||

|

1

l

i

I

l
'!

WW“!

I
‘
f

E 1i

 

THE iNFiUENC‘E C??? DC-GMA‘HSM AND RIGEMYY
GEN PLEfigSQ-i‘éfi-NG

Thesé: {For $315 Degree a? M A.
MlCHiGA’N STATE UNIVERSITY

Robert Paul Beach
@964

. "3V 2 9 2001

 

 

 

THE INFLUENCE OF DOGMATISM AND RIGIDITY ON REASONING

ABSTRACT

To test the relationship between synthesis and analy-
sis and induction and deduction with dogmatism and rigidity,
Rokeach's Dogmatism Scale and the Gough—Sanford Rigidity
Scale was administered to 3&1 undergraduates at Michigan
State University. The 20 students scoring highest and
the 20 students scoring lowest on the Dogmatism Scale
were selected for individual testing. Both groups of
subjects were further divided into two groups of ten each
on the basis of their scores on the Rigidity Scale.

The first task consisted of a set of twenty-five
analogies emphasizing induction, and another set of
analogies emphasizing deduction. The subject's per-
formance on this task was measured in terms of the
relative amount of time Spent on the preparation period
of the problem and on the solution phase of the problem
as well as the number of errors made. The second problem
was the Denny Doodlebug Problem with separate measures of
analysis in terms of the time required to overcome the
beliefs of the problem and the number of beliefs the sub-
ject could overcome by himself, and synthesis measured
in terms of the number of minutes required to solve the
problem after each of the beliefs was overcome.

The general conclusions of this study are that: (l)
in a problem solving situation in which the primary task

of the subject is to integrate various beliefs of component

parts of the problem Open subjects are superior to closed
subjects; (2) in a problem solving situation in which the
task of the subject is to analyze or discover the parts
of the problem, non-rigid subjects are superior to rigid
subjects; (3) there is a tendency for subjects who excel
in synthesis to also excel on inductive reasoning; (b)
there is a tendency for subjects who excel in analysis

to do well on problems emphasizing deduction.

/ .
Appr oved Mk/égidaz

Date 7%; as; we!

THE INFLUENCE OF DOGMATISM

AND RIGIDITY ON REASONING

by

Robert Paul Beech

A THESIS
Submitted to
Michigan State University
in partial fulfillment of the requirements
for the degree of
MASTER OF ARTS
Department of Psychology

196M

Page 32

Table XIII.

BREATA

Kean Solution Time for the Denny Doodlebug

Problem Following the First, Second, and

All Three

Beliefs.

 

 

Group First Second All Three
Belief Belief Beliefs

High Dogmatism-High Rigid 21.83 15.95 8.65

High Dogmatism-Low Rigid 28.36 24.48 14.90

Low Dogmatism-Eigh Rigid 19.22 12.95 5.19

Low Dogmatism-Low Rigid 11.31 8.31 1.86

High Dogmatism 25.095 20.215 11.795

Low Dogmatism 15.265 10.630 3.525

High Rigid 20.525 14.45 6.920

Low Rigid 19.835 16.395 8.uoo

Page 39 Chi Square =.226

Page #0 Chi Square = 1.034, p less than .15 one tailed test

Page 41 Chi Square = 1.096,

Page 42 Chi Square = .510

Page b# p less than .70

Page 44 Chi Square = .083, p less than .80

Page 50 In Table XXXIV, the time taken to solve the Doodlebug

Problem after all three beliefs had been overcome for
the Nonrigid group = 2.6

INTRODUCTION.

METHOD. .

RESULTS .

DISCUSSION.

SUMMARY .

REFERENCES.

APPENDIX.

TABLE OF

CONTENTS

Page

LIST OF TABLES

Table
I. Comparison of Means and Standard Deviations on
the Dogmatism Scale. . . . . . . . . . . . . . .
II. Comparison of Means and Standard Deviation on
the Rigidity Scale . . . . . . . . . . . . . . .
III. Comparison of the Mean Number of Errors made
on the Problems Emphasizing Induction. . . . . .
IV. Analysis of Variance of the Errors made on the
Problems Emphasizing Induction . . . . . . . . .
V. Comparison of the Mean Preparation Index on the
Problems Emphasizing Induction . . . . . . . . .
VI. Analysis of Variance of the Preparation Index
on the Problems Emphasizing Induction. . . . . .
VII. Comparison of the Mean Number of Errors made on
the Problems Emphasizing Deduction . . . . . . .
VIII. Analysis of Variance of the Errors made on
the Problems Emphasizing Deduction . . . . . . .
IX. Comparison of the Mean Preparation Index on
the Problems Emphasizing Deduction . . . . . . .
X. Analysis of Variance of the Preparation Index
on the Problems Emphasizing Deduction. . . . . .
XI. Analysis of Variance of the Errors made on
Problems Emphasizing Induction and Deduction . .
XII. Analysis of Variance of the Preparation Index on
the Problems Emphasizing Induction and Deduction .
XIII. Mean Solution Time for the Denny Doodlebug
Problem Following the First, Second, and
All Three Beliefs. . . . . . . . . . . . . . . .
XIV. Analysis of Variance of the Time Required
to Solve the Denny Doodlebug Problem after
the First Belief . . . . . . . . . . . . . . . .
XV. Analysis of Variance of the Time Required

to Solve the Denny Doodlebug Problem after
the Second Belief. . . . . . . . . . . . . . . .

Page

16

18

2h

25

25

26

27

28

28

29

30

31

32

33

33

Table

XVI.

XVII.

XVIII.

XIX.

XX.

XXII.

XXIII.

XXIV.

XXV.

XXVI.

XXVII.

Analysis of Variance of the Time Required
to Solve the Denny Doodlebug Problem after
All Three Beliefs. . . . . . . . . . . . . .

Comparison of Mean Number of Minutes Taken
to Overcome the First Belief, the First
Two Beliefs, and All Three Beliefs . . . . .

Analysis of Variance of the Time Required
to Overcome the First Belief . . . . . . . .

Analysis of Variance of the Time Required
to Overcome Two Beliefs. . . . . . . . . . .

Analysis of Variance of the Time Required
to Overcome All Three Beliefs. . . . . . . .

Analysis--The Number of Beliefs Overcome
by the Subject in the First Fifteen Minutes
of the Experiment--Rigid vs. Non-Rigid . . .

Analysis-~The Number of Beliefs Overcome
by the Subject in the First Fifteen Minutes
of the Experiment--Open vs. Closed ... . . .

Synthesis and Induction-~The Time Required
to Solve the DD Problem After All Three
Beliefs Had Been Overcome and the Number of
Errors Made on the Problems Emphasizing
Induction. . . . . . . . . . . . . . . . . .

Synthesis and Induction--The Time Required
to Solve the DD Problem after All Three
Beliefs Had Been Overcome and Preparation
Index on the Problems Emphasizing Induction.

Synthesis and Deduction--Time Required on
the DD Problem after All Three Beliefs Had
Been Overcome and Number of Errors Made on
the Problems Emphasizing Deduction . . . . .

Synthesis and Deduction-~Time Required on
the DD Problem After All Three Beliefs Had
Been Overcome and Preparation Index on the
Problems Emphasizing Deduction.. . . . . . .

Analysis and Deduction-~Comparison of Number
of Beliefs Overcome in First Fifteen Minutes
of DD Problem and the Number of Errors Made

on Problems Emphasizing Deduction. . . . . .

Page

34

35

36

36

37

38

39

ho

bl

#2

1+2

’43

Table

XXVIII.

XXIX.

XXX.

XXXI.

XXXII.

XXXIV.

XXXV.

XXXVI.

Analysis and Deduction--Comparison of Number of
Beliefs Overcome in First Fifteen Minutes on DD
Problem and Preparation Index for Problems

Emphasizing Deduction. . . . . . . . . . . . ...

Analysis and Induction--Comparison of the Number
of Beliefs Overcome in the First Fifteen Minutes
of the DD Problem and the Number of Errors Made

on the Problems Emphasizing Induction. . . . . .

Analysis and Induction--Comparison of the Number
of Beliefs Overcome in the First Fifteen Minutes
of the DD Problem and the Preparation Index on

the Problems Emphasizing Induction . . . . . . .

Correlations Between Induction, Deduction,
Analysis and Synthesis . . . . . . . . . . . . .

Ability to Analyze: Comparison Between Rigid
and Non-rigid Groups and Between Closed and

Open Groups on Mean Number of Minutes Taken

to Overcome the First Belief, The First Two
Beliefs, and All Three Beliefs . . . . . . . . .

Ability to Synthesize: Time Taken to Solve

the DD Problem after the First, Second, and
Third Beliefs were Overcome by Rigid and
Non-rigid Groups and by Closed and Open Groups .

Synthesis: Mean Time Taken to Solve the DD
Problem after the First, Second, and Third
Beliefs Had Been Overcome. . . . . . . . . . . .

Analysis: Mean Time Taken to Overcome the
First Belief, the First Two Beliefs, and
All Three Beliefs. O O O I O O O O O O O O O O O

Page

’43

M4

nu

45

’49

50

51

51

ACKNOWLEDGMENTS

The author wishes to express his sincere appreciation

for the time and effort so willingly given in the comple-

tion of this study
committee.

In addition,
Dr. Donald Johnson
and for the verbal
deduction, and for
preparation of the

Special thanks for

to Dr. Milton Rokeach, Chairman of the

the author wishes to convey thanks to

for the use of his serial-exposure box
analogies emphasizing induction and
his constructive criticisms in the
manuscript. To Dr. Bertram Karon,

his invaluable assistance in the for-

mulation of the analysis of this study.

THE RELATIONSHIP OF DOGMATISM AND RIGIDITY ON REASONING

There is a general distinction made by psychologists
and others interested in the study of problem solving and
the process of thinking between reasoning from a part to
a whole, and reasoning from the general to the particular.
The former frequently is referred to as inductive reasoning
and the latter as deductive reasoning. Clarke, for example,
states that "The term logic. . .is usually employed in a
more restricted sense. . .means either the process of making
generalizations (inductive logic), or the process of drawing
from assumptions inferences which are implicit in them but
which are not obvious (deductive logic)(1929, p. 77)."
Similarly, Miller differentiates between the two forms as
follows, "Deduction is that form of thinking in which an in-
dividual (case) which is problematic is interpreted and con-
trolled by referring it to some concept, or law, which is,
for the time being at least unquestioned. Induction is that
form of thinking in which a concept or law, which has become
problematic, is reconstructed through an investigation and
analysis of individuals." (1915, p. 233) More recently
Vinacke describes the same two processes as: "Deduction,
crudely defined, is reasoning from the general to the par-
ticular, or, more Specifically, from stated premises to a
pr0per conclusion; induction, similarly, signifies reasoning

from the particular to the general." (1952, p. 76) The

distinction between induction and deduction appears to be

primarily linked to the nature of the problem more than in

the person attempting to solve the problem. If the problem

is one which requires the application of a known principle

or generalization to a Specific case, then the thought

process is deductive. If the problem is one for which an

example has been provided and the problem solver must find

the general rule which is Operating, the reasoning is inductive.
The question then arises whether there are individual

differences in problem solving which are Specific to the kind

of problem being worked upon or are persons who are capable

of one kind of reasoning also capable of the other equally

well? Further, if there are individual differences with

reSpect to the ability to solve inductive or deductive problems,

what other variables are associated with these individual differ-

ences? One technique which was employed to study such re-

lationships was factor analysis. Of the factor analytic

studies on problem solving which attempted to separate these

two functions, the results have not been conclusive. Thurstone

factor analyzed 57 tests taken by 250 college students and

identified an induction factor, a restrictive thinking factor,

and a deduction factor (1938). However, a similar factor

analysis of 60 tests taken by 710 eighth-graders found only

the induction factor (Thurstone and Thurstone, 19u1). Additional

factor analytic studies by Holziner and Harman failed to

confirm the separation of inductive and deductive reasoning

(1938). One of the most complete studies of reasoning
using the method of factor analysis is that done by
Guilford and others at the University of Southern
California (Guilford, Comrey, Green, and Christensen,
1950; and Guilford, Green, and Christensen, 1951). It
was hypothesized that the ability to graSp a system of
relationships in its totality, or to see trends in a
series of objects, to identify a relationship in a
variety of settings would constitute inductive reasoning
ability. The reasoning ability tested in syllogisms was
felt to be deduction. Factor analysis of 3% tests of
283 subjects identified a general reasoning factor of
prime importance while deductive‘reasoning was found in
tests of reasoning, inference, syllogisms, and false
premises, but the tests which were used to define this
factor were multiple-choice and true-false form. No
general induction factor was identified.

Another approach has been to analyze a the process
of problem solving into various phases. Dewey described
five such separate stages in the problem solving sequence
as follows: "(1) a felt difficulty; (ii) its location
and definition; (iii) suggestion of possible solution;
(iv) development by reasoning of the bearings of the'
suggestion; (v) further observation and experiment

leading to its acceptance or rejection; that is, the
conclusion of belief or disbelief." (1910, p. 72) From

such an analysis of the problem solving process it is

possible to formulate different methods of studying

the reasoning process. First there is the possibility

Of constructing problems in which the various phases may

be measured separately, such that an objective record may
be made as the problem solver passes through each successive
phase. Another alternative which is also based on the
analysis of the problem solving process into its component
parts consists of constructing problems in which the diffi-
culty of the problem lies primarily in one phase. Perfor-
mance on this kind of a problem may be compared with
performance on another problem for which the difficulty
lies in some other phase.

One of the major benefits of using a method based on
the analysis of the parts of the problem solving process
into separate phases is that it is possible to take into
account the fact that in most problem solving behavior,
both an inductive and a deductive Operation are involved.
Johnson in referring to problems of logical relations
states that:

"Forming concepts, principles, and patterns

of personality traits may be called inductive tasks

in that the particulars are organized into patterns...

But we do not observe these cognitive patterns di-

rectly; we test for the pattern by asking the thinker

to use it in some particular way, that is, by a de-
ductive process...problems that are called deductive,

e.g., the syllogism, always include an inductive step

because the particulars of any problem have to be
organized before any conclusions can be produced."

(1955.p- 2M3)

Following Dewey's analysis of the problem solving

process, reason may be thought of as the solution of problems

by logical Operations. More specifically, the Operations
which the reasoner goes through in order to solve a par-
ticular problem may be thought of as follows: first, the
reasoner organizes the problem situation and the wanted
solution as well as he can. He then produces reSponses
and finally judges the acceptability of these reSponses.
Induction is the process of organizing the particulars of
the problem situation into a conceptual pattern. An in-

ferenceisofthe form. If thus and so then .

 

Deduction is the process Of fitting a solution pattern
to the inference thus organized. It is reasonable to
assume that individual differences may be found at any
point in the sequence of logical Operations. Consider
the following analogy:

large is to small as expensive is to
fine, post, run, time, cheap

 

First the relationship between large and small

must be discovered. Second, expensive must be placed

 

into this relationship with a place for the other term.
Third, the reasoner tries out the offered solution words
to see which one best fits the search model. If a measure
of deductive ability is desired, it is necessary to con-
struct items in which the third step is suffiently diffi-
cult that some will fail at this point, while steps one
and two are so easy that everyone will be successful at

these points. If a measure of induction is desired, it

is necessary to construct items such that SOme will fail

on the first steo, that is in establishing the relationship,

while there should be relatively little difficulty in

{1
CI.
(0

'U
1;)

8-7)

two and three. (Tohnson, 1355, o 33,

:‘d

Johnson, in an attempt to study the determinants of
preparation time, constructed a block of 25 analogies
with the first pair of words more difficult than the
second pair, hence emphasizing induction. Another block
of 25 analogies was constructed with the second pair of
words more difficult than the first pair, emphasizing de-
duction. Using a technique of serial eXpOSUFC of the
analogies, he found in connarix'lg preparation tirne between
the two types of analogies hat log mean preparation time
was greater for the analogies emphasizing induction, and
log mean solution time was greater for those emphasizing
deduction. Similarily, a time index of relative difficulty
of preparation differentiated the two types of problems
shgnificantly (Johnson, 1960).

In further inv-stigation of the method of serial
analysis of inductive and deductive reasoning using various
formats for the presentation of the analogy, the preparation
index, obtained by dividing preparation time by the total tine
to get the preportion of he whole time that is spent on

the first period,differentiated inductive from deductive

problems thus Sipporting in general the distinction between

these forms of reasoning, and the use Of serial presentation
Of verbal analogies as a technique for studying these dif-
ferent processes (Johnson, 1962).

While the technique of comparing performance on differ-
ent problems which emphasize different phases of the reason-
ing process has been useful in the study of the problem
solving, a number of studies have been done in which com-
parisons were made between the subjects' performance on
different phases Of the same problem. This has been one
of the ways in which Rokeach has attempted to study the
diStinction between Open and closed subjects. In discuss-
ing the Doodlebug Problem he states that one of the reasons
the problem is so difficult is that "the subject must first
overcome not one but three currently held beliefs, and re-
place them with three new ones. This is the analytic phase
of the problem. . .He must then organize them together, or

integrate them into a new system. This is the synthesizipg

 

phase of the problem." (Rokeach, 1960, p. 173) The Problem
is constructed so that it is possible to obtain five separate
measures of the thinker's ability to analyze, all of which
test in one way or another how long it takes the thinker

to overcome the separate beliefs of the problem. Synthesis
refers to how fast the thinker can integrate new beliefs into
a new belief system. When the thinker finally states the

solution to the problem, it is the end result of a prior

process of synthesizing activity. In the Doodlebug Problem,

there are three separate measures of synthesis or inte-
gration.

Another dimension in which the distinction between
analysis and synthesis is profitable is in the study of
perceptual tasks. Witkins and others have indicated that
peOple differ in the extent to which their perception is
analytical. This dimension of individual differences
has been called "field-dependence-independence." The
tendency toward an analytical or global way of perceiving
characterizes a person's perception in a wide variety of
situations, making for marked individual self-consistency.
(Witkin et. a1. 1962, p. 58). The implication would appear
to be that differences in analytic and synthetic thinking
are general personality factors which are related to a
number of other variables. From what has been mentioned
above concerning individual differences in inductive and
deductive reasoning ability, it might be reasonable that
such differences would also be related to some general
personality dimensions. One of the earlier personality
variables which seemed to be related to a person's ability
to reason was authoritarianism as measured by the F scale.
Analysis of the ideological as well as the clinical material
has suggested that ethnocentrism is related to stereotypy,
rigidity, and concreteness in thinking (Adorno et. a1. 1950,
P- 280). ROkeach, in an attempt to expand the measurement

of authoritarianism to general authoritarianism develOped

the Dogmatism scale which tested for general authoritarianism

Of both the right and the left (Rokeach, 1960). With
reSpect to the ability to solve abstract problems, Rokeach
differentiates between the difficulties which arise due to
the subject's resistance to change belief systems which is
measured by the Dogmatism scale, and difficulties which are
primarily due to the subject's resistance to change a single
belief as measured by a Rigidity scale such as that of Gough
and Sanford (1952).

The referent of dogmatic thinking is within a total
cognitive framework of ideas and beliefs organized into
a relatively closed system; while rigidity refers primarily
to the difficulties in overcoming single sets or beliefs
encountered in attacking, solving, or learning Specific
tasks or problems (Rokeach, 1960, p. 183). This dis-
tinction between Dogmatism and Rigidity has been related
to the above mentioned distinction between analysis and
synthesis by Rokeach and others (1955). Persons who
Score high on Dogmatism have difficulties with synthesis,
while persons who score high on Rigidity have greater
difficulty in analysis. These results have been replicated
using the entire range of scores on the dogmatism scale
(Fillenbaum and Jackman, 1961), although certain a5pects of
the problem used to test these differences, namely the order
of presentation Of the hints used in the Denny Doodlebug
problem have been questioned (Lyda, L. and Fillenbaum, 1960),
indicating that the construction of the problem used is Of
great importance. The differences found between Open and

Closed subjects who were either high or low on rigidity has

10

not only been found to Operate on conceptual problems such

as the Denny Doodlebug problem, but has also been related

to perceptual problems following Witkin (1950). Rokeach

and Levy designed problems for which separate measures of
analysis and synthesis in perception were possible which

were found to be related.to dogmatism and rigidity. (Rokeach,
1960) Similarly, Huberman has reported differential behavior
of extremely high and low dogmatic subjects on certain Kohs
block tasks (1961).

After having found dogmatism and rigidity to be re-
lated to analysis and syntheSis in cognition and in per-
ception, it would seem profitable if deductive and inductive
reasoning could be related to analysis and synthesis. Upon
examination of the processes which are measured by Johnson
(1962) it is apparent that those problems which emphasize
deduction do so by making it difficult for the subject to
apply a general rule while in the inductive problems, it is
difficult to produce the general rule or to state the re-
lationship for Which the two words presented are examples.
The process Of deduction is similar to analysis in that the
subject is required to break the existing structure of the
problem into its component parts in order to determine
which of the alternatives Offered indeed fits as a part
in the relationship, and which of the alternatives do not.
Induction and synthesis if not altogether the same process
at least share the common feature in that they are both
concerned with putting the various parts of the problem

together, or taking what has already been given

11

the various parts of the problem. Rokeach, in Speaking
about the distinction between these two interrelated
aSpects of cognitive functioning states:

"There are many ways of talking about these two

aSpects: the resistance to change of systems of

beliefs; rigidity and dogmatism; the analysis

phase and the synthesis phase in thinking and

perceiving; the overcoming of sets and the inte-

gration of sets; the separation of an item from

a field and its reconstruction into new fields.

Perhaps we are also dealing with the process of

deduction and induction." (Rokeach, 1960, p. 289)

If the above description Of Johnson's measures of
deduction and induction as being similar to analysis
and synthesis in which the subject must make a judgment
concerning a relationship between concepts holds true,
then rather Specific hypotheses may be formulated re-
garding the relationship between dogmatism with synthesis
and induction and of rigidity with analysis and deduction.
It is to be expected that those persons classified as
dogmatic, that is resistant to change of belief systems,
should have greater difficulty solving problems empha-
lsizing induction than persons classified as Open or less
resistant to change Of belief systems. Further, it is
anticipated that those persons who are classified as being
rigid Should have greater difficulty in solving problems
emphasizing deduction than would persons classified as
being more flexible. Third, it is expected that the ability
to do well on problems emphasizing deduction should be re-

lated to analysis as measured by the Denny Doodlebug problem.

Similarly, it is expected that the‘ability to do well on

problems emphasizing induction should be related to the

ability to do well on conceptual synthesis.

Hypotheses. The Specific hypotheses of this study are

 

as follows:

Concerning induction:

1. The Open group should solve inductive problems with
fewer errors and faster than the closed group.

2. The rigid group should not differ from the non-rigid
group with reSpect to this variable.

Concerning deduction:

1. The Open group should not differ from the closed
group with reSpect to this variable.

2. The rigid grOUp should make more errors and take more
time to solve problems emphasizing this variable than
should the non-rigid group.

Concerning synthesis:

1. The Open group should be faster than the closed grOUp.

2. The rigid grouP should not differ from the non-rigid
group with reSpect to this variable.

Concerning analysis:

1. The Open group should not differ from the closed group
on this variable.

2. The rigid group should be slower and discover fewer Of
the beliefs without outside help than the non-rigid

group.

The relationship between analysis and deduction should
be demonstratable in that those who do well on analysis
should also do well on deduction.

The relationship between induction and synthesis should
be such that those who do well on synthesis should also
do well on induction.

NETHOD 13

Penulation. The subjects used in the eXperiment came

 

from a pOpulation of over 300 students enrolled in either
introductory psychology, education, or political science
at Kichigan State University during the summer of 1963.

'1"

4: .1, O

Ufl

mati~m Scale. he degree of Open.ess-closedness

(w-
L1

for each individual was measured by Rokeach's Dogmatism

U)

cale Form E1 which can be found in the Appendix. The

(:1

ogmatism Scale (U0 items) along with the Cough-Sanford

" scattered

Rigidity Scale (22 items) and 2h "filler items
throughout the scale to disguise its purpose was adminis-
tered to the students in a classroom setting. To each of
the questionnahwistatements the subjects reSponded indi-

cating how strongly they agreed or disagreed by means of

the following scale: +1 - I agree a little, +2 = I agree

I!

)n the whole, +3 = I agree very much, -1 I disagree a

little, -2 = I disagree on the whole, -3

I disagree

very much. The subjects were informed that the question-
nairerwas part of a larger study and that they might be con-
tacted later. The subjects were told before the question-
naire was passed out that if they did not wish to partici-
pate they could leave, although none Of the students did

in fact leave. The following instructions were printed

at the tOp of the questionnaire:

 

1For a full discussion Of the construction of this
scale, see Rokeach (1960).

11

Rigidity Scale. The degree of flexibility in the

 

ability to change single beliefs was measured by the
Gough and Sanford Rigidity Scale which was mixed in with
the Dogmatism Scale. The items of this scale can be
found marked Rigidity in the Appendix. The instructions
for the entire questionnaire were the same for all ques-
tions, subjects indicating strong agreement with the item
by marking +3 and indicating strong disagreement with the
item by marking -3. The referents in this scale are for
Specific tasks or habits rather than total belief systems.
Rokeach has pointed out Hun;dogmatic and rigid think-
ing are discriminable processes not necessarily indepen-

dent processes. The correlation between the Dogmatism and

\

Rigidity scales rangesfrom .37 to .55. (Rokeach, 13*0, p.
193). The correlation between the Dogmatism and Rigidity
Scales for the subjects used in this eXperiemnt was .7
which is artificially high due to the process by which the
subjects were selected for the eXperiment. The fact that
these are two discriminably different processes has been
supported by the results of two factor analyses, both in-
dicating that the Dogmatism and Rigidity Scales are measur-
ing essentially independent psychological dimensions (as
reported in Rokeach, 1960).

Subjects. Twenty students with the highest scores
and twenty students with the lowest scores on the Dogmatism

Scale were asked to come into the laboratory for individual

15

testing. They were told that they had been selected on
a random basis in an attempt to know something more
about the general pOpulation of students who were
given the questionndre in class. Three of the high
Scorers and two of the low scorers were not available
for individual testing either because they were only
attending the first five weeks of the summer session,
or refused to make an appointment for individual test-
ing. These were replaced by the students with either
the next highest or lowest score. The two groups were
then further divided into high and low scorers on the
Rigidity scale. The subjects were selected in a we such
that the eXperimenter did not know which of the resulting
four groups the subject belonged to until after the in-
dividual testing was completed.

To quij dealing with negative scores, a constant
of 160 was added to the Cogmatism score after algebrai--
cally summing the responses. Hence, scores on the
Dogmatism Scale could range from +¥O for complete dis-
agreement with each statement to +280 for complete agree~
ment to all statements. The range actually Obtained was

from +85 to +l?~.

\

As indicated in Table I below, low scorers on
Dogmatism differ on the average about 62 points from the
high dogmatic scorers. Low scorers on Qigidity differed
from high scorers on the average of slightly under five

points on dogmatism.

Table I--Comparison of Means and
the Dogmatism Scale

Standard Deviations of

 

Group

High Dogmatism--High Rigidity
High Dogmatism-~Low Rigidity
Low Dogmatism-~High Rigidity
Low Dogmatism--Low Rigidity

High Dogmatism--combined
Low Dogmatism--combined

High Rigidity--combined
Low Rigidityevcombined

N

10
10
10
10

20
20

20
20

Mean

182.7
177.8
120.2
115.5

180.25
117.85

151.u5
1h6.65

Std. Dev

11.186
8.108

13.315
12.u21

9.835
12.762

3h.223
33.550

 

The scores on the Rigidity Scale were treated in

a similar fashion in that a constant of 88 was added

to each score to avoid dealing with negative scores.

The possible range of scores after the constant has

been added is from +22 to +15“, although in the sub-

jects tested the obtained range was from +57 to +120.

Table II below indicates that low dogmatic subjects

differ from high dogmatic subjects on the average of

about 17 points.

Low scorers on Rigidity differ from

high scorers on the average of approximately 23 points.

Thus the precedure was successful in separating the

groups on Rigidity but the relationship between the

Dogmatism and the Rigidity scorers is also quite strong.

16

Table II--Comparison of Keans and standard Icviation

on the Rigidity E‘cale

 

 

Group N Teen Std.fcv.

. - o - o 0 0 fl 0 K . 0
High 9 gmatism-—Uigh Rigidity 10 10?.6 v.15i
High Cogmatism——Low Rigidity 10 86.7 7.25'2
Low' llxjnatism--Tfigfl1 Rigiditj' 10 92.2 7.800
Low Bogmatism-—Low Rigidity 10 69.6 8.072
High Dogmatism-—combined 20 09.15 13.639

. . . (‘1 r'\ ' I

Low Dogmatism-—comb1ned 2O 00. O 14.1‘0

o a c ' o o ' C’
High Rigidity--comhined 20 100.00 11.2ld
Low Rigidity-—combined 20 78.15 12.0U5

 

Trocedure. In the individual testins situation, the

. ,
\
'9

 

students were given two kinds of problems to solve. First
the :ndsjects wordnvi<n1 a set (Hizinalogies cmnfiuisizing
induction and 'deduction. ’ . After having completed
all of the analogy problems, the subjects were presented
with the Eenny Doodlebug problei in an attempt to obtain

a measure of conceptual analysis and synthesis.

For the first set of problems, the subjects were
asked to solve 50 analogy problems 25 of which were con-
structed with the emnhasis on induction and 25 of which
with the emphasis on deduction, all 50 were given in an
irrCSUlar order. The material was divided into two parts
and exno;ed serially behind a half-silvered mirror. The
preparation emposure cons’sted of the first pair of words.

The second exposure included 'hc first word of the second

pair and five numbered words as alternative solutions.

Both eXposures were controlled by the subject and timed
electrically. When the subject finished studying the
first pair of words, he turned a switch which turned off
the first orposure and turned on the second. We then
chose one of the solutions and pressed one of the five
numbered buttons, thus registering his choice and ending
the second exnosnre. Each subject received a score for
'induction and ‘deduction. .. ' based on his perfor-
mance on the total set of problems emphasizing analysis
and synthesis. The scores recorded‘were: the time spent
on the first part of the problem referred to as the prep-
aration period; the time spent on the second part of the
problem or the solution time; and the number of errors
made on each set of problems, in terms of the number of
incorrect solutions.

Aboaratus. The serial—exbosure box consists of two

‘

 

chambers separated by a partition and separately lighted.
The side toward the subject includes a half—silvered mirror
about 7 x 9 inches, and on the back side is a holder for

5 x 8 cards positioned so that when the light in the left
chamber is turned on only the left half of the card is
visible. Similarly, when the right chamber is lighted,
only the right half of the card is visible. The verbal

analogy is presented on a card as follows:

19

1. fine
2. post
large small echnsive 3. run
4. time
5. cheap

The answer to each problem is a number from 1 to 5,
which is eXpressed by pushing one of five apprOpriately
labeled buttons. A system of interval timers, relays,
and clocks permit the subject to view each half of the
problem as long as he wishes while a record of the time
spent on each part of the problem is obtained. The subject
is not permitted to return to the earlier part of the prob-
lem once he has switched to the second part. The eXperi—
menter made a record of the amount of time Spent on each
part of every problem as well as the answer chosen by the
subject. The primary assumption involved is that the sub-
ject in his effort to solve the problem Operates serially
on the material that is presented serially. While the
material on the left half of the card is eXpOSCd to view,
he has no problem to solve but is getting ready for the
material presented next, that is preparation. The activity
during the next period involves the selection of a solution
and pressing the apprOpriate button to indicate the choice
(Johnson, 1960).

Instructions to the subjects.

 

Side one:

"These problems are all in the form of analogies, such as;
light heavy little big

The relation between the second pair of words is the same

.r‘
.L

as the relation betwe ea tlie irst
fire hot
Your problem is to find a
to :_ic.e--a hat has to fire. You
were.- Tina list of Wfive words
1he first pa 1r of words 1.111 apnea
box. fiaea you have discovere'

move the toggle switch below

TN

Side two:

I! .1

...-oooa!"£1100?{
words besi:‘e it in the r:ig}1t
to the left side, so study the
you move the “witch. y'u
twist: in ..nd iioléi‘ iJel
as the word of ”

at the wor rd n

i

I I\
..a ....4

a
5;

G

7§L

main“. 1
..LJ.\.« in

the button
your choice .

r "V

bl

were then
w’

The subjects
acquaint then

given
to

1. large Sta 1 .
CXpensive 1. fine 2.post

 

 

2. year month

week l.first 2. roux
3. :uick automobile

Baldwi: l.violin 2.1i
b. knife cut

shovel l.scratch 2.d
The oodiebug Problem.

fter the aLLjitts Lad Cdupl

deductive problems, they were ask

kind of problem involving an

C
'3‘ '."\ -7
1. .1 k-. ...-1’

a world of his ovn. The subjects
1cm was not an easy one, but that

which coulc be rea:hcd by good 10
conditions of the problem, the 91
problem were th.n real to the sub

been gone over, the CXperimentcr

nirile lie traiixed so 'thrzt {J10

Iiian know*idn3thcr cu‘xiot

word tha
trill

the relation
to tile rl-(511toooooooooooo"

the

irst
iave

gical

another

t relation

the correct

ha. the

search for
idr: of"the
0'] t1’105dl

left 5
‘aetxree

r 111 tiie

numbered column of
You cannot returi
pair of words be fore
found he right word
with the same number

I‘

1our practice tri ls

O.

ith the procedure.

I;

,.run 4.time 5.cheap

’7
j.

11 r

1d ‘oot ..la}"5.nirH3

ght 3.piano +.knife 5.cale

F

13 3.lift 4.pour 5.sh—rpcn

:

eted the inductive and

a

ed to solve different

inary bug who lives

were told that the prob-

the re wa s a

solution

analysis. The

tuation, and the specific

ject. fifter the problem

asked the subject to think

echrincntcr coulc let
rrect. Tha suLject was.

ndar

allowed to use scratch paper and ask questions at any

time. Questions were answered by repeating the part of

the conditions, situation, or problem which were relevant
to the question. After the subject had worked for fif-
teen minutes, he was asked if he had a solution. If the
subject had no solution, or if the solution offered was
incorrect, the subject 'as read the first of three hints.
At the end of twenty minutes, the subject was asked again
if he had solved the problem and if not, he was given the
second hint. At the end of twentj-five minutes the sub-
ject was again asked if he had a solution for the problem
and if the subject did not have the correct solution, he
was given the final hint. The subject was then allowed
twenty minutes more to work on the problem, if he could

not solve it he was given the solution. When the subject
had either solved the problem or had been given the answer,
he was thanked for his Gooberation and asked not to tell
anyone else about the problems used in the exneriment
because the eXperimenter wished to test additional peeple.
During the time the subject was working on the problem,

the exoerimenter wrote down all questions asked and comments

made by the subject in order to determine at what time, if

atall, the subject discovered one of the hints by himself
and how he made use of the new information which was avail-
able to him at any given time. The conditions, the situation,
and the problem are given below as well as the three hints

or beliefs which the subjecteither discovered by himselfi

(O
30

or which were read to him at the specified time intervals.

The Confiitions

 

Joe Voodlebng is a strange sort of imaginary bug.
He can and cannot do the following things:

1. Vie can jump in only four different directions,
north, south, east, and west. Uc. cannot jump diagonally
(e.g., southeast, northwest, etc.).

2. Once he starts in any direction, that is, north,
south, east, and west, he must jump four times in that
same direction before he can switch to another direction.

3. He can only jump, not crawl, fly, or walk.

U, We can jump very lar;: ji>td1003 or very small
distances, but not less than one inch per jung.

5. Joe cannot turn around.

The Situation

 

[Joe has been junping all over the place gettixg some
exercise when his master places a pile of food three feet
(jirectly west of him. Joe notices that the pile of food
i.S a little larger than he. As soon as Joe sees all this
f‘ood, he steps dead in his tracks facing north. After all
T115 erercise Joe is very hungry and wants to get the food
3.5 quickly as he possibly can. Joe examines the situation
and then says, "Darn it, I'll have to jump four times to
Set the focal."
firing ?roblem

Joe Doojlcbug was a smart bus anj he was dead right

Q

VLt1 his conclusion. Why do you sunpose Joe Doodlebug had

J.

t30 take four Jumps, no more and no less, to reach the food?

10
\.J

The hints which were given are as follows:

1. The facinf belief. "I'm $0115 to give you a hint:

 

Joe does not have to face the food in order to eat it.
(Repeat hint.) CI, T'll give you five minutes more."

2. llna<fiircctirni heliirf. "T'1J_;jive yetizLiother*liint:

 

Joe can jump sideways and backwards as well as forwards.
(Repea hiit) T'll give you five minutes more."

3. The movement helici. "Let's read the problem avain.

L)

 

(The et<perimeritcr reads the entire problem to the subject.)
Now, here is the last hint: Joe was moving east wnen the
food was presented. (Repeat hint) You have twenty minutes
more."

Specific measures of analysis for the Doodlebug prob-

lem are as follows:

1. Time taken to overcome the first belief.

2. Time taken to overcome the first two beliefs.
3. Time taken to overcome all three beliefs.

3. Kumber of beliefs overcome without outside help

by the time the first belief is given as a hint.
. Number of beliefs overcome by the time the second
belief is given as a hint.

\n

1

Three measures of syntaesis obtainable from the
Iloodlebng problem are:

1. Time taken to solve the problem after the first
belief is overcome.

a. Time taken.ta solve the problem after the second
belief is overcome.

3. Time taken to solve the problen after all three
beliefs are overcome (R Reach, 1050, pp. 175-176).

/

Induction.

 

The hypothesis that Open subjects would perform

better than closed subjects on the problems emphasizing

induction waile there would be no difference between the

rigid and he non-rigid groans is supported both by a

comparison of the errors made, and of the preparation

index for these problems.

The superiority of the low dogmatic subjects on the

problems emehasizing induction is apparent from an exam-

ination of the mean nunber of errors and mean orenaration
L _,

index.

Table III below presents the average number of

errors made by the subjects in the various grouos. It

rnay

be seen that the closed subjects, on the average, make

about two more errors than do the Open subjects, while the

I

rigii

and non-rigid subjects differ from each other by

less than one point.

 

Table III Comoarison of the Tean Number of Errors made
on the Problems Emohasizing Induction.
Fear} Fur-1501‘ Of 1788.0 TU‘anI‘ Of P’ean yumber Of
Errors made by Errors made by Errors made by
1e Rigid CrOUps the Eon-Rigid Combined Groun
Grouns
Close 8.3 9.3 8.8
open 6.7 6.0 6.8
Combined 7.5 8.1 7.8

 

1»)
UK

Table IV below gives the summary table for the analysis
of variance of the errors made on the 25 inductive problems.
The probability levels resorted are in terms of a one-
tailed tes as he direction of the differences obtained
had been predicted in advance.

Table IV. Analysis of Variance of the Errors made on the
Problems Emphasizing Induction.

 

SOURCE 5.5. d.f. 3.3. F ratio D
Dogmatism 30.0 1 30.0 3.81333 ( .035
Rigidity 3.6 1 3.6 .43315 ....
Dogmatism t Rifidity 1.5 1 1.6 .19251 ....
within 299.2 35 8.31111

Total ‘ 3fi¢.b 39

 

Table V below gives the average preparation iadex for the
induction problems. As indicated by an analysis of variance

0

01 the preparation index for these problem2.the Open grOUp
has a lower preparation index on these problems indicating
that they are Spending relatively less time on the pres-
aration phase of these problems.

Table V. Comparison of the Tean Preparation Index on the
Problems Emphasizing Induction.

fean Probaration H.an Preparation Tean P.I.
Index of Rigid Index:of Non- Combined
Group Rigid Group
Closed .Q7) .353 .¥56
Cpen .397 .311 .Uou

COmbined .U38 .332 .335

Table VI gives the summary table for the analysis

 

of variance of the breearat'on index on the inductive

. \q \‘ ‘

problems.

Table VI. Analysis of Variance of the FreaeratIOi Tndei
0:: Chi? Triflslesn; Tarfl1a¢‘izi:*“ "13'H:ticx:.

bTURCQ . . e.f. .‘. 3 ratio w

u ‘: ' y j 'l I f I I V
Dogmatism .Ojdkffl l .Ofibwi 3.04354 <:.O)
r‘. I ~. .. — f ./ , r
nigidity .0003’0 1 .00030 .OIBJ ...
7‘ I ‘ t P-
begmatism x .00 000 1 .00400 .31F/ ...

Rigjiditj’
iii thin . '4'

\A
k x
O
O
O
\ J
‘J\
O
H
0
LA
\x)

Total ..'

‘3
\1
(I)
O
U
J
L)

J

 

he hypothesis that rir'd szbjects would not her-
form as well as the non—rigid subjects on the nroblems
emphasizing deduction and that there would be no differences

a“
L

between the oven and closed subject

in

ailcd to be confirmed.
A comparison of the errors made and of the prefiaration index

for these problems indicates that differeeces between the

rigid and the non—rigid subjects does not approach statis-
4.

.ical significance. furthermore, the differences between
the Open and the closed subjects, while not sthtistically

Significant, is g eater “has the differences found between

0 o 1 1 1 o _ o 1 ‘ Q ~
the rigid and non-rigie susjects.

he fact tlat the rigid subjects do not differ free
the hon—rigid subjects on the problems emdhasizing de-

duction and that there is no sisnificant difference

I»)
{.z‘\

between the Open and closed subjects is evident from an
examination of the mean number of errors and mean prep-
aration index for these problems. Table VII below pre-
sents the average number of errors made by the subjects
in the various groups. It may be seen that the closed
subjects make an average of one more error than the
corre5ponding Open subjects. The difference between the
rigid and non-rigid subjects, however, is considerably
less than one.

Table VII. Comparison of the Mean Number of Errors made
on Deduction.

 

Mean Number of Mean Number of Mean Number of
Errors made by Errors made by Errors made by
the Rigid Groups the Non-Rigid Combined GrOUPS
Groups
Closed 6.0 6.2 6.10
Open 5.1 5.0 5.05
Combined 5.55 _ 5.6 5.575

Table VIII below gives the summary table for the
analysis of variance of the errors made on the 25 de-
ductive problems.

The probability levels reported for the differences
due to Dogmatism are given in terms of a two tailed test
of significance as no differences between these groups

hadbeen anticipated.

Table VIII.

Analysis

of Variance of the
the Frobl ems Emphasizing;

13
C0

Errors
Deduction.

made on

 

J

SOURC

L—

4
.1

Dogmatism

Dogmatism x

Rigidity
Within

Total

(1')

11.025

0
1'.)
\A

10
n.)
\.t

210.500

{0
(J
H
\l
\1
\JK

\,.)
ON

\0
I)

T.S. F ratio p
11.025 1.«-8551 (.25
.025 .00327 ...
.225 .00625 ...

 

Table II below gives the average preparation

for

the

these problems, the differences
subjects approaches atatistical

ences between the rigid

Table IX.

the problems emphasizing

analysis of variance of

deduction. As

and the

index
indicated by
preparation index for
between the Open and closed
significance but the differ-

non-rigid subjects does not.

Comparison of the Mean Preparation Index on
the Problem Emphasizing Deduction.

 

 

“can v.1. Vein °.T. Tean P.I.
Rigid Ton-Tigid Combined
Open .3 O .337 .3133)"
Closed .2 1 .334 .3125
Combined .3.05 .3355 .328
Table K gives the summary table for the analysis

of variance of the preparation

index on the deductive

i.)
3

problems.

 

Table X. Analysis of Variance of the Preparation Index
on the Problems Emphasizing Deduction.
SOURCE 5.5. d.f. “.S. F ratio p
Degnatism .00961 1 .00a61 1.99792 ( .25
Rigidity .00225 1 .00225 .h6777 ...
Degmatism x .00733 1 .00782 1.62993 ...
Rigidity
Within .1733u 36 .oou91
Total .lOBOM 30

 

The fact that the Open subjects tend to do better on
all of the problems, and the fact that there is a differ-
ence between the kinds of problems may be seen in a
comparison of all groups on all problems. Table XI below
gives the sunmary table for the analysis of variance of

“

the errors made on both the induct ve and the deductive

Po

kinds of problems. Table ”(II {fives the summary table
for the analysis of variance of the preparation index for
both sets of problems. It may be seen that Dogmatism

differentiates the groups at a statistically significant

level on the errors made and asproaches statistical

U7

ig‘jnificanee for the preparation index. There is a
statistically sigzificant difference between the kinds
of problems both in terms of the number of errors made

and the preparation index. The levels of statistical

significance reported are in terms of a two tailed test

as no Specific prediction had been made concerning the
performance of the subjects on all problems or the differ-
ences in the difficulty of the problems.

Table KI. Analysis of Variance of the Errors made on
Troblems Emahasizinp Induction and Seduction.

 

SOURCE 3.3. d.f. T.Z. 1 ratio 9
nomntlr. ”’ 5125 1 “F "12' “ 20 05
go.“ of“ ‘tU.J ..J K.) ~j ‘VOIU < 0 J
Rigidity 2.1125 1 2.1125 .22 ...
Dogmatism x 1.5125 1 1.5 25 .16 ...
Rigidity
Error (a) 50.5500 35 9.7375
Problems 99.0125 1 93.0125 22.Uo <:.01
.‘ n I at p- l "’
L smatism x ¢.)13) 1 5.5125 1.02 ...
Troblcms
Rigidity x 1.5 25 1 1.5125 .3'" ...
Problems
Cogmatism x .3125 l .3125 .07 ...
Rigidity x
Troblems
/ i l
Error (b) 153.1500 36 3.4208

(N

Total

C.

\A
H
C I)
\1
\A
K.)
\J

 

 

m .. if 'r' 1 ,. ' c. 1‘ T? - n 1 ~. __ _ .- ,
lable A i. uuu13Jlo oi .arianee 01 tae lreaaratioa
. ’1 n ‘ ‘ h A 1—- 1— ~-- . PO . \Np "~ . ‘
Inie_ 0n tae -10aiens -ughasisihg Induction
.. 1 ‘a ,1 ' .
aia‘xl JC‘JUCt‘l-O‘1.
"j «\va". (*1 ‘f 1 I" 1' 'V T" '
E‘xlw-sv‘.~ u . 1. (3.1. . _ . .1 . .1 r0..th ,_)
”"9 ...“ _' .... .' 6’)" “fin"
uO|Jnt;C~~t .1. SiAA . O Y 1‘ ~'¢' l 0 0 “ J “-4 3 O 27 < I 10

I
I—Jo
‘J
F...
(L.
F.
(‘1‘
”5'.
‘3
O
O
O
H
”D
J
J
L.)
O
\J
O
O

Dogfiatisn x .01153 1 .01153 .35 ...
gigidity
sa’e r
Error (a) .4:ss5 go .01352
, (3 O! (
I-‘reble:.1s .2287)... 1 .223“ 59.2.- (.01

t._.
O
J
3
(+-
F4
‘0
D
O
C 3
F4
H
O
(.3
9
H
H
l O
l J

I'.‘ ' ("-3. 1 ' +- v' nf‘fin't final) ”'3’:
-ulglaib)r (L .'\I'\l 54~J. 1 O JV-wa—vl OJ.) . . .
*3 ‘ “fl
LTOSlCMS
T“ IRN‘. _>o ..‘ nfi - n h ‘(j
,'JO'\JAA-C~tl.:3;}1 .‘L . Onv/l ' .OJOJ1 .CU . . .

-,-3 n ”‘0 1
.‘.l-.l(1

“H.
’1
O
H
A
0'4
I ..J
H
4"."
"x 3
k .3
( J \
O
(3
K J
\\
\Q

Total .91932

\I
\D

 

r“+!\r‘r-t. r. ml
l‘AVLLV...‘ l_._‘. ...!

10 hypothesis that the Open subjects

 

should be sunerior to the closed subjects and tlat the
rigid subjects should not differ from the non—rt
subjects is.supported by all three measures of synthesis
on the Benny Doodlebug problem.

Table XIII presents the mean solution times of the
four groups cf subjects separately as well as for the
his-.1 and 101: dogmatic subjects and the ri.f-:id and non-

id subjects is consistent for all tLree measures as

.‘

predicted while the difference between the ricid and non—

\d
(J

rigid subjects is negligible.

Table XIII. Fean Solution Time for the Penny Doodlebug
Probler Following the first, Secon., and
All Three Beliefs.

 

GrOUp First Second All Three
Belief Belief Eeliefs
T o o o I o '
high Dogmatism-High hlgld 20.82 15.74 8.6%
High Dogmatism-Low Rigid 20.37 23.99 14.15
Low Dogmatism-High Rigid T.02 12.95 5.t2
Low Dogmatisnhlxnr Tieid 11.61 €.b0 1.95
e1~h :ohnatism 21.595 19.515 11.,95
LOV Dogmat'sn 5.? 10.675 3.d20
High Rigid 19.92 1n.8u5 7.165
low fiigid 3.99 15.9»; 8.200

 

Table XIV below gives the summary table for the
analysis of rarianee of the time required to solve the
problem after the first belief had either been overcome
by the subject or had been give: to the subject in the
form of a hint. It may be seen that there is a sin-
nificent difference between the high dogmatic and the low
dogmatic subjects. Yo such difference exists between
the rigid and the non-rigid subjects. Vowever, there is

a significant interaction between Dogmatism and Rigi

(L

'ity
indiCating that the degree to which dogmatism determines

the subject's performance for this measure of synthesis

is influenced by the degree of rigidity for that subject.

kl)
\

111 m

2 e of Lie lime fiequirel to
jenny Ceadlebfiv Problem after tne

‘ n
x

G
I)
(7‘
,4
(D
l
id
. 1
'4
O
7
i
i—J
V
L
-f
O
“5
,4
1
{J
'1
L..—
:3
O

 

‘v-s --fi.-‘y-\ -- -.

I-
l
...J
C.
.1
r4-
\<
’3
t J
H
O
J
O
1...;
O
O

U
0
fl
5
r-f-
p.40 f.
’r
‘4
\ 1
k1
D
\.l
O
i—J
\JI
\'\
.3

O
1
D
.0

0
IA
J
O
[—4

/(Q mgr)

a
c-L
:3
(J
J
‘\
J
J
J

 

 

s
E)
d
U
H
0
J1
A

v N
.4
‘i
J
,.
ad

1
f...
I‘»
k $
4
.0
('f‘
,1
O
.0
rod
.4
.4
J
H.
v
c
r.
L.
(Id
f—l
\J
U)
P3
0
h’
d-
:1
O

with.tfiu3 tine requirwus‘ta -olve tin: eroble: sifter the

ence between the coon and the closed ”rouge while there

. \J
. ‘ 1 -I u r\ . 1‘ _ ‘ I 1 _ d ,' V _ _ ’1 .fi ‘ _o N yo ’3
is no suca u1.fereace setueen t)? 1131d and no —r1g1

“ o P‘ ' ‘. "‘ \ ~ 5 . - v ‘ : J- : -‘- u
grouns. L.avn, thelc 1s a 2:3n1f1ea2t interact101

between dogmatism and riridity.

’TablcsffV. .knalyhyis of“Vari;ntee CH7 the TTbne ihequirwxi to
Felve the ienny Toodlebng Problem after the
Lecont Telief.

I

'oA

 

r.)
L)
.1
('1
o
. I
.-
f“)
o
o
L.
o
"1
H
t.)
,1.
0
v‘

in
O
['2
E)
rt.
id.
(0
\l
\J
(0
\JJ
21\
H
\l
.\J
O
I J
\ )
Ch
H
\I
\1
’\
O
O
H

-’.. . O "' - 4-! :—
Dogmatism x 37c.225 1 379.225 5.52 <’.0)
Lligixiit3r

... . Q Q f
tithin 3039.330 2 .3 0

‘Total 2222 222 2:

filTI lei
to

t]
-8 ~

Table

'1‘. '

' v - . - . . N ‘ r T) ' '-
al).;i: oi u;ruzztce of ‘tne ..ine .mxpiiree

iolve he Tee‘v “oodlebuj Frohlen after
l_'NJree "Pfi‘cfs.

 

 

SSCRC; §um of d.L. Team P p
Seuares Square ratio
. ff ,a( a [I ‘f H.. _ O
Do’fxzzatism 59).:7; 2~ 1 39.73625 11.01601
Rigidity 10.712?5 1 10.71225 .2] ...
eogmatism x 22%.00625 1 228.0062; 19.51605
R’jidity
.. C“ (K. ’ -' r? "‘73"
‘ thln 1C1?.1)CBOO 3’) you/‘41:!)
Totgl 2655.15775 3?
Analveie. The hypothesis that the rigid -ubject$
would he slower and discover fewer of the beliefs Ly

themselves

Ivoulxl be

jeets we

required 0 di

b

Table *TII gives

conceptual analysis in terms 0

time required

'3
V

It may b

1.

the closed
for the
it is

belief

'_\
L4

rid tile

1': .1. L11

ncni—ri

tlia

seen
suoj
second

onlgr as

a woulfi

scever the

a

to overcome

4 L

-‘ ,—
enat. file d

v

(-0

zion-Ifigjid

to be large enough to be

the hypothes

4%.r1
LJLU

gzwau;)e

tile

(:“P'

lare

from):

1 'V“

”Al-

ary

4"

l
4.

th
orze, tlfO

ifference

e

on

f‘

-..‘J‘.
‘. LLL

of

subjects While there

Open and closed sab-

the time

the eata concerning

e average amount of

5!,
Au

, a1 all hree beliefs.

between the and

Open

and becomes larger

but even for the third

between the

"3

T‘

\Q.

irst belief nd

eigafi7icaut.

ereaee between the rigid
See“ that it consistently

takes the rigid srbje eets lonjer to overcome each of the

beliefs, nartieiilarly for the second and third beliefs

fable KVTI. Congarison of Teen Number of Finutes Taken
J

to Tvercome the Tirst Telief, and the First
Two Teliefs, and All Three Beliefs.

 

Crown :irst First Two 111 Three
Zelief Eeliefs Eeliefs
High Bogmatis: Tigh “igid 9.72 15,60 ”2.90
High Dogmatism-Low Xijid 3.51 3.39 31.93
a v o o o 1"
Low Cogmatism-gigh .igid 9°89 1€,15 23.91
Low Costaat" rn—Low igid (5.1:; 0.1;; 15 .90
High sor.atism C.115 1/,C05 22.b15
Q a
Low Togmatism c.165 12.800 19,§0)

Y_. ‘ ‘ j" m 0 . \ p Q -
high i_id1t} 7.9” 15.275 23.H05
Low ‘.T1Ci 3 7.49 10.9? 18.?15

Table VVIII presents the summary table of the analysis
of variance 0; the tiztie reqvired for either the sub
to discover the first belief for himself or to have it
given to him as a hint. It is apparent that the d'ffer-
ences between the risid and the non-rigid subjects fall
short of being statistic lly significant and that the
differences between the Open and closed subjects is
extrernely small. The hyoothesis is not confirmed by

this data.

\ .J
U!

to
.\

Table XVIII. Analysis of Variance of the Time Requ1red
to Overcome the First Relief.

 

SOURCE Sum of Kean F o
Squares‘ d.f. Square ratio
Dogmatism 9.025 1 9.025 .2 ...
Rigidity 53.82: 1 53.824 1.60 <f.2
Dogmatism x 12.321 1 12.321 .37 ...
Rigidity
Within 1213.186 36 33.6906
Total 1280.356 30

 

A comparison of the time required to overcome two
beliefs gives strong support for the hypothesized difference
between the rigid and non-rigid subjects. Table XIX gives
a summary table for the analysis of variance of the time
required for the subject to discover the first two beliefs
or have them given to him as hints. Clearly the differences
between the rigid and non-rigid groups is statistically
significant while the differences between the Open and
the closed subjects remains extremely small.

Table XIX. Analysis of Variance of the Time Required to
Overcome Two Beliefs.

 

SOUR E 3.3. d.f. N.S. F ratio p
. 11.2u025 1 1u.2802 .“4 .
Dogmatlsm 5 "
Rigidity 235.52025 1 2&5.52025 7.62 <f.01
‘03??t}sm X 30.n5 2 1 30.45025 .9u ...
Rigidity
.Within 1160.51900 36 33.93553
1
Total 1450.76975 39

 

then the time required to overcome all three beliefs

is analyzrad , the hypothesis is supported at a statistically

37

significant level, ut none-the-less is complicated by
the fact that the differences between the Open and
closed subjects is larger than before. Further, for the
first time, there is an interaction between the two
variables which, although not significant, indicates
that the way if which rigidity Operates to differentiate
the different subjects' performance on this tasx may be
influenced by the degree to which the subject is dog-
matic. Table KY below presents the summary table for
the analysis of variance of the time required to over-
come all three beliefs.

Table XX. Analysis of Variance of the Time Required to
Overcome All Three Beliefs.

 

SOURCE 5.5. d.f. M.S. F ratio p

Dogmatism 63.051 1 63.051 1.985 .( .25

Rigidity 201.601 1 201.601 6.3u76 <’.025

Dogmatism x 123.904 1 23.00fl 3.901 .<.10
Rigidity

Within 1143.37 36 31.760

Total 1531.926 39

 

A further method of measuring conceptual analysis is
to consider the number of beliefs which were overcome by
the subject on his own before the eXperimenter gave any
of the beliefs in the form of hints. Table YKI below
gives the chi square analysis of the number of beliefs

overcome by the subject in the first fifteen minutes,

b.)
C0

dividine the subjects in terms of their rigidity scores.

In order to avoid cx§ected frequencies of less than five

r...)

n each cc 1, the number of subjects wao overcame two

t-J e

and three beliefs were combined. It is quite evident

that the groups differ 1? ChClr ability to aralyze, .
this difference heirs statistically Significant beyond

the .01 level of cenficence and in *Le direction pre-
dicted, J‘hat the non-rinie subjects overcame more beliefs
than do the rigid subjects. ’Towev~r, when the Open and

closed subjects ar- compared in terms of the number of

O

beliefs overcome in the first fifteen minutes of the
esperimcnt, as in Table XXII, the differences between

cant.

F...

the eroups are small and insignif

Table XXI. lnalysis——The Yumber of Eeliefs fivercome
by the Subject in the First Fifteen Tinutes
of the IXperimeht.

 

Group 0 Zeliefs l Belief 2 Beliefs 3 Beliefs

Rigid 7 11
TRMi-rigid

\1
H
O\ 0
O\ (J

('0

Chi Square1-2 12.60
d of. = 2
p less than .01

 

 

-ates correction for continuity has been applied
to all chi square analysis p esented.

’20
//

lnalysis-—Th Tumber of Beliefs fvercome
3y the Subje t in the First Fifteen
iinutes of the “

theriment.

DJ ..d I

Group 0 Teliefs l helief ? T)el"-efs 3 Teliefs
Open 7 7 3 0
Closed 7 5 3 6

ChJ_ 3(quIT‘ = .‘33l

d.f. _ 2

 

that the s c

5!)

glwrfliesis :Nie liki‘ction. Tflie hjjxjthesi

”.-. .__._.

 

who do well on. synthesis {lion :1 also do well on indiction

by the data. Table YEIII below nresents

was not confirmed

J)

the observed frequencies and the ch' Square analysi- of
the time required to solve the TJen-1y Seoulebug nroblezxz

after all three beliefs had been overcome and the number

of errors made on the problems emnhasizing induction. Tn

f1
rt
13‘
J
3
*fi
r—Jo
.4
(D
F?”
b J‘
0

order to avoid eXnected frequencies 194'
data was analyzed in terms of a two by two table L01:
‘ed “fter l ?C minutes. It may be seen that while
the differences ohteined are not statigtically significant,

they are in the direction predicted.

, Synthesis and Induc ion-—The Time quuired
to iolve the T? Problem After 311 Three
“eliefs Ted Seen Svereome And The Tnnber

T".

of Trrors Tade on the ~roblems fmphasizing
Tnduction.

 

inntes Teonired to Tolvc V“
m fifter ill Three ?elicfs Ted Teen

“e l: n I’...,‘-,.. ,
0—1. 9 3-;. “ ‘- .T; a an; above

lumber of 0-7 G j 2 7
Errors made ?—+ j 5 3 3

011 Trmhteixive
Troblems

Chi :Jc l'irc = 2.31

I}
1.

d.-. 1
w less ta

:1
l-' ~.1

.0? one tailed test

u

y -.

 

Table TYTV below presents the observed frequencies
and the chi snuarc test of the relationship between the
number of minutes required to solve the Kenny Foodlebu:
problem after all three beliefs hid been overcome and
the preparation index for the nroblems emphasizinfi

iiuhiction. beihi the relatficnrfliin bctixwmi'flie two

measures of synthesis in in the direction fredicted but

W

ails to achieve statistical sihnificance. is in the
above Table YTIIT, the table was condenrei into a two
by two table in order to avoid espected frequencies 0

lesc thfin IiVO.

 

Tkfijle ‘T’IV' :3nithewsis (11” H‘dtw**~rr~-— “1c Te’ve 1‘eqtv?rebi
to fouve the T7 Frohlew After 111 Three
ielie¢s Bod bee? {vercowe one ¥Ubjects
jcorinf Itove and Felov the Vefiian Preferetion
‘neex on 1*v"“rohleww:Tenifies%2‘:jj Tnfiuction.
Tfirher of “in“tes .equzrefl L "elve P?
treble? “pter Ell Three D91'1er Wed ‘eew

7UiflvctiIW? 0—1 ’V‘ 2—? TV7 “—§.T”3 t anelrflrove

T“ro!ilexw3
'Prevvrretjxni lmeleV' 91 R ° 7 5
h n

\——
-
-—‘

3.

anex above wl e

(‘1
.3.
'J
.1
O
'l
«A
D
J
J

" 1
(A o ‘ 0 : —L
o less tee} ,7 one te‘]ed test

”“"‘e “e5*”me1‘ o

l ‘ . -1- .L ~._~ a . ~ \¢

t3r* “”WT“1H‘O€ 11<e<

- 4" . v “ -- I . '1 ’L" I l 1 ‘< - "“ V . rw ' P “
COITIIC‘\,910;1 LPt'HO '-‘1 13:1.110v7'127 Huh 53': {410316 tl‘lC lC‘xC-L

which no such relationship is erodieted reveals n “17-

nificent differences. Table ”V” below bresents tbs “'1

telu, of the number of errors rent on the treblens ee—

7fiieeizjflyj fled u3tlon. szzilerlyy 7fubl: TTTTT glxwx‘ txe

(iii severe :uuxlysivt<3€ sywtluxiie and (hxhzctLOI 1L;Lai the
_ -° W 1 o .v« *,.1 L'-- . P,- ‘\ ° u

*7re.x1-et14nr l .Iex C). t h .tgluet,Lve .ro.file. . _, l3 t‘le

'“11 :mpzfire t/fiLlCS :flsove, tre ;::al)w 13 ot‘ tee :7iia 1“ :ud*

 

 

 

 

 

 

 

aha“ N- --9 ‘- ‘ ‘1 ‘1 .. o . ‘ 7" --‘ ‘ ‘1. l- I r"- . f‘ a“ - o ‘ 1
L 3-410 ‘- a. O 'J} :.\L:C ' \J (I. ‘xn ..'\‘-A {CV L‘D-l-“, L |-\. vi ‘3‘ rnfiij '3‘]
- ' .- “‘ ‘ . ~. 1 l -' _{\ _ ‘1‘ n -l '11. . ~ . _‘ r '. p ‘ - ‘5 ‘ 1
L13 .- . r\3~_L‘j VI ..A tkr‘l ..~ .L -i.';‘\,‘\: _-\‘1 ..CL t) -r.d A
-‘ " “ \ .1‘, ‘ ~ ~\ ~“ ‘ 5“ H 'A. ~ ‘ »\ ‘ ‘
c.. tereo e edd n “er 0; rzors h 3 ed
L‘ fl 5, '~ rs ‘- ~~ T: 1 — - , w v" -‘ l . ‘
c1e 1oll “, ~‘1 -n:ng ”CAUCLLUu.
-.--..7-o m.—‘*-'.’_‘m'-
...1. . , r‘ x -- ; . .. w“... - .... .\ .1 J- - - -1 ., "n’
.1 .-..' \fl‘ ‘JL 1 ‘llet-J -.L~-‘: I, . L _'. u“) ~.'\)_i_ \ .3
f' . ‘ 1 ‘ "\ ‘ ‘ r‘ '_ > A 1 -v “ 1 A‘ a ‘ w“ ' o f‘ ‘ 1 r ‘
.Jul . L_tnr rll ”lac fillCLi ml Heex
V.' "\ sfi C 13 ~11) .
1 me c ’7 (NW '. -' fi‘ ’ V1 ..1 ,
0- . . - -.-/' v - -/' . ‘1 _, U \L 11.4 L~ lji) \' J
"V‘ \f 1“
\. n ‘ r 0.1.
‘ _- r q .0 (7 : f3 3"
drl‘ql‘:) _ \_K£ \ ». -_) ‘7 .4) J
, 11" VI 1 J— . - / 'l '1 r) M
0‘] 1‘L.- ‘tc l T.\ (7‘ + ‘ t H, [I
‘r‘(‘ ‘1 ("
~5‘ ' l N
::L ~qwere z .b/
,; P _ 1
'- o - I ‘ ‘-
‘1 — L! 1. L" ' - - .- .l t- -. -
, 1.05, t1 .1 .10 L.-.O to: 19.. .0531.
m’-~I-—‘——O -
m“? mrt'vr-y- I: _ 9 \ 0 _ ,1 x . ' ' "W ’ '—
uole .h. -. ”33tee215.(1:. Tedzetzona«TW 4‘..eQUIred 0e ‘49
.N" ) Y‘1 II A _ Q r\‘ hr 0 f‘ ‘ 11' ‘
. roslew reter .ul ,Jre\ ,ochls .a? con
r‘- - . V 7‘ 1 ': 1- : J- _ W. at -. r re .
exe'eo e .3 men‘ctu corn)» .po.e 1‘J 010v
+3 ' -1 ° ,. . n ._ ‘ .. 1. ' , . 'm - . I . r‘. ‘1 -
.lze e ,'1:1 o‘ee ; Tait 'o 1 .d’>\ ()9 t we rCHchLnT
1 . -1 o '_ . _ . . ‘
vfi~1euL/1‘ eflnetLOJ.
—— n
~a , "\ *0 " I v v 4 " \ . L v- V -‘
u.;?1 o_ 34tes LeQ‘Lre! to olxe J.
7* ’ ' 1 , t, ’\ p _ ,‘ 1 1 1 1‘1 I 7~ ’ _ I" ‘ 1‘ 7 " ‘
_roele. “.ier .Ll .nree 01L“;S an ~03.1
‘K'Tr‘cij ‘l 1‘ .
n ‘I (30 f) ’3 r ( .' r n (1’ , .3 1|
‘- — .‘ . ,. u_/ g ‘) L“) Q ('1? \) i'.‘1(‘ p.30 ‘rfi
w-x K ‘1, ,' 1' ‘—' r f‘,“ u-I /: 3-! r3 I'fv
rre a 1‘303 “ole: ./M) c J , f
- 1 r" 1’ f“
. r‘. '~‘~v- 71 1‘ «- v'r‘ f) [1’ I. r) "\
.L .3~V-L ‘- -\.)I ~ j‘) ‘4 I) 3 -1
F‘.‘ o I Q ’ ‘1 'q
u‘ Ll —\lL‘.k‘\:~O : . I 'L.—
.‘ P
“I o l- c = 1
~ 1 1 .1; *7 r.- v . .° 3 .
) les teej .{/ tee tallCU teat
1 r v --.. - r x 1‘ “ . " ‘\ L m‘ - 1 ‘ ‘1‘ r - :-~ .u ‘- ~
‘Pfjj ,5 n ‘ '4zu,tlow. lne IU’MfitIC“‘ tlit ~Afl35eeas
-‘ - ~ - L r r‘ 1 e. . - a, 1 r - " u ‘ -‘: ‘ . .
:10 co 5311 or 1n0;ys1~ noutd 4.30 .0 R011 on flevuctJOT
.-q 4— ‘ AA . .... ‘ .‘I, , _‘ _7 r _ ' i ’ 1 ‘_. 1‘: F-‘r- -' ‘.’I"TT" -
”Cu Tet eon;11 ‘. by two teta. 3510. _.‘fT eel '?!r
' ° __ , - , Y .' .
belxw' "1\%) the rhiaeetefi.€recnw‘de1rhs and tJna cql smvtaro
, , . . t‘ ‘ _ 1-1 , “fl _ . V.‘ _ . _ .., , , _‘ 1 .
:L111}~.'s e~ :30t*1 one \re ;:.;t::w1 1 vlex :uxs tare 311<3rs
.. . Lu , ~ , ‘ .1. ° '- 1 ‘ ' v I . ., 1 r‘ 1- r ,e H
new“ on ‘re :mect:on nrooteud v1t1 {we nuaoer o. eelxets
vr‘ 9“ - - . L h . ~ ‘ I‘ ' '1‘ - \ v . '4‘ I * A‘ o 1 q‘ ' ,‘ ' N f1 1‘ r.‘
oneteo e 1: tWe LerL ijteea .LH”&C£ or two e1Hy do leowv

l 3

1+3

problem. In both comparisons, those who do well on
deduction indicated by either a high preparation index
or few errors, also seem to do better on analysis. How-
ever, in neither comparison does the relationship between
deduction and analysis reach statistical significance.
As in the previous chi square analysis, for purposes of
statistical evaluation, the table is reduced to a two
by two contingency table.
Table XXVII. Analysis and Deduction--Comparison of

Number of Beliefs Overcome in First Fifteen

Minutes of DD Problem and the Number of
Errors Made on Problems Emphasizing Deduction.

 

o Beliefs l Belief 2 Beliefs A11 3 Beliefs
0-5 Errors U 7 2
6 & above 9 6 Q 2

d.f. - 1
p less than .20 one tailed test

 

Table XXVIII. Analysis and Deduction-~Comparison of Number
of Beliefs Overcome in First Fifteen Minutes
on DD Problem and Preparation Index for
Problems Emphasizing Deduction.

 

0 Beliefs l Belief 2 Beliefs All 3 Beliefs

P.I. below .325 u 8 3 5
P.I. above .325 9 5 3 3
Chi Square = 1.823

d.f. = 1.

p less than .10 one tailed test

 

By way of comparison to the relationships previously

tested in Tables XXVII and XXVIII, Tables XXIX and XXX

- o ' 1‘ I '* - v, . . ~ 1, '4' x \
or“_n't t.3 Olaorncd -rocHDRClo: bl; soc Ch ,analc
r ~u n v - v v— 1 o . J . u- I 4* . . .. 1 . 1 . ......
anal}'. s o- lhl 's Klijl 11(RK3L107. o TWTlltl'Viul
.. ‘ _ - J- 1 7 . 4_1 ‘ 0 . O a c
L‘ lrod'c 0‘ jot con thV“ variables 32' it is QHLtC

clccué that tfllore isszlo oxfi<iowce inr tfwr 0YJ¢*;X:"0 of‘

:3ucfl a: I‘ol;.ti:)n.fili~l. .ls 163*" tllo (oasn: tfllozl c:ylgv1r'.lg
33*5111C‘t1031 3.11:1 C‘.L".IL1.:,"".:!..‘TI, the number 0.." subjects who over-
came either two or thrlc beliefs were co Bland for the chi
Sfpylre :21al;1sis. is no llirocfiiion Tyli LuCWl predicated, {filo
two tailed level of signiflcince is reverted.

Table .‘ITIITC. .4‘.n;llyr-:is 3.313; Inductiorb—Cellimrison of‘ ‘110.

Tumbor of Beliefs fivercono in the Tirst
Fifteen Vinnto: of rho CD Problem and the

Number of frrors made on the Problems.

.-1,,: ' .. ,
33Als_21wf Tn.

 

O
"J
O
...:
re.
’9
*5
a:
f.)
d
’3
H
O
W
(J
r:
0
H
0

f3 All

Ecliefs

K»)

.J
(A
Q

(“J 1'" _‘
O-o -rlors
r r
n '7 L . ‘

er alaoxro

13 LN

{
\.l
I

D « ‘ffi
CLIL ,r'U.11 o = .13 ?
d.f. 3 '3

~.— . .I
p less thaJ . J

 

Table TI‘CFI. .‘1.;~:;l)-'si5 and Inclvctlon--"301:1p..rison of tile
’\

sumbcr of ’clicfs Overcome if the Pirst
Tiftoon Xinutos of the DD Problem and the
Prcjaration Index on tie Problems Emnhasizing

Induction.

 

'3
'3
O
H
L—
.)
’3
J]
H
H
1‘)
..A
O
'3
I.)
-'J
(J
H
r...
(J
H)
uh
’r‘
H
H
\J
U
f“:
H
A.
’)
H)
J]

7‘1 ‘1‘ .. 1, - an :r r (7‘ r) K
,.. Q A. . 1|.J<;‘..L.‘3‘;b . ./' Q J -. ‘_‘ k)
n 1 . Q!) r r’ If 1. m
"I: .I . a-DO V0 . Jr‘s) \J‘ .r J

J
FA
0
Ln
r:-
If
m
0

‘ J
\I
\A

 

1&5

Another way in which the relationships between analysis,
synthesis, induction, and deduction may be indicated is by
means of the correlation between these variables. Table XXXI
below gives the correlations between the preparation index for
both inductive and deductive problems, the errors made on both
sets of problems, analysis measured in terms of the number of
beliefs overcome in the first fifteen minutes of the Denny
Doodlebug Problem, and synthesis measured in terms of the time
required to solve the Doodlebug Problem after all three beliefs
have been overcome.

Table XXXI. Correlations Between Induction, Deduction,
Analysis, and Synthesis.

 

P.I. P.I. Errors Errors Analysis Synthesis

Induc- Deduc- Induc- Deduc-

tion tion tion tion
P.I. Induction 1.00 .73* .03 .14 -.10 .08
P.I. Deduction 1.00 -.07 .09 -.01 .09
Errors Induction 1.00 .u3* -.01 .22
Errors Deduction '. - 1.00 -.25 .28*
Analysis 1.00 -.28*
Synthesis 1.00

* significant beyond the .05 level

 

From the above table, it may be seen that the preparation index
measures correlate suite highly although the mean preparation
index for inductive problems was found to be higher than the
mean deductive preparation index. Secondly, the correlation be-
tween the errors indicates that those who make mistakes on one set
of problems also tend to make mistakes on the other. The corre-
lation between analysis and synthesis indicates that doing well
on one phase of this task tends to be associated with doing

well on the other phase which is most likely due to the corre-
lation between dogmatism and rigidity reported earlier. It
should be noted that both of the error measures tend to corre-
late with synthesis through only deduction does so significantly.

Perhaps this reflects general problem ability.

DTQCVigTCT

The subjects who participated in this study were
chosen specifically on the basis of dogmatism and only
secondly were they further divided into high and low
rigidity groups. Due to the high correlation between
these two variables, it is not at all surprising to
find that when high dogmatic subjects on the average
score 17 points higher on the rigidity scale than do
the low dogmatic subjects. It would be advantageous in
an extension or replication of this study to equate the
groups beforehand so that there would be no difference
between the high and low degmatic subjects in terms of
their average rigidity score. Another alternative would

be to test a random sample of the entire range of both
the dogmatism and rigidity scales which would increase
the extent to which the findings of any study such as
this may be generalized. A second difficulty in the
present study concerns the difference in the difficulty
of the problems emphasizing induction and deduction.

The assumption is made that difficulty on these different
kinds of problems lies primarily in either the preparation
or the solution phase. It could be that the differences
found between the high and low dogmatic subjects on the
inductive problems is due to some general problem solving
ability. Iowever, this does not seem reasonable in view

of the fact that significant differences were not found

[-
l‘w
‘d

between these groups on the deductive problems.

In spite of the above liuitations, the findings
of this study generally sUpport the assumption that
it is possible to devise tests which separately measure
he ability to integrate beliefs or to analyze‘~ the
parts of a problem, The results of this study sUpport
the first hypothesis that Open subjects solve inductive
problems better than do closed subjects.‘ ThiS'is indicated
both in a comparison of the number of errors made and
of the preparation index for these problems. Open
subjects made fewer errors, and had an average prep-
aration index for these problems. Open subjects made
fewer errors, and had an average preparation index
lower than the closed subjects indicating that they
spent less time on the preparation phase of the problems.
However, the predicted differences between the rigid and
non-rigid subjects was not found on the problems emohasizing
deduction.

The measures of analysis a d synthesis used for the

!

Denny Doodlebug problem give stronger support for the
differentiation between dogmatism and rigidity. All three
measures of synthesis indicate a difference between the
Open and the closed subjects with no such difference bet-
ween the rigid and non-rigid subjects. However, the inter-
action between dogmatism and rigidity is also significant
for all three measures. While the comparison between

rigid and non-rigid subjects regardless of the dogmatism

’48

score is not significant, within either the high or the
low dogmatic group there is a difference between the high
and low rigid subjects. Particularly with the measures
of synthesis which include the time required to solve
the second and third beliefs, the measure of synthesis
includes some measure of the ability to analyze; thus
the interaction between these two variables. The four
measures of analysis presented generally suPport the
hypothesis that rigid subjects will not be as capable of
distinguishing the parts of the problem or the separate
beliefs to be overcome as will the non-rigid subjects.
Considering the time required to overcome one, two, and
all three beliefs, it is clear that in each case the
difference is in the predicted direction although not
statistically significant for the first belief. The chi
square analysis of the number of beliefs overcome in the
first fifteen minutes of the problem indicates a differ-
ence between the rigid and non-rigid subjects while no
such difference between the Open and closed subjects.

The results of the analysis of the relationship between
synthesis and induction were not significant. However, the
trend observed was in the direction predicted, that is those
who excel on synthesis tend to do well on induction.
Similarly, the results of the analysis of the relationship
between analysis and deduction were not significant but were

again in the direction predicted. .An analysis of the relation-

ship between either synthesis and deduction, or of analysis

Hf\
... 1

and induction for waich no trend had been predicted
indicated no statistically significant relationship be-

tween these variables. Tt ray be that if the relationship

between induction and synthesis could be tested on a larger
sample using the entire range of atilities found in a ran-
don sample of the pOjulation that the trends asparent in
this study would be brought to light.

The comaatibility of the findines on he Penny

loodlebug Trotlem reported for this .XHerimeit with

O...
L

earlier work done usiné the same problem is of interest

\,

as this Wives an indication of what changes and simi-

(_.>
larities may be found with Specific changes in the

administration of the uroblem. Table KIQIT below gives

_ I

the findings from a similar study reported by iokeach

contrasting dogmatic thinking with rigid thinking

(2 Reach, 1950, p. 188).

Table YYYIT. Ability to Analyze: Comvarison Tetween
?igyid zine! Text-rig id (IrOLnis {MIC TTOtixaen
Closed and 7pon Crouns on ”can Yumber
of Tinutes Taken to Cvercome the First
Telief, the ?irst Two Peliefs, and All
Three feliefs.

 

1’ it]

71%
o d

‘H'J‘:

we All Three

Groun First T
' efs :eliefs

r-‘Cf‘

. . f F, (I
Tigid 3.“ 1;.) 1-.0
W . o I

.on-rigid 5.? 13.1 1 .4

0
FJ
O
an
O
(L
\1\]
(3 c
H
-J ‘J
l—I
‘ J \D

15.; l

It is seen that the closed and Open groups do not differ

from each other in analytic thinking while the rigid group

usually takes longer to analyze than the non-rigid group.

The conclusions are sinilar to those reported from Table

XXXIV. and the times are reasonably similar. Certain

differences appear when comparing the results of these

two studies on their measures of the ability to synthesize.

Table below gives the results from Rokeach's study.

Table XXXIV. Ability to Synthesize: Time taken to Solve
the Doodlebug Froblen after the First,
Second, and Third ?eliefs Were Cvercome

By Rigid and Vonrigid Crouns and by Closed
and Open Groups.

 

GrOUp After First After Second After Third
Belief Belief Belief
Overcome
Rigid 1h.5 7.2 3.2
Honrigid 15.1 6.6 2.8
Closed 15.2 7.7 3.3
Open 13.“ 6.1 2.4

 

While the general conclusions that the Rigid do not differ
from the Nonrigid and that the Closed take longer to solve
the problem than do Open subjects, the mean solution times
are considerably shorter than those reported from the
present study. However, it should be noted that whether
or not a solution is reached in the study reported by
Rokeach, the session was terminated 30 minutes after the

problem was given to the subject while in the present

study, the subject was permitted #5 minutes to solve

the problem. In another_study done by Rokeach and
Vidulich (as reported in Rokeach 1960), in which the
subjects were given MO minutes to solve the problem,
more compatible times are reported for synthesis in
Table KKKV below while lower mean times for analysis
(Table XXXVI) are reported. This second difference
would appear to be due to another difference in adminis-
tering the problem as the hints were given at the end of
10, 15, and 20 minutes.

Table XXXV. Synthesis: Nean Time Taken to Solve the

Doodlebug Problem After the First, Second,
and Third Beliefs Tad Been OverCOme.

 

Group After After After
First Belief Second Belief Third Belief

Closed—memory 21.9 16.0 11.3

Spen—memory 13.2 9.0 4.2

 

Table XXXVI. Analysis: Mean Time Taken to Overcome the
First Belief, the First Two Beliefs, and
All Three Beliefs

 

Group First Pelief First Two All Three
Beliefs Beliefs

Closed-memory 3,8 9.7 13.3

Op n-memory 3.9 9.1 14.0

 

The general conclusion which comes from such coupar-

isons is that the differences between dog.atic and rigid

thinking are consistent,

found reflect differences

but the Specific differences

in the testinr situation.

\n

Ix;

53

rt I x v .'\_ 7‘ \r

U\,4.‘.4&Lh

To test the relationship between synthesis and analysis
and induction and deduction with dogmatism and rigidity,
Pokeach's Dogmatism Scale and the Cough—Sanford Rigidity
Scale was administered to 331 undergraduates at Tichigan
State University. The 20 students scoring highest ans
the 20 students scoring lowest on the Dogmatism Scale
were selected for individual testing. Poth groups of
subjects were further divided into two groups of ten each

¢

on the basis of their scores on the Pigid

J

~1e.

Ho

ty Sc

9

The first task consisted of a set of twenty five
analogies emphasizing induction, and another set of

analogies emwhas'zing deduction. The subject's per-

p—I

formance on this taS' was measured in terms of the
relative amount of time spent on the preparation period
of the problem and on the solution phase of the problem
as well as the number of errors made. The second problem
was the Benny Doodlebug Problem with separate measures of
analysis in terms of the time required to overcome the
beliefs of the problem and the number of beliefs the
subject could overcome by himself, and synthesis m asured
in terms of the number of minutes required to solve the

problem after each of the beliefs were overcome.

Ln

The hypotheses tested were:

I. The Open group should solve inductive problems
faster and with fewer errors than the closed
groun. The rigid group should not differ from
the non-rigid group on this variable.

This was supported by the data.

II. The non-rigid groue should solve deductive
problems faster and with fewer errors than the
rigid group. The Open group shoulJ not differ
from the closed group on this variable. There
was an insignificant trend in the Opposite
direction.

III. The Open group should solve the synthesis part
of the Denny Doodlebug problem faster than the
closed group. There should be no difference
between the riqid and the non-rigid groups on
this variable.

This was supported by the data but there was
also a significant interaction between dog-
matism and rigidity.

TV. The non—rigid subjects should overcome more
beliefs without outside assistance and faster
than the rigid grOUp. There should be no
difference between the Open and the closed
groups.

In general, this was subported by the data.

V. There should be a relationship between the
ability to solve deductive problems well and
the ability to do well on analysis.

A non-significant trend in the direction
predicted appeared.

VI. There should be a relationship between in—
duction and synthesis such that those who
do well on one task should do well on the
other.
A non-significant trend in the predicted
direction was observed.
The general conclusions of this study are that: (l)
in a problem solving situation in which the primary task
of the subject is to integrate various beliefs of component

parts of the problem Open subjects are SUperior to closed

subjects; ('2) in a problem solving situation in which the

\n
\A

task of the subject is to analyze or discover the parts
of the problem, non-rigid subjects are superior to rigid
subjects; (3) there is a tendency for subjects who excel
in synthesis to also excel on inductive reasoning; (4)
there is a tendency for subjects who excel in analysis

to do well on problems emphasizing deduction.

 

 

 

 

 

 

Adorno, T.W., Frenkel-Erunswik, Else, Levinson, D.J., and
Sanford, 9.3., (1950). The authoritarian :esonality.
Yew York: Harper.

Clarke, a.G. (1929). The Art of itraight Thinhing, A Primer
of Scientific Tethod for social Ingyiry. few York:
D. Appleton-Century Company.

Dewey, John, (1910). How We Think. I'Tew York: D.C. Heath
and Co., Publishers.

Edwards, A.L. (1960). ngeriucntal Wesign Tn Pcycholooical
Research. ‘l'ew York: Holt, Ileinhart, and inns-ton.

Fillenbaum, S. and Jackman, A., (1051). Dogmatism and
anxiety in relation to uriblem solving-—an extension

of Rokeach's results. J. Abnorm. Soc. Dsychol., 63:

0rd, J. .,
P.P.. (195
abilities:

Guilf

 

Comrey, A.L., Green,

and Christensen.

T“: T:‘
1\.‘.’

O). A factor-analytic study of reasoning

 

 

I. Uynoth0505 and descrififlon Of tCStFo

 

Reports from the Psychological Laboratory,
Los hngcles, University

Cailford, J.P.,

of

R

Green,

.F., Christensen,
A factor-analytic study

To. 1,
Southern California.
P.R.. (1951).
of reasoning_abilities: II.

 

.
11 1“
1“»4.‘

.ihistration
Reports from the

no

(A.

of tests and
r3sychological

.L

1“

alysis of results

Laboratory, 30. 3.

Los Angeles, University of Southern California.
Cough, H.G., and Sanford, R.¥., (1952). Rifiidity as a

psychological variable.

University

Assessment
Harvey, O.J.,

playing.
11a7 (07).

Holzinger, K.J.
two factor

Huberman, J.

of Rokeach's study on dogmatism.

of California,
and Research.

Unpublished manuscript;
Institute of Personality

(1963). Cognitive determinents of role

 

Technical Report 20.
, and Harman, H.H.,

analyses,

Psychometrika,

Contract Jonr

3.

(1938). Comparison of

3: 35-60.

 

(1961). A re-examination of certain ascects

{Tn publ i shed . A .

Thesis, University of British-Colombia.

56

/
pf/

/’

IV

’4

57

Johnson, D.E., (1955). The Psychology of Thought and
Judflwent, Yew York: Harper and Brothers.

 

Johnson, D.H., (1960). Serial analysis of think'ng. finn.
H.Y. Acao. Sciences. 91: 66-75,

 

Johnson, D.Y., (1961). Formulation and refermulation of
figure concepts. The Amer. J. of Psychol. 7U: M18-U2b.

 

Johnson D.T. (1962). Serial anal sis of verbal analos
v 9 y t.
problems. J. Educ. Dsychol. 53: 86-88.

 

Johnson, 9.x., and Hall, E.R. (1961). Organization of
relevant and irrelevant words in the solutiOn of
verbal problems. J. Psychol. 52: 99—10%.

 

Johnson, D.r., Lincoln, R.F., and Hall, E.R., (1961).
Amount of material and time of preparation for
solving nroblems. J. Psychol. 51: h57—Q71.

 

Laffey, J.L., (1957). A theoretical and empirical
analysis of loyalty' to and defection from a
belief system. Unpublished H. A. Thesis. Tichigan

State University Library.

Lyda, L., and Fillenbaum, 8., (1960). Dogmatism and problem
solvin5--an examination of the Denny Doodlebug problem.

EcGovney, W.C. (1953). C gmatism and rigidity as determirantsc
of cognition. Unpublished 7.A. Thesis. Michigan State
University, Library.

filler, I.E., (1915). The Psychology of Thinking. Iew York:
The Xacmillan Company.

 

Oram, A.. (1957). Some determinants of the formation and
modification of new belief systems. Unpublished T A
Thesis. Xiehigan State Vniversity Library.

Rokeach, N., (195%). The nature and meaning of dogmatism.
Psychol. Rev., 61: 191—20u.

 

Rokeach, ‘3, C1960). The (Huni end Close] Tjrui. New Yorlu
Basic Books.

 

Rokeach, 1., FcCovney, W.C., and Denny, Y.?., (1955).
A distinction between dogmatic and rigid thinking.
J. Abnorm. Soc. Psychol. 51: 87-93.

 

Swanson, T.S., (1958). Problem solving as related to ‘
dogmatism, depersonalization, and familiarity with
chess. Unpublished TEA. Thesis. ITichigan State
University Library.

58

 

Thurstone, L.L., (1938). Primayy mental abilities.
Psychometric Monogr. Y . 1, Chicago, Fniversity
of Chicago Press.

Thurstone, L.L., and Thurstone, T.C., (1931). Tactorial
studies of intelligence, Psychometric Tonogr.,
Yo. 2, Chicago, Yniversity of Chicago Press.

 

 

Vidulich, R.Y., (1956). The integration of multiple sets /
into a new belief system. anublished ?.A. Thesis.
Tichigan itate T’niversity Library.

Vidulicn, 3.1., (1060). A re-cxamination of the distinction
between dogmatic and rigid thinkfirmg Unpublished Manuscript.

Vidulich, R.‘., Adams, 3., and Creenbaum, 3., (1960). ?elief
Congruence and Cognitive Isolation in “aired Associate-
Learninw. Paper read at Southeastern Psychological

Association, Atlanta, Georgia.

 

Vidulich, P.I., and Kaiman, W.P., (1961). The effects 0f
information source status and dogmatism upon conformity
behavior. Unpublished Manuscript.

Vinacke, W.F., (1953). The fsvchology of Thinkinh. Yew York:
NcGraw—Till Book Company, Inc.

 

T
‘l
L

 

'alker, H.N., and Lev, J., (1953
a

). Stetistocal Inference.
‘En: Yorfiiz Holi:, Reiiniart, 1

rd Vfiniston.

Witkin, U.A., (1950). Individual differences,in ease of
perception of embedded-figures. J. Fersonolity.

 

Within, H.A. (1962). PyycbOIOSicel differentiation: studies
of develonemnt. Sew York: Wiley

 

 

59
APPENDIX

Analogies Emphasizing Induction

1. ink cuttlefish sting bee

2. visit invade friend enemy
3. sturgeon caviar hen egg

h. ascend decend up down

5. pregnant meaningless full empty
6. inauguration completion infant adult
7. retina cochlea eye ear

8. demise cadaver birth baby

9. statue sculptor book author
10. oxygen tank milk bottle
11. focus periphery center circle
12. total section book chapter
13. progeny ancestry forward backward
1h. numismatics hobby waltz dance
15. naive 50phistieated simple complex
16. flask stepper bottle cork
17. noxious injurious large big
18. tabby Spaniel cat dog

19. senility adulthood evening day

20. surplus sufficiency more enough
21. jury convince target aim
22. automatic V machine sour acid
23. regard eager danger afraid
2U. criminal prison bird cage
25. anecdote relate meat carve

10

\-J

O\

I.)
(Q

10
b.)

10

.(I‘

to
\J\

1u1a105fix7s thm.asi7ixm~

(‘T‘x‘
yard
write

round

shoe
mea
horse

bottle

kin
Ln \
LJ

(‘9
t.)

double
key
win

(“”0

La

beginning
foot
letter
square
011t
txvo
arm
foot
vugar
cot
milk
Small
least
fork
ground
for
thick

fail

Deduction

m nocular

safe

drama
tightrOpC
acQuire
omega

swo rd

seal
circular

creditor

protein
herbivorous
V'C in
perennial
increase
groom
heavenly
entry
diverge

solvent

60

binocular
combination
asset

eject
audixvzce
acrobat
dispose
alpha
dagger
envelope
cubical
debtor

(h:et

elbow

bulb
carbohydrate
carnivorous
'blood

anrnlal

converge

i31rfl:rtw3t

PSYCHOLOGY QUEST IONNA IRE

’1

10.
11.

12.
13.

11.

15.

16.

The following is a study of what the general public thinks and feels about a number
of important social and personal questions. The best answer to each statement below is
your personal opinion. We have tried to cover many different and Opposing points of view;

 

you may find yourself agreeing strongly with some of the statements, disagreeing just as
strongly with others, and perhaps uncertain about others; whether you agree or disagree
with any statement, you can be sure that many other people feel the same as you do.

On the IBM answer sheet which you have been given, mark each statement in the space
provided according to how much you agree or disagree with it. Please mark every one.
Blacken spaces +3, +2, +1, -1, -2, or —3 depending on how you feel’in eadh case, using the
following numbers:

 

+3 I agree very much. -3 I disagree very much.
+2 I agree on the whole. —2 I disagree on the whole.
+1 I agree a little. -1 I disagree a little.

Are there any questions? Please make no marks on this booklet.

A person who thinks primarily of his own 17. It is only when a person devotes himself to

happiness is beneath contempt. an ideal or cause that life becomes meaning-
ful.

I an often the last one to give up trying

to do a thing. 18. If I could get into a movie without paying
and be sure I was not seen I would probably

The main thing in life is for a person to do it.

want to do something important.
19. In this complicated world of ours the only

Everyone at times thinks about things too way we can know what is going on is to rely
bad to talk about. on leaders or experts who can be trusted.
In a discussion I often find it necessary 20. I never miss going to church.

to repeat myself several times to make sure

I am being understood. 21. There are a number of persons I have come

to hate because of the things they stand for.
There is usually only one best way to solve -
most problems. 22. I like to know some important people because
it makes me feel important.
lost people Just don't know what's good
for them. 23. There is so much to be done and so little
time to do it in.
It is not always easy to tell the truth.
- 24. I usually maintain my own opinions even
In times like these, a person must be though many other people may have a differ—
pretty selfish if he considers primarily his ent point of view.
own happiness.
25. It is better to be a dead hero than a live
I prefer work that requires a great deal of coward. ‘ '
attention to detail.
26. I do not like everyone I know.

A man who does not believe in some great
cause has not really lived. 27. A group which tolerates too much differences
' of opinion among its own members cannot
lost people get angry sometimes. exist.for long.
I'd like it if I could find someone who 28. I find it easy to stick to a certain sched-
would tell me how to solve my personal ule, once I have started it.
problems.
29. It is only natural that a person should
I often become so wrapped up in something have a much better acquaintance with ideas
I am doing that I find it difficult to turn he believes in than with ideas he opposes.

my attention to other matters.
30. I gossip a little at times.
Of all the different philosophies which ‘

exist in this world there is probably only 31. While I don't like to admit this even to
one which is correct. myself, my secret ambition is to become a

great man, like Einstein, or Beethoven, or
I dislike to change my plans in the midst Shakespeare.

of an undertaking. ‘
32. I do not enjoy having to adapt myself to
new and unusual situations.

-2-

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46.

47.

48.
49.

50.

51.

52.

Even though freedom of speech for all groups
is a worthwhile goal, it is unfortunately
necessary to restrict the freedom of certain
political groups.

Some people vote for men about whom they
know very little.

If a man is to accomplish his mission in
life it is sometimes necessary to gamble
”all or nothing at all."

I prefer to stop and think before I act even
on trifling matters.

lost people Just don't give a "damn" for
others.

Once in a while I laugh at a dirty Joke.

A person who gets enthusiastic about too
many causes is likely to be a pretty "wishy-
washy” sort of a person.

I try to follow a program of life based on
duty.

To compromise with our political opponents
is dangerous because it usually leads to the
betrayal of our own side.

At times I feel like swearing.

If given the chance I would do something of
great benefit to the world.

I usually find that my own way of attacking
a problem is best, even though it doesn't
always seem to work in the beginning.

In times like these it is often necessary
to be more on guard against ideas put out
by people or groups in one's own camp than
by those in the opposing camp.

I do not read every editorial in the news-
paper every day. -

In a heated discussion I generally become

so absorbed in what I am going to say that I
forget to listen to what the others are
saying.

I am a methodical person in whatever I do.

Once I get wound up in a heated discussion
I Just can't stop.

I think it is usually wise to do things in a
conventional way.

There are two kinds of people in the world:
those who are for truth and those who are
against the truth.

Iy table manners are not quite as good at
home as when I am out in company.

Ian on his own is a helpless and miserable
creature.

The United States and Russia have Just about
nothing in common.

55.

56.

57.

58.

59.

60.

61.

62.

63.

65.

66.

67.

68.

69.

70.

71.

72.
73.

74.

75.

76.

-3-

In the history of mankind there have probably
been Just a handful of really great thinkers.

I always finish tasks I start, even if they
are not very important.

The highest form of government is a democracy
and the highest form of democracy is a
government run by those who are most intelli-
gent.

I would rather win than lose in a game.

The present is all too often full of unhappi-
ness. It is only the future that counts.

I often find myself thinking of the same
tunes or phrases for days at a time.

Unfortunately, a good many people with whom
I have discussed important social and moral
problems don't really understand what's
going on.

I have a work and study schedule which I
follow carefully.

Fundamentally, the world we live in is a
pretty lonesome_place.

I find it hard to make talk when I meet new
people.

It is often desirable to reserve Judgment
about what's going on until one has had a
chance to hear the opinions of those one
respects.

I usually check more than once to be sure
that I have locked a door, put out the light,
or something of the sort.

The worst crime a person could commit is to
attack publicly the people who believe in
the same thing he does.

I have never done anything dangerous for
the thrill of it.

In the long run the best way to live is to
pick friends and associates whose tastes and
beliefs are the same as one's own.

I believe that promptness is a very impor-
tant personality characteristic.

lost of the ideas which get printed nowadays
aren't worth the paper they are printed on.

I am always careful about my manner of dress.

It is only natural for a person to be rather
fearful of the future.

I always put on and take off my clothes in
the same order.

Iy blood boils whenever a person stubbornly
refuses to admit he's wrong.

When it comes to differences of opinion in
religion we must be careful not to compro-
mise with those who believe differently from
the way we do.

.e

mam—,3. 2
1".D'
«a BB

3m USE WU

 

HICHIGRN STRTE UNIV. LIBRRRIES

I III III II lllll ll
9 8

 

312 31044 4625