vocational guidance, the
work of social secretaries, deans, adviser, and others who direct
students' choices of schools, studies and careers is obvious."

The impression gained from all these investigations is that human
nature is not a medley of capricious capacities which vary from
year to year, but rather a fairly consistent combination of abilities
throughout life.



Educational Significance of Sex Differences. If we may judge
fairly at the present time concerning the nature and amounts of
differences between the sexes in mental characteristics, it would
seem that the differences are so small in native intellectual abili-
ties that they are almost wholly negligible in the education of boys
and girls. That boys and girls ought to be educated differ-
ently may very probably be desirable, but for reasons other than
differences in ability. The professional, business, and domestic
life of men and women makes it necessary to have different train-
ing for boys and girls. But so far as the native abilities involved
in school work are concerned, boys and girls might as well pursue
the same courses from the first day of school to the last.
X Popular vs. Scientific View of Sex Differences. Probably more
fallacious psychology of sex has been spread abroad by novelists
and journalists than has been disseminated on any psychological
question of popular interest. Occasional and extreme differences
in individuals of either sex have been seized upon and exaggerated
by descrijitive phraseology and represented as though they were
the normal divergences between men and women. Up to less than
two decades ago, there was practically no scientific knowledge of
the nature of sex differences available, and the statements of popular
beliefs about such differences were hardly exaggerated by the sort of
differences implied in the Sanscrit myth of the creation of woman.

"In the beginning, when Twashtrai came to the creation of woman,
he found that he had exhausted his materials in the making of man, and
that no solid elements were left. In this dilemna, after profound medita-
tion, he did as follows: He took the rotundity of the moon, and the
curves of the creepers, and the clinging of tendrils, and the trembling of
grass, and the slenderness of the reed, and the bloom of flowers, and the
lightness of leaves, and the timidity of the hare, and the vanity of the
peacock, and the clustering of rows of bees, and the joyous gaiety of sun-
beams, and the weeping of clouds, and the fickleness of the winds, and
the softness of the parrot's bosom, and the hardness of adamant, and the
sweetness of honey, and the cruelty of the tiger, and the warm glow of



fire, and the coldness of snow, and the chattering of jays, and the cooing
of the kokila, and the hypocrisy of the crane, and the fidehty of the
chakrawaka, and then compounding all these together, he made woman
and gave her to man. But after one week, man came to him and said:
Lord, this creature that you have given me makes my life miserable.
She chatters incessantly and teases me beyond endurance, never leaving
me alone; and she requires incessant attention, and takes all my time up,
and cries about nothing, and is always idle; and so I have come to give
her back again, as I cannot live with her. So Twashtrai said: Very well;
and he took her back. Then after another week, man came again to him
and said: Lord, I find that my life is very lonely since I gave you back
that creature. I remember how she used to dance and sing to me, and
look at me out of the corner of her eye, and play with me, and cling to
me; and her laughter was music, and she was beautiful to look at, and
soft to touch; so give her back to me again. So Twashtrai said: Very
well, and gave her back again. Then after only three days, man came
back to him again and said: Lord, I know not how it is; but after all I
have come to the conclusion that she is more of a trouble than a pleasure
to me; so please take her back again. But Twashtrai said: Out with you.
Be off. I will have no more of this. You must manage how you can.
Then man said: But I cannot live with her. And Twashtrai replied:
Neither could you live without her, and he turned his back on man, and
went on with his work. Then man said: What is to be done? For I
cannot hve either with or without her. (Thomas, Source Book of Social
Origins, p. 512.)

Such popular beliefs have been in part justified by the probability
that many obvious differences are due to the work, and the result-
ing variation in experience and environment, of women as con-
trasted with those of men. Thus men know more about business,
politics, current events and machines because their occupations
bring them much more in contact with these things; but it does not
follow that women could not, or would not, know as much about
them if their occupations were as much concerned with them.
Women know more about cooking, social events, and household
utensils because their occupations bring them much more in con-
tact with them; but it does not follow that men could not, or would
not, acquire as much knowledge or skill in these directions if their
occupations required it.

The differences between the sexes are probably quantitative
rather than qualitative. Both men and women have the same re-
flexes, instincts, and capacities with the exception of certain as-
jiects of the sex instinct. These are probably similar in the main


and dififer chiefly in their manner of expression. The differences
due to sex life and the rearing of children, with the consequent
differences in occupations and experiences, will account for many
of the superficially observable differences between men and women.

What are the differences that have been scientifically measured
and compared? In order to produce a complete picture of mental
differences between men and women it would be necessary to
measure each trait in a very large number of persons and to com-
pare the measurements with regard to both the averages of the
abilities and the manner of the distribution of each ability. This
has been done in part only with a few traits and only upon small
groups of persons.

Differences in Average Amounts of Mental Abilities. There
are two methods by which abilities of two groups may be compared.
Either we may state the actual average or median of each group,
or we may state how many members of one group reach or exceed
the average or median of the other group. The latter method is
preferable in many respects to the former in that it makes possible
a comparison of groups of various sizes and indicates the relative
differences more nearly true to fact. The two methods may be
illustrated in the case of a memory test consisting of the oral pres-
entation of ten words at the rate of one word per second and of
asking the subjects to record immediately the number of words
remembered. He ma> then state that the number of words
remembered on the average by men was 6.9 and by women 7.2.
Or we may state that 43.6% of men reached or exceeded the me-
dian of the women. The latter method of comparison represents
probably more true to life the amount and kind of difference or
similarity that actually exist. The differences, hastily inferred
from a comparison of averages only, would lead to the conclusion
that in regard to memory women are distinctly superior to men.
The implication would be that all women have a memory superior
to that of men, whereas the fact is that the number of women
having a memory superior to that of men is really small and that,
in these few women, memory is better only by a very small shade.
If 43% of men reach or exceed the median of women, it means
that if the 7% of women having a slightly superior memory were
omitted, the remaining 93% of the women would have a memory
ability identical with that of the men. A difference of 7% in the
distributions between two groups is represented by the curves in



Figure 28. The difference is so small that the groups could hardly
be distinguished.

By the method of amounts of overlapping in the distribution
of one group over the other, the following results have been ob-
tained from students in the University of Wisconsin in a series of
tests on memory as just stated, on perception consisting in the
cancellation within one minute of as many of a certain geometrical
figure as possible, on motor ability consisting in tapping with a
pencil upon a card as rapidly as possible for thirty seconds, and on
mental addition as described elsewhere. ^

Fig. 28. — Distribution curves representing a difference of 7% between the
medians of the two groups.

Percentage of men reaching or exceeding the median of women.

Perception of geometrical forms .

Memory of words

Motor ability

Mental addition

193 men
55 men
25 men
21 men

200 women
77 women
50 women
46 women


In the interpretation of these percentages of overlapping it
must be remembered that if 50% of one group reaches or exceeds
the median of the other, it means of course that the two groups
are identical in ability and distribution. If the percentage of men
reaching or exceeding the median of the women is over 50% it
means that the men are superior by the number exceeding 50%.

Helen Thompson WooUey made a series of tests as indicated in
the following table upon twenty-five men and women at the Uni-
versity of Chicago, on the basis of which Thorndike has computed
the following percentages of men reaching or exceeding the median
for women:

^Experiments in Educational Psychology, revised edition, chapter i6.



Percentages of men reaching or exceeding the median of the women. After
Woolley as computed by Thorndike ('14, Til, p. 178).

Reaction time 68%

Tapping. . .' 81%

Sorting cards, speed 14%

Sorting cards, accuracy 44%

Thrusting at target 60%

Drawing Hues 72%

Threshold of pain 46%

Threshold of taste 34% (22)

Threshold of smell 43%

Lifting weights 66%

Two-point discrimination 18% (43)

Memory (syllables and learning) 32% (46)

Ingenuity 63%

In a similar comparison made on the basis of 100 boys and 100
girls from results obtained by Gilbert, the percentage of boys
reaching or exceeding the median of girls was as follows:


Percentages of boys reaching or exceeding the median of the girls. After
Gilbert ('94) as computed by Thorndike ('14, III, p. 182).

to 14 years 15 to 17 years

Discrimination of weights 48% 58%

"colors 39% 58%

Reaction time : . . . 57% 76%

Resistance to size-weight illusion 55% 68%

Rate of tapping 64% 73%

Thorndike ('14, III, p. 183) reports measurements in which the
comparison of the percentages of boys reaching or exceeding the
median of girls for persons 8 to 14 years old, were as follows:


Associative tests, oppositcs, addition, multiplication, etc 48%

Perception, A-test, etc 33%

Memory of words 40%

The writer has made comparisons in the case of school subjects
on the basis of abilities measured by means of tests and scales.
Speed of writing was measured in terms of letters written per
minute. Quality was rated by the Thorndike scale. Attainments


in the remaining subjects were measured by the author's tests in
these fields. The following percentages of boys, reaching or ex-
ceeding the median of the girls, were obtained:


Speed of handwriting, about iioo boys and iioo girls 47%

Quality of handwriting, " iioo " " iioo " 39%

Arithmetical reasoning, " 1250 " " 1250 " 60%

History, " 429 " " 526 " 72%

Geography, " 447 " " 472 " 48%

Figures of a similar sort computed by Thorndike ('14, III, p.
183) on the basis of teachers' marks showed the following percent-
ages of boys reaching or exceeding the median of girls:


High school pupils

EngUsh 41%

Mathematics 57%

Latin 57%

History 60%

College students

English 35%

Mathematics 45%

History and economics 56%

Natural sciences So%

Modern languages 40%

The difficulty with many of the measurements is that they are
based on too small a number of persons. Comparisons based on
twenty-five persons from either sex may be indicative but not
final. Summarizing, we may say that women and girls are supe-
rior in sensibility, in memory, in most forms of perception, in
quality of handwriting, and linguistic fluenc v. It is interesting
to note in this connection that in the survey of mental-test
results ^ the women excel in twelve out of fourteen tests which de-
pend chiefly upon linguistic fluency. Thus the females excel in
speed of reading, both oral and silent, in amount of information
given in describing an object or in making a report, in the genus-
species test, in the number of words thought of and written per
minute, in the part-whole test, in the opposites test, in memory
span for words, in memory for logical- verbal material, in the word-

' Given in the various chapters of Whipple's Manual of Menial and Physical Tcsls.





building test and in the Ebbinghaus completion test; while the
males excel in the rate of association and in the sentence building
test. Apparently the popular belief in the greater linguistic fluency
of women is not without foundation. Men and boys are superior
in motor capacities, such as tapping, quickness of reaction, in
arithmetical reasoning, and in resistance to suggestions as indi-
cated by the size-weight illusion and the use of suggestive ques-
tions in testimony. The two sexes seem to be approximately
equal in associative processes and in most school subjects. Th^
amounts of difference, however, are very small. This is particu-
larly true of all the traits that have been measured in a sufficiently

•J 10


Fro. 2Q. — Comparison of general intelligence of boys and girls as measured
by the Stanford revision of the Binet-Simon tests. After Terman ('16, p. 72).
The numbers along the vertical axis are intelligence quotients as explained in
Chapter VII.

large number of persons to make the comparisons safe. Any
differences lying between 40% and 60% of the number of either
sex reaching or exceeding the median of the other are practically
negligible. If 60% of one sex reach or exceed the median of the
other, it means that 10 persons in a hundred of the one sex, are by
a small amount superior to the other. Differences larger than this
have been established with a fair degree of certainty practically
only in the case of one large field of capacities, namely, that of
motor abilities. Differences in nearly all other respects in which
comparisons have been made on large numbers of persons are
almost entirely within the limits of 40% to 60%. Terman found
in measuring the general intelligence of nearly i,oool>oys and girls



by means of his revision of the Binet-Simon tests that for the ages
of five to fourteen girls tend to be very slightly superior to boys
and that after fourteen they are practically equal. His results are
set forth in Figure 28.

It seems a likely interpretation that motor superiority has been
carried over to include intellectual superiority as well. For centu-
ries women have been considered intellectually inferior to men.
They were thought to be incapable of acquiring anything more
than an elementary education. It has been only since the middle
of the 19th century that co-education and women's colleges have
been generally established. Intellectual inferiority has probably





1 ihr.


(-• 1

a 1

'\i \

— — —Women

^ /kl \ rr\/\

V \


.-1 Ci CO tH

1 t 1 1

"-1 oj 03

CO t- 00 Ci

1 1 1 1 1

^ in CD (^. ca

9-10 -
10-11 -
11-12 -
13-13 '

Fig. 30. Range of ability of men and women in color discrimination. After
Henmon ('loj.

been inferred chiefly from motor and muscular inferiority and from
the conditions of a narrower environment and dependency due
to the bearing and rearing of children. The inference and belief
of intellectual inferiority is apparently unfounded. This conclu-
sion may be fairly drawn both from the specific psychological
tests that have been cited and also from the recent successes of
women in the acquisition of higher education.

Difference in the Range of Variations in Abilities. Besides
comparing the average amounts of any given ability in the two
sexes, we may compare also the range of abilities from the lowest
to the highest in the two sexes. Such comparisons have been made
in a few traits and the general inference has been that the range of
abilities is wider among men than among women. The distribution
of the abilities in the geometrical perception test made upon 193


men and 200 women mentioned in a preceding paragraph, was
as follows:

Scores: 2-3 4-5 6-7 8-9 lo-ii 12-13 14-1S

193 Men 4.5% 15.4% 33.9% 21.9% 15.4% 56% 2.8%

200 Women 3.2% 20.8% 38.9% 21.6% 10.7% 3.9% 0.8%

Thus in the extremely high ability of canceling 14 to 15 geo-
metrical figures in one minute, there were 2% more men than
women, and in the lowest ability of canceling only two to three
geometrical figures, there were 1.3% more men than women.
Comparisons of this sort can be made safely only on large numbers
of individuals, and consequently there is as yet little reliable
material available.

The ratio of female to male variability has been computed by
Thorndike ('14, III, p. 194) on the basis of tests of memory, re-
action-time, discrimination of length, opposites, and cancellation
made by Gilbert ('94) upon 100 boys and 100 girls of each age
from 6 to 17. The average ratio in all tests for the ages of 9 to 12
was found to be .92, for the ages 13 to 14 1.025, and for the age
of 15, .97. Girls were slightly less variable at all ages except 13
and 14. In a test of color discrimination Henmon ('10) also found
a slightly larger variability among men than among women as
shown in Figure 30.

The author made a comparison of the range of abilities in history
and geography as measured by his tests in these subjects, and
found the following distributions:

History, 8th Grade
Percentages of boys and girls attaining the various scores

Scores: o-io n-20 21-30 31-40 41-50 51-60 61-70

Boys 4.2% 9.3% 15.3% 17.0% 13.2% 12.2% 11.5%

Girls 6.2% 22.7% 22.7% 16.4% 12.8% 9.4% 6.0%

Scores (continued): 71-80 81-90 91-100 loo-iio Total

Boys (continued) 9.3% 6.6% 2.1% 0.4% 288

Girls (continued) 2.3% 2.6% 0.9% 0.0% 352

Geography, 7th Grade

Scores: o-io 11-20 21-30 31-40 41-50 51-60 61-70 71-80

Boys 06% 2.8% 4-7% 8.4% 5-3% 12.2% 14.7% 14-7%

Girls 03% 1.6% 56% 10.0% 10.0% 12.5% 18.7% 18.7%

Scores (continued): 81-90 gi-ioo loi-iio 111-120 121-130 Total

Boys (continued) 10.3% 8.1% 9.1% 6.9% 2.8% 320

Girls (continued) 10.0% 8.7% 7.2% 5.9% 3.1% 322


The variability of boys in the case of history is somewhat larger
than that of the girls, whereas in the case of geography it is sub-
stantially the same.

ThornJike ('14, III, p. 195) has given the range of ages of boys
and girls in the third year of high schools in Chicago, Philadelphia,
New York, Detroit, Fall River, Los Angeles, Lowell, and Worcester
as follows:

Age 13 14 15 16 17 18 19 20 and over Total

Boys 7 92 594 1246 1203 572 103 67 3974

<^irls 4 73 562 1351 1289 554 120 34 3987

There are about twice as many boys as girls at cither 13 or 20
or over.

In support of the general belief that the range of general abilities
is wider in men than in women may also be cited the fact that in
the history of the world most of the great geniuses have been men,
and also the statistical fact that male idiots and criminals at the
other extreme of the distribution curve considerably outnum-
ber the female. The fact that the great geniuses of the world
have been men rather than women would accordingly be explained,
not on the basis of lack of opportunity, but mainly on the basis
of greater exceptional ability. The theory seems plausible but
has been proposed rather in advance of a convincingly wide range
of experimental data. If it is true, it would mean that according
to the perception test the one or two per cent most gifted individu-
als are men and the i or 2% least gifted individuals are also men,
that of the next 10 or 12% of most gifted individuals approximately
two-thirds would be men and one-third women, and likewise of
the next 10 or 12% least gifted individuals at the other extreme,
about two-thirds would be men and one-third women. For the
remainder of the distribution the number would be practically
identical. The facts should not be interpreted as implying that
men as a hile are superior to women, but would mean simply that
only the one or two exceptional persons in a hundred would be
superior to the most gifted women. The remaining 96 or 98%
would be largely identical.


Problem. In a certain obvious sense, the entire native equip-
ment of any human being is inherited. The various capacities and
the relative amounts of them with which a person starts in Hfe
are derived from the cells from which the individual originates.
The differences among these original cells, even when derived
from the same parent, are assumed to vary with regard to any
potentiality according to the normal distribution curve about
the central tendency of that particular parent. Stalks of corn
grown from seed taken from the same ear will vary considerably
from one another because the seeds themselves, even from the
same ear, are different, but yet they will vary around the general
type of the parent stalk. It is therefore obvious that children of
the same parents will not be absolutely alike but that they will
vary about the central tendency of their ancestors. The specific
problem is not: Are mental traits inherited? but rather: How much
do children of the same parents or ancestors resemble one another
in the amounts of different traits possessed, and in the manner
in which the various traits combine? To what extent are abilities
in school work inherited? To what extent are the wide ranges of

Online LibraryDaniel StarchEducational psychology → online text (page 5 of 41)