-_—

IRISH DAIGRANTS » THEIR INTELLIGENCE AS RATED BY TESTS
i , 7
Lawrence Kolb, Senior Surgeon and Hs. A, Vonderlehr, P. .« Surgeon,
, United States Public Health service

The work reported here is part of a research problem on the intelli-~-
gence Of persons who apply for permission to emigrate to the United States
from iuropean wis iia It was done because the law provides for the
exclusion of mental defectives and it was necessary to devise some more
_ accurate measures of defect than the standards established on American ,
people, ehiefly on sched children, provide.

The sobering influence of responsibility. has aleve eaves those
charged with the examination of immigrants from the error of assuming
that a test standard found valid for children or one group of adults
could be applied with justice to persons of any age and from different
enviroments. Allowance has always been made for the effect of educa tion
and environment on test results, but the allowance has never been

measured or definite, and one of the problems is to make it more definite.

a

Seops of Work

The general plan of the work in Dublin was the same as that already
rae J ned :
reported for Naples. Each case was studied not only from the standpoint
a ilne
of teat ability but as to schooling, reading ability, reading habits,

occupation, place of residence, size of home and size of family.

Ean

ale
ine

‘eae a hee
2a

A total of 788 persons were oxamined at the American Consulates in
Dublin, Cobh and Belfast. ‘The distribution was as follows:

Men Women
Dublin 328 347
Gobh 1 0tCO

Belfast 0 55

These people were applicants for immigration visas who appeared at
the consulates by appointment in order to undergo the medical and other
examinations necessary to secure a Visas The work was done in Dublin
between beater, 1928 and August, 101, and in Belfast and Cobh during
the last six months of this period. |
| The Belfast and Cobh work was done in order to get a comparison

between the people applying in ‘those cities and in Dublin, The Cobh
| cases were given the Ferguson Form Board and Kohs Plock Design tests
and the Belfast cases were given the Ferguson Form Board ‘ana Pintner
| shidinnbends tests. There was little difference between the test ability
of the Cobh and Dublin cases, The results are, therefore, combined in
one table to show what applicants from the Irish Pree State do with the
Ferguson and Kohs test. The Belfast cases tested higher and the results
are given in separate tables to show the approximate expectation of women
from northern Ireland. |

All of the tests used in Ireland were not given to the entire group
of 788 cases, ‘The Belfast and Cobh cases were given three tests of
proved value merely for the purpose of checking against the Dublin
gases, and some tests used at first in Dublin in the examination of 70

men and 70 women were discontinued later on and others substituted for
a Ato 7 i

ose

theme Also a group of Dublin cases was given the geometric figures and
drawing designs from memory test because some deviation from the standard
methods had been ‘ok Se giving these two tests to the general Groupe

: The cases were uniform as to education and environment throughout :
the three years during which the work was carried on, but this was not
depended on entirely to insure that seme selective factor did not
operate at times to vary the ability of the groups tested, All the
groups were given some test or tests that the general group received
and their comparative ability was checked against these results, ‘The
Ferguson Form Board test was used as the checking test for all groups
except the special ous for the geometric figures an drawing designs
from nenory tests. For this special group the checking test was the
ball and field test, wees :

Perenun who apply for visas at the consulates in Ireland do not
have great distances to travel, and for this reason fatigue does not
influenee test results. Apprehension over whether - not the visa will
be granted affects the ability of some applicants, and this must be born |
in mind in considering individual cases, but the effeet on the gencsral
group is so slight that it may be disregarded.

Selection of Cases for Examination

During the entire course of this study except a part of the last
year the mumber of persons applying for visas in Dublin ranged from
about 40 to 125 daily. Adults wore selected from among these at random
and given the special examination unless they happened to have more than
a common eahent education. The method used to insure a random selection
was to ¢all the name from the top of a list or take the person who
happened to be sitting on the end of a bench used for waiting applicants.
“40

There was, however, a slight selective influence in that during the
period when about 25 per cent.of these subjeets were examined, catia
excluded for illiteracy did not reach the medical section, and it
seneunet that 11 men and 10 women who were examined in Dublin had ayyear
er move of secondary sehoehing, "

For the purpose of this work a person with higher education was one
who went a year or more to a secondary school or college. Only about
three and one-half or four per cent of Irish applicants at Da&blin had
enuh eteentien ont Weber partial axelusion foun She specially examined
group was in part balanced by the exclusion of the few illiterates who
would have been examined if they had reached the medical section.

Persons with higher education wore not examined for two reasons.
First, they are ovviously ast defective and what they can do with tests |
has very little practical application to the otadine up of persons
suspected of defect. Second, it was desired to examine in different
Buropean countries, groups of adult applicants with a we or less
uniform é@duecational background so that valid comparisons could be made.

The method of selecting cases in Cobh and Belfast was the same
as in Dublin. :

The Teste |

The tests were taken from various sources, especially from the
Stanford Binet series, but experience had shown that some Irish immigrants
failed to do many of the simple Stanford tests and also that the verbal
and mathematical tests at or above the 12 year level were especially
difficult for nearly all of them. The Verbal difficulty was in a slight
-5-

ae 7 Pip Aeceg
a Pek A, paiane Su
Oe ae eh

measure due to the giving of the ‘tests 7 there is enough
difference betwoon the Irish and Ameri rie: Snglish language, especially

in the case of poorly educated irish, to cause one being questioned by

      

| an american to strain for the meaning of some words. This may be

misunderstood, as the special attention directed to catching the meaning

 

of words is siphoned away from the main point, which is to comprehend
the sentence. This introduces an uncertain factor into test performance
that 16 1s desirable to avoid. — , ,

Yor the reason above given, only one Stanford verbal test ae high
aé the l2«year level was given, and some of the others were modified
with the idea of bringing them more into line with the Irish environment.
After 70 men and 70 women had been examined, all the purely verbal
tests were discontinued and the Pintner Non~language test, the Porteus
Maze test and the Kohs Block Besign test were substituted for them.

The List of tests follows. |

(1) Simple questions of general information.

(2) Mutilated pictires (Stanford Binet, yoar 6)

(3) Writing from dictation (Stanford Binet alternate, year 8)

(4) Differences (Stanford Binet,year 7, modified for Ireland)

Gredit given for correct anawer if four of five are correct.
(a) Fiy and watterfly
(>) Cat and dog
(¢) Stone and ege
(da) Wood and glass
(e) House and stable
(5S) Similarities (Stanford Binet, year 8, modified for Ireland)
Gredit given if three of five are answered correctly
(a) Wood and coal
(b) Potatoes and onions
(c) Horses and cattle

(4) Bread and meat
(e) Gigarettes and pipes
~in-

(8) Comprehension, 2nd degree (Stanford Binet, year 6, modified
for ‘Treland). Credit given if two of three are answered correctly.
What mst you dos |

(a) Se is de websted hak oie weds 4 aaa
(bo) If you find that your house is on fire?
(s) if you are going some place end miss the train?

(7) Comprehension, Srd degree (Stanford Binet, year 8, modified
for Ireland). Credit given if wo or three are answered correctly.

What ie the thing for you to do:

{a} When you have broken something which ahold to someons else?

(b) When you are on your way to (work « school ~ church) and
notice that you are in danger of being late?

(ec) If someone hits you without meaning to do £t?

(8) Comprehension, 4th degree(Stanford Binet, year 10, modified
for Ireland). Credit given if two or three are answered eorrectly.

(a) What ought you. to say when someone asks your opinion about a
person you don't know very well?
(>) What ought you to do before beginning something very importhnt?
(e) se Son we judge a person more by what he does than by what
saya?

(9) Weaving of abstract worke (Stanford Binet, year 12, modified
for Ireland}. Credit given if three of five are answered correctly.

{a) Love
(b>) Justice
(¢) Revenge
(a) Pity
(e) Lazy

(10) Absurdities (Stanford Binet, year 10, modified for Ireland),

Credit given if four of five are correctly answered.

(a) "A man eaid: °2 know a road from my house to the city which is
- down hill all the way to the city and down hill all the way
: back home'’*,

(>) “An engine driver said that the more earriages he had on his
train the faster he could go. |

(c) "Yesterday the civic guards found the body of a girl cut into
18 pieces. ‘They believed that ske killed herself."

(4) “There was a railroad accident yesterday, but it was not very
serious. Only 48 people were killed.
oFe

(e) "A bicycle Pider, being thrown from his bicycle in an accident,

(12)
(12)

struck his head against a stone and was instantly killed.
They picked him up and carried him to a hospital, and they
do not think he will get well again."

Reading and report (Stanford Binet, year 10).

Arranging weights (Stanford Binet, year 9).

2 (15) Giving value of stemps (Stanford Binet alternate, year 9»
Irish stamps used).

(14)
(15)
(16)
(17)
(18)
(19)
(20)
(22)
(22)

(23)

(24)
(25)
(26)
(27)
(28)

Ball and field test (stanfora plies. 3 years 8 ond 12).
Identifying geometric figures (Mullans Learning test).
Counting 20 to 0 by ones (Stanford Binet, year 8).

Counting 20 to 0 by twos.

Copying diamond (Stanford Binet, year 7).

Drawing designs from memory (Stanford Binet, year 10).

Drawing swastika design from nenerte |

Naming day, date month ana yoar (Stanford Binet, year 9).
Naming days of week forward (Stanford Binet alternate, year 7).
Naming cays of week backward.

Nening months of year forward (stanford Bhaet alternate, year 9).
Naming months: fs year backward,

Repeating digits forward.

Repeating digits backward,

Definitions; (Stanford Binet, years 5 and 8, modified)
What is a:
Chair needle
Table Ho mamsmemee SOON
BOAt Wcxmmmmunee OTe
SOldier «-nm—.-<«§ balloon

 

 
Sd

(29) Opposites (three samples given + man, good, day)
What is the opposite of:

 

 

Boy long
Gold : high
sick poor

 

(30} Additions.

(31) Subtractions,. ?

(32) Gube imitation test (fullans moves).

(33) Cube imitation test (Pintner modification).

(54) Mare and Foal test (original).

(85) Gwyn triangle test,

(36) Healy construction test A.

(37) Healy construction test Be

(38) Ferguson Form Board test.

(39) Pintner Nonelanguage test.

(40) Porteus Maze test.

(41) Kohs Block besign test.

No description of any of the atanfora Binet test will be given here
because they are well known to all who are interested in intelligence
testing. The seme rule will be followed in the case of other comnonly
used tests that appear in this series.

: All tests were given and scored by the standard methods as deserived
in "Termanis Seamer it the Intelligence” and elsewhere, except as
folkows and as indicated in the text from time to time:

Kohs block sedaus test: Three trial designs were given in each case :

before the test proper was started.
-9-

Pintacr Meenlangeege test: Explanations were given in minute detail
for each of the six divisions of this test so there could be no misunder-
 gtanmding as to what was required.

A lead pencil was used with all tests shauna pen or pencil was

required.
Edueation and Environment

Experience with adult domi grants has shown that uehooking and
environment influence their test ability, capecially the ability 80 perform
certain kinds of Neotine The school history and reading habits as well
as the environment of each subject and of the group as a whole were,

: therefore, studied. |

 Bwery child in Ireland 19 supposed to go to school and all exeept
“one in this group of 788 claimed to have attended school for more than
two yoarss The one exception, a young woman, attended for two years and
eould sign her name, but could not oteis Another young woman had just
learned to reads All of the others 41d better, but some wend poorly.

Seme Irish ehildren start school when three or four years of age;
some go ten years or more and do not get beyond the third grade. The
poor progress that they make is not due entirely te lack of ability to
learn and to inefficien! schools, A large proportion, capesialiy in the
rural districts, attend very irregularly because of work at home, the
indifference of parents, or because the nearest school is several miles
distant from their homes. cS

The distance of the sehools wan apparently not the important factor
with the cases reported here, as some from cities and toons as well as _
many who lived close to rural schools did not attend regularly.
@l0-

the reading ability of each subject was tested by having thom read
the paragraph given on the Reading and Report fest, Stanford Binet seale,
year 10, as given belowe

“New Yorke September Sth» = A fire last night burned three houses
near the center of the city. It took some time to put it out. ‘The less
was fifty thousand dollars, and seventeen families lost their homes, In
saving e girl who was asleep in bed, a fireman wae burned on the hands,”

This was scored as follows: ,

‘Reads well =- Time no more than 15 seconds regardless of mistakes.
Time from 16 toe 20 seconds inslueive with no more than
one mistake.

Reads fairly well « Time from 16 to 20 seconds inclusive and more
than one mistake,
Time 21 to 55 seconds provided there are no
more than three mistakes. 3

Reads poorly « Time between ‘21 and 36 seconds inclusive with more
than three mistakes,
Any time in excess of 35 seconis whether or not there
are mistakes,
Time as above was recorded on 842 women who were conseentivelysexa

 

ined with results as recorded below.
Dublin and Cobh cases 187 Belfast eases 55

Reads well 75e2 per cent Reads well 86.5 per cent
Reads fairly well 21.4 " . Reads fairly well 12.9 "% *

Reads poorly 54 * ° Reads poorly it .*. *
‘Reading habits as well as the amount and kind of schooling apparently
exert an influence on intelligence test ability. I+ is, of course,
obvious that reading habits are determined in part by native ability and
in a given case that does not read or reads very little, poor native
ability may be the cause, but where only about 30 por cent of a Broup of
presumably normal persons read daily, poor education and environment must

be largely responsible.
~ll-

Table 1 shows the schooling and reading habits of the various eroupé.
It is noted that nearly all of these people read some, but that only
about SO per cent of the entire group read daily. It is also of interest
that the “men ie slightly more schooling and do more reading than the
a :

(Take in Table 1)

fable 2 shows the size of homes, size of families and place of
resiéence (eity, small town, or country) of Irish 1
of family refers to me number of brothers and sisters including those
that have died, Parents and the subject himself are not included in the
‘Sumber. The size of home refers to the number of rooms without the
kitehen in the parents’ home, For the purpose of thie work a city isa ~
place with more than 2,500 inhabitants, a omall town is a place with
hetwoen 1,00 and 2,500 inhabitants. ‘The country (rural) is any place
with less than 1,000 inhabitants. — |

oe "(Take in Table 2)

A large proportion of Irish lemigrants work on farme, but many of
the farm workers live in or near Villages or towns and have considerable
contact with other people. Practically every one of them has access to
at least a weekly newspaper and the majority have relatives in the
United States. |

Wany of the jemigrants from Belfast come from near the Pree state
border and airfer very little from the south of Ireland emigrants. The
Belfast women had slightly less schooling than the Dublin ani Cobh women,
but got into a higher grade at school and read more. ‘They also lived in
slightly larger homes, The South of Ireland emigrant has an average of
6e6 brothers and sisters, making an average family of 0.6 members, ‘The
average home of the entire group has three roome and kitchen.
oli-

fest Results
Questions of General Knowledge.

A set of 36 simple questions of general information dealing mostly
with every day matters of common knowledge was — those given to 70
nangent 70 women examined in Dublin.

The questions are as follows:

Tells time by the ¢leck?
No. seconds in minute?
No. minutes in hour? :
Ne. hours in day?

Nos days in week?

No. months in year?

‘No. days in year? a
Approximate length (meter or year),(correct if cuessed with’ 6 6 inches).
Height of door?(correct if given within 10. inches).
Name seasons Of the year.

Identify three coins, Irish.
Identify three postage stamps, Irish,
Prom what is leather made?
Name three kinds of birds.
Name three kinds of animals,
Name four vegetables,
Name four fruits.
Name three kinds of trees.
When are potatoes planted?
When are potatoes dug?
Name threa rivers,
Name three cities.
Gelors of country's flag?
Capitol of country?
Name of the ruler of countrys
In what countries are the following Cities located:
London.
Varsaw
Paris
New York
Berlin
Dublin
Chisago
Prague
Rome : :
What-eeean is crossed when going to America?
lie

Percentile scales of correct responses to these 36 questions were
made. The median score for the men was 52 and for the women $1, The
questions most frequently missed were the location of the cities, Only
a few knew the location of Copenhagen, Prague or Warsaw, and many
missed Berlin.e Paris was sometimes located in England, London in Paris,
and Berlin in New York or America, The height of the door, the length
of the year, the capitol and ruler of the country, naming three rivers,
giving number of seconds in a mimute, number of minutes in an hour and
number of days in a year are questions ‘that were missed next after the
cities in order of frequency. ie do not believe that these questions
are or sufficient practical value to justify their use in testing the

intelligence of Irish immigrants. A few, like the height of the door,are
too indefinite, others are too easy and the fact that so many missed the
cities indicates poor bhnen tiek rather than lack of native intelligence.

Table 3 shows the percentage we persons whe succeeded on 16 of the
Stanford Binet tests, These tests were given end scored according to
Terman's method, As previously stated, some of them were modified to
suit the Irish environment, and others were changed, The modifications
made no appreciable difference, but the changes probably afteet the
results somewhat and they will be deseribed,.

The important changes are as follows:

Differences (stanfora Binet, year 7). Two questions were added to
those used by Terman and eredit is given for pasaing the test if four
of the five questions instead of two of three, as in the original series,
are answered sorrectly. The Terman questions and method would have
given 100 per cent of passes for this test. The question that caused
the test as given by us to be harder was the difference between a cat
and d0g6«
wie

The form of the question was, "What is the difference between a fly
ani a butterfly?", ete. The question with percentage of correct responses
are given below. :

Males Females

(Fly and butterfly 92.9 82.9

Stanford (Stone and egg 100.0 94.5
(ood and glass 92.9 ——- 9866

: (House and stable 9507 100.0
Additional (cat and dog S209 30.0

| Similarities (Stanford Binet, year 8). ‘The three last questions in
the Stanford Binet series were replaced by four new questions and credit
is given for passing the test if three of the five, instead of two of
the Zour as in the original series, are answered correctly, It is
believed that the @hanges made the test easier, The form of the
questions was, “In what way are wood and eoal alike?" etc.
‘The questions wh pereentages of correct responses follow,

Males Females

Stanfora - Wood and coal a 97 el  91e4
_————— onions 98.6 94.3

Horses and cattle 90.0. 90.0

Substitutions’ (Bread and meat 94.3 9403

, Bee Ps ae sGiqavettes end pipes 9666 97% el

Meaning of abstract words. Two of the words in the Stanford series,
charity and envy were replaced by two others, love and lazy. it is
believed that the change made the test easier. The words with nicaan tages
of correct responses follow: | |

Males ' Females

(Justice 60.0 $85.7

Stanford (Revenge 7269 47el
(Pity 6547 68,8

Substitutions (eve | 61.4 70.0 —
(Lazy T4e5 6547

(Take in Table 3)

+> peel
o1i-

Seventy per cont or more of adult Irish immigrants do these tests
up to and including the nine year Stanford level, but both men and
women fail on two of the four ten year level testa. The women also
fail badly on the 12 year level meaning of abstract words, but an
acceptable number of men pass. These sousta confirm a general observa~
tion of those who\ have examined incendie Irish immigrants, that they
do the Stanford tests fairly well up to the nine year level and then
begin to fail rather badly except in repeating digita, Thare axe,
however, other seattering successes at higher levels, and the average .
Irish immigrant will make a test age of about 1} years on the modified ©
Stanford tests. ae | ,

‘Drawing designs from memory is an especially hard test for these
people. It was given by Terman's method with only ten seconds exposure,
but the necessity of carefully looking at both desizns was pee Bee
Another group was given this test with 20 seconds exposure and in
addition to the usual instructions the examiner nce his finger .
around each design and ¢slled attention to the form by saying “this is
the way it goes". The method resulted in inereasing the successes of
the men by SO per cent, ana of the women by 100 per cent (see figures
under Swastika design), but the women still failed to reach the nine
year level in accordance with Terman's requirement that for a test to
be put at a certain age level 65 per cent or more of persons of that
age should pass it. |

Performance of this test is apparently influenced a great deal by
schooling, but that form schooling is not the only factor is shown
by the fact that the men with slightly less s¢chooling than the women

do such better with the test. The environment of the Irish males is
ol6-

probably more favorable to the development of faculties necessary to
the performance of it. ‘That sex itself is not responsible for the
difference is shown by the results in Norway. The women there did as
well as the men. :

| In Norway, England and Germany more then 75 por gent of all
groups succeeded with the drawing denigne from memory test. In Italy
the performance Was not as good as in Ireland,

The Irish men did better than the Irish women on 14 of the 16
tests in Table 5. ‘This is of especial interest because the women had
slightly more schooling. Formal schooling is, therefore, eliminated
as a cause of the guperior male performance and the question arises
whether the male environment or innate superiority of the male sex is

responsible for it. We are inelined to believe that after the school
| years, Irish males have in general a more stimulating environment than
females and the difference in test ability noted here and in the tests
that follow is environmental. |

Table 4 shows the percentage of persons who succeeded in five
simple tests, ‘The two tests, naming days of the week and months of
the year forward, were given and scored by the Stanford method, using
both cheeks and time. One mistake was allowed in naming the montisof
the year backwards |

These teste have been used a great deal in the examination of
imuigrants beeause they in easily given and call for a type of inform-
tion that is within the reach of every adult. The results here seem to
indicate that they are too easy, but in a group of £11iterate women in
another country only $2 per cent said the days or the week backward and
only 45 per cent said the months of the year backward.

a (Take in Table 4)
#lL7=
The Swastika Design

The Swastika design was exposed on a eard for 20 seconds after
which the card was taken away and the subjcet required to draw the |
design from memory. This teat was given to 70 men and 70 women
demoed ta tely after the Stanford Binet designs (modified). The Swastika
is more difficult to draw. ‘The results are of interest chiefly in
showing that the men are superior to the women on this as on other
tests. The percentage succeeding on the opantord Binet (mo@afied ) and
the Swastika are shown together for comparative purposes. |
; | ‘Males Females
‘Stanford Binet designs

Traced with finger and exposed 20 seconds 8741 64.5
Swastika design Ee : °
Bxposed 20 seconds 77 ol 571

Swastika design

 

ee

 
-1S<
Definitions

Ten words wore selected, three from the Stanford Binet five year
level, two from the Stanford Binet eight year level and five from other
‘sources. ‘The Stanford method was followed in asking the meaning of the
words and in scoring the replies according to whether the definitions
were given in terms superior to Use, in terms of use or failure,

The words together with the percentage of subjects answering in
terme of use, superior to use or failure are given in Table 5 and the
percentile distribution of answers in terms of use and supsrior to use
is given in Table 6. oe. |

(Take in Tables 6 and 6)

The ability to define in terns superior to use ‘. plaeed 7
Terman at the eight year level and at Least 65 per cent of his ‘subjects
suceueted with 50 per eent of his four words. it is of interest to
note that only about 25 per cent of male and female adults in thie
series of eases succeded in so defining 50 per cent of ton words. Their
failure illustrates: the difficulty that adults with a rather poor
"education and who a9 not read regularly havewith sueh tests as defining

a : a

Opposites
fen simple words were selected for this teste They were;

Three practice wartt (am good and day) were used to explain to the

subject what was required of him. The technique was as follows: "What
is the opposite of man? ‘that word has the opposite meaning to the word
Ee

~10-

man", If the question was answered correctly, good and dog were given
in the same way. In the ease of failure on any of the three words the
examiner supplied the correet answer and explained it, After this the
ten words were given without further explanation. The percentage of
eorrect responses for each worlh is een in Table 7 and the percentile
scores in Table &, | :
(Take in Tables 7 and @)
The ete maintained their position ahead of the women on the opposites

test, and it is of interest that ten per cent of the women failed to

give the correct response to more than six of the words.
Ball ané@ Field Test

The Ball and Field test is treated separately because it was given
to a larger number of persons than the other Stanford tests and it is zg
desired to make some comparison with it and a few other tests between
age and occupation eens The test was given to the group of 70 males
and 70 femles that @ook the other Stanford tests and to a special ¢ roup
that was given the geometric figures and Drawing Designs from Memory
teste | The latter group made ono+tenth of a grade leas in school, a
aifforence too small to have any effect on test results, The enviven-
ment was also the same and there was no variation in tost peiak wi
between thon. The results on the Ball and Field test for the total
of thermen and women and for subdivisions of the total groups are

presented in Table 9,
#200

Neither sex did this test adequately at the 12 year level. Only
20.7 per cent of the women succeeded and there is no satisfactory
‘explanation why the men with less schooling-should de so mich better,
It is possible that their environment may have given them greater
familiarity with the type of problem involved. ‘the only alternative to
this is that men are superior to women in the kind of ability required
for the solution of the test. ee /

The better performance of the unskilled workers as compared with
the skilled was probably an accident due to the small numbers in the
aan

A skilled worker for the purpose of this article is a clerical
worker or one who follows some trade or profession. |
| (Take in Table 9)
Geometric Figures Test (Mullane Learning Teat)!2)

 

- Mentality of the Arriving Alien. Publie Health Pulletin 80, 1917.
. Ppe 45 * aDe ;

 

This test was taken by the original 70 memaand 70 women and by'a

special group similar to them in all respects, 2

‘The material for the geometric figures test consists of a card
with 20 geanetric figures on one side and three on the other. ‘The
subject was allowed to look at the 20 figures momentarily, the three
Cigures were then exposed and the examiner daid, "Look at theee three
figures carefully, they are among those on the other side, when I turn
the card over you must find them there". The card was turned over ro
ten seconds and the examiner onid, "Point out the three figures". If the
subject succeeded the test was discontinued, if he failed, a second,

and if necessary, third trfal was given with exposure of the three
~21-

figures for ten seconds on each trial. ‘The problem was to find all |

three figures on the same trial and the card was turned for the next

trial as soon as a mistake was made.

The original group of 70 men and 70 women are given separately for

the Geometric figure test.

The original group did slightly better and

the men of both groups did decidedly better than the women.

“The results are given in Table 10.

(Take in table 10)

Repeating Digist Forward and Backward

This test is used at several different age levels in the Stanford

Binet scale. It has been used extensively in the examination of

immigrants because it is easy to give ‘and interpret and has been found

to be as reliable as any single tost of Stanford seale.

The series of digits given below was used and the rate of

pronouncing them was approximately seven in five seconda.  Suecess

with one line in o series of three was counted as success for the number
of digite in that line and the next higher series was immediately given
The test was stopped after four lines in succession were failed,

Repeating digits:

Porward  Beekward
641 374859 92 471952
$52 621746 387 §8S294
857 #276859 629 752638
4739 92685164 6528 4162503
2654 7251895 4837 3826475
7261 %857201 8629 ) 9452837
31750 486. 72534696 31879
42837 40853762 #69482
98176

82795482

52961
Roe

The percentile seores for number of digits repeated are given |
"im Table 11. | |
(Take in Table 11)
Irish immigrants repeat digits better than they m rform other tests
except certain construction tests. Sixty per cent of both men and women
repeat seven forward. This pereentage almost satisfies Terman's —
require:ent for the 14 year level. Sixty per cent of the men repeated
- five digits backward, which percentage almost satisfies the requirement
for the 12 year level. 3
We have observed that English speaking immigrants repeat digits
pilasivets better than iumigrants speaking other languages, and it is
the only test in which the Irish performance equals or approaches the
Norwegian or German performance.

Computation Testa « Additions and Sub teactions

The computations listed below were given in regular order
beginning with the easiest, Yor the additions the form of the question
was, "How many are 6 + 7", or "4f you add 6 and ? together how many 7
does it make"? Yor the subfractions the form was, "{f you take 7 away
from 20, how many are left"? In the case of the two problems 10-(2+4)
and 20+(S+4) the form was, “If you add 2 and 4 together and take the
sume away from 10, how many are left"? The percentile scores for

problems correctly solved are given in Table 12.
E
E
g
g
g

6+7 16 = 4
12 + 13 20 «7
15 + 16 32 = §
2+3+4 25 « 12
o+¢ 5+ 8 22 « 5
6+ 8+ 9 ; 23 «+ &
12 +13 + 10 50 + 11
B+ea+6 10 = (2 + 4)
iv?*iS+0+ 4 20 =~ (3 + 4)

The hardest problem was 17+ 13 + 9 + 4. More than 70 per cent
ef both males and fonnles got each of the other 17 problens correct.
‘Additions and subtractions are used more than any other tests in the
primary examination of imaigrants to find whether or not they are so
backward as to yountes. a thorough examination to determine their status
as to mental deficiency. If the anawers to such sums as 6 + 7, 12 ¢ 18,
15 + 16 are not prompt and accurate, the case is suspicious, but most
“often turns out not to be defective. One person in the series of 140
eases that took these tests failed 6+ 7, and only three per cent failed
the other two sums, but only one was found to be mentally defective.

The ability to perform even such simple computations as are given
here depends largely on schooling and the environment that follows the
period of achooling. ‘This is shown by the table below giving the
percentage who sueceeded on one of the additions and subtractions in
Dublin and Naphes together with the schooling of the groups tested.
The dit Peruneee are believed to be due almost entirely to schooling.

 

 

 
onke=

Percentage Succeeding on the Two Computation Tests,

 

 

Males Females
Dublin Naples Dublin Naples
Years sehooling 7.5 Sed 8.7 3,5
Grade reached Sed $5 4e5 Se8
Number of cases 70 131 70 154

Per cent of success
17 *+13+90+4 67.1 55.5 44,3 29 2
50 + 11 90.0 87:8 let  Sde2

More difficult arithmetical problems like the arithmetical
reasoning test, problem of the enclosed boxes test ani the ingenuity
test in the Standard 14 year, 16 year and 18 year series are too

Aiffioult for Irish Free State and Italian imnigrante and in Non
Tnghish speaking erentwries the language factor militetes against their
eons | ie | |

The simple computations used here together with some others were
given to SO of the first subjects examined by one of us in Oslo. They
wore then discontinued because all the subjects did all the problems
ouverts. The same type of performance holds good for prospective
imaigrants examined in Germany and England.

_. (Take in Table 12)

All of the misgelleneous tests that have been described are useful
in the practical work of examining iemigrants, but none is significant
when standing alone, They have, therefore, been combined into two
point scales in order to increase their value by expressing the scores
made on them as one figure for each seale. 3 7

Nineteen Stanford Binet tests constitute stale No. 1 and 10 | |
miscellancous tests ineluding two Stanford dow, drawing designs from
memory and definitions, that were so far changed as to lose their value

as Stanford tests, constitute Table 2,
The Stanford tests hive been put into a separate scale because they
have a recognized ombined value and also so that they may be used either
in the point seale or as a part of the full Stanford series of tests scared
by the regular age level mothod. | :

The only reason for making a point seale out of these tests is that,
as given by us, with practically all of the higher tests omitted, they
- Gannot be accurately seorred by the age level method. ‘The median test
age of our group on the Stanford scale would, we think, be about 12 years,
and, as previously stated, our Stanford tests with the exeeption of
repeating digits and the meaning of abstract words do not earry above the
ten year level. |

in construeting the seales, value has sehe given to each test on the
‘pants of what the women did, begause their performances on teats varies
more from standard measures than the performance of =, and we em

a that any increased accuracy that may result from weighting the scores

should be applied to them, |

4 store based on either the men or women or on the two groups combined
would give satisfactory comparative results » but so many factors influence
the test performance of adult immigrants that no unit of mines ean
accurately express the intelligence of any groupe A measure to be at all
just must be comparative and comparisons are obtained Sereneh —
the resulte by the percentile method. |

Some weightine of the scores is necessary in order to get into the
atales sueh useful tests as repeating digits, additions, geometric figures,
etc, without unduly stressing the value of ‘ines teste as would be done _
if each addition oun digit® repeated was given a value of one and so
on with the other tests. |
 

The seales and the method of constructing them are given in
Gopendia 1, ant Vas pereenttle ebevee of ‘he enh Oak wenne Oh cee ale
are shown in Table 13. ‘The maximum score possible on scale 1 is 33 and
on seale 2 is 85.

Tod bo bbch f é ie

The Cube Test (Pintner Modification)

The Binet block cubes for the arranging weights test were used. The
examiner sat opposite the subject and gave the 12 movements devised by
Pintner. ‘The Pintner tae thoa *) was followed, but more definite instruetions

 

(1) Paintner, Ret The Standardiz ation of Knox's Cube Test: Psychological
Review, Vol. XXII, No» 5 (1915). pps 377=401e

were given. After the subject was told to “wateh carefully and do what

I do”, he was also told to"touch the bleck first that I touch first,

touch the block second that I touch second”, and 90 one If failure on

the first line appeared to be ae failure to enseretand the problem, it was
egunted a failure, but the line was given over again after the tuptmetions
were eebentete The test was continued until five “ six Lines were

failed one it was obvious that there would be no more sugcesses,

The is Pintner movements or lines for the cube test are as follows:

kst move 1254 7th move 1352436
2nd “ (12343 8th “ 14524
sma * = 2342 9th * Isle
4th " 1524 10th * 4145124
Sth * 44352 . lith * 182413
6th * 44ns 28th * 4142341 .

The percentile scores for the cube test are shown in Table 14.

Onde WAbLY
 28F=

The male median of six cerreect lines corresponds to a test age of ten

years and the female median of five correct lines corresponds to a test

age of seven, accordint to Pintner and Paterson's ™ norms, These medians

 

(1) Seale of Performance. Tests, Pintner and Paterson, p. 157.
are emerier to the Italian and inferior to the English, German and
Norwegian median as worked out by one of us on this test,

The 70 subjects first examined were given the cube test according to
the first six moves used by Mullen'®) which follows. |

 

(2) Public Health Bulletin / 90, %» Hs Mullan, 1917. ps 40.

 

ist move 1425 4th move 4251
Rnd * 1542 Sth * S24)
Sra” =: 1524 6th “ 2415

The median of both males and females by Mullan’s method was five
‘Zines correct. The Pintner modification is edicts as the range of
ability that may be expressed by the six Mullan moves is more limited.
The cube test has been one of the favorite tests for the onetinan
| tion of danigrentes The fact that the performance of some groups differs
widely from American norms does not detract from its value if the expecta-
tion of the groups tested is known and it is used with a battery of other
testa.

The Mare and Foal Test

The original test devised by Healy was used. It is important to bear
thie in mind in studying the scores as the test is now commonly given with
the omission of the four geometric pieces at the top of the picture making
£4 quater to parfdims ‘hia, the Pintnsr Meditionsten'?), wan wend by ne

 

(1) A Seale of Performance Tests, by Pintner and Patersone pe 26.

 
~28—
of us in the examination of immigrants in Italy. Therefore, the scores made
there should not be compared with the Irish scores.

The test was given with the pieces scattered beside the board at
random and withent allowing the subject to have a previous view of the
completed board. He was told to put the pieces in as quickly as possible. .
The time limit was dive minutes on this as on all other form board tests.
The test was given ened times in succession. If a subject failed he was
not shown the solution before starting the subsequent trial, but the
blocks already in place were dumped out and he was told to try again.
the pareentile scores for the first, second, and third trials are
_ @hown in Table 15. |
, (Take in Table 15)

The Gwyn Triangle Teat

This test was given immediately following the Mare and Poal teste
The pieces were placed beside the board at random, not in the usual way
with the four triangular blocks at the top, and three trials in
succession were given @s with the Mare and Foal test.

This test resembles the triangles portion of the Mare and Yoal test,
especially in that there is a similar space to be filled by similarly
shaped pieces. It follows that doing the Ware and Foal test three times
gave practice for the Triangle test. This, together with the fact that
one of the triangles was not turned over in the usual way made the
Triangle test much easier for these subjects and the results are not
Comparable with any published norms. It is of interest, however, that
in spite of the easy way in which the test was given to the Irish, a
(eh —

group of Norwegian women examined by one of us did better on the first
trial than the Irish women did on the third trial which was nearly their
sixth trial. The Irish first trial scores(approxinately their third trial
aecres) were, however, better than the German first trial SGOreSe

The results are shown in Table 16.

Healy Construction Tests A and B

The Healy Gonstruction test aang Healy Construction test 3 were

 

(1) A Manual of Individual Tests and Testing, Bronner, Healy, Lowe and
Shinberg. PDs 112115.

given in the order named immediately following the Gwyn Triangle test.
The Healy A was presented in the usual way with the five pieces arranged

irregularly at the sides and the Healy B was presented with the semilunar okies

space pointing away ffom the subject and the pleses arranged irregularly ae
at the side of the board so that no two pieces belonging together were in
juxtaposition. On the first trial of each of these tests the pleses were
removed out of sight of the subject so that he did not get a view of the
completed boards. If he failed to complete the cante he was not shown the
solution before starting the subsequent trial or trials, but the blocks
already in place were dumped out and he was asked to try againe In these
cases the subsequent trials were really a continuation of the first

‘trial. ‘This had no effect on the median scores, because there was such a
large proportion of tiret trial successes that the median for the subsequent
trials was not involved, but the method must be born in mind if the second
and third trial results on the lower pertentiles are compared with results
secured by one of us on Italian, German, tnglish and Norwegian iomigrants.
In these countries the subjects who failed were shown the solution before

starting another trial.
: ~ Sie
The Irish resudts are shown in Tables 17 and 18.
: (Take in Tables 17 and 18)
The first trial median of the men on the Healy A gives them a test

age of between 15 and 14 by Pinitner and Paterson's norns'*?, and between

 

(1) Seale of Performance Tests, Pintner and Paterson. pe. 123,

14 and 15 by the norms of Lowe, Shimberg and wooa'®) , By the latter

ee

(2) Further Standardization of Construction Tests A and B. Journal of
Applied Psychology, 1924. pp. 324-338.

“norms the men have a test age above 16 on the Healy B test.

The norms of Lowe, Shimberg and Wood for the Healy A and B tests were
established in Boston on persons of different races and nationalities, the
majority of whom were recruited from the ranks of juvenile delinquents
and it is probable that the groupewas inferior to groups upon whom American
norms are usually established. This may explain why these are the only
tests in this entire series on which the Irish males test at or near
normal.

The men maintain their superiority to the women on both tests, but
it is noted that the difference between them diminishes greatly on the
oecond twin) and alse’ Aiagppenrs on the thirds |

The discriminative value of the second and third trials on these two
tests depends on the intelligence of the person examined « For bright peovole
the teste, especially the Healy As lose practically all their discriminative _
‘value by the second traal and on the third trial it is doubtful whether
onyikins but the rapidity of motor reactions is tested. For duller people
tthe second and to a less extent the thira trial are valuable indications
of learning ability. This is especially wee in the case of those who

complete the tests, but perform poorly on the first trial or who fail and
 

are shown the solution before starting subsequent trials, Under these
Gircumstances a very good second trial performance is of more significance
than a first trial fat luve, but continued failure or poor performance on
the second and third trials is an indication of low intelligence.

The fact that some narmal people fail while some defe¢tives complete
one or both of the tests on the first trial indicates that the results
should be read with @aution.

Tests of Large Seope

The tests in the series used in Ireland that are large enough in
scope to insure that the performance of them has considerable significance
when standing alone are the Ferguson Form Board test, the Pintner Non-

language test, the Porteus Maze test and the Koh's Block Design test.
The Ferguson Form Board Test

The Ferguson Form Board test is the most reliable single test that
‘has been used to examine immigrants, because it can be used without
language and is so constructed that accident does not influence the results
to any great degree. “Defective people do not score well on this test and
& poor score always means a poor mind, but a poor score does not necessarily
mean defect as the results in certain countries show, The results must be
read along with the results on other tests. ? |

The test consisting of six graded form boards each prawiinies six

spaces filled by blocks was given an@ scored sasutitne to the sethod in.

use at the Judge Baker Foundation'?? , The seoring in use there, referred

 

(1) A Manual of Individual Mental Tests and ‘Testing. pp. 126-127, Bronner,
Healy, Lowe and Shimberg,.

 
-32-

to here as the Shimberg scoring, is supplemented in this work by the
original Perguson scoring and by a percentile table showing the number
of boards completed, The Shisberg scoring (see Appendix %) is more
Widely used and we believe it is more satisfactory than the Zerguson
scoring. The Shimberg scoring does not d@iseriminate between the upper
ranges of intelligence in a group of bright people, because a large
2 proportion of them make the maximum score, 60 points, By the Ferguson
scoring the maximum score, 50 points, is practically hemenatade to reach

: and ita discriminates somewhat but not sufficiently between different
grades of higher Suvelidounce,

The Shimberg seore and age equivalents according to the Meehan
Shimberg male norms follows: ae

 

Age 9 whe hs MM BH
Points 12 18 2 30 36 42 48 &

 

 

The test is not standardized for age below the nine and ato ve the
16 year lémels, but two months are allowed for each point euteoen 12 and
4. In thie article the same ratio is followed in estimating test ages
or soreene who make more than 34 or less than 12 points, A person who
naked only six points is eredited with eight years of mental age and so
on. This extension of the scheme upward and downward is not a proved
abeuséin siantentiaasion. but it is approxtna tiie so and is used here for
eonvenience, |

Ferguson'*) aid not standardize his results with age, but presented

 

(1) Ferguson, Gy O. (1920), A Series of vorm Boards, Journ. Exp. Psy.,
: 728s,

them by school grade. His medians from the first grade to the second year

high school ranged from 6 to 16 points. These two extremes of grades usually
~Sde0

correspond aoe 6 and 15 and this should be borne in mind in reading the
Ferguson score table given here.

The pareentile seores as shown in Table 19 are given for the Dublin
and Cobh men and women, the Belfast women, and for susmbdivisions of the
Dublin and Cobh men and women according to age, sex and ¢coupation.

| | (Take in Table 19) |

The median score 36 made w the general male group is equivalent to a
test age of 15, while the median for the women is equivalent to a test age
of 10 years and 10 months. ‘The women have more schooling, so the difference
in favor of the males must be either an aneixemmentel or natural sex
difference. The skilled men with a test age of 13 vue ami 10 months are
: superior to any other group. They test one year and four months higher than.
the farmers. the difference is probably due in part to the selective
@afluence of occupation and in part to the stimulating influence of skilled
occupations on wutteranaiins test ability. It is noted (Table 2) that 54.5 ;
per cent of the skilled live in cities. | |

The male 55 « 44 age group tests three and one-half years lower than
the 25 - 34 age group. More than 84 per cent of both groups live in the
country and practically the same proportion of both groups, 13.5 per cent
against 12 per cent, were skilied workers, This percentage rules out
differences in residence and occupation as causative factors, but the 35 « 44
age group lived in slightly amaller homes and had less schooling than the
younger groupe There may then have been an intangible environmental
difference both selective a atimalating, In any event it is obvious
that the 11 years difference in median age vetween the two groups could
not of itself cause the three and one-half years difference in test

ability. In this connection it is noted that there is little difference |

 
wGhe

between similar female age groups (See both Shimberg and Ferguson tering)
The 45 + 60 age group tests lower, but their munbers are too small to be

of definite significance. _ |

: The differences between the groups is doubtless due to various factors
and they suggest thatconclusions from test results in indivicual cases
should be drawn with caution, but the fact that the ten percentile of the
best group here, the skilled workers, is equivalent to a test age of only
nine years indicates that the group as a whole is dull. They test decidedly
lower than similar groups in some other Hurepean countries, An examination
of the table for boards completed furnishes an illuminating contrant to the
Norwegian results. Out of 363 an and women of the same relative social
‘goalie examined in Oslo only four fated comple toy em the six boards.
Winety~nine per cent of Norwegians and 15 per cent of the Irish completed

six boards.
‘Pintner Non-language Test

This test consists of six sections that call on five different types
of mental fesponse for solution, A pencil is used, but the only formal
education required is the ability to write the digits. Experience with
immigrants has shown, however, that near illiterates are at a disadvantage
in attexpting to solve some of the problems presented by the Pintner test.
ALL of the Irish group who took it ee been to school and could read and
write and all but two men and two women claimed to indulge in more or less
_ reading. | |

The total number of points that it is possible ‘ make on the Pintner
Non«language teat is 601. | |

The Irish results are shown in Table 20.

(Take in Table 20)
“S55

The Dublin groups maintain their relative standing with the men ahead
of the women, and the skilled men better than all others. The 25 - 34 age
groups are again the best age groups, It is of interest that the men
surpass the women although the women have more schooling, but the most |
striking thing about the results is that the Belfast women are decidedly
better than any croup of Dublin men, although the Dublin skilled men had
more schooling and got inte a higher grade in school.

The Belfast women make a higher test age @n the Pintner test, but the
Dublin men and all Dublin groups score higher on the Tobeion test. The

| table below is of interest. |

 

ERE ert a ENS AIEEE NAIA REIS ETE. Pit CT SIERO i ERR oe AS a Tt

 

 

Skilled workers (Beifast Females
ia, ne
oe Average years anahiie 8.9 8.5
Gane OSs Average grade reached : 6.9 Bd ee
| Ferguson Form Boards median test age ig yre. 10 mathe li yre. 2 mnths.
Pintner Non~language median test age 10 * 10 mC UE

 

The table suggests strangly that the male environment favors the
development of faculties necessary to perform the Yergusén test much waive
than it does the Pintner ee, The sane thing is shown by comparing the
Dublin males and females on the two tests. The test age of the Dublin
males is two years and two months higher than the females on the Ferguson
test, but only ten months higher on the Pintner test. The relative |
difference between the Pintner and Ferguson test performance of bright

| people is not great. Female immigrants, largely domestic servants and the 5
like, examined by one of us in Oslo made 15.86 years on the Pintner and
16.50 years on the Ferguson.
— B6~

The Pintner Non-language test is a valuable test for showing up poor :
ability in persons who have had schooling and of especial interest is the
fact that the best two problems, reversed drawings and picture reconstruction, —
@me always done very poorly by thon. Many of the Irish group could not get
the idea of the picture reconstruction problem even after several eaplesee
tions and demonstrations. From the standpoint of test age equivalents as
Pintner test apparently rates the Irish too low, but this is no
practical handicap to its use among them if the percentile method of scoring | : Z
is used. |

Porteus Maze Test

The graded series of mazes age 5 to 14 was used and the test given
according to the method deseribed by Porteus 47, This is a valuable test

(1) Guide to the Porteus Maze Test, by S. Ds Porteus, Vineland, MN. Y.,
fraining School, 1924. Pubs Now 25.

for work with imnigrants because of its non-language character. It is
supposed to be a test of temperament as well as of intelligence. Our
observation has been that bright people are handicapped somewhat by the
temperament factor, but that duller persons take their time and as a rule

_ go into closed roads only because of dullness, The very dull and illiterate
test lower than normal, many of them failing to score, mediocre people test
slightly higher than normal and bright people test lower than normal, because
the test doos not carry into high ranges of inteliigence and they have no

means of | compensating for a careless mistake.
The median seore of the Irish is higher on this than on any of the four
comprehensive tests, but it differs only slightly from the Ferguson median.
The ratio of difference between scores on the two tests holds true for most
of the subgroups throughout the entire percentile range, This differs from
results secured by one of us in Italy where the Porteus median was also
higher than the Verguson median, but in the lower per¢entiles of some of
the groups that were tested the Ferguson seores were higher than the ;
Porteus scores. | |

The Porteus Seat suite the temperament as well as the educational
background of Irish immigrants and it is fair to them.

The test age of the various groups arranged by the percentile method
4s shown in Table 21. |

| (Take in Table 22) — | :

The men test ‘i. years higher than the women on the Porteus test and
the 25 « 54 age arene wainteina its superiority to other age groupse The
farmers do as well as the best occupation groups They make one and one-
half years ~ ‘than they ao on the Ferguson test, and five years more than
on the Pintner Non-language test. The Porteus Sane test seems to be
especially suited to bring ont the full ability of Irish farmers. They
are not confident of themselves, so work slowly and avoid pitfalls.

‘The Irish test closer to other immigrant groups on the Porteus than
on most other ate The simplicity of the test, the limited opportunity
it gives to bright people and the temperament above referred to are offered
as explanations, The table below illustrates the points that have been
stressed.
 

 

Males * Females |
Norway Ireland Norway Ireland
Porteus Maze test age «45.0 = :18..5 15.0 11.5
Ferguson Form Board test age 16-66 13.0 16.5 11.83

Pintner Non-lancuage test age . 15.88 963 15.87 9.0

The number of Norwegian: cases ranged from 158 males on the Porteus
- Maze to 208 females on the Pintner Non-language test.

The Kohs Block Design Test

In the Kohs Block Design test’? a6 colored cubes one inch in diambter

 

(3) Tutelligence Measurement, by 3, Ce Kohs, 1912.
and all painted alike are used, ‘he subject makes colored designs with
these cubes similar to 17 graded patterns that are presented to him en
cardboard cards. |

| The test was given by Koh's method with the exception that all subjects
were allowed to do three practice designs, This would not affect the
results with subjects who are able to do the test well, but the extra |
‘ictnabetone and practice made it easier for those reported on here,

| Pifty-five of the women and 19 of the men who took the test were
examined at Cobh. Both the Cobh and Dublin cases were given the Ferguson
ont ont made equal scores on it. Both groups are similar in origin to
the groups that took all the other tests in Dublin, but it is noted that
they, especially the men, are eredited with slightly more schooling.

The results are shown in Table 22.

(fake in Table 22,
The median score of the men on the Kohs test is 386, test age 11.16 yeas
and the women 998, test age 10.25 years, It would seem that the preoceupa~ -
tion of women with colors and the adjustment of patterns and Aish in
connection with making and repair of clothing should give; them an occupa~
tional advantage with the Kohs test. If there was any such advantage it
served only to reduce the distance neta the performances of the sexes.
There were 19 skilled workers among the females, including students, clerks
and dresamakera. their neaian, ma 57, test age 11,08 years. It was rather
surprising that a large number of the women did not get beyond the ninth

design up to which point only four of the cubes Were used, The teat was

always discontinued after five designs in succession were misaed.

The 25 ~ 54 age group did not do better on this test ae on the others,
but the group is rather small and it happens that all the skilled workers
fell into the younger group. |

Because the Kohs results are so much lower than the American norms, the
results secured by one of us working with adult immigrants in two other
countries ave given in the table below. he 15 ~ 24 age group is used for
the Dublin and Naples females as they are more uniform in age and education —
than the general groups. | Bs

,  Kohs Block Design Test

 

 

 

Males te |
; Stuttgart Dublin Naples Stuttgart Dublin Naples
Number of cases § 60 8) oe os ye 50
Median score 81 38 ~ @h 74 22 23

Median test age 14.42 11.17 =613642 0S 15,92) = 10S‘
~A0u
Education, Environment and Sex

There are two outstanding findings in the resulte that have been
shown; the superiority of the men over the women, and the poor scores made
by both groups as Ccompsred with standardized scores and scores made by
some other immigrant @YOUDS.

Education must be excluded as a cause of the male seeentaeety. as the
women had slightly more schooling and better reading habits (Tables 1 and 2)
than the men. They also lived in slightly larger homes, There is no
question of the selection of males better than the average, Both sexes
Were selected at random; they were the type of immigrant that have come
from Ireland for years; practically all ef them had vela tives living in
the States and they originated in the same type of home environment, The
better showing ef the nen must, therefore, be eniaantet for by a natural
superiority of males over females or by a more stimulating male envi ron-
meht. |

It is well known that girls test as well as boys; this is suggestive
but not detaineive proof that women have as ints native test ability as
men, It is true, however, that the environment of the sexes is fairly
uniform in chilahooa, but diverges with advancing years so that from 15
years onward the males have, as a rule, a much more stimulating contact
with the outside world. By work, play and social contacts their minds are |
- G@eveloped in various directions denied to females and we believe such
environmental development is responsible for the superior test ability
of the Irish males. The difference noted between the sexes is then an
“dllustration of how environment may affect test ability. On the four
tests of larger scope, the Pintner Nonelanguage, Kohs, Ferguson and Porteus
the difference ranges from 9 months to 23 years with an average ‘difference
of 1.58 years.
 

=41-

General Ability of the Croup

wrbellg

Poor environment and poor education are not - responsible for
the poor performance that the various groups made on the bitte Their
general mental ability is below the average. The Stanford tests wore
done quite well by the mon, but these tests, except repeating digits, were
taken from Low age levels, ‘The use of most of them in another country was

discontinued because no subject failed to do them correctly.

There is no question of the Irish group having verbal or mathematical —

ability higher than the ability shown in the tests that were used. We
have tried the higher Stanford tests of this type on Irish immigrants and
have found that they could not do theme They do better, as do immigrant
groups in all acanakee: on the so-called spatial relations tests of the |
performance type than they do on any other type of test. Ye, therefore,
think that performance tests come nearer to expressing their general
mental ability than other tests, ‘Some groups of immigrants may have
 gpeetal onrak ant ma thema ties. ability, but such abilities do not
develop to any great extent in the common school groups beeause they are
not exercised to the same extent as is mechanical ability.
x Because poor education handieaps some immigrants in the performance
of Verbal tests and language handicaps others, the comparison of different
groups with one another ant of all the groups with American standards must
be made by non~language testse .
The Ferguson form Board test as performed by the males is believed
to give & txuer picture of the general intelligence of the Irish group then
any of the four tests of large scope. The median test age of the general
. group of auden oh this test was 15 years, The skilled male workers surely
are not handicapped by this test and their median test age on it is 13.83
we

years, Ten per cent of the general group of men have a test age 5,83 |
years or lower and 10 per cent of the skilled group have a test =a of nine
or lower on the Ferguson test.

: The Porteus Maze test gives both men and women a slightly higher test
age than the Ferguson and it may be that the scores on it represent more
accurately than the Yorqnius scores the true sientaly age of Irish immigrants,
but,as previously stated, the Porteus test seems to favor mediccre people
and to handioap the very dull and the bright. Nevertheless, the score
nade on it must be sabsthorse in judging any case.

Ten per cent of the males made 8.17 years or Soe on the Kohs and 7.12
years or less on the Pintner Non-language test. These are valuable tests
but in the practical application of them the south of Ireland immigm nt
must be culled more by the percentile scores than ow the test age scores.
These tests may Gall for a higher type of ability or for a different type
ef ability than the Ferguson and Porteus tests, but we have, for practical |
purposes, anenel that they present some special environmental handicap
to the group uneee consideration. It is of interest, however, that the
ten percentile of the Belfast females with no more education than the ;
Dublin males make 9.25 years on the Pintner test and the ten pereentile
ef a group of Norwegian males, mostly from rural districts, examined by
one of us makes 12,75 years on it and that the woret Norwegian in a group
of 376 males ain ten years,which is higher than the Irish median.

- When all things are considered, it seems apparent that the true
median men tal. age of the ereep reported here is not above /3 years. This
is not Guinan with the 15.15 years mefe by the White Draft Army on the
: Army tests'1), The Army tests were much harder than the Bereusoyy verted

 

(i wu "Memoirs, National Academy of aeiense, Vol. 15, De 790.

 
mee

tests, mathematical tests, and Sates of the type of the Pintner his
language upon which the immigrants do so badly were used, and the teats
were not given individually, as were these, but to groups so that one or
the subjects must have tested poorly because of distefling factors
inseparable from group testing. =

‘The Criterion and Measure of Defect

All of the groups reported here are helieved to have a true median
mental age of about 15 in spite of the faet that the women test lower on
all of the tests and the men test lower on most of them. Environmental
handicaps account for the differenee and they should be discounted in some
way so that all the tests can be used as no one test ean be depended upon
to be reliable in any given case,

The variations due to environment and education was overcome by one
ef us in the case of Italian imsigranta by adopting a standard group of
people and a eaahees test shown to be fair to that group. I> was then
assumed that ali groups Were equal in native ability to the standard
group and that the per¢entile scores on all tests meant the same in terms
of mental age as the stores on equivalent percentiles of the standard
teat.

The Ferguson Form Board test was adopted as the standard test for
the Itddians besause it seemed to be the fairest test for them. For the
same reason it is adopted as the standard test for Ireland. In both
places it gives more uniform results than any other test and a higher

seore than any test of large seope except the Porteus.
wth wo

The general group of men is adopted as the standard group in Ireland
as it was in Italy. In Italy the difference in median test age on the
Ferguson test wetueun this group and the illiterate women of the same age
was 4el years, In Ireland the difference in test age between the men and
women ts 2-18 years. In both eases the men and women are presumed to have
equal mental age in terme ot native mental ability and the seores of the
women are referred back to the scores of the wn to get the correct mental
avez that is, a mental age corrected for environment and education,

Likewise the percentilesseores of all the tests for all the groups are
referred paek to scores on eqnivalent percentiles of the men's Perguson
test score for the corset mental ABCs For instanee, the first trial seore
of the 20 percentile vomen en the Mealy B test is failure. This is veteusih
back to the 20 percentile men's Ferguson seore which is 17, mental age 9.83
“years. It fe inferred from this that a female Irish immigrant whe fails
on the Healy & may Yave a mental age of 9,85 years. A woman who makes a
‘geore of 13, seer age 8.53 years, on the Kohs test does not have a mental
age of 8.33: years, nan correct mental age is found by observing that
store 15 falls at the 20 percentile and thet the 20 percentile mon's
Ferguson score (shimberg) is 17, test age 9.83 years, The correet mental
age of this women is then 9,85 years, The same procedure is followed in
the casos of ali of the tests by observing from the tables what percentile
the subject's score falle on and by erediting him with a mental age at
leastagqusl to what ‘6. see percentile on the men's Ferguson ‘iat score

shown,

§
e
—45~

In order that low scores may be conveniently located, the percentile
scores from 10 to 0 of both men and women are given in Table 23 for the
Ferguson test, Ferguson Boards completed, Pintner Nonelanguage test,
Porteus Maze test, Seale No, 1, and Seale No. 2, and the gquivalent mental
age of each percentile according to the men's Ferguson test wee is given
in the margin. The low scores for the women only are given for the Kohs
test, because too few men took the Kohs to permit a separation of scores
below the 10 percentile. ae

| (Take in Table 23)

Finding the corrected mental age does not yet settle the question

of who under the imnigration law is aefective. Mental defeet is not a well
defined concept. Ton per cent of the people of this group have a corrected

| mental age slightly below nine years and some people in institutions for
the fecbleminded in this country test as well or better, Nevertheless,

ten per cent of the Irish group are not defective, 3

The tables given here refer to groups from average environments and
with a given average amount of schooling. In judging individuals, ;
allowances must be made for deviations from these averages because
environment and eheecSion eat thiols influence test ability on all kinds
of tests. It happens then that a defective person may test a year or
more higher than one who is not defective.

‘Only in very extreme cases ean the diagnosis of mental deficiency
be made on one test alone. It may be necessary to give to one Irish
immigrant whose intelligence status is in doubt all the tests listed here
as well as the reguier Stanford seale. It ie helpful to find hie median

percentile rank from the pergentile scores that he makes on the tests listed

 
~a6=

in Table 239 a3 was deseribed in another paper for Italian immigrants.
A As an exemple of the method, an Irish male immigrant who ean read and

write seores 11 = 2 ~ 98 ~ 9 « 17 = 12 and 1 on the Borguson, Ferguson

 

Bearcs completed, Pintner Non-language, Porteus Maze, Scale 1, Seale 8 alee a :

Kehs tests respectively. An examination of Tabel 23 shows that these.
seores fall at the follewing pereentiles, 10 + 2+6+284 23 = 0.6
The median of these igs 2, and the man ia mentally defectives he proof
of this is not altogether because his median falis so low as the two
percentile, but because on all the tests his scores are very low, The

usual ease will have a much higher score on one or two tests and this
often throws doubt on the cipenetes eis 5

Dull or poorly Lettered people scatter a ereatjon jatelligunts seta;
for this reason a “unifora rule of measure is likely to be misleading even
: if the subjects are aneasen tis? uniform as to environment and education,
poe if mistakes are to be avoided the test results must be considered
individually as well as collectively.

Based on our snperimes with Irish immigrants, some observations on
the significance of test results and the relative value of tests are given
below. Our Seale Nos 1 made up of 19 Stanford tests has no advantage
over the Stanford seale for use with Irish immigrants, ‘The seale is
anand for mewine, in what percentile certain dull persons fall, and this
is very important, but for general use, Terman's Stanford seale is more
satisfactory because higher tests are used and the scores mete on it enn

| be compared with standard scores.
fn

4 Stanford score of 10 years, Te2e 62.5, is not associated in Irish
adult imaigrants with mental deficiency, as we know it, and a Stanford
score of 9.5 years is seldom associated with it. In the rare cases in,
which the association occurs the individual comes from an curieeinna’:
better than thie averages ‘. Stafford score below nine is always suspicious
and a Stanford score below eight means edtae sae in ye all cases,

Qerth- Crlatrne Welt gore &
The Stanford scale a as reliable for use “— Irish immigren ts

 

as any test that we used, : } them, but errors |
would be made if it waco welaly Gabeniel upon. The results are influenced —
by education, but the tests at low age levels fit well with the kind of
education that these people have. |

- In judging the Ferguson Form Board results the number of poards
eompleted as well as the ostent seore should be considered, A Blow me
‘working person may complete four boards and make only a seore of four,
mental age seven years and eight months. It is doubtful whether any _ ee
person of the type considered here who completes four boards is defective. .
Table 1 shows thatten per cent of the men and 40 per cent of one group of
women completed aie three in the allotted five minutes for each board.
Nevertheless, it is a faet that seme defective people complete three
boards. this is merely another illustration of the tnreliability of single .
tests as exact measures of intelligence. If the seores are low on all
the tests the case is clear, and an Irish immigrant whe tests as low as”
eight years on all of them is defective. Sight years is slightly lower
than the four percentile men's Ferguson test score, which is used as a
atendard, wat 1¢ does not follow that four percent of Irish fasigrente

are mentally defective. Many will test as low as eight years on some of
ale
the tests, but few will test so low on all of thom. It is the median
test age that is important. If the median test age for all the teats is

eight or lower, the case is low enough to be considered defective unless
there is a high score nine and one«half years or more on the Stanford

? stale or Pintner Nonelanguage test or more than three Ferguson Boards

are completed.

A seore by an adult Irish p t+ of 175 (test age eight years and
three months) Fad to our minds proof that he is not mentally defective.

An examination of Table 19 shows that the 10 per@emtile Irish man makes a

score of 109, test age seven years and two months, and the 10 percentile

Ttish woman makes a score of 89 (test age six years and nine months)

on this test. These and higher scores on the Pintner Non~languagg test

are in our experience often associated with mental deficiency. Your per
cent of both moniand women have a test age below seven on the Pintner test.

These are very low scores, but when referred back to the four percentile

men's Ferguson seore it is seen that the correét mental age is eight
“years ‘and two months. Nevertheless, it is to be borne in mind that

though these people had been to sehool and could wead and write, they
did not do the test as well as the average Amertean seven year old ; public
school child. :

A mentally defective male Irish imsigrant may score as high as 1
years on the Porteus Maze test, but this does not often eeaet A
female aay score 6.5 or 7 years and not be considered defective. Such
Gases are, caret. very rare. This test is 80 easy that all scores
velows: ten are regarded with suspicion. Their significance depends @n
the schooling and reading ani writing ability of the subjects tested.
=6.0—

ott
ee onli te

y

s
x

ca test age of 865 on the Kohs Block Design test has not, in our
experience, been associated with mental deficiency in Irish immicrante,
Some females make below seven years and are not defective according to
our present conception of defect. It can be almost positively stated
that if an Irish fomale of the type reported on here carries the test
beyond the first nine designs she is not defective,
In View of the number of tests given to the Irish group, nothing

is to be gained by sino a special seale of the performance testa,
Mave and Foal, Gwyn Triangle, Healy Construction A, Healy Neisgtenntion a
and Pintner @ubes. If a decision cannot be reached without these testa,
a combined score for them would not help the selection, but observation
of the person taking them and the separate scores would be of assistance.
As previously stated, the learning ability ie revealed by the second
trial performance of the first four is important.

| To eumariz’ the general observations on the meaning of the test
results as applied ‘o thane enneisive immigrants, the variation in teat
ability between different groups, the low scores of all the groups as
compared with established undasie and the testing of many presumably
normal persons at or below the feebleminded level, suggests that for the
practical purposes of diarnesisc a wide variety of tests should + used

and the results interpreted with caution and conservatism.

Taek
=50~

ad

J 7 Summary and Conclusions

Seven hundred and thirty-three prospective imaigrants from the —
Irish Free State and 55 from Northern Ireland were given intelligence
tests in Dublin, Cobh and Belfast.
he age Of the subjects ranged from 15 to 45. They lived in home
with an average of approxima tely three rooms and kitehen, and more than
85 per cont were from rural districts, The average size of the families
from which they came wae approximately 9eb.

All subjects haé been to school for an average of about eight years
and all but one could road; the average grade reached was 3.9 for the
men and 4.5 for the women. ee

The 25 + 34 age group tested higher than other age groups and

- skilled workers tested higher than unskilled workers and farmers.

The Northern Ireland women had only two tests. On one, the Ferguson
Form Boards, they tested four months higher than the Free State ‘womens
On the other, the Pintner Non-language test, eae tested three years and
five months higher than the Free State women. 3

All groups tested low; the median test age of the general group of
men ranged from 9,64 years on the Pintner Nouslanguage teat to 13.5 on
the Porteus Maze test: The median test age of the general group of
women (Free state) ranged from 8.81 years On the Pintner Nonelanguage
test to 11,5 years on the Porteus Maze test.

On four testa of large seope the difference in test age between the

men and women ranged from 9 months to 245 years with an average of

1.58 years. ‘The men tested higher than the women although the women had
Slightly more schooling. ‘The superior performance of the men is judged
to have been caused by a more stimulating environment,
-5l-

All Free State groups except the skilled workers who are superior
kibimtd , 7
to the others are judged to have about equal native mental ability.

 

The true median mental age (the corrected mental age) of the Free
State group arrived at by making corrections for education and environe
ment is judged to be 1S years. The true median mental age of the lower
ten per cent of this group ranges from eight years and eight months te :
seven years and four months. . ne.

. The corrected mental age of all the Free State groups except the
skilled workers on all tests is found by adopting a standard group of :
subjects and a standard test that brings out the full mental ability of
the group. ‘The percentile rating of any group on any test is then
referred back to the score of the equivalent pereentile on the standard
test. The mental age indicated by this score is the corrected mental
age and it is usually higher than the test age.

The standard group is the general group of men and the standard
test is the Ferguson Yorm Board test. ,

<P
Any of the tests may give erroncous results in individual cases and

in order to avoid errors all of them should be used when mental deficiency

is suspected,