‘ ~ K":
.L“: ‘4'."‘1 .

'0’. I:
.73” 0‘1

.
.
u
o

it":

 

”.’l
.y‘)

I

.' " "
{I :i.

V

.\ 5‘
. n
o .
n-o
at».
I. ‘9-
. '6 It
(a
V .-
1

_‘ ,1 I
.
pf. I
.. o 9

......., a.
nu .
m. r.

.

a an.

wasus

 

——~—-—

c‘i
1!»:
L [B R A R 3‘
:‘flichiqumrc

,fl,‘ University {J

L

32"..-

V'm‘k

 

ITEM VERSUS CONFIGURAL ANALYSIS OF ACADEMIC ACTIVITY
PREFERENCES IN RElATION To VERBAL AND
QUANTITATIVE ABILITIES
By

Muhammad S. Sajid

A THESIS

Submitted to the College of Conmnmication Arts of
Michigan State University of Agriculture and
Applied Science in partial fulfiJJJnent of
the requirements for the degree of

MASTER OF ARTS

19 59

ACKNOWLEDGMENT

The writer feels indebted to Dr. Louis L. McQuitty
without whose help and guidance it would not have been
possible for him to undertake the study.

He also wishes to express his gratitude to Dr. H.
Kumata for his encouragement and valuable suggestions
throughout the study.

The researcher is grateful to Mr. James C. Lingoes
for his help regarding the programs for the Mistic,
without which it was not possible to complete the study
within the available time.

The author wishes to thank Dr. Albert E. Levak,
Training.Advisor, Ford Fbundation Pakistan Project, for
all mechanical facilities; and to Miss Wilma J}‘Wbodford
for her most sincere and continuous help and c00peration
in preparing long, tiring tables and in proofreading.

*Hi—fi-‘Wfi—X%

ii

 

 

{a

ITEM VERSUS CONFIGURAL ANMsI‘SIS OF ACADEMIC ACTIVITY
PREFERENCES IN RELATION TO VERBAL AND
QUANTITATIVE ABILITIES

BY

Muhammad S. Sajid

AN ABSTRACT

Submitted to the College of Communication Arts of
Michigan State University of Agriculture and
Applied Science in partial fulfillment of
the reqlirements for the degree of

MASTER OF ARTS

1959

{I
'1

Approved by . I ‘, . 4” I, 41.55 " 1" 7%!— a/ «L
, 7

ABSTRACT

Problem

The purpose of this study is to examine differences in patterns of
responses to the items of the.Academic.Activity'Preference Inventory by
the freshmen who scored high on verbal but lOW'On numerical.items of
the College Qualification Tests versus those who scored high on numerical
but low on verbal items of'the same tests at the time of admission to
Michigan State University, September, 1958.

The study also undertakes the comparison of the configural and the

item analytic results.

Reziew of Literature
Clinical psychology had two dissimilar heritages-dynamic psychology

and psychometric methods. In harmony with the latter it has stressed
objectivity; in sympathy with the former it has focused on patterns of
behaviour .

Clinicians faced serious problem when.patterns of responses were
neglected in favour of linear models by the psychometrists. Hence the
former turned to projective techniques in assessing configurations.

This move made the psychometrists aware of the seriousness of the situ-
ation and consequently they broadened the capabilities of their tradition
by Showing that configurations could be objectively assessed. Zubin,

Mechl, Gaier, Lee, McQuitty, etc., are some of the'pioneers in this field,

iv

who claim that the configural approach has unique predictive value
which item analytic approach lacks. Both these methods have been

applied and compared in this study.

Ergcedure

.A group of 82h freshmen of Michigan State University who had taken
both the College Qualification Tests and the.Academic.Activity Preference
Inventory in September, 1958, constituted the fpopulation‘ of this study.
The subjects were classified into two groups,.A and B, on the basis of
their verbal and numerical scores. Group.Aa consisted of 16h students
who had high verbal but low numerical scores; and group B had 176
students with numerical but low verbal scores. Each group was further
subdivided into Au A2, and B1, 132, respectively. The subgroups A1 and
Bl'were used as experimental sample and.A2, Bz'were treated as cross-
validation sample. The data were exposed to both the item analytic and

the configural methods.

ConcLusions and Recommendations
The configural results were better than the item analytic results,

but not at any significant level of confidence.

The study was restricted to the first forty items of the.Academic
Activity Preference Inventory which has 275 items. The prospective
researcher is advised to select sets of analytically suited and con—
figurally suited items out of these 275 items. This would put him in
a better position to see the correct picture of the relative merits of

the two methods.

. . _A
\ 7 , ,
h J \ 7
' s >( " ‘j ‘ - A r—
. . ‘ ,_— I 7 > ‘ I
0 ‘ ’ 7‘
I
I o
_ ,
V 7‘
> , 7 ,
Q
._. . A — L R b >
a L 7‘
.

F1

 

An experimental design of this kind stresses the necessity of a
theoretical approach toward the preparation of the configurally suited
items. This would be a great help to the researcher who spends a

great amount of time in selecting such items.

vi

I.
II.
III.
IV.

V.

VII.

TABLE OF CONTENTS

Page

INTRODUCTION
PRoBLEM
ASSIEPTION AND SCOPE or STUDY
BRIEF DESCRIPTION OF THE METHOD

DECRIPI‘ION OF HIE ETSOOOOOOCOO0.000000000000COOOOOOCOOOO

smYOOOOOOOCOOOO00.00.000.000.0...OOOOOOOOOOOOOOOOOOOOOOOO

SubjectS................................................

Items...................................................
Item MflYSiSooooococoa-o.cocoon...0.0000000000000000...
Agreement MYSiSooo0.000000000000000.no00000000000009.

WY Am CONCIDSIONSOOOOOOOOOOOOOOOOOOOOOOOOOOOOO0......

BEILIOGMYOOOOOOOOOO;OOOOOOCIOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOOO

APPmDmOOOO...0.0000COOOOOOOOOOOCCCCOOO0.0...OOOOOOOOOCOOOOOOOOO

Vii

12

E

17
17
18
21
29
33
36

38

n . 0
Q 0 R
o c o
O C a
i c t

TABLE
I.

II.

III.

IV.

V.

VI.

VII.

VIII.

IX.

I.

II.

XII.

LIST OF TABLES

Intercorrelation of Every Item with Every Other Item in
Group.A. (Subjects Having High Verbal and Low Numerical

ScoreS)..................................................

Intercorrelation of Every Item with Every Other Item in
Group B. (Subjects Having High Numerical and Low Verbal

ScoreS)..................................................

Matrix of Differences: Cell Entries of Table I Minus
cell Entries Of Table II...cocoacoco00.000000000000000...

Types Developed Through Linkage.Analysis from Table III
(Matrix 0f“DifferenCGS)00000000000000.0000.0000000000000.

Matrix Of Reciprocal PairsoOQOOOooooooooooooooooooooooooo

Sums and Averages in Descending:0rder of Columns of
Table III (Matrix Of Differences)........................

Matrix of Items of the Largest Column-Sum in Table III
(Matrix Of Differences)..................................

IMatrices of.All the Items Appearing in.E&ght Types.......

Matrix of Items Having Largest Entries in Their Columns
in Table III (Matrix Of,DiffereIlceS).QOOOO’OOOOOOOOOOOOOO

Two-x Two Tables of Thirty-Eight Items Showing Cell

Frequencies..............................................

Ranked ChisSquare Values for ThirtyrEight Items..........

Specimen Pattern of Response (Produced by Mistic)........

viii

Page

38
39

h0

h2

h3

h5
h8

h9

51

I N n
( '. o
, _ :
a A A
a ‘- 6

\§.\

v. t .
b 4 1
0 Q .

 

I. INTRODUCTION

The present study investigated a problem which has received scant
attention: the differences in patterns of responses between students
who score highly on verbal items but do poorly on numerical items versus
those who score highly on numerical items but do poorly on verbal items.

From the very beginning of man‘s serious intellectual efforts, to
understand human behaviour, both philosophically and scientifically,
there has been at least some concern with the significance of patterns
of responses, and one of the persistent theories has been that of
typology; By studying the works of the psychologists in any period from
the pre-Socratic to the present, it is quite common to run upon phrases
which deny the possibility of explaining wholes by'a study of their
constituent parts. Mach (13) supports this theory by an example that
the arrangement of lines in geometrical figures causes the emergence of
different totals which are reported as squares, rectangles, diamonds and
so on. This led him to resort to the doctrine of "sensations of Space,"
sensations which, while not pointing directly to the elements of the
original experience, must be taken jointly'with them if the tstructured
total! is to be explained. Etkin (ll) reports that the mumel and plant
kingdoms are classified in a manner which reflects that characteristics
have different predictive indicants depending on the combinations in

which they occur.

"Clinical psychologists have been surprisingly ahistorical" (15).
Little has been written about the development of clinical psychology.
In part, this neglect is due to the clinical psychologists being very
busy during and following the Wbrld War II. Young (15) remarked,
"Making history on every hand as we are, we have a notion that we some—
how have escaped historyu“ However, by tracing the history of clinical
psychology and by going back to the turn of the last century, it be-
comes evident that its origins are to be found in the dynamic and
psychometric traditions in psychology. The latter, one of the headwaters
from which clinical psychology sprang, was a part of the scientific
tradition of the nineteenth century and stressed objectivity; 'Whenever
a clinical psychologist insists upon objectivity and the need for
further research, he is, intentionally or otherwise, showing the influ-
ence of this tradition. Going from Galton through Binet and Terman, it
is evident that they always had a respect for quantitative measurement.
Similarly Cattell along with Thorndike and'Wbodworth stressed dealing
with individual differences by means of statistical analysis.

The other major source of influence (dynamic psychology) con-
tributing to the growth and development of clinical psychology was the
thinking and writing of James, Hall and their associates, also known as
the "Boston group." .Although they could in no way be labelled clinical
psychOIOgists, their thinking was much closer to clinical psychology and
to progressive psychiatry than was Titchenerts structural.point of View.

Their main interest was to understand human personality through the

 

 

 

 

patterns of his behaviour. The emphasis of modern clinical psycholo-
gist on patterns of subject‘s responses, in understanding his behaviour,
is an evidence of the influence of dynamic psychology (IS).

Louttit (5) states that the interest of the clinical psychologist
is in the subject considered as a physical, social and psychological
being in the matrix of his environment; and the understanding of the
individual depends upon the knowledge of the clinical.psych010gist of
the physical, emotional, educational, social and psychological factors,
related to the individual, as a.wh9l§,

.Allport (1) quotes that the clinical approach is absolutely
necessary for the investigation of personality as a whole, for a true
picture of personality cannot be pieced together. It is an organismic,
and not an additive, total.

To summarize, in the language of MCQuitty (7), "clinical.psychology
has two dissimilar heritages-dynamic psychology and psychometric
methods. In harmony with the latter it has stressed objectivity; in
sympathy with the former, it has focused on patterns of behaviour."

Clinical psychology encountered a serious problem, however, because
psychometrics tended to neglect patterns of responses in favour of
linear models. The clinicians realized that too much emphasis on psycho-
metrics restricted their discipline and that each individual clinical
psychologist should demonstrate to the bordering professional dis-
ciplines and to the lay public that clinical psychology had a useful

contribution far more valuable than psychometrics alone. Hence, the

 

clinicians accordingly turned to instruments such as projective tests,
that assisted in assessing configurations. The period in which pro-
jective methods were developed was pervaded by revolt against atomistic
tradition of the early experimental psychology. Atomistic research
began with the attempt to analyze psychological phenomena into elements.
Opposed to this viewpoint is one which has various names--global,
holistic, organismic or field theoretical. Lewints typologcal concepts,
Mlportts personalistic psychology, Murray‘s organismic theory and the
dynamic7approach of Maslow, differ somewhat in conceptualization, but
unite in emphasizing the importance of totality and wholeness of
personality and of patterns in understanding human behaviour (5,110.
Here clinical psychology has shown a willingless to sacrifice its birth-
right of objectivity to its interest in patterns.

However, as psychometrics was about to lose one of its most thriv-
ing, valuable and renowned offspring, it has broadened its perSpective
and capabilities by demonstrating that configurations can be objectively
assessed. Gaier and Lee ()4) point out that one of the more promising
trends in present day psychometric research is an increasing interest
in methods of evaluating patterns of test scores and test re5ponses.

In clinical, vocational, social and educational psychology, there is a
growing agreement of opinion that taking account of interrelationships
among test items will improve the efficiency of prediction. Zubin

asserts that total score may conceal as much as they reveal. A total

score may carry considerably less diagnostic significance than a direct

and detailed analysis of the responses p§;_§§, The authors (Gaier and
JLee) provide arguments that consideration of response configurations
will yield more fruitful results with higher degree of predictive
utility than obtainable by the traditional additive methods. At least
one important research conducted by McQuitty (6} on psychological well-
being has concluded that mental hospital patients differ from community
persons primarily in terms of their patterns of responses. He points
out that since the appearance of the Wbodworth Inventory during'WOrld
'War I, the psychologists have been trying to investigate a definite
problem: whether or not, on the basis of carefully constructed in-
ventories, they can classify accurately even such widely different
subjects as the mentally ill and the mentally healthy persons. The
investigators did not meet such success because of two uncontrolled
problem areas: (a) what inventory test items to be tried out, (b) what
method to be used in assigning tweightst to item responses for the
assessment of psychological'wellabeing.- The test constructor in this
field has greater difficulties than the experimentalist who has two
uncontrolled variables and does not know which one is responsible for
his results; whereas the former, instead of merely having two uncon-
trolled variables, has two uncontrolled classes of variables, and does
not know to which to attribute whatever success he has achieved.
McQuitty, since 1935, and more recently his students, have carried a
series of systematic studies of personality inventory items and methods

of weighting responses on them in the assessment of psychological

well-being.1 One of the conclusions that McQuitty reached is that the
mentally ill differ from mentally'healtny in response patterns (6).
This is an evidence in favour of the claim that configurations can be
objectively assessed, and this is the meeting ground of clinical
psychology‘s two dissimilar heritages——dynamic psychology and psycho-
metric methods.

Cattell (2) insists that psychologists should study the meaning
and effects of the total personality configuration rather than of more
levels in specific variables; and the importanoe of the one and indi~
visible total configuration cannot be overestimated. He criticizes
those techniques which specifically deal with effects of configurations
but relegate the pattern to intuitive assessment rather than to explicit
mathematical treatment. He proceeded further and developed hp and other
coefficients of pattern similarity; Cronbach and Gleser (3) also
developed methods of profile similarity. McQuitty (7) criticizes all
these highly developed pattern analytic methods such as those mentioned
above, for assessing profile configurations rather than patterns of
responses to individual items. "In the profile approaches responses to
individual items are used to yield total scores on several variables;
and the configurations are isolated in terms merely of patterns of

standings on scales, i.e., on linear continua. Thus, they are methods

 

1For other methods of personality assessments (e. g., T Method,
H Method, NH Method, MH Method, etc.) developed by McQuitty during his
long continuous research, see (6).

for studying data ordered to linear continua; and data that do not fit
are discarded."

Zhbin (16) has pointed out that such information may be lost in
thus allocating data to linear continua. IMeehl (12) has Shown that it
is theoretically possible for responses treated configurally to have
predictive efficiency which they lack when treated individually; For
instance, an objective history of vigorous athletic participation at
high school level, would argue in favour of masculinity in the male.
But such a history in a male of 35, without heterosexual experience,
living with his mother and 1sponsoring‘ boys* clubs, would give an
indication of the latent homoerotic component. Hence, patterns of
responses have unique predictive value. Meehl’s paradox, as he calls
it, is recognized by mathematicians. They take account of it in their
definitions of independence by stating that 1the propertyBo is said
to be completely independent of properties B,, B2,....Bn if two condi—
tions (necessary and sufficient) are satisfied: (i) B5 is independent
of every property Bl, B2,...Bn taken separately, and (ii) B0 is inde-
pendent of the logical product of every group of properties selected
out of B1, B;3,..Bn (7).

In short, in the field of personality measurements, recent
research indicates the possibility of getting higher validity by using
patterns of responses rather than total scores for prediction. In this
area, the "differential method" has been used often. It takes into
account summative individual differences. (Differential weights are

assigned to individual test items and a summation of scores on various

 

 

r 1' I'ull

 

items is used as a predictor of personality). But zubin (l6) feels
that this had not led to fruitful results. He advocates the "integral
method” which focuses its attention on similarity between individuals.
He is against the traditional assessment technique of personality
inventories, because he maintains that the pattern which produces the
score is itself more important than the summative score on the inventory.
An average does not serve the purpose in judging the individual because
it is not possible to know how it is composed. Two'subjects may get the
same average score by receiving different scores on individual items.
Though both of them may have the same average score, they are not
“equivalent in their structure." zubin says that some personality
Specialists are interested in the totality of personality irrespective
of the complex interrelationships of the variables which make up the
personality. Other Specialists, like clinicians, social workers, etc.,
are interested in the individual variables comprising personality.

.A golden mean would be to group individuals into families or types.

The method used is to find out individuals possessing "similarly inte-
grated characteristics in a given set of variables and, after the sub-
groups of similarly structured individuals are discovered, the patterns
of characteristics that make them similar can be isolated and further
studies can be undertaken in other Variables of the individuals in each
sub-group . . . . The primary tool in this procedure is a technique for
discovering similarities between individuals." This type of classifi-

cation is a kind of typology where the individuals are classified, on

the basis of similarity, into different types. The general criticism
of typological methods that they put individuals into pigeon holes

that do not fit them cannot be raised against the method of zubin where
individuals are permitted to group themselves into whatever constella-
tions they may exhibit in common. "It is an operationally determined
personality'pattern.“

Thus, zubin (16) in his agreement score (number of test items on
which two subjects agree in their responses) has laid a foundation upon
which it is possible to formulate a pattern analytic method for classify“
ing subjects in-terms of major pattern of responses to individual items
of a test. However, he did not develop the method in any general sense.
iMcQuitty (7) developed a comprehensive procedure for classifying persons
in terms of their major patterns of responses.

"In agreement analysis, the responses may concatenate in any fashion
whatsoever: they are not restricted to linear continua; the method does
not order the data according to any preconceived model. Rather, it
classifies the subjects in terms of those patterns which include the
greatest possible number of responses for each. These are called
predominant patterns; and the data are ordered in terms of them.
Responses that do not fit these patterns can be used later to reclassify
the subjects in terms of less predominant patterns if it seems worth-
while? (9).

The present study is planned to investigate some differences in the

type of thinking between those students who score highly on verbal items

10

but poorly on numerical items (i.e., having high verbal ability but low
numerical ability) versus those who score highly on numerical items but
poorly on verbal items (i.e., having high numerical ability but low
verbal ability). The reSponses of the students are scored configurally
and McQuitty‘s agreement analysis is applied in the form of a computed
version developed by Lingoes.1 Also a comparison has been made between

the results obtained by agreement analysis and those by item analysis.

 

lJames C. Lingoes is a graduate assistant in Psychology at Michigan
State University. His version has not been published. It gives results
similar to McQuitty‘s original analysis.

A

II. PROBLEM

The purpose of this study is to examine differences in patterns of
responses on selected items of J‘uolaisl Academic Activity Preferences
Inventory (AAPI) in two groups of freshmen (1958) who were selected on
the basis of their performance on the College Qualification Tests (CQT).
One group scored high on verbal items and low on numerical items, while
the other group scored high on numerical items and low on verbal items
of the COT. Two approaches will be used to study the above differences:

(a) an item analysis, and (b) a configural analysis of the data.

 

1Dr. A. E. Juola, Evaluation Services, Michigan State University.

12

III. ASSUMPTION AND SCOPE OF STUDY

The rationale for selecting the two groups of subjects is based
on the assumption that those who have high verbal ability but low
numerical ability Lhi__._r_1k__ differently from those who have low verbal
ability but high numerical ability.

The present thesis investigates two hypotheses:

a. Students who have high verbal but low numerical ability
have response patterns different from those who have high
numerical but low verbal ability.

b. The configural approach has unique predictive value which
an item analytic approach lacks.

IV. BRIEF DESCRIPTION OF THE METHOD

Before we outline the research design it would be helpful to
describe MbQuitty-Lingoes machine agreement analysis briefly. This
method takes into account the pattern of responses of one individual and
looks for that individual whose pattern of responses is most like that
of the first individual. .After classifying and combining these two
individuals, it brings in that individual whose pattern of responses is
most like what the first two individuals had in common, and classifies
and combines this third individual with the first two individuals.

In this study this process was repeated to the tenth level, i.e., those
ten individuals were classified and thereby combined together whose
patterns of responses had most in common. This procedure is carried
out for each individual in turn. Overlaps in patterns, i.e., the

presence of the same individuals in the patterns, are later eliminated.

V. DESCRIPTION OF THE TESTS

A. College ggalification Tests?

The College Qualification Tests (CQT) Form B are designed to serve
colleges in their admission, placement and guidance procedures. There
are three tests in this series:

Mal Test (0115;): This is a fifteen minute test of vocabulary,
containing-75 items. It is an efficient measure of the verbal
ability.

NumericaliTest {CQE-Nz: This is a thirtyrfive minute test contain-
ing SO items on arithmetic, algebra and geometry. It measures
skill in handling numerical concepts.

Information Test (09:13:21: This is a thirty minute test composed
of 75 items from the fields of science and social studies.

It measures the student's background.

Scores on the Verbal, Numerical and Information tests are summed to

yield the CQT Total scores.

The CQT. are administered to freshmen seeking admission to Michigan
State-University as a measure of their general academic aptitude. The
present study takes into account the first two scores only, i.e., '

verbal scores and numerical scores.

 

lThe Psychological Corporation, 522 Fifth Avenue, New York 36,
New York.

15

B. cademic ct'vit eference nvento

This inventory was constructed by Dr. A. E. Juola, Evaluation
Services, Michigan State University. The assumption is that the follows
ing item classification areas are in one way or the other related to

academic success:
1. Study'Orientation.

Haphazard versus systematized, planned, efficient use of
time in school. Mechanics of study, (e.g. reading the
introduction and summary of each chapter first and then
reading the chapter, or reading in the order given in
the book-«introduction, main chapter, summary) is not
covered.

2. Adjustment.

Self-confidence, morale in academic setting, feeling
secure in school.

3. Ultra-academic Edsel.

Dedication to ultra-academic ideal and high scholastic
motivation-real bookworm, puritan scholastic motivation.

h. Academic Ideal.

High scholastic motives and values. .Academic activities
are most important but not all important.

5. Socio-Economic Class.

Items portraying values which differentiate the lower
classes from higher classes in areas somewhat removed
from school (e.g. semi-academic recreational areas).

6. Achievement Motization.

.An obsessive desire to go ahead, to get good grades,
apparently due to some internal or external very
strong'urge.

16

There are 275 items in all which are liberally scattered over these six
(somewhat overlapping) areas. Each item has four possible and equally
correct answers. For instance, item LL is "Discussing books with friends."
On the scoring sheet, space 1 is to be marked if the individual very
definitely likes the activity; space 2 is to be marked if the individual
feels a mild positive reaction to its space 3 is to be marked if the
individual feels a mild negative reaction to it; and space )4 is to be

marked if the individual very definitely dislikes the activity.

17

VI. STUDY

me
.A group of 8h2 freshmen of the Basic College of Michigan State
University (1958), who had taken both the CQT and.AAPI constitute the
'populationt of this study; Out of these 82h freshmen a random sample
of 127 males and 96 females were selected to determine the distribution
of scores on verbal and numerical items of the CQT. Scattergrams were
plotted between verbal and numerical scores on the CQT separately for
each sex. Median scores for verbal and numerical items were h? and
3h, respectively for males, while ho and 22 for females. These criterion
scores were used as a basis for classifying Subjects as high or low in
verbal and numerical ability. Out of the 127 males there were 2h
(about 20%) who were high on verbal and low on numerical items according
to the above criterion, i.e., they had scores equal or greater than h?
on the verbal items and scores equal or less than 3h on the numerical
items. Based on the results of the selected sample, the following
groups were selected from the population:
A" High verbal ability, low numerical ability.

H§l2§3 h7+ (on verbal items), 3h- (on numerical items).

There were 70 males in.the population who satisfied

this conditien.

Females: h6+ (on verbal items), 22— (on numerical items).

There were 9h females in the population who satisfied

this conditien.

B. High numerical ability, low verbal ability.

18

Males: 3h+ (on numerical items), hé- (on verbal items).
There were 86 males in the population who satisfied this
condition.
Females: 23+ (on numerical items), hé- (on verbal items).
There were 90 females in the population who satisfied this
condition.
The above two major groups (A and B) were randomly divided within each
sex into two equal sub-groups, the first of which was designated the
experimental sample, and the second of which was called the cross~
validation sample. These groups are detailed below:
Elperimental Sample

1. Al - High verbal ability, low numerical ability,
35 males, h? females - 82

2. B1 - High numerical ability, low verbal ability,
h3 males, hS females - 88

Cross~validation Sample

l..Az - High verbal ability, low numerical ability,
35 males, h? females 8 82

2. B2 - High numerical ability, low verbal ability,
_ h3 males, hS females 2 88.

ltems
Although there are 275 items in the AAPI, not all of them could be

analyzed because of machine and time limitations incident to the use of
even high speed electronic computers. The present program for corre~
lational matrices are restricted to 38 variables on the computer used
(1.6. Mistic). Since the various items of this inventory have not been
grouped according to the rational categories described (under "Description

of Tests") any selection of 38 items was assumed to be as good as amy

19

other for the purposes of this study. A frequency count was made of
all the 275 items of the.AAPI for the experimental sample. The first
38 of these which met the criterion of being answered in the same way
by less than 80 per cent of the subjects (N :- l70), were selected.

The Meehl paradox (12) shows that the items which yield the best
configural differences are those which intercorrelate differently in the
two groups of subjects, such that if we subtract, the difference would
be relatively large. Therefore, the intercorrelation of every item
with every other item was calculated for A1 and B1 separately. This
process yielded two matrices of intercorrelation, one for A1 and the
other for B1 (see Tables I and II, Appendix). The matrix of B:L was then
algebraically subtracted from the matrix of A1. The new matrix was
called matrix of differences (see Table III, Appendix). In order to classify
the items into a number of types or clusters of differences, McQuittyts
elementary linkage analysis (8) was applied to the matrix of differences.
This analysis is a method of clustering. It can be used to cluster any
objects which have distinctive cluster—characteristics. Linkage is
defined as the largest index of association which a variable has with a
composite of all the characteristics of the members of a cluster
(consequently as shown in Table IV, Appendix, every variable is assigned
to a cluster in terms of its highest index of association). Cattell (8)
recognizes the importance of cluster method by stating that it reduces
an almost endless variety of variables to a comparatively small number

of representative variables.

In this study, the application of elementary linkage analysis to
the matrix of differences yielded eight types (Table IV, Appendix).
Some of the types did not yield highly interrelated clusters and involved
very few items. Hence, in order to select the items which may yield the
best configural differences further investigation was made by applying
the following methods:

1. Sum and average of each column in Table III (Table of Differences)
was calculated. Matrix of the first sixteen1 items having the
largest column-sum of Table of Differences was prepared. Sum
and average of each column of this matrix was calculated and
ranked.

General Mean (of all the sixteen columns) I .1518
Mean of the first thirteen largest columns a .1595
(See Table VII, Appendix).

2. The highest entry in each column of the matrix of differences
vas marked. The first highest entry was examined. It obviously
yielded two interrelated items. Every time the list of the
items was checked and the duplicates were eliminated. This
process of examining the entries and pooling the non—duplicate
items was continued till such time that there were sixteen

selected items on the list. Matrix of these sixteen items was

 

1The figure of "sixteen" was maintained throughout these four
methods, because there were eight types and therefore eight reciprocal
pairs (highly interrelated items). In order to have a fair comparison
between the items obtained through the types and the items obtained by'
other methods, the number of the items was to be kept constant, in re-
lation to their suitability to the configural approach.

21

prepared. Sum and average of each column of this matrix was
calculated and ranked.
General Mean of all the sixteen columns - .1513
Mean of the first thirteen largest columns - .1593
(See Table IX,.Appendix). I

3. Matrix of the eight reciprocal pairs1 (appearing in eight types-
See Table IV, Appendix) was prepared. Sum and average of each
column was calculated and ranked.
General Mean of all the sixteen columns - .1309
Mean of the first thirteen largest columns - .136h.

h. Matrices were prepared for the sets of the items appearing in

eight types. Sums and averages of all the columns were calcu-

/
13.th 0

General Mean - .1588

(See Table VIII, Appendix).
It was clear from the results of the above methods that: (a) the
averages went down if more than thirteen items were considered, and (b)
method 1 gave the best items. Hence, the items which were used in this
study were numbers 6, 11, 15, 16, 19, 22, 23, 2h, 31, 32, 35, 39 and ho,

as obtained from method 1.

tom s‘s

At this stage it was considered advisable to expose the data to

 

¥A and B are said to be reciprocal pairs if A has its highest
correlation.with B, and B has its highest correlation with A.

22

item analysis for the purpose of testing hypothesis 2 (comparing the
item analytic and configural results). It has been mentioned that each
item on AAPI has four scoring categories. Therefore, for the Mistic
facility, each item was divided as nearly as possible to the median in
relation to the number of responses to each category. For instance,

on item number 6, number of responses to category 1 was 133 to category
2, 703 to category 3, 613 to category h, 26. Hence, the line was drawn
between the first two and the last two categories, and the responses to
categories 1 and 2 were called 1, and those to categories 3 and h.were
called 0. Chi-square was calculated for all the thirty-eight items.
The results are given in.Tables,X.and.XI (Appendix). Those thirteen
items which were to be used in agreement analysis (i.e., 6, ll, 15, 16,
19, 22, 23, 2h, 31, 32, 35, 39 and ho) were ranked according to Table
VII in one column, and were ranked according to their corresponding
values of Chi-square in another column. Then Rho was calculated to see
whether or not the two sets of items for item analysis and for agree-
ment analysis were selected independently.

Ranked According to Ranked.According to the

 

Item Table VII Correspondlpg Value, Table ll
19 ' 13 1 '
ll 10 2

6 11 3
140 8 b.
15 5 5
2h 2 6
23 7 7
22 h 8
35 9 9
39 6 10
32 3 ll
31 l 12
16 12 13

 

23

Rho I: "-1.6 (P > .10)

That is, the two sets of thirteen items were selected independently for
the agreement analysis and for the item analysis. Or, in other words,
those items which were likely to yield pattern differences were not
necessarily those likely to yield item analytic differences. In addition,
r was calculated on all the thirty-eight items based on the rankings

from item analysis (See Table II) and from method 1 described above (See
Table VII) and was found. to be zero. This is further evidence that the

two methods for selecting items were satisfactorily independent.

2h

 

 

Ranked According to Ranked.According to
Item Table VII Table X;
h 31 1
29 21 2
27 15 3
l9 l2 1:,
11 h S
25 23 6
9 l7 7
6 8 8
h0 9 9
15 ll 10
12 19 ll
21 38 12
30 33 13
10 30 1h
36 29 15
2 27 16
2h 2 17
S 26 18
13 37 19
23 3 2O
38 22 21
22 7 22
33 10 23
28 32 2h
35 6 25
3 2h 26
8 2O 27
20 18 28
37 3h 29
1h 35 30
18 36 31
17 28 32
39 13 33
3h 25 3h
26 16 35
32 S 36
31 1 37
16 1h 38

r - -.000
This further confirmed the results obtained previously.
3 The first thirteen items giving the highest values of Chi-square

(Table XI, Appendix)'were selected for the item analytic approach.

They are items: 1;, 6, 8, 11, 12, 15, 19, 21, 25, 27, 29, 30 and to.

The subjects of A1 and B1 were scored on these items in such a way as

25

to maximize the difference between the groups in favour of high scores

for group A1. The following distributions of scores for the two groups

were obtained for: (a) the thirteen most significant items, (b) the
four most significant items (i.e., items 1;, 29, 27 and 193 p < .05)

and (c) the three most significant items (i.e., h, 27, 293 p < .01)

Frequency Distribution Number 1 (13 items)

 

Scores A1 B1
1 O 2
2 1 8
3 2 13
LL 3 13
i; ll 17-1-30 - La 22
6 13 10
7 1h 12
8 20 h
9 l2 3
10 6 0
ll .9. .1.
82 88

Frequency Distribution Number 2 ()4 items)

 

— Scores A1 B1
0 h 16
1 17 _21_+31 :52 1:1
2 22 21
3 18 5
h 2.1. .5
82 88

26

Frequency Distribution Number 3 (3 items)

 

- Scores A1 B1
0 12 3h

1 25 374-13 - 50 hl

2 21 8

3 .21; .5

82 88

For each frequency distribution the cut-off point was selected
which allowed for the maximum difference in scores for the two groups.
These empirically determined cut-off scores were 5, l and l for the
three frequency distributions.

Using the same scoring system each subject in the cross~validation
sample (Ag and B2) was scored and corresponding frequency distributions
were made. These were:

Frequency Distribution Number h (same 13
items as in ED 1)

 

 

Scores A2 B2
2 O 3
3 6 6
h 8 7
5 _ 20 3112.56 - 9o 16
6 15 20
7 17 22
8 9 12
9 6 1

10 O 1
ll .1. .9.
82 88

27

Freqiency Distribution Number 5 (same h
items as in ED 2)

 

Scores A2 2 B 2
O 2 2

__ l _ 28__30+69_!_99 17
2 37 E2

3 15 22

h .2 .5.

82 88

Frequency Distribution Number 6 (same 3
, items as in ID 3)

 

Scores A2 B,a
O 12 6

1 50 62+h3 - 105 32

2 19 36

3 .2: .1

82 88

Applying the cut-off points determined from the experimental sample

fourfold tables were constructed. Below are presented these tables

as well as the corresponding tables on the experimental sample.
Results on item analysis on first 13, first it and first 3 items

vide Table II, presented in fourfold tables.

 

l terns

High verbal 65
Low numerical

A 30 95

High numerical
Low verbal 4'1 fl ~25

82 88 170

Correctly assigned :- 6S+58~123~72%
CR3- a .63 (not siglificant)

items
High verbal
A Low numerical 61 Bl 92
High numerical
B Low verbal .21 .51 .18..
82 88 170

Correctly assigned . 61+57-118-69%
.CR - .37 (not significant) .

(3 items)

High verbal 145

A Low numerical 13 58

High numerical
B Low verbal ‘31 '15 “ll—2'
82 88 170

Correctly assigned . hS+7S-l20-7l%
CR -- 2.7L,”-

*_—

28

Cross-validation Sample

1 1.3 items I
A2 B 2 Sum
h8 56 101;
.311 .32. .22
82 88 170

Correctly assigned an h8+32~80~h7%
CR '- l.79 (not significant) ,

ELM).
52 69 121
.19 .12 19.
82 88 170

Correctlyakgssignedu ~52+l9~7l~h25§
CR ' 3092.

11.122112).
20 h3 63
J22. 1+5 1.21
82 88 170
Correctlygssigned . 20-PhS-6S-38z
CR '3 3.10

l'I‘he results are compared by McNemar‘s Critical Ratio Formula

(19a) McNemar, Q.
Sons, 1955.

Psychological Statistics. New York: J. Kiley &

 

29

The above 2 x 2 tables show that the most significant items did not
hold up on cross-validation. In fact it is to be noted that there was
a tendency for the items to discriminate between the groups in the
reverse direction. In the cases of 3 and )4 most significant items we
obtained significantly poorer classification than can be expected by

chance (P < .001).

Agreement Analygis
Agreement analysis was applied to the scores of the subjects of

A1 and B1 on 13 items (discussed above i.e., 6, ll, 15, 16, 19, 22, 23,
2h, 31,3 32 , 35, 39 and 10). First their patterns were prepared on
Mistic (one pattern as a Specimen is given in the Appendix). All the
patterns within each group, A1 and B1, and then between both the groups,
A1 and B1, were compared. There were some duplicates within each group
but there was none between the two groups, However, all the duplicates
were dropped. This left 1411 patterns of responses in A1 and h2 patterns
of responses inBl. Subjects of A1 and B1 (experimental sample) and
those of A2 and B2 (cross-walidation sample) were scored on the patterns
of A1 and B1 on the Mistic. Each subject of A1 and B1 was then classi—
fied in terms of the patterns. This process discriminated between the
good and bad patterns. (Good patterns were those where most of the
subjects were correctly classified and bad patterns were those where
most of the subjects could not be correctly classified). All those
patterns where the ratio of wrong classifications tattotal classifications

was equal to or more than 13).; were dropped. This eliminated L11; patterns

 

30

of the 86. The subjects ofA2 and B2 (cross-validation sample) were
‘ scored on the basis of the remaining h2 patterns ofAl and B1
(experimental sample; 23 patterns in.A1 and 19 in B1). Each individual
was assigned toAl or B1 depending upon whether or not he made the
highest score with.Al or B1, If an individual of A2 could be assigned
to.Al, he was labelled as "correctly classified," if he was assigned
to B1, he was labelled as "incorrectly classified." Similarly an
individual of‘Bz was "correctly classified" if he could be assigned to
B1: otherwise "incorrectly classified." This yielded h7 correct
classifications and 35 incorrect classifications in.A23 h6 correct
classifications and h2 incorrect classifications in B2. The configural
approach yielded results whichralthough'were not reliably different
from chance when applied to the cross~validation sample, were, nevertheless,
in the expected direction.

The following fourfold tables were made to compare the results

obtained by the agreement analysis and the item analysis:

31

Cros s-Validati ong Sub j ects

Configural
A B Sum
Item Analytic B 60 h? 107
3 Items
A 38 63
E73
CR1 - 2.68%
A B
B 17 32
Item Analytic
h Items
A 72 89
CR - 3.9L?
A B
B 27 39
Item Analytic
13 Items
A 62 h2
CR - 1.81 (not significant)
A B
A _
B2 12 )16 88
Actual , D
A2 87 35 82

CR - .80 (not significant)

 

”The results are compared by McNemar‘s Critical Ratio formula (19a)
McNamar, Q. Psycholoa’ca; Statistics. New York: J. Wiley & Sons, 1955.

32

It may be noted that the critical-ratio in the case of the 13
items is not significant, but in the other two cases it is significant
at 1% level of confidence. In general, the results obtained by the
configural approach are better than those by the item analysis, but the
fact, that the item analytic results are poorer than those which could
be obtained by mere chance, makes this slight superiority unreliable.

The configural results were compared with the results which could
be expected by mere chance. The former results were superior to the
latter but not significantly.

The item analytic results which were obtained in this study were

unusual, nevertheless, they were checked thoroughly.

33

VII. SUMMARY AND CONCLUSIONS

The present study investigated the differences in pattern of
responses to selected items of the.Academic Activity Preferences
Inventory by freshmen who scored high on verbal items but low on numeri-
cal items versus those who scored high on numerical items but low on
verbal items of the College Qualification Tests. The study also showed
the comparison between the results obtained by item analytic method and
those by agreement analysis.

McQuittyeLingoes machine agreement analysis was applied to dif-
ferentiate two categories of people. In our present study we have
assumed that the students who have high verbal but low numerical abili-
ties have patterns of reSponses different from those who have high
numerical but low verbal abilities. Since they were taken to be two
categories of people, agreement analysis was applied to differentiate
them.

Three hundred and forty freshmen were selected out of 82h, who had
both College Qualification Tests and Academic Activity PTeference
Inventory in September, 1958, Michigan State'University, on the basis of
their verbal and numerical scores. Group A was formed of 170 freshmen
who had high verbal but low numerical abilities. Group B had 170
freshmen who had high numerical but low verbal abilities. Each group

was further subdivided into two equal subgroups. These subgroups were

3h

called.Al,.A23 B1, B2. A1 and Bl'were taken as the experimental sample
andA2 and B2 as the cross~validation sample. Their responses on the
AAPI were subjected to item analysis and agreement analysis. The re-
sults obtained by these methods were compared and the following
conclusions were drawn:

1. The results of both the approaches did not support the
hypothesis significantly that the patterns of responses differ
as a function of high verbal and low numerical ability versus
high numerical and low verbal ability}

2. Item analysis showed significantly poorer classification on
cross~validation sample in cases of 3 and h most significant
items chosen item analytically.

3. The difference between the two approaches is significant in
the cases of 3 and h most significant items, but is not
‘significant in case of 13 items chosen item analytically.

h..Although the configural approach is slightly better in general,
the fact that neither approach yielded better than chance
prediction does not allow us to assess the merits of one
method over the other.

However, the prospective researcher is recommended to prepare the
matrices of all the 275 items of the AAPI and construct thereby matrices
of differences. Then he would be in a'better position to select
configurally suited items. Similarly all the items should be exposed

to item analysis. This would give him a correct picture of the relative

merits of both the methods.

35

An experimental design of this kind stresses the necessity of a
theoretical approach toward the preparation of configurally suited
items. If a theory could be developed through which items suited for
configural method could be prepared, it would facilitate the situation
tremendously by saving the time of the researcher that he spends in

selecting such items.

l.

2.

5.

6.

10.

12.

36

BIBLIO GRAPHY

Allport, G. W. Personal Documents in Psychological ,Sgience,
Soc. Sci. Res. Coun. Monog. 19112.

Cattell, R. B. "rp and Other Coefficients of Pattern Similarity."
Psychometrika, 11:, 279-298, 19119. 4

Cronbach, L. J., and Gleser, G. C. "Assessing Similarity Between

Profiles." Psyphol. 13:311., 9;, h56-h73, 1953.

Gaier, E. 11., and Lee, M. 0. "Pattern Analysis: The Configural
Approach to Predictive Measurement." Psychol. Bull” 1Q, 1110-1118,
1953- .

Louttit, C. M. "The Nature of Clinical Psychology," Psychol, Egg”
164 3633-389: 19390 -

McQuitty, L. L. “Theories and Methods in Some Objective Assessments
of Psychological. Well-Being." Psyphol. Monog., 6_8_,No. 11.1 (Whole No.

.385), 19511-

McQuitty, L. L. "Agreement Analysis: Classifying Persons by
Predominant Patterns of Responses ." Br, J. Stat. Psycholq 2, 5-16,

1956 .

McQuitty, L. L. "Elementary Linkage Analysis for Isolating
Orthogonal and Oblique Types and Typal Relevancies ." Educ, Psychol.
Measmt. 11, 207-229, 1957.

McQuitty, L. L. "A Pattern Analysis of Descriptions of ‘Best‘
and ‘Poorest' Mechanics Compared with Factor Analytic Results."
Psyphol. Monog, 1;, No. 17 (Whole No._hho) 1957.

McQuitty, L. L. "Job Knowledge Scoring Keys by Item versus
Configural Analysis for Assessing Levels of Mechanical Ebcperience.

Educ. {sygholg Measm . I 18, 661-81, 1958.

Mchitty, L. L. nMaximum”Minimum Hierarchical Analysis ." (Unpub-
lished MSS) . '

Meehl, P. E. "Configural Scoring." J_._ Consult. Psychol. 11;,
3.65-1.71, 1.950 0 .

37

13. Murphy, G. Historical Introduction to Modern Pchhology.
New York: Harcourt Brace and Company, 1919.

lb. Sargent, H. "Projective Methods: Their Origin, Theory and
Application in Personality Research.“ Psychol. Bull. , L24
5793, 19145- -

15. Watson, R. I. “A Brief History of Clinical Psychology."
252911010 B11;., 50-, 321-146, 19530

16. Zubin, J .- "A Technique for Measuring Like Mindedness ." J. Abnorm.
.erOCJ PsthLO,.33J $8-16, 19380 .

 

 

 

 

 

 

 

 

 

 

 

 

 

38
TABLE I. INTERCORREIATION OF EVERY ITEM WITH EVERY OTHER ITEM m GROUP A. (SUBJECTS HAVING RICE VERBAL AND Low NUMERICAL SCORES)
terms 2 3 h 5 6 8 9 10 ll 12 13 117115 16 l7 l8 19 2O 21 22 23 211 25 26 27 28 29 30 31 32 33 311 35 36 37 38 39 110
2 0316 1797 ~1608 ~2007 1513 1069 1050 «2016 0278 1765 055171, 0252 ~1319 2616 01017 1232 ~1335 2212 -1212 2762 1835 ~h071 2632 1088 2392 ~0561 12993-1268 07817 0122 2805 1509 1522 0870 01195 1180 0965
3 0316 02311 ~0511 ~016h 1311 0732 1111 0018 ~0671 ~1997 ~112l -0086 1691: ~0128 @1210 0511 011711 ~1826 00116 00811 ~039Ll ~07LL7 ~~1696 1760 ~1135 ~l5h6 ~110h {0158 £100 ~0029 -OJ.19 ~05211 "1978 0532 0966 ~O907 ~0577
2'1 1797 02311 ~1790 ~0120 ~0598 01171 —0361 ~2h27 ~0330 1781 21117 ,-2999 -OO3l ~135l 0.1611 ~~3:602 ~l2l8 1927 0870 2201 0158 2069 2287 11830 2521 —0965 3035 9-1878 ~1611 11113 2087 -O256 0372 2657 21115 1199l 071111
5 ~1608 ~0511 ~1790 ~09h9 1101 0557 «0059 1821 2262 0561 —02757 0106 0012 18911 31121 03113 0096 0858 11h7 ~0h61 0176 15119 ~lh22 0217 ~10h8 2082 ~0336'. 0160 2030 ~18h5 ~206h ~0903 0h67 0667 ~1220 ~ll68 1283
6 ~2007 -016h =0120 ~09h9 ~0388 ~0893 ~0568 1703 ~1030 0078 ~2112 01125 0719 —0619 1509 ~059h 1659 ~1352 1722 ~0178 ~0111 0570 0676 ~2755 ~0686 0855 ~0233 3.19217 0531 -0177 0272 1088 0763 ~0665 ~0967 {330371510
8 0 ~0 88 2 2 118 -0 ~02 8 0281 10‘ 3306 0078 0657 15211 0111111 95211 07811 2129 21167 1597 ~0960 1266 0308 1311 0039 1079i 0379 1358 ~1561 0685 -0717 ~0607 01186 ~1066 0870 ~0697
9 31869 883—2 3897? 8:55;} 0893 2532 53 —098:LL* @888 @6150 2136 2619180531 0062 -0556 —2252 0715 ~2177 —0025 ~255h 117h 0361 0978 11116 1752 31752 «1988 O957'Vf—28h5 -lhh5 0226 1151 -0172 —012O 0761 -0287 1503 0886
10 1050 1111 ~0361 -0059 —0568 11811 -0981 0557 0587 ~0938 ~1086; 2582 0978 1902 0776 1091 0537 ~~068.9 0763 @1156 2585 0691 0351 0392 "1099 ~1896 0800.} 1525 1083 0010 1383 0612 0216 -ZLll5 1381 731311 0617
11 4016 0018 .2772? 1821 1703 43535 @670 0557 1715 {3502 ~2087 I 21116 1188 ~0066 ~0060 2807 0170 --lh29 1501 ~3005 -0259 33117 *3031 ~~1798 ~2216 11129 ~2000 f 28211 2153 ~0626 ~3872 ~0281 -0hhl ~1378 ~13OLL "2680 ~l6lll
12 0278 —0671 -0330 2262 ~1030 ~0258 —0610 0587 1715 ~1370 093h 51987 ~0215 0003 0873 0887 21211 ~1209 ~0273 ~11060 0037 1111 ~1897 ~0967 ~1305 1563 23113»; 282h 3622 1910 0183 1338 ~1h97 ~2228 h02h 0818 —06h6
—- . - 0 0 88 0586 ~1093 171111 ©th 0220 ~3535 1891 05211 1518 1095 ~1539 1872 11211 25611 05113 3931 7-0972 0121 ~0356 0710 2105 3285 2l8h 0813 267h 2h07
11131, 8588—1321 2137 83675 £31: (18337 26319 31882 388; 89311 0988 9 82162 ~0117 01107 0219 ~2766 ~08h9 ~003h 2021 1822 0711 ~0826 2731 2907 11195 and 2236; ~3310 0370 0963 20711 —0232 0722 1025 01199 11208 1198
15 0252 ~0086 -2 6 2 82 21116 -:L 8 0586 ~2162” 1291 1202 ‘0669 17119 ‘0727 0236 11127 06511 2375 0691 05118 “0777 ~0753 1552 ~07113 314119 1527 —088)1 —1025 11196 0998 0236 ~1163 ~338h 01169
999 01116 01125 330 0531 5 9 7 7 _ __ __ . __ __ 66 _ 6
16 -l3l9 l69h —0031 00172 0719 0078 0062 097A 1188 ~0215 ~lO93 -0117 11191 0290 05111 00511 1186 0152 ~0526 1319 2676 0289 ~0596 3368 ~0638 2255 1172 0915 1358 2808 ~Ol76 0868 0379 O h DEED 01162 11103
17 2616 -0128 _1351 18911 @619 0657 —0556 1902 ~0066 0003 17811 0770721202 0290 6201; ~0131 —0288 0352 1712 0812 2761 07113 1965 1099 @182 01:7L 1067 {0622 1990 0797 1896 2729 017:: 03:: -062 ~02: 1:9:3
- .. ~ ’~066 '0 620 ~1629 0276 0615 23111 02113 0793 ~0789 1273 0395 ~07 l9 ~OOl9 74-0252 0939 0566 1696 0037 11 12 «010 039 12
18 0hh7 —01h0 01611 31121 1509 15211 2252 0776 {£60 8883 3288 3%? 1789 0535'): —01'1_'LLL ~1629 —0935 03911 1621 ~1823 0768 0107 0652 --1956 ~0300 —0h1h ~12h0‘ 1,15 1565 0161 ~0590 0517 0815 0125 0866 ~233h ~0387
19 1232 ~0511 ~3602 ~0383 ~059h Ohm 0715 1091 O7 7 ‘ —08h9 ._O727 1186 .0288 0276 .0935 -o981 0883 ~2115 0980 29115 ~19h9 ~1636 ~1363 0700 ~2628 772 80 2976 2666 {31511 Ohhh ~0732 ~2998 0252 ~2862 -3803
22 {1223; £825; :ng23 88958 $358 8581: Z8825 8889? 3829) {21289 13189;. ~003112 0236 0152 0352 0615 03911 ~0981 1201 2029 1190 2077 1913 1227 2761 02811 118115 §—1185 —0721 1120 2019 ~0655 0951 1732 ~096h 2016 :21700
22 ~1212 00h6 01170 11117 1722 2129 ~255h 0763 1501 ~0273 052h "202111827 -0526 1712 23111 1621 08113 1201 - 1036 1339 2001 ~OlOO ~~1581 9:58 078: 0159»; 2830 2191 ~1329 ~0556 —0615 0787 12:: 3:: 37:; ~09;
- .. ~ 2 02 ~182 —-211 2029 1036 2061 ~2065 2855 0530 293 030 2707 ;. .1270 ~1333 ~331 0997 -0320 ~0137 1 -
28 318368—8398 8388 81196 3111? 12.897 8363381518 «312395 883? 115198 8:37:12: (23% 26:72 C213761 0793 0768 0988 1190 1339 2061 9258 1868 0877 1528 1328 .1558 .0380 3220 ~0275 2021 1791 1257L 162.2 0&8? 3822 $1818
- . ~2077 2001 ~2065 ~025h ~31100 ~~1797 ~0902 01 2 -2 32 2760 1163 057h ~2118 0290 -O 2 -l -O 9 - - ‘
25 ~h071-07h7 —2069 15119 0570 ~0960 0978 0691 33117 lulu 1539 ~0826 0691 0289 ~07h3 ~0789 —Oh07 2985 O 2855 11168 ~3h00 0992 3957 63m 3283 4765 —0000 {3217 3369 0711 1605 3067 01139 26514 1548
26 26 2 ~16 6 22 ~11 22 0676 1266 11116 0351 ~3031 ~1897 1872 273.1 05118 ~0596 1965 1273 0652 ~l9h9 1913 ~030 _ _- __- _
27 1088 1760 11838 0217 ~2755 0308 1752 0392 ~1798 ~0967 11211 29079750777 3368 1099 0395 —:L956 ~1636 1227 ~~1581 0530 0877 1797 0992 M 21152 3796: 225:5 2673 1007 237: 3:: 0:1 9:13;: 31:: (2):: if): :13;
.. .. .. ‘- — 2 61—0 2 128 —0902 3957 2 2 07 3 9 ~2h75 ~0196 032 3 1 1 0 1
2: {21568 {1351318 8585 "2882 “888: 8839 3988 311896 80212; 1323 851613 ~fii9ifi 792552 32888 818? 8918 383.8 88788 02811 0832 0386 —13568 0162 ~03L§8 ~3Z98 fig? 26 ~1726 1539 21709 ~1957 —2170 ~0372 0&2? 2&2: 3232 38212: “$58
— - .. " - 2110 ~2628 1815 0159 2707 1358 ~2h32 323 295 3 9 ~17 ~2588 —0920 -0951 3722 170 l 3
30 1299 -1392 3035 ~0336 —-0233 1079 0957 0800 2000 13113 3931 2236170783 1172 1067 “0019 l _ _ __ __2 _2 1 _2 88 ~20 08 008 ~2h28 11137 -2507 -3020
a as: om .1. 92a ~29 1525 2823 2992-3310 29 0915 ~92 155 5555 555 -155 5555 5555 555.555.0155 .555 5a 5555 .58.; 51.; 05.5.1.1.
32 07811 {0700 ~1611 2030 0531 1358 ~1885 1083 2153 3622 0121 0370.4 1527 ~135h 1990 0939 15 5 97 61 {)2 5 05711 .0217 2378 0326 ”1957 "0951 2 13 1373 1086 23211 18% 0125 2811 0753 {3229
35 5558555 558555555555553555555515555 555555 5558555555 85255555555558555555555597 -21. 3n gg 353; gig
3 2 “0 " ‘ ~ 1 0290 0711 0961 1851 ~0377 170 3087 2h69 23211 1390 011 097
x -- 232 14196 0868 2729 0037 0517 Ohhh —0655 ~0615 -0320 179 ..
35 I509 ~052h —0256 ~0903 1088 —0717 ~0172 0612 —0281 1338 2105 —0 __ 0 1 0 8 ‘01 1657 ~062h 1605 @035 06116 ~079LL 1566 0308 0158 18% 0219 0161 0715 01180 0300 £255
36 ~1522 ~1978 0372 0867 0763 ‘0907 '0120 0216 "08813 :33ng 333i (£6225 8:232 8668 8358 1282 (93:63:25 3998 19732 1801i 1885 1160 ~1652 3067 3196 1528 2886 3559' -2h28 ~1391 0125 1552 0970 0715 01188 26113 311011
37 0870 0532 2657 0667 £665 OM) 0761 "1115 "137 ‘ I 6 866 0252 ~0961 ~06h1 ~168h 0388 ~0195 01139 21160 0973 -0609 0236 ""1537 16110 2811 2128 31181 01:80 0888 1575 91419
38 OM95 0966 2m; ”1220 ‘0967 “-1866 {1:83 33128 {23688 88:18 (266178 2288" 23388 812182 3229 ~8392-63331; ~2862 2016 07117 11135 ~0002 «3053 2651: 3399 11030 9828 if 1«2507 0289 0753 1537 0206 0300 2:83 1&5 1269 1299
l~1168~00 070 - - 5 t _ _ 8— 0.-~ - -02-02 0 09
88 8928’ 333; 83:8 1283 ~1818—0697 0886 0617 ~161h ~06h6 21107 11911 01169 1103 1299 1628 ~0387 ~3803 1700 2917 0755 0819 1618 15158 31117 173 111111 .. 3020 289h —0229 1271 hl 55 3
__ __ 4 7 ", 4
Note: Decimal points are omitted. , 1 .3

 

1

f
. $

 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

39
TABLE II. INTERCORREIATION OF EVERY ITEM WITH EVERI OTHER ITEM IN GROUP B. (SUBJECTS HAVING HICH NUMERICAL AND LOW VERBAL SCORES)
_ _ f 1 111:
Items 2 3 1 5 6 8 9 10 11 12 13 11‘ 715 16 17 18 19 20 21 22 23 21 25 26 27 28 29 30 .7 31 32 33 31 35 36 37 38 39 10
__ . 7 7
2 -0960 2390 —2O59 ~2088 2321 1056 0107 —0225 ~0121 1325 11821-0017 0550 ~0009 0677 1253 ~1328 0860 0359 1157 2321 ~2321 3117 2087 2219 0735 2272 ~1966 0818 1558 3781 0209 ~0310 1771 0300 3191 3012
3 0960 ~0122 0318 0138 0691 ~1075 0160 ~0887 1323 0192 0128 1 0982 -0833 0621 0169 0366 0020 0073 —1112 1927 0515 0902 1231 0911 —0817 0716 —0367 -0260 ~0101 11011 0661 0555 ~0152 0338 0795 1012 0179
1 2390 ~0122 ~0716 ~1178 «0895 0676 ~0111 0091 ~1176 2119 27821-0522 -0111 1179 0138 ~0558 ~2179 2018 ~1228 —0177 1823 ~1082 2915 5005 2581 ~1019 3218 ~1310 —0899 0868 2667 1279 1368 2510 ~0912 1091 1371
5 ~2059 0318 —0716 0733 ~1971 ~2181 0516 0507 1056 ~0188 0193 ~1869 —0626 2162 1878 0669 0099 -0130 0119 ~1712 ~0590 1981 0892 ~1170 ~0122 1079 0337 1812 0128 -0736 ~1191 ~1131 0619 -0563 ~1150 ~1381 ~1687
6 ~2088 0138 ~1178 0733 0110 ~2110 ~0126 ~0270 ~0115 0197 ~1519 _—0681 1557 0218 0000 0012 2871 ~1735 1207 ~1719 ~2030 2888 ~1379 1111 ~2539 -0286 ~2263 #0965 ~1173 0853 ~3251 ~1121 ~1311 ~2696 ~1231 ~2575 ~1826
' 2 — 2 0 —0 68 0 2 201 1097 1919 3 15 0819 0015 1079 ~2360 1083 0081 2835 1938 0522 0828 1119 —0822 —0312 1123. 97-7193 0827 1191 ~0851 -0101-1108 1760 0109 0000 1251
8 8888 88878 8888 88.821 ~811qu 2912 291 5811518 08.61 0858 1280 07121—0118 0833 0613 0218 1128 0905 1070 ~1936 1362 1959 ~2113 2290 1529 3001 0371 1553' * 71806 0932 2132 2987 ~1085 ~1989 1165 1332 1667 1711
10 0107 0160 —0111 0516 ~0126 —0059 ~0119 ~0285 0730 0159 03113 0319 —0085 1183 —0091 1309 ~0525 -0621 0069 03811 ~~1088 0196 ~0913 ~0968 ~1557 ~0137 0377 . .0000 ~0005 ~1552 ~1061 ~0176 0676 -03117 0551 ~0178 0651
11 0225 30887 0091 0507 ~0270 ~0968 ~1161 0285 2122 ~2200 M101'? 0175 1503 0161 -0271 -0359 1573 —0118 1039 0317 ~1702 3161 ~1390 1178 0581 0712 -0716." .2370 0216 —0073 0632 1082 0936 0659 2523 —1016 —0235
12 -0121 1323 ~1176 1056 -0115 0721 0250 0730 2122 ~0535 70375 1139 ~1591 0350 0177 ~1713 1800 -1185 ~3685 1221 ~1861 1031 ~1588 ~1080 1059 3256 ~1665 1775 1893 0171 0372 0582 —0311 -1001 1198 -0516 -0710
-2 00 —0 22211 , 208 --1 8 21 06 0965 0196 0397 1910 3721 19511 0727 2652, 0118 0771 0560 1210 1895 1595 1521 0783 1932 0037
1137 1388 83188 81488 8188 518819? 818897" 8816.8 8888 1817 08378 2221 1.333 .8881]; 813138 8888 2188—1887 086; 0188 1833 1211 ~1518 3855 3366 1619 ~1328 1123 .0190 0603 0126 2251 0073 0519 3157 —0192 1036 1660
‘ 11 —O 82 «1822 ~1869 ~0681 1919 0118 0319 01175 ~1J139 0079 ~{1161 2822 0871 -1719 1313 -1999 0609 3328 1762 -0891 1319 ~1333 0021 ~0271 «0531 ~1236 45-1150 ~2167 ~1051 —-2023 1739 0839 -o527 ~2092 -0611 1098
18 8888 8833 -0111 —0626 1557 3515 0733 —0035 1503 ~1591 0261 ~078h 2822 0598 0065 1283 ~1757 0511 0710 0781 1025 1007 ~1138 0103 ~0722 ~1161 ~1259 50533 -1686 0135 ~1105 0512 0810 0015 0081 ~1166 0299
17 «0069 0621 1179 2162 0218 0819 0613 1183 0161 0350 1391 011110 0871 0598 67115 11111 ~~1273 1963 0819 ~0806 1779 1321 1236 2219 0122 0977 1358 -0133 0292 ~1529 0502 -0291 0738 1172 ~0378 -0133 0015
0 ’ .‘. - 1 01 ~0 1 0875 ~0255 2128 1939 0272 2325 06611 0675 1528 ~2007 1200 ~1150 «0190 0975 0291 0270 0060
18 0677 0169 0158 1888 8088 C18LLB 8888 8888 3888 881713? 8888 215188 13288 (1888 $1.88 -0111 -0111 8888 1808. 168; 1885 0123 0225 1632 0139 0620 0135 -0856.' “2223 0550 0591 0118 0001 ~1275 1187 ~1585 1112 1556
19 1253 “0366 @558 O 9 O 79 25 15 3 1800 —1998 —1017 1—1 .—1 5 ~12 1195 ~0758 ~3917 0167 ~1573 -1713 2890 701189 ~2588 ~0789 0062 ~1096' 21030 0509 —0703 0910 1393 -0502 -2173 —0217 —1888 ~2606
:3 "3823 8883 "83:8 .8328 4887;; ”12888 8888 :8821 —01:1?_8 ~LL85 2517 0069 027933 0277377 1988 1513 1001 ~3917 0523 0180 1071 ~1175 0895 3518 1113 1172 32g5 170000 ~1176 ~1121 1659 -0132 1210 0711 0107 2100 2219
22 0359 --1112 --1228 0119 1207 0081 ~1936 —0069 1039 ~3685 0637 31162 1 3328 0710 0819 0197 1609 0167 0523 1096 ~1096 1636 339511 0115 077: ~~12312 "008:- 31137 -1765 0132 0365 2752 0110 016158 18:1 319:9L +377:
7. - —O 80 10 6 0623 0733 1776 1619 093 1103 ~02, .7-70803 —0289 0189 0612 2186 ~0993 011 10 3 09 313
23 1157 1927 0177 "1712 ~1719 2835 1362 0388 9382? 88811 8191968 181:8: «886321 $888 8888 88878 8888 881718 1871 ~1086 0623 ~1203 3869 2115 0896 ~2JD9 3698‘ +1511 —0387 0131 2861 ~1221 —0732 1123 -0111 0675 1705
88 8881 8888 8888 {195881 "8888 89588 8888 “8896 8861 1031 -0397 ~1518 13819 1007 1321 0255 0225 2890 ~1175 1636 0733 ~1203 ~1117 —0210 «1616 07% ~216gf~ '110191 —0521 41890 ~3016 —0050 02128 12296 0271 ~2671 @383
r " H _ _ . —' .. 8 08 —0 1776 3869 "111117 3522 1072 ~02 190 _-,:‘_-1998 1075 ~0998 3181 —-1617 -08 2 271 ~11 7 3197 31
:6 3857 12% 555555551315 (181278 818883 3888 18188 8888 8872831 8888 8888. 8888 88:88 12.88; 8:888 28888 3588 08:85 1619 2115 —0210 3522 2720 ~1150 311627 {#1182 ~0579 0919 3068 1217 1881 3173 1711 1116 3560
~ 1113—0 6 0 0896 ~1616 1072 2720 1718 3327 7-000 —0069 0633 5318 0716 1172 1518 1019 3361 1898
28 2219 ~0817 2581 ~0122 ~2539 0822 3001 ~1557 058:1 13;: 88788 {139% 28878388121 88278 8888 "8:63; "85/2; 1172 0881 193% _2109 70763 3258 ~1156 1713 17515-19773 0837 _2372 0750 07759 1169 _27762 01771 13777 00277
29 0735 -0716 -1019 1079 —o286 ~0312 0371 -0137 07 3 6 2 1123 3 8 0661 -0856 ~1696 3275 0085 ~0282 3691 ~2167 1906 3162 3327 -0751 2--~-g7'~"—0137 ~2006 0352 3968 —0171 1612 3236 ~0162 1371 2632
6 218 0 37 "2263 1123 1553 0377 "0716 "1665 2 5 "1236 "1259 13 5 .. —- —- 0 0 13 ~0137. 6 1 0000 0 01 ~1501 2321 0757 0897
30 2272 '03 7 3 3 2 0 175 0118 0190 ~1150 .0533 {3133 0675 2223 1030 0000 ~1137 —0803 1571 0191 1998 1182 000 9 7 2031 067 ~09? 9
31 "1966 '0260 "1310 1812 .0965 -1933 ”1808 8888 08:86 1893 0771 0603 ~2167 ~1686 0292 1528 0550 0509 -1176 ~1765 0289 ~0387 0521 1075 ~0579 -0069 0837 ”2009] 52031 0171 -0193 0569 0100 0078 ~05611 ~092l1 0390
32 “0818 ~0101 {3899 0128 ~1173 @827 £93 1 6g 3' 0 ~1121 0132 0189 -0131 ~1890 ~0998 0919 0633 ~2372 0352 ”370676 0171 0966 0772 0935 1310 0191 0507 0171
33 1558 1101 0868 0736 0853 1191 2132 ~1552 "0973 “9871 0560 "012' 71958 '0135 “1529 "2007 059” “O7 3 65 .6355 0612 2861 ~3016 3181 3068 5318 0750 3968' 240197 0193 0966 0231 -0857 1682 1789 3751 2118
31 3781 —0661 2667 ~1191 ~3251 ~0851 2987 ~1061 ~0632 0372 1210 2251-: ~2023 ~1105 0502 1200 ~0118 —0910 33.133 2752 2186 ~1221 —00 50 _1617 1217 07% 0759 ~0th 1.0000 0569 0772 ~0231 2139 $222 «3385 0631 ~1337
35 0209 0555 1279 ~1131 ~1121 —0101 ~1085 0176 1082 0582 189g 3%? 1739 0512 02915 £1158 $33151 .3853 1210 0th £993 ~0732 0228 @5772 1881 1172 1169 1612:, {0901 0100 0935 33857 2139 33816 0252 1085 0697
36 "03116 "0852 1368 0889 2.211% ”fig "fig; 888? 8255312355 £21 3157. 2823?? 383165) (£182 39?]; 1187 -2173 0711 0158 0611 1123 —0296 1271 3173 1598 ~2162 3236,, 171501 0678 1310 1682—0222 -0816 0803 2820 1931
37 1771 0338 2589 “‘0 3 " 9 ' 351w ' — _ 1 10 0111 22016231552 2 -0561 0191 1789 -0385 0252 0803 0581 1091
22 2222 2222 3223 2232 3333 3353 3555 555555 55551555 .555 5555 555 5551315 355 51515555551555.5551 5151 1115 355 117 171.555 .091 057 31 105 187. 1 5 155
- 81 —-2 0000 1 7 ~0 - 2 "‘ " .. 8 8 0021 232; 0 218 -1 06 1931 109 33
if) 8888 (13%; 11113191 ”£87 11582 1251 17117 0651 0235 0710 0037 1660 11098 0299 0015 ~0060 1556 ~2606 2219 «0772 3138 1705 2683 2231 3560 19 0897 0390 171 1 337 97
Note: Decimal points are omitted.

 

9“

D‘J‘

V.
H ._ w W
m h * H
n
-

fl I

_

h v
., r.
_

In. ha ‘. u

. fl . 1"
..\ ;‘ M,fl., V
m ,Vm.fi ,
fl? , A

V‘* ‘ ‘
1 fiw _ n
R f. a A. n

m ._ M

 

 

 

 

00

 

 

 

 

 

 

 

 

 

 

 

 

 

80
”L 31 35 36 37 38 39
1 I 29 30 f 31 32 33 76 0900 0195 201.11 1026
E I 9 28 ’7“ ll 9
mTRIES1OF TABL 25 26 27 2 113.6 0979 1300 1526 01911 0171 1% 0630
7 60 0 97
NUS cELL . 23 211 «0698 l 525 1079 3357
ENTRIES 0F “BLE I M ‘ L 9 20 21 22 99 0173 1299 0973011111 1299 107g 8580 1535 0982 12313? 0070 0213 (22312
1 : cEELL _ m 8 l 0815 O9 0830 073:, 0712 027 508 01 6 2272
DIFFLRENCES . 11 16 17 1 01186 1750 0816 0318 0213....‘0588 109 0873 0 11 2031 02 7
MATRIX OF 15 1571 1305 16119 2930 0063 OOSLL ' 682 1902 l 23 2512 21 . 870 19118
TABLE III. 12 l3 1.1 21 0007 1352 88 1883 0939 8 0668 0175 6 1003 06737 1 2008 0676 35 1 1279 1h75 O 0828
11 7 685 0230 00 11511 1899 1-1 8 8 1865 09 7 0 1887 092 2030 .2889 38 0813 050 011 1619 01611 11
9 10 x 2 9 1869 2 01115 O l 1698 2 7 6 01132 053 1853 1.1111 .- 85 2752 15 1869 O7 . 2956 003
6 8 28' O 9 2 07119 0309 1111 0961 009 8 1251 079 2055 1911* 1111 2312 21 8 1872 0913 60 0768 0830 1379
1 1+ 5 0399 ©th 09 71068 25 7 2830 02711 30 03 1288 099. 1 1919 2310 2188 0351 90‘ 39 0513 190 11111 1088 on 3827 1668
1 2 9 6M3 11.79:L 1805 0693 2077 0113 h 1012 00 8 0515 157 , 0.8 1181 0596 8%? 88 1562 2 ]_ 1377 2037 136M 006h
100818 2987 0 19911 1 0568 15 3 212 O3 3 11182 O 3 011118 2359 0,1525 10 2110 080 227 011711
0808 51 0905 668 06351 285 05811 0836 l 68 03111 0223 01123 ., 0553 3 6 1.186 l 0
01151 0081 620 1807 09 518 11116 0 88. 2 8 0867 1509 299 2210 03 8 3121 1.11111 60 01158 1759 811-1 011511 1937 81 0189 075 030 07112 237
2 0656 9 9 58 0297 02 605 13111 1 275 05931 0162 1569 0 272 1095 0 820 3633 089 621 2976 2 83 0322-10 0500 0210 2132 0891 0 829
2 1276 101111 13 8 0 2738 O 3 0885 O . 1357 31167 2500 01113 1 025 0832 0 23 0117 l 3 23611 19 m . 0650 O916 0159 0173 0929 27110 0
9 0656 1677 O 7 227 01112 197 0 0060; ~ 0871 1199 0218 1062 0 0282 3352 lb 83 0309 011 127931320 1089 0177 0159 0763 1928 11011
2 0593 0859 101111 0828 1 011331 0982 175 0.679 1119 087.0 03 1311 811 1898 O3 0610 01811 - 3:800 0233 0998 02113 0679 0139 2511
, 6 1907 ' 6' 3359 O 3166 in 3812 52 8 2600 86 0813 g a h 0170 26 01131 h 013M 1
5 91151 0302 1358 1977 0380 12113 292 0860 087 1129' 22 3 5 0530 02111 8 03211 0028 099 11112 003 259 2576 00 203 311899 399 2 23 1229 O3 283 1120 050 8
6 0081 1 0870 0828 0532 O 2 01113 1397 , 119111 031 03117 0296 263 26 021.13 0553 927 0692 11211 O 8 0279 2086 0 11.111118 0332 926 13911 3023 O 0 011 0126 168
620 (329/ 8 12117 0380 2 08h 07 1698 10701 05118 1379 863 1537 06 359 0008 OS 628 1881 079 811 1091 2113 ”1189 1698 23 1358 0290 9 311116 19113
8 0808 O 0205 273 2 1283 953 8112 O“ 835 1309: 53 0119 l 0 98 0103 1 1108 3266 O 812 2965 00 0790 0318-. 0 01196 1187 1062 21451 1197
87 1807 0605' 010- 0M9h 0 O 7 507 1350 03 h 5288 7 73 1901 01. 0781 0 20 0602 Ar ,2 0589 2573 8 0513 2090 0h69 O97h
9 29 0951 0217 1973 01133 60 01113 01107 12361 O 667 0033 030 1136 1272 03 266 2103 37 2067 0729 ll 1 068319927 01133 100 0230 0825 00811 0519
10 0683 2518 13111 0982 08 0835 . 21001 0 31 1080 0 2 13 0389 l 1808 0982 1 0986 081 81", 0708 1015 6 0786 09119 8 1020 1371 6 2115
0905 208 0885 1698 238 7 1331 03 1229 9 0893 1155 15“ 0519 03» 2167 33 9 0523 02 9 1213 269
11 179i 1 11116 1 876 1397 1309 l 1‘ 08 011.119 0985 1611 82 05311 817 0300 32., 1250 1 0360 2 05116
99LL 1750 0 2 1070 507 100 1 1 O3 1 1525 1017 00 0980 l 0638 O9 - 07511 030 367 03 7 2383
12 0399 8 Crib-9 0275 60 1907 11.1 9 11.1 05128 0 06671133 8 CELL 0898 211.7 8 1191 0632 0952 OSYLL 88 15701M85 6 11181 0191 3 1h 2727 011-7]- 286
1805 096 88 0593 00 0879 2263 19 1370 1358 310331 030 1185 1771 807 0012 337 23 0055 1860 21 13118 31 0211123987 395 2506 0856 19 0729 0877 l
13 011120 0693 0635 O? 1109 1357 6 1 1059 0.315 03117 0353 003 1 OM49 05111 0177 O 6 0676 05,142 27 0902 1818 23 6 0322 2016 J .2 J 1111 3550 0355 015 2389 2963 69 0382 1065
11L 0928 -068 21177 2285 0838 31167 97 01119 0530 0301111080 229 1525 1185 0177 293 0678 2209 OIL? 0365 OBSLL 169 0817 298.9 13.01167 1207 011111 08143 32“) 0852 1356 Ogoé 0511-3 0683
15 0299 8527 0113 0588 88 0162 1199 * 0296 0119 2118101136 1 0985 1771 2936 0060 2835 1089 1659 23369419511 3 211611 0928 0 2117 120” l 1017 01413
3 8 O 7 870 12111 1863 5 . 1, 2 2 29113 898 0607 0678 2 1079 632 07111 ; 7, 6 1687 85 2328 7 07119
16 1869 h 2830 056 6 2500 0 66 2630 0798 117 8 1611 0 12 0676 — 1238 133 01 1538 O 25 02651525 9 0781 01 6 1916 027 0160
.. 2885 07 9 1509 15 9 0218 31 211 1537 ‘0373 03 9 2117 00 0060 09119 211 8 07111 09 78-0767 1075 0015 025 076 0666
17 11 15113 0635 01113 2 11103 03 826 0103: 1286 0893 05112 2209 11138 1953 15 7 85 0090 137* = 1 01128 11129 526 0120 O 171 1158
0309 027 12 0636 1272 199 00211 O 559"1901 17 3378 0119 21135 09119 2530 28 L12 09511191 011 1105 O 211 0750 2 82
18 0230 301111 1.0 2 1836 25 1311 0113 l A 1808 JD 1 2723 85 1332 53 0578 26 - 0307 19 5 1963 00 8 1309 11
01115 003 121 1095 00 62 31112 ; 1108 2103 0082 119 0902 03 21101 19 0 21.2275 012? 2920 083 0320 039 3917
20 0007 1899 0091 12 8 0515 2210 O 3352 528,11 9 9 0527; 32 8 0781 2067 053 0980 11160 121 1696 1089 2885 0578 86 09753059 1086 1303 0211 3087 0596 096 22011 1213 3
21 1352 1698 099 0188 08110 3 1898 099 0892 7 062 0729 1155 0952 23 0632 071“ 0 26112 214 75 2251 1637 1010 0 0058 20 9 6 0703
1188 0368 633 lhh 1122 - ; 881 08112 1582 1817 22 1659 0925 009 0532 O9 19* 0295 0331 190 17 02h
22 1571 8 1251 1571 21 1598 3 0117 0383 8 1121111 2965 SL129 5711 13118 O3 0817 will 1377 0967 5 0596 215. ;; 0899 0959 1215 23 2212 11117
23 1305 0939 l665 ‘ 2 2310 1118 1111 0562 l 8 0113 2690 -. 0088L 0602 11 05119 063 1570 02”“ 29 1958 2569 O7 01128 0127 ‘ 8 7 0899 162“ 107 1192 3866 O 3 0952
O 0223 611 0610 .2086 109 8 0683 O3 0 1185 01m 3507 29 0307 2146. 19,10 52 16% 1531 0177 11173
25 1750 2930 0598 058 16111 11111 8 2800 23 01811 0086-. ' 21131 031 27 0708 125 511 3965 1 211611 0781 in 19011 2920 O 7" 087 1900 15 1076 1978 0315
26 0815 6 0175 16 7 66 oth OAS 17 1693 0813} OM93 MB 0089 O9 015 2M6? O7 550 Olhh 0185 0015 0836 217 [13 0058 0959 1192 1531 0998 0675
081 853 21 1759 O7 22 1279 114899 11.1, 8 0589 l 1128]. 3 113 0928 256 1105 0014—6. 0596 1215 0130 15 067
27 0999 63 0929 l 0351 2359 23 12811 03 1120 385°- 11 0332 199 3369 0301 1 0355 08 2110 2328 O 0526 1963 0 ‘ 9211 2069 866 0228 O3 11 3
0318 00 003 11111 1 0596 on 1511 101—19 0233‘. 399 2573 01133 6 0360 019 506 3015 0 21117 1916 00211 032... :0: 17 0639 3 85 0177 099 067
28 0173 00511 1 .030 00111 1525 O 29 0650 2 23 2326 1008 078 3367 2 8 0852 27 0120 -~ 2200 23 0837 07 3 0675 3
6 0830 0673 2 12 1039 1937 17 10170 29 8 2 01196 119 0523 0856 23 9 6 12011 0 7 03978.3.10887 02116 22111 0952 1117 119100
29 129 0737 0213 682 2889 23 0513 1088 0916 1089 .. 8 1229 139 8 0513 O9 289 03117 2 63 135 076 0750 ~ 2611 1213 11117 0925 85 1171118
. 1 9 23 ll 7 0230 O 6 119M 9 o7h9 O 1 1309 83 0703 5975 009
30 0973 0568 l 2008 2185 0553 2381 00 01771 09 0326 30 2090 21 05h 6 1606 666 217 2 17 328A 37272 3
8 011111 1902 8 1562 0189 05 02113 283 1358 825 10 99 07 9 10117 0 68 1182.129. 8807 51029
3]. 069 1299 0712 6 2752 190 2111-11-1 32110 6 0210 0159.. 0159 01131 O 0290 1062 O 1213 2727 O7 0382 DELI-3 01a) ll .' ' 1 fr? 8 119067 3
* 1 3 210 5353
32 1602 275 1109 067 1536 1872 088 0801 0756 1690 OlZBW 63 0679 1120 1 OM69 1371 2696 OM71 0677 1065 0683 Ohl 01331 36101 60
A 6 1075 O 0873 3523 813 0913 l 1377 118 663 2132 07 11 0908 2”; 09711 0081* 2383 1286 115615352112 1
.33 3—1—3 051115 0580 508 2512 O 1869 01160 037 1227 0 10929' 0139 050 0126 34116 0519 21115 5 120561 L560? - .1
38 0979 1079 1535 0182 2111 0503'- 07011 0768 2 L1 0030 0691‘ 2720‘ 1928 Ol3h 1888 19113 1197 50629 57222 58142 _ ‘
3? 0:961? 01911 01er (310:0 0267 187% iii 2956 311—399? 0321: 2370 0:2: 118521117315 37357 311598 11 I
7 0311 ‘ 35 w
0171 3357 0213 2272 0 8 0828 O 11 311197 ..
38 0195 0897 0316 1911 55386 1119 ,.
20 2077 19 38995 38569 1195211 7
7 ‘ .
38209 3957
"t’bedc
_‘~ Th decimal points are Oml
Note: e

 

\ A

x

254
”Jun“ A

‘0 VI. “ fl

TABLE IV. TYPES DEVELOPED THROUGH LINKAGE ANALYSIS FROM TABLE III
. (MATRIX OF DIFFERENCES) _ . .

 

 

Type I ‘ o 18
, 30 .2V 29
° .36 P .31; ° 21
Type II
.3h 39

1h 42.) 19 / 3

32 22 .30 5
I'ype III J40 014-0 / 6

Type IV 17
- 7
.30 alt/L;
35 .32: 38 E .38 ll $.30 27 E.26 13
Type V 10
.35 .36 l 2 6
16 -<-—-—-. 21* ‘V‘LL 36
1.30
7
.29 29
VI
We 6 :33: 3h/ 3
Type VII 2
. .301 3
9 fi 25

Type VIII ‘ .29

 

i

LI

.8380 98 mafia 850% BE

ll

«mpoz

soma I maesaoo pmomhwa :mmphfinp pmnam mo nmmm
moma I mEBHoo 503m 05 a .8 59: 88:3

 

 

SHEEN HHEHHHHH HHHE NHHEHHHNHHHN >finpmflox§m
mmHH 34H 83 $3 92 :mHH mmoH ~me omnH 8NH wmmH NQHH SNH RHH 93H NmHH .. 2
onH SH? mNHmH :33 :38 88H mmém Smmm wmmHm HHNmH :mHoN €me SmmH omHmH wmomm 38H .. H309
83 080 5% 08H 88 ago 58 HEN 88 $8 H80 :30 80H $8 1.8 mm
3% 4me 50m 88 38 mHom 8mm mHmo 88 28 $8 38 HHS «Hmm mSH mm
ammo 38H 08H $8 88 38 mmmo 80H mNNH ammo E8 $8 NSH 8mm Ema :m
Sac 50m 88H $8 8mm :mmH- $8 88 24H m8: 8% 30H 83 $8. 88 Hm
08H 83 $8 $8 mmmH 88 ES 88 use HmmH :mHH 88 .3...” mmom cam 0.0.
$8 28 88 $3 mmmH 38 mmmH ammo 58 £00 goo $8 HNHm OHS 90H mm
83 mHom 9.50 am? 88 $8 an: HmHH 8R 0on Smo @me mmmH mHmH ammo a
SS comm $8 $8 EOH NmmH wmfl 8mm 88 mqfl Q80 8% £8 HRH EmH mm.
HEN mHmo mqu 88 88 N80 HmHH 2mm mNNH 030 Ram 08... 38 wmoo E8 mH
R8 88 mNNH 93H «£0 £8 8% quN mNNH HmmH $8 mRH H30 ammo Emu 0H
88 28 38 83 waH 800 88 wOHH was HmmH 88H Ewe $8 32 80H mH
Homo $8 :8 8mm 4a: N80 $8 @000 3% Sec 80H. 89 58H ammo 88 :H
:30 wmmo 88 20H 88 $8 mmmH 6% 08m QMH Bmo momH 88 mmmo :mmH NH
80H HHS mgH mmoH fiHH HmHm wmmH £8 98 H$o goo SH 88 SNH SwH m
$8 «Hm... mmmm $8. mmom OHS mHmH HRH «moo ammo 8: name memo SNH mono o
28 EOH mqmo 4H8 ommm 30H ammo QmH 38 53 $8 $8 :mmH 88H mono m
mm mm :m Hm om mm Hm mm mH 0H mH 4H NH m o m mEPH

 

'
III!

 

 

 

ltl'
IL

‘l‘llfl
III,

M!

 

lull

 

IL

mdem A4oommHomm mo NHmadz Ab mqmdy

113

TABLE VI. sums AND AVERAGES IN DESCENDDIG ORDER OF COLUMNS OF IABLE III
(MATRIX OF DIHERENCES) .

 

 

 

 

Number Item Column Sum Column Average
1 31 60210 15811
2 Zh 581125 1538
3 23 57222 1506
h 11 55386 11158
5 32 53538 11409
6 35 51029 13143
7 22 50629 1332
8 6 1195211 1303
9 140 119100 l292

10 33 M9067 1291
11 15 b.8521 1277
12 19 118102 1266
13 39 1171118 12h8
lb 16 h7315 12115
15 27 175615 1200
16 26 115601 . 1_200
1? 9 1135811 13M
18 20 112021 1106
19 12 111911 1303
20 8 81608 1095
21 29 111331 1088
22 38 M985 1079
23 25 110561 1067
211 3 39577 10112
25 311 38607 1016
26 5 38569 1015
27 2 38309 1008
28 17 37357 983
29 36 37272 981
30 10 37177 978
31 LL 36995 97h
32 28 362112 95h
33 30 36101 950
3h 37 35975 9h?
35 1h 35668 937
36 18 314598 910
37 13 3M9? 908
38 21 33870 891

 

Mean of the first sixteen items - 13113 7
Mean of the first thirteen items a- 1373

 

Note: Decimal points are omitted.

8386 8.8 mpfiog 85809 882

 

mama .- mgaoo 8me3” 83.5.8. 8.5“ .8 88:
Emma I 88380 58.5mm 08. .58 H0 885 8886

 

 

 

 

H85 8 NH EH HHH H E E HH HE E HHHM Hun p M H 1.8268
H8H 88 8H 82 88 :mmH mmHH 88 A88 08H A88H 8mH 8.: 8H 88H 88
EH8 8H8 8H8 8H: 988 :88 88H 38m mH8m 88mm 888 898 H88 8.8 H88 888 .. H889
80m 88 88 88 Sam 98 88 8me mmmm E8 m8H 8HH 88 RmH 8H8 0:
86m 88 88 mHmH 8mm 89 80 E8 88 88 88 88H 38. 18H .88 mm
88 88 NmmH 68H 80m 88 8mm 88 8mm 88 S8 88 88 88 «Hmm mm
88 88 88H 88 88H 88 H80 88 8mm H88 88 88 2.8 88 88 mm
18m mHNH 68H 88 88 88 82 88 83 88 88H 88 8% 8H 88 mm
Sam 88 88 88H 88 HmHH 88 8mH 88 8mm 88 88H 88 88 88 Hm
88 SS 88 88‘ 88 Ha ommm mmmH 80H 88 :8 88m 88 8.8 8H 8
88 88 8mm 88 82 88 emmm 88 89” 88 88 88 88 88 88 8
8me E8 88 88 88 8% mmmH 88 88 88 HmHH 88 88m 38 mHmH 8
8mm :8 88 8mm 80H 88 86H 8H 88 88 8mm 88 89H 88 HRH mm
88 88 8mm 88 88 8mm 88H 88 88 080 N80 8me HomH 88 mHmo 8
m8H 88m mHmo mm8 88H 88 8H8 88 HmHH 8mm 88 88 88 88 88 8H
83 88 88 38 88 88 88 88 88 88 88H mmNH HmmH 8H8 88 OH
88 28 88 88 8mm 88 88 88 88 88H HRH 88 HmmH H8H 8HH mH
amH 18H 88 88 88H 88 88. 38 88 8mm 88 88 B8 H8H m8H HH
8H8 88 N88 88 88 88 1LHmH mmom mHmH HRH mHmo 88 88 8HH m8H o
3 mm mm mm mm Hm 8 8 8 mm 8 mH 8H H H w 288
888888 .8 898v HHH 889 E 28-2888 5888 m8 8 888 8 H88 .85 88

U

I
‘
1
‘ ‘
‘ J
‘
‘
I ‘
A
n

TABLE VIII.

MATRICES OF.ALL THE ITEMS APPEARING IN

EIGHT TYPES

 

 

25022;;
Items

12
23
33
2O

18
21

Total

Expo II

Items

Total
M,
GM

12

S28h
0322
2381
032a
0296
002a

23
528h

2989
3550
05h2
1017
2209

30

0322
2989

1303

'0932

0683
1570

33
2381

.3550

1303

3369
2573
0301

20

O32h
05h2
0932
3369

1771
2936

18

0296
1017
0683

2573
1771

0898

21

002u‘
2209
1570
0301
2936
0898

 

8631
1233
1010

lb

52h8
0122

15591

2227

19
52h8
3hh6

7799
111k

39

0122
3hh6

 

5370
1790

1959

15

h899
399u
1901
0629
2285

869k
2898

31
h899

0295
3967
3917
1682

3568
1189

32

399h
0295

3965
328h
1902

13h77

1925

22

1901
3967
3965

21h5
0998

’987u

no

0 629
3917
328k
21h5

2970

7238
lOBh

2285
1682
1902
0998
2970

 

- 13708

2285
2158

1&760

2h60

22h0

2163

lth0 12976 129h5

2158

9837
16h0

7938
113h

 

Continued

1|

TABLE VIII - Continued

Total

35

3866
0801
1535
3023
0256
0210

38
3866

3827
3357
OSOh
07h9
0030

11

0801
3827

2518

0530
2976
1698

h

1535

3357
2518

2830

0175
0668

3023
OSOH
0530
2830

1120
0353

27

0256
07h9
2976
0175
1120

2600

0210
0030
1698
0668

0353
2600

 

9691
138k
1372

16

3701
1059
3h67
0679

0h31.

12333

1762

2h
3701

3633
03hl
2963
2389

12350

176h

10

1059
3633

12h3
0768
0060

11083

1583

3h67
03hl
12h3

1279
0501

8360
ll9h

37

0679
2963
0768
1279

1531

7876
1125

36

0&31
2389
0060
0501
1531

 

9337 13027

1556
1358

2171

Bh
3523

3523
1111 2920

7163
1190

29

2920

 

h66h 6003
1555 21h8
1060

u061
677

6831
1139

7220
1203

5312
886

5559
79h

Continued

H

V

TABLE VIII - Concluded

 

 

Total

2129 211;

Items

26
28

Total

3121
2987

25
3121

1750

2987
1750

 

6108
2036
1706

2930
0318

h871
1629

26

2930
2885

h737
1579

28

0318

2885

 

3228
1083
1363

5815
1938

3203
1068

Over-all Mean 8 1588

 

Note: 'Decimal points are omitted.

87

4i

 

118

63920 6.8 35.86 Hmfiomc

«opoz

 

 

 

HHHH NH HHS HHH > HHH H HH B HE EH E E H. HM EH 182ng

68H NcmH ecmH cmmH cm: HccH cmcH HmcH Hm: cmmH chH HHmH :cNH 62H «RH 53 n 2

$ch mmcfi :85 :30 SEN icon Rmcm 28m @6ch HRH.“ 33H mmcfi cmNmH $15 33m 83H .. H38
mSc mmmc mcsc 18mm 5% chH 3mm 33 EmH :cHH amcc $16 68 a? £8 c:
memo comm mem scam ammo mmsc emem mHNH Hmcm mch mmmc Hmcc zwac swam cmmc mm
mmmc 8% «RH ccmH 1.ch chm comm 3mm mHmc cmmc Sac mch cmsc Hccc cccH mm
mcsc SQ NmmH ammo 50H fiHc cmmm Hcfi Rec 2mm 28 $3 Hcmw mmmc NcmH mm
:mmm :cmm cccH ammo memo econ qzca cmmm chH Nmmc :mmm mmmc mmsH smmH mwcH mm
mem wwcc pmcm emcH mmwc :mmH pogo comm coco wczH mac: cccm mch cmzc mmmH Hm
ccmH mac chm 133. econ :mmH can mad HmHH Hcem comm ammo cmcH 3: 33 :N
$8 $3 ccmw cmmm :42 $16 cmfl 88 3.9 mch ccHH cccc 5mm mmmm 9:6 mm
qum mHmH Ncmm quH cmmm scam mmqm coco NHcc ccmH HcmH mmmH «Hzm Nose ammo mm
33 HEN mHmc mmac chH ccac HmHH 3mm NHcc RS 86 2mm cmcm 6ch cHNc mH
.3: mch cmmc 3mm Nmmc cfiH HcR mch chH mNNH HRH Scc mRH mHmc mmcH 6H
mmcc ammo Sac 88 4QO mac: 8% chH HcmH 63c HRH HccH memo HaH SNN mH
63c Hacc $8 $9” mmmc comm ammo cccc mmmH clam $8 39 mcMH 92 man 4H
68 1:3 cmec Hcmm mm: 39 cmcH 13mm «Hem cmcm $2 czmc mcmH 34c 28 NH
R2 Sam Hccc mmmc NRH 4ch 3H mmmm ~ch 6ch mHmc daH chH 35 Eco HH
emcc cmcc cccH NcmH cccH mmmH mmcm 06c «Sc 38 $2 0% mmfi mfic memo 2
2 mm mm mm mm Hm 1a mm NN mH cH mH fl NH HH cH 23H

 

 

GEES ac EREV HHH Ema E

995900 Emma 2H gezm egg UzHDfim mafia” mo E942 .NH Ema.

TABLE I. m X 'IWO TABLES 0F THIRTY-EIGHT ITEMS SHOWING CELL
FREQUENCIES ‘ .

W

 

 

Item Cell Frquencies
Number .A B _ 0 D

2 5h 63 28 25

3 16 lb 66 70
h 39 13 h3 75

5 5h 53 28 35

6 h3 35 39 53

8 h8 55 38 33
9 35 h9 h? 39
10 19 28 63 60
11 50 no 32 08
12 h9 63 33 25
13 52 51 30 37
1h h3 h3 39 h5
15 53 67 29 21
16 53 58 29 30
17 h? 53 35 35
18 h6 52 36 36
19 21; 37 61 51
20 3h no 88 08
21 56 51 26 37
22 56 56' 26 32
23 23 29 59 59
2h 38 06 0h h2
25 32 23 5O 65
26 51 53 31 35
27 39 18 03 70
28 28 3h Sh 5h
29 20 h? ‘ 62 21
30 09 an 33 hh
31 25 26 57 62
32 35 38 h? 50
33 27 25 55 63
3h 38 03 uh h5
35 . 39 38 h3 5O
36 66 66 16 22
37 h5 h5 37 M3
38 25 31 57 57
39 25 29 57 59

no h2 56 no 32

 

TABLEiXI. RANKED CHI-SQUARE VALUES*'FoR.1HIRTYaEIGHT ITEMS

 

 

Item Chi-Square
h 19 .9 75
29 13.778
27 12.803
19 2.395
11 3 .502
25 2.659
9 2.372
6 2 .255
20 2.195
15 2.178
12 2.122
21 1.526
30 1.259
20 1.183
36 .252
2 .211
22 .382
5 .359
13 .325
23 .277
38 .223
22 .227
33 .222
28 .200
35 .172
3 .170
8 .137
20 J35
37 .111
12 .096
18 .056
17 .051
39 .032
32 .030
26 .010
32 .007
31 .000
16 .000

 

*Yates correction for continuity applied.

TABIJE XII. SPECD’UEN PATTERN 0F RESPONSE (PRODUCED BY MISTIC)

51

 

 

012
011
010
009
009
008
007
007
006
222 193 192

010
0N1
0N2
OJN
0J0
0F6.
0+5
0+6
0N3
0+7
mowm08123m1L11

220

208

230

165

166

195

167

 

52

ACADEMIC ACTIVITY PREFERENCE INVENTORY

Items

1.

2.

Studying during free hours in the day, so as to reduce the evening‘s
load. -

Believing that my parents would sooner have me work than go to school.

3. Discussing books with friends.

11.
5.
6.

7.

8.
9.

10.

13 .

114-0

15.
16 .

17.

Going to parties where couples are expected to pair off.
Staying away from school activities in which I don‘t do well.
Going along with a chairman’s decision rather than starting a fuss.

Horking on tasks for long periods of time, without interruption or
diversion.

Having friends who are inferior to me in academic ability.

Cutting classes when I need to cram for a test.

Learning to repair such things as the radio, sewing machine, or car.
Considering studying as important as work I will do later.

Participating in a discussion that is exceptionally logical, precise,
and coherent.

Pretending that I agree with a teacher after I see that he has his
mind made up.

Giving up on a problem rather than doing it in a way that may be
wrong.

Feeling that examinations measure what I have learned.
Feeling that examinations measure what I know.

Changing my answers on examination questions.

18. Doing more constructive things than studying.

19.

Going to school.

20.
21.
22.
23.
2h.

25.
26.

27.
28.
29.
30.
31.
32.

33-

35.

36.
37-

38.

53

Relying on specific class assignments to spur me on to accomplish
things.

Keeping to a regular schedule, which means working'When I don‘t
really feel like it.

Believing that teachers, on the whole, are fair in the ways they
grade.

Spending a good deal of my time on activities which are amusing but
of little practical value.

Preparing for examinations by first taking time to arrange the
facts I must learn in some logical order.

Reading great novels written in the past.

Searching continually for the source of difficulty in a problem
until I‘ve located it.

Wbrking in science and mathematics rather than art or music.
Trying to develop a sincere interest in every course I take.
Laughing at a dirty joke every once in awhile.

Reading books which stress adventure.

Sitting around and thinking.

Giving all my energy to whatever I happen to be doing.
Spending some time to get "warmed up“ to the task of studying.

Believing that my parents regarded going to school as important as
working.

Setting a goal as to how much material I will cover before each
study'period.

Fixing things around the house.

Looking'up things in original sources in order to find out for
myself.

Completing assignments if they are boring and dull.

 

”'7117777777' 1177777777725