close

Вход

Забыли?

вход по аккаунту

?

A CONTRIBUTION TO THE STUDY OF ATTITUDES AND ATTITUDE ESTIMATES

код для вставкиСкачать
INFORMATION TO USERS
This material was produced from a microfilm copy of the original document. While
the most advanced technological means to photograph and reproduce this document
have been used, the quality is heavily dependent upon the quality of the original
submitted.
The following explanation of techniques is provided to help you understand
markings or patterns which may appear on this reproduction.
1.T h e sign or "target" for pages apparently lacking from the document
photographed is "Missing Page(s)". If it was possible to obtain the missing
page(s) or section, they are spliced into the film along with adjacent pages.
This may have necessitated cutting thru an image and duplicating adjacent
pages to insure you complete continuity.
2. When an image on the film is obliterated with a large round black mark, it
is an indication that the photographer suspected that the copy may have
moved during exposure and thus cause a blurred image. You will find a
good image of the page in the adjacent frame.
3. When a map, drawing or chart, etc., was part of the material being
photographed the photographer followed a definite method in
"sectioning" the material. It is customary to begin photoing at the upper
left hand corner of a large sheet and to continue photoing from left to
right in equal sections with a small overlap. If necessary, sectioning is
continued again — beginning below the first row and continuing on until
complete.
4. The majority of users indicate that the textual content is of greatest value,
however, a somewhat higher quality reproduction could be made from
"photographs" if essential to the understanding of the dissertation. Silver
prints of "photographs" may be ordered at additional charge by writing
the Order Department, giving the catalog number, title, author and
specific pages you wish reproduced.
5. PLEASE NOTE: Some pages may have indistinct print. Filmed as
received.
Xerox University Microfilms
300 North Zeeb Road
Ann Arbor, Michigan 48106
■B-2,0,8£6
LD3907
,G7
Sappenfield, Bert Reese.
1941
A contribution to the study of atti.S25
tudes and attitude estimates.;.
New
York, 1941.
lOp.1.,191,39 typewritten leaves,
tables, forms.
29cm.
Thesis (Ph.D.) - New York university,
Graduate school, 1941.
Bibliography: p. 1-3 at end.
AG7977
Xerox University Microfilms,
Ann Arbor, Michigan 48106
T H IS D IS S E R T A T IO N HAS BEEN M IC R O F IL M E D E X A C T L Y AS R EC EIVED .
L ibr ary
N. Y. U otv :
4
A COHTRIBUTIOH TO THE STUDY OF ATTITUDES
AUD ATTITUDE ESTIMATES
Bert Reese Sappenfield
A.B., De Pauw University
M.A., New York University
Hew York University
April, 1941
Submitted in partial fulfillment of the requirements
for the degree of Doctor of Philosophy at Hew York University
Approved
Advisor
ACKNOWLEDGMENTS
The writer wishes to express his gratitude to
Dr. Edwin R. Henry, to Dr. Douglas H. Fryer, and to
Mr. George Rudy, who through their guidance and
oo-operation made this study possible.
HO
TABLE
OF
CONTENTS
Chapter
I
II
III
IV
V
Page
Introduction
Researches Yielding Data on Comparative
Attitudes of Religious Groups
3
Researches Employing Techniques Suggest­
ive of that Used for the Present Study
8
The Problem
13
Methods and Procedures
19
The Communism-War Questionnaire
20
The Birth Control Questionnaire
24
The Abbreviated C-R Opinionaire
28
Personal Data Blank, Newspaper Check­
list and Menace Checklist
33
General Problems Relating to Administra­
tion of the Questionnaires
36
The Subjects
39
Definition of Scores on the Attitude
Questionnaires
46
Statistical Treatments
51
Co-operation of Respondents
58
Comparative Attitudes of Religious Groups
63
Summary
VI
1
Differences between Groups in their
Judgments of the Attitudes of Typical
Catholics, Jews, and Protestants
70
71
Extent of Agreement among Catholic,
Jewish, and Protestant Respondent Groups
Regarding the Attitude Positions of
Typical Catholics, Jews, and
Protestants
72
TABLE
OF
COUTEFTS
Chapter
VI
(oont'd)
Page
Comparisons of Group Estimates of
Attitudes with Expressed Attitudes of
Catholic, Jewish, and Protestant
Respondents
80
Comparative Variability of Estimates of
Catholic, Jewish, and Protestant
Attitudes by the Different Respondent
Groups
89
Comparisons of Variability in Group
Estimates of Catholic, Jewish, and
Protestant Attitudes v/ith the
Variability in Expressed Attitudes of
Catholic, Jewish, and Protestant
Respondent Groups
96
Summary
VII
104
Inter-Condition Comparisons and Relation­
ships between Self-Attitudes and
Attitudes Attributed to Different
106
Groups
Inter-Condition Comparisons of Mean
Scores:
Comparisons among the
Attitudes and Estimates of the Atti­
tudes of Typical Catholics, Jews, and
Protestants, for each Respondent
Group
107
Inter-Condition Comparisons of Vari­
ability Measures:
Variability Com­
parisons among Attitudes and Estimates
of Attitudes of Typical Catholics,
116
Jews, and Protestants
Correlations between Inter-Condition
Sets of Scores:
Relations between
Attitudes of Respondents and Attitudes
Attributed to Typical Catholics, Jews,
and Protestants; Opinion-Distanees as
Related to Conservatism
125
Summary of Inter-Condition Findings
136
TABLE OF CONTENTS
Chapter
VIII
Page
Reliability Studies and Evaluation of Items
in the Abbreviated C-R Opinionaire
Summary
IX
158
Studies of Factors Related to Groups
Thought to be "Menaces", and to Newspapers
which the Respondents Claimed to Read
160
Summary
X
141
Summary and Conclusions
173
175
Implications of the Research Findings
186
Recommendations for Further Research
189
TABLES
Page
1
2
3
4
5
6
7
8
9
10
11
12
13
Comparative Distributions of Religious
Preferences in Different Samples
41
Pereentage Distributions of Birthplaces of
Respondents, their Fathers, and their
Mothers, for each Sample
42
Percentage Distributions of Ages of Respondents
in each Sample (Age Stated as Age at Last
Birthday)
44
Number and Percentage of Respondents of each
Religion in the C0mmunism-War Sample who
Refused to Respond under Given Conditions
59
Number and Percentage of Respondents, of each
Religion, in the Abbreviated C-R Sample who
Refused to Respond under Given Conditions
61
Self-ScoresJ
Inter-Group Comparisons among
Mean Attitude Scores
64
Self-Scores*
Inter-Group Comparisons of
Variability in Attitude Scores (in Terms
of Standard Deviation Values)
67
Catholic-Scores* Inter-Group Comparisons
among Mean Attitude Scores
73
Jewish-Scores* Inter-Group Comparisons among
Mean Attitude Scores
75
Protestant-Scores* Inter-Group Comparisons
among Mean Attitude Scores
77
Comparisons of Mean Catholic Self-Scores with
Mean Catholic-Scores of All Groups
81
Comparisons of Mean Jewish Self-Scores with
Mean Jewish-Scores of All Groups
84
Comparisons of Mean Protestant Self-Scores
with Mean Protestant-Scores of All Groups
86
TABLES
Page
Table
14
15
16
17
Catholic-Scores*
Inter-Group Comparisons of
Variability in Attitude Scores (in Terms of
Standard Deviation Values)
90
Jewish-Scores* Inter-Group Comparisons of
Variability in Attitude Scores (in Terms of
Standard Deviation Values)
92
Protestant-Scores*
Inter-Group Comparisons of
Variability in Attitude Scores (in Terms of
Standard Deviation Values)
94
Comparisons of Variability in Catholic SelfScores with Variability in Catholic-Scores
of All Groups (in terms of Standard Deviation
Values)
97
18
Comparisons of Variability in Jewish Self-Scores
with Variability in Jewish-Scores of All
Groups (in Terms of Standard Deviation Values) 99
19
Comparisons of Variability in Protestant SelfScores with Variability in Protestant-Scores
of All Groups (in Terms of Standard Deviation
Values)
101
20
Catholics*
Inter-Condition Comparisons among
Mean Attitude Scores
108
Jews*
Inter-Condition Comparisons among Mean
Attitude Scores
111
Protestants*
Inter-Condition Comparisons
among Mean Attitude Scores
113
Catholics*
Inter-Condition Comparisons of
Variability in Attitude Scores (in Terms
of Standard Deviation Values)
117
Jews*
Inter-Condition Comparisons of
Variability in Attitude Scores (in Terms
of Standard Deviation Values)
120
Protestants*
Inter-Condition Comparisons of
Variability in Attitude Scores (in Terms
of Standard Deviation Values)
122
21
22
23
24
25
\
TABLES
Table
26
27
28
29
30
31
32
33
Page
Group Comparisons with Respect to InterCondition Correlation Coefficients*
Abbreviated C-R Opinionaire Scores
126
Group Comparisons of Correlations between
Algebraic Opinion-Distances and the Scores
from which the Differences were Derived*
Abbreviated C-R Opinionaire
130
Group Comparisons of Correlations between
Self-Scores on Abbreviated C-R Opinionaire
and Algebraic Opinion-Distances (not in­
volving Self-Scores)
132
Group Comparisons of Correlations between
Self-Scores on Abbreviated C-R Opinionaire
and Absolute Opinion-Distances
134
Reliability Estimates for the Abbreviated
C-R Opinionaire under Different Conditions,
Computed b y Three Methods
142
Comparisons between Mean Scores on First and
Second Halves of Abbreviated C-R Opinionaire
145
Comparisons between Mean Scores on (l) Radical
Items and (2) Conservative Items in
Abbreviated C-R Opinionaire
147
Contribution of Abbreviated C-R Opinionaire
Items to Total Score
149
34
Inter-Item Tetrachoric Correlation Coefficients*
Frequency of Sign Reversals
152
35
Inter-Item Tetrachoric Correlation Coefficients*
Frequency of Correlations of Various
Magnitudes (Without Sign Reversals)
153
36
C-R Opinionaire*
Group Comparisons of InterItem Tetrachoric Correlation Coefficients,
Involving Item Pairs with Correlations of
*60 or Above for Catholic Group
155
C-R Opinionaire*
Group Comparisons of InterItem Tetrachoric Correlation Coefficients,
Involving Item Pairs with Correlations of
*50 or Above for Jewish Group
155
37
TABLES
Page
Table
38
C-R Opinionaire*
Group Comparisons of InterItem Tetrachoric Correlation Coefficients,
Involving Item Pairs with Correlations of
*50 or Above for Protestant Group
156
39
Biserial r'st Scores on Abbreviated C-R
Opinionaire versus Particular Menaces
Checked (Items Possibly Conservative)
161
Biserial r's* Scores on Abbreviated C-R
Opinionaire versus Particular Menaces
Checked (Items Possibly Liberal)
161
Tetrachoric Correlation Coefficients
between ’’Menaces" having Positive Biserial
r's (equal to 2 ffbis r) with Scores on
Abbreviated C-R Opinionaire
163
Tetrachoric Correlation Coefficients
between "Menaces" having Significantly
Negative Biserial r ’s with Scores on
Abbreviated C-R Opinionaire
163
Biserial r'st Total Number of Menaces
Checked versus Particular Menaces
Checked (Significant Correlations)
165
Biserial r's*
Total Number of Menaces
Checked versus Particular Menaces
Checked (Non-Significant Correlations)
165
Biserial r'st Relation between Number of
Menaces Checked a n d Particular Newspapers
Reported as Read Once a 'Week
169
Biserial r'st Relation between Scores on
Abbreviated C-R Opinionaire and Particular
Newspapers Reported as Read Once a Week
169
40
41
42
43
44
45
46
47
Biserial r'st Relation between Total
Number of Newspapers Reported as Read at
Least Once a Week and Particular Newspapers
Checked
171
APPENDICES
Appendix
Page
A
Bibliography
1
B
The Communism-War Questionnaire
4
C
The Birth Control Questionnaire
8
D
The Abbreviated C-R Opinionaire
11
E
Soores of Catholics, Jews, and Protes­
tants on ’'Attitude toward Communism"
and "Attitude toward War"
14
F
G
Self-Soores and Jewish-Scores of
36 Catholics
14
Self-Scores and Catholic-Scores of
64 Jews
15
Self-Scores and Catholic-Scores of
10 Protestants
17
Self-Soores and Jewish-Scores of
22 Protestants
17
Self-Soores, Catholic-Scores, JewSoores, and Ptotestant-Scores on
"Attitude toward Birth Control"
18
Soores
of 12 Catholics
18
Soores
of 15 Jews
18
Soores
of 14 Protestants
19
Self-Scores, Catholic-Scores, JewishScores, and Protestant-Scores on
Abbreviated C-R Opinionaire
20
Soores
of 63 Catholics
20
Soores
of 151 Jews
22
Soores
of 55 Protestants
26
APPENDICES
Appendix
H
Page
Distributions
Score Distributions of Catholics, Jews,
and Protestants on "Attitude toward
Communism" and "Attitude toward War"
28
Jewish Subjects
28
Catholic Subjects
29
Protestant Subjects
30
Attitude toward Birth Control
I
28
31
Catholic Subjects
31
Jewish Subjects
31
Protestant Subjects
32
C-R Opinionaire Soores
33
Catholic Subjects
33
Jewish Subjects
34
Protestant Subjects
35
Inter-"Menace" Tetrachoric Correlation
Coefficients: All Groups Combined
36
Catholic Group:
Inter-Item Tetrachoric
Correlations— AbbreviatedrC-R Opinion­
aire
37
Jewish Group:
Inter-Item Tetrachoric
Correlations— Abbreviated C-R Opinion­
aire
38
Protestant Group:
Inter-Item Tetraohoric
Correlations— Abbreviated C-R Opinion­
aire
39
/
Chanter I--Introduction
During the past ten or twelve years, hundreds
of papers have been published on various phases of
attitude research*
The popularity of this field suggests
its importance in the judgment of those who seek a clearer
understanding of socio-psychological phenomena.
House (10)* states,
As
"...the behavior of people is largely
determined by what they think other people think and
intendf in other words,
social behavior is largely a
process of the interaction of attitudes” .
Allport ll .
contends that "whatever stability society may have is
clearly due to the similarity of men's attitudes".
The
foregoing statements point out how great a significance
has been attached to attitude research, and thus may
account, in part, for the great number of studies which
have been reported.
The writer has compiled a bibliography of over
400 items, concerned with the application of attitu.de
questionnaires to varying groups, usually students in
psychology or sociology classes.
However, it is not the
aim here to review the extensive general literature.
Mention will be made only of those studies which have
* See Bibliography, Appendix A
yielded data on the comparative attitudes of different
religious groups, and of the few studies which have a
suggestive, though a none too definitive, similarity in
technique to that employed in the present research.
The aim of the present study, as will be pointed
out in greater detail in the following chapter, has been
to investigate attitudes of Catholics, Jews, and Protestants
considered as separate groups, and to compare these self­
attitudes with the opinions the different groups express
regarding the attitudes of Catholics, of Jews, and of
Protestants in general.
The specific technique employed
has been that of requiring subjects to respond to attitude
questionnaires according to (l) their own attitudes,
(2 ) their opinions of the attitudes of Catholics,
(3) their
opinions of the attitudes of Jews, and (4 ) their opinions
of the attitudes of Protestants.
3
Researches Yielding Data on
Comparative Attitudes of Religious Groups
The studies to be reviewed here have provided
evidence on the relative conservatism of Catholics, Jews,
and Protestants in their attitudes with respect to various
issues.
Allport ( 51) attempted to measure the radicalismconservatism and the prejudices of 375 Catholic, Jewish,
and Protestant students by having them respond to a
seven-page questionnaire.
The items of the questionnaire
had been validated by resort to expert opinion--that is,
each item had been assigned a value of 1 (most radical,
to 4 (most conservative) on the basis of the judgments
of 25 professors of social science.
Jews were shown to
be lowest in prejudice and relatively radical.
Catholics
were relatively radical, as compared with Protestants,
while Protestants were high in prejudice and conservatism.
Vetter iXl), on the basis of a questionnaire
study, concluded that Jews tend to be in the extreme
liberal and radical groups.
Nelson
reported data from Porm K of the
Lentz C-R Opinionaire, administered to students in 18
colleges and universities in different sections of the
country.
Catholic, Methodist, Presbyterian, Seventh
‘
i
Day Adventist, and United Brethren institutions were
represented.
Data were interpreted by institutions rather
than by grouoings based on stated religious preferences.
The Catholic institution was found to be "far more con­
servative than any other institution included in this
investigation".
Sukov and Williamson (33) administered the
Minnesota Scale for the Survey of Opinions to 1,324
freshmen at tv e University of Minnesota.
On the economic
conservatism section of the questionnaire Jewish men and
women scored more radical than non-Jewish men and women,
though the differences were not statistically reliable.
Harris and his associates ( ^ ) found Catholics
more conservative than Protestants, and Protestants more
conservative than those who stated no church preference.
The latter difference was statistically reliable.
The
subjects were students in elementary psychology classes
at Purdue University.
Harper's "Social Study" was the
measure of liberalism-conservatism employed.
Droba (&), incidentally to the standardization
of his war attitude scale, found the mean score of Catholics
to be somewhat more favorable to war than was the mean
score of Jews.
Certain Protestant denominations ^.Lutherans
and Episcopalians) scored more favorable to war than did
Catholics, while other Protestant denominations ^Baptists
5
and Disciples of Christ) favored war even less than did
Jews.
However,
the groups were small (F=7 to 3 3 }t and
no measures of the significance of differences were
reported.
Merton
(IS)
reported results from administra­
tion of a scale for measuring attitudes toward the negro,
(constructed by MacCrone) to 679 college students in
classes in sociology at Harvard, Radcliffe, Pennsylvania
State College, Tulane University, and Louisiana State
University.
Catholics were most conservative (having
attitudes least favorable to the negro).
Jews and students
stating no religious preference were most liberal ^most
favorable to the negro), while Protestants were between
the two extremes in attitude.
These relative positions
were the same for the three religious groups in eacu of
the institutions sampled.
Most of the differences were
statistically reliable.
Carlson (3 ) administered five Thurstone
attitude scales (Prohibition, God, Pacifism, Communism,
and Birth Control) to seniors in the University of
Chicago.
The scales were given to students to be filled
out at leisure.
individuals.
Completed scales were returned by 215
Jews were most liberal or radical on each
attitude continuum.
Catholics were most conservative
on all attitudes except toward prohibition.
Protestants
were intermediate on all attitudes except toward prohibi­
tion, on which they were most conservative.
Carlson
concluded that ” ...an undergraduate’s present religious
affiliation, or his early religious training, may be a
more important factor than the student’s sex or later
school training in determining his or her attitude on
these social questions”.
Jones ( ||), as a part of a follow-up study of
the same students during four years in college, reported
the relative standings of Catholics, Jews, and Protestants
on five Thurstone attitude scales (V/ar? Hegroj God,
Influence on Conduct? God, Reality? and Church).
In both
bis follow-up group (N = 69) and a group consisting of
all seniors (K = 126), he found Catholics most conservative,
Jews most liberal, and Protestants intermediate on all
scales except that measuring attitude toward the negro,
on which the relative positions of Catholics and Protes­
tants were reversed.
Most of the researches reviewed were directed
upon other problems than the study of differences in
attitudes of religious groups, and only incidentally
provided data on this problem.
Yet trie results of the
studies have, with reasonable consistency, shown Jewish
7
groups to be more radical or liberal than other religious
groups, and have shown Catholic groups to be more con­
servative than other religious groups.
Protestants have,
in general, been shown to occupy an attitude position
intermediate between that for Jews and that for Catholics.
While mean differences in attitude positions in several
of the individual studies have been small and statistically
unreliable,
the consistency of results indicates that
a combination of samples would probably yield statistical
reliability.
t
Researches Employing Techniques Suggestive
of that Used for the Present Study
The technique used in the present study, as
previously mentioned, required members of a given group
to respond to a questionnaire in accordance with their
own attitudes on certain issues and again in accordance
with the way they believed members of other groups would
respond.
Ho identical technique has been employed, in
the knowledge of the present writer, by other investi­
gators.
However, a few studies have been reported which
made use of techniques having a suggestive similarity
to the present one, either in aim or in detail (though
not in both).
Spencer (22) in his book, Pulera of Conflict,
has described a new approach to personality measurement,
designed to overcome the defects of the more usual person­
ality inventories, which involve self-re jorts of isolated
behavior items that are pre-evaluated as having a given
significance in and of themselves, re.-;ur:.less of responses
to other items descriptive of behavior.
His assumption
was that a response to a particular item cannot have a
constant value, but must be evaluated in terms of the
respondent’s own conception of the importance of the
particular item of behavior.
The method used by Spencer
9
permitted the derivation of a score "not in terms of a
prejudging of isolated behavior, but rather in con­
junction with the subject*^ own evaluation of the behavior,
his appraisal of his parents' ideals regarding it, his
rating of his parents in terms of it...and, lastly, in
terms of his rating of his associates on the variable".
Specifically,
the individual was required to responu to
a series of items, not once but six times, each time
with different instructions-*
according to (a) his own
beliefs, ideals, and aspirations,
his mother's beliefs, etc.,
father's beliefs, etc.,
(b) his belief as to
(c) his belief as to his
(d) his mother's actual behavior
in the situations mentioned,
(e) his father's actual
behavior, and (f) his associates' actual behavior in
regard to the variables.
While Spencer's method was not
concerned with attitudes or with group membership, it was
one of the few instances in the literature wherein subjects
were reou.ired to answer a riven questionnaire in accord­
ance with instructions to respond as they believed other
persons might be expected to respond.
Whistler and Remmers
have reported on the
construction of a generalized attitude scale, which
might be interpreted as having an aim somewhat similar to
that of the present study.
In details the technique was
to
quite dissimilar.
The generalized attitude scale described
had as its aim the comparison of persons or groups with
respect to given variables, such as personality character­
istics, favorableness or unfavorableness of the environ­
ment, quality of experience, and others not reaaily
classifiable.
Three types of statements were used*
X-Y statements, I-He statements, and 7/e-They statements.
For example-t
WX had fewer unpleasant experiences than Y did."
"I have fewer unpleasant experiences than he hoes."
"We have fewer unpleasant experiences than they did."
As an example of the way in which the scale
was meant Tot be used, 'Whistler and Remmers (30) reported
a study in which 150 men and 149 women students in
psychology classes at Purdue University were asked to
compare "We, the Americans of Today" with "Pioneer
Americans", "Pre-War Americans", and "Americans 30 Years
Kence".
Thus, res oor.sts
.-re in ter-& of comparisons
bet eer. t,'. e subjects * own group and other groups.
authors’ name for the scale,
and Group Morale",
The
"A Scale to Measure Individual
indicates that their aim was mainly
that of interpreting the relation between self-concepts
which certain individuals and groups held, as compared
with their concepts of other groups.
In this respect it
would be possible to suggest a relation between the aim
of Whistler and Remmers1 method and that of the tech­
nique of the present research.
The only investigation, known to the writer,
requiring subjects to respond to an attitude question­
naire according to their own attitudes, and then accord­
ing to what they believed to be the attitudes of another
individual, was that of Winslow (3 0 #
A questionnaire
was constructed from the following types of items;
questions on attitude toward the negro and foreign pol­
icy, from Likert's attitude scale| questions on religion
and current economic policy, from H a l l ’s questionnaire;
and questions on present government policy, from recent
newspaper and magazine editorials.
The scale was con­
structed according to Likert's method (IS), that is, by
assigning values of 1 to 5 to responses ranging faaom
strong agreement to strong opposition.
Eighty-six
students in general psychology at Brooklyn College were
given the questionnaire, with instructions, first , to
answer the items as they themselves believed, and,
second, to answer the items as they believed a self­
selected friend of the same sex would respond,
jhach
of the 86 students gave a similar questionnaire to
the particular friend he had in mind when answering
the questionnaire under the second condition.
The
data were interpreted in terms of correlations between
pairs of friends' actual ppinions, and correlations
between estimated and true opinions.
In general, the
true opinions and
correlations between^friend's estimate of opinions were
higher than the correlations between actual opinions
IX
of pairs of friends.
The above mentioned report by Winslow is
important, perhaps, not so much because of the actual
results obtained as because of the suggestiveness of
the technique.
Such a technique might be extended
to the study of self-attitudes, and their relation to
the concepts which members of a given group express
regarding attitudes of another group, differing in some
socially or culturally significant aspect.
The
investigation presently to be reported in detail has been
concerned with such an extension of this technique.
(3
Chapter II— The Problem
The general problem at the basis of the present
study was that of applying a technique, desoribed in the
following ohapter, for determining the oharaoter of
attitudes attributed to members of groups by virtue of
their group membership.
The attitudes aaoribed to members
of given groups should be an important determinant of
behavior toward members of these groups.
Because Catholic,
Jewish, and Protestant student groups were available for
study, the attitudes of these groups and their estimates of
the attitudes of typical Catholios, Jews, and Protestants
were chosen for study.
Specific questions to which answers were sought
can be stated as follows:
1. Is a teohnique whioh depends upon responses
to attitude questionnaires in terms of the respond­
ent's idea of how members of some particular group
would respond a pratiotable technique from the
standpoint of co-operation of respondents?
2. What are the relative attitude positions
of student groups classified aooording to stated
preference for one of the three religions— namely,
the Catholio, the Jewish, and the Protestant religions?
3.
Do
the student groups classified on the
basis of religious preference differ with respect
in­
to the variability in their expressed attitudes?
4.
Do the student groups classified by
religions agree with one another in their estimates
of the attitude positions of typical Catholics,
Jews, or Protestants?
If not, in what respedts
do the groups disagree?
5* Are the student groups more, or less, con­
servative than the attitudes they estimate for
typical individuals with the same religious
affiliation as their own?
How do the estimates
of typical Catholic, Jewish, and Protestant
attitudes differ from the expressed attitudes of
Catholic, Jewish, and Protestant students?
6. How do the student groups of the different
religions compare with respect to individual
differences in their estimates of Catholic, Jew­
ish, and Protestant attitudes?
Do the groups
differ sufficiently in the variability in their
estimates to justify concluding that stereotypes
regarding the attitudes of particular religious
groups are more marked in any given student group
than in any other?
7.Are individual differences more, or less,
marked in the estimates of Catholic, Jewish, or
15
Protestant attitudes than in the expressed attitudes
of Catholic, Jewish, or Protestant students?
Are
individual differences in the former sufficiently
less marked than in the latter to justify con­
cluding that stereotypes exist regarding the
attitudes of any particular religious group?
8 . If the student groups differ in their
attitudes from their estimates of the attitudes of
tynical members with the same religious affiliation,
which other religious groups, if any, do they
estimate to have attitudes equally as conservative
or as radical as their own?
That is, with the
typical members of which religion do the different
student groups identify themselves in attitude
position?
9. Are individual differences within each
group more marked with respect to exoressed
attitudes, or with respect to estimates of
attitudes of typical Catholics,
Jews, or Protestants?
What evidence is provided for the existence of
stereotypes by the comparisons of the individual
differences in attitudes and estimates?
10. Are the individual attitudes of the student
groups classified as to religious preferences more
related to individual estimates of the attitudes
of typical members of their own religion or to
estimated attitudes of typical members of some other
religion?
Are the individual estimates of attitudes
of typical members of the student's own religion
more related to the individual estimates of
attitudes of any given group than to the individual
estimates o f ‘attitudes of any other given group?
11. Is an individual's conservatism related
to the difference between his estimate of attitudes
of tynical members of his own religion and his
estimates of the attitudes of tyoical members of
some other religion?
12. Is an individual's agreement with con­
servative statements, or his disagreement with
radical statements, more important in determining
his score in general conservatism (as measured by
the Abbreviated C-R Opinionaire described in the
following chanter)?
Which of these types of
responses is the more imoortant in determining
the scores based on the respondent's idea of how
typical members of given religions would respond?
13. Is the reliability of an attitude questionnaire--specifically, the Abbreviated C-R Opinionaire--higher or lower when respondents give
responses in terms of their own attitudes than when
17
they give responses in terms of their ideas of how
tynical members of the different religions would
respond?
14. Which type of iterns--radical items or
conservative items--is the more highly related to
total scores on the Abbreviated C-R Gpinionaire?
15. Are the terms "radical" and "conservative"
as apolied to items of the Abbreviated C-R Opinionaire properly descriptive of the items?
That is,
are ra .ical items ocsitively inter-correlated,
conservative items positively inter-correlated,
and radical-conservative pairs of items negatively
correlated?
16. Is the organization of radical and con­
servative attitudes the same for the different
student groups classified on the basis of religious
preference?
Are items which are highly inter­
correlated for a given group likewise highly inter­
correlated for the other grou.ns?
17. Which groups, noted for their political,
social, or economic activities (among those listed
in the Menace Checklist, described in the following
chanter) are more likely to be considered "menaces"
by conservative than by radical students?
Which
groups are more likely to be considered "menaces"
by radical than by conservative students?
18. Are items in the Menace Checklist which
are either positively or negatively correlated wi th
conservatism themselves highly inter-correlated?
19. Is there a significant correlation between
the number of items checked in the Menace Checklist
and general conservatism?
Is there a significant
correlation between number of iters checked in the
Menace Checklist and the number of items checked in
the Newspaper Checklist (described in Chapter III)?
Is there any evidence that the number of items
checked in both checklists can be accounted for
in terms of a tendency to check indiscriminately
many or few items?
20. Which of the New York City newspapers a re
more likely to be read by radical than by con­
servative students?
The methods by which the data were obtained
and analyzed in order to supply evidence in answer to
the above questions are presently to be described
(Chapter III).
Chapter III— Methods and Prodedures
This chapter is devoted to the description of
methods and procedures used in obtaining and analyzing
the data.
The first four sections deal with the descrip­
tion of scales and questionnaire materials employed,
the
directions for administering these materials, and the
methods of scoring.
The next section gives selected
personal data descriptive of the groups of respondents,
and provides an indication of the extent to which the
groups were comparable with respect to data common to
all groups,
Pollowing this, a section is devoted to
general problems relating to administration of the
questionnaires.
Two final sections provide a general
description of the types of scores obtained, and of the
statistical procedures employed for the analysis and
interpretation of the data.
20
The Communism-War Questionnaire
The earliest (1939) questionnaire employed
in the present study was built around scales to measure
attitudes toward communism and war.
These attitudes
were chosen for study, because it was thought that com­
munism and war were issues about which there might be
some misunderstandings between religious groups.
It
was believed that Jewish and Catholic groups would not
differ appreciably in their own attitudes toward these
issues, but that they might think they differed con­
siderably.
The Communism-War questionnaire in its complete
form is presented in Appendix B.
This consisted of items
taken from the scales of the Thurstone series, to measure
Attitude toward Communism and Attitude toward War.
Form
A of the scale to measure Attitude toward Communism,
prepared by L, L, Thurstone, and Form A of the scale to
measure Attitude toward Tfifar, prepared by R. C, Peterson,
were combined into a single sequence of items.
Items
1 , 2, 3, 4, etc., of the combined scale consisted of
item 1 from the war scale, item 1 from the communism
scale, item 2 from the war scale, item 2 from the com­
munism scale, etc.
The combined scale, therefore, con­
tained 40 items, alternately taken from the scales to
;?/
measure Attitude toward War and Attitude toward Com­
munism, so that the two original scales were completely,
though not separately, represented.
The first four
items of the combined scale were*
1. Under some conditions, war is necessary to
maintain justice.
(Scale value = 7.5)
2. Both the evils and the benefits of communism
are greatly exaggerated.
(Scale value = 5.4J
3. The benefits of war rarely pay for its losses
even for the victor.
(Scale value = 3.5)
4. It is only the ignorant and incompetent who
want communism.
(Scale value = 2.2)
The combined scale was presented under two
conditions*
first, with instructions to the respondent
to give his own attitude, and,
second, with instructions
to respond as he thought either a member of the Catholic
Church or a member of the Jewish Congregation would
respond.
Whether the subject, under the second condition,
responded as he believed a Catholic would respond, or
responded as he believed a Jew would respond, depended
on which of two forms of the questionnaire he received.
Instructions for the first condition were
as follows*
This is a study of attitudes toward war and
communism.
Below you will find 40 statements of
different attitudes toward war and communism.
Put a check mark ( ✓ >
statement.
if you agree with the
Put a oross (X) if you disagree with the
statement.
If you oannot decide about a statement, you
may mark it with a question mark.
This is not an examination. People differ
about what is right and wrong in these issues.
Remember that nobody will know what you write,
since your name is not on the paper. Express
your own attitude, even if you think it differs
from that of your friends. Please indicate your
own attitude by a cheok mark when you agree and by
a cross when you disagree.
Instructions for the second condition, on the form
designed for Jewish respondents, were as follows:
This is another study of attitudes toward war
and communism.
On the next page you will find 40
more statements expressing different attitudes
toward war and communism.
This time please mark
them the way you think a member of the Catholio
Church would mark them, if he were expressing his
sinoere opinion about the statements.
Put a cheok mark ( / ) if you think a Catholic
would agree with the statement.
Put a oross (X) if you think a Catholic would
disagree with the statement.
If you cannot decide about a Catholic’s
answer to a statement, you may mark it with a
question mark.
This is not an examination. People differ
in their opinions about what a Catholic would think
is right and wrong in these issues. Please mark
the statements as you sincerely believe a Catholic
would mark them. Indicate your own idea of what a
Catholio would think by a oheok mark if you think he
would agree, and by a cross if you think he would
disagree.
Instructions for the second condition, on the form
designed for Catholio respondents, were the same as those
23
immediately preceding, except that "Jewish Congregation"
was substituted for ,rCatholic Church* and "Jew" sub­
stituted for "Catholic**
The Communism-War questionnaire was scored separately
for attitude toward communism and attitude toward war.
Scoring was done in the manner recommended in the re­
spective Instruction Sheets edited by L. L. Thurstone
That is, the score on each scale consisted in the
median scale value of items checked (agreed with).
High
scores on these scales indicated attitudes relatively
more in favor of the issues.
The Birth Control Questionnaire
It had been assumed in the construction of the
Communism-War questionnaire that it would be impractic­
able to ask members of a given religion to respond as
they thought members of their own religion would respond,
since it was thought respondents might refuse co-opera­
tion under such a condition.
The Birth Control question­
naire was constructed, to be used with a small group,
to determine whether co-operation would be given under
this condition.
Birth control was chosen as the issue
to be studied, since it was thought to be an issue upon
which there might be misunderstandings between religious
groups.
Issues such as communism and war seemed too
emotionally charged to elicit honest opinions in 1940.
The Birth Control questionnaire is reproduced in
Appendix C.
Form A of the scale to measure Attitude
toward Birth Control, prepared by K. A. Y/ang and L. L.
Thurstone, was used in its original item sequence.
The
first four items, with their respective scale values,
were *
1. Birth control is a legitimate health measure.
(Scale value = 7.6)
2. Birth control is necessary for women who must
help earn a living.
(Scale value = 7.4;
3. The practice of birth control may be injurious
physically, mentally, or morally.
(Scale value =
3.3)
25
4. We simply must have birth control.
value = 10.5)
(Scale
The birth control scale was presented under
four conditions of instruction, requiring the respondent
to mark statements*
(b)
(a) according to his own attitude,
according to the way he thought a typical Catholic
would mark them,
(c) according to the way he thought a
typical Protestant would mark them, and (d) according
to the way he thought a tyoical Jew would mark them.
The scale under each separate condition of instruction
was reproduced on a separate page.
The pages were
stapled together so that trie four conditions were pre­
sented in the order listed above.
The instructions under the first condition
were as follows*
This is a study of attitudes toward birth
control.
Below you will find 20 statements
expressing different attitudes toward birth control.
Put a check mark ( ^ ) if you agree with the
statement.
Put a double check {</</) if you strongly agree
with the statement.
Put a cross (X) if you disagree with the
statement.
If you cannot decide about a statement, you
may mark it with a question mark.
This is not an examination.
People differ
about what is right and wrong in this issue.
Remember that nobody will know what you write,
since your name is not on the paper.
JUxpress
your own attitude, even if you think it differs
X (a
from that of
own attitude
double check
a cross when
your friends. Please indicate your
by a check mark when you agree, by a
when you agree emphatically, and by
you disagree.
The instructions under the second condition
were as follows:
This is another study of attitudes toward
birth control.
Below you will find 20 statements
expressing different attitudes toward birth control.
This time please mark them the way you think a
typical member of the Catholic Church would mark
them, if he were expressing his sincere opinion
about the statements.
Put a check mark ( ✓ ) if you think a typical
Catholic would agree with the statement.
Put a double check
) if you think a typical
Catholic would strongly agree with the statement.
Put a cross (X) if you think a typical
Catholic would disagree with the statement.
If you cannot decide about a typical Catholic's
answer to the statement, you may mark it with a
question mark.
This is not an examination.
People differ in
their opinions about what a typical Catholic would
think is right and wrong in this issue.
Please
mark the statements as you sincerely believe a
typical Catholic would mark them.
Indicate your own idea of what a typical
Catholic would think by a check mark if you think
he would agree, by a double check if you think he
would agree emphatically, and by a cross if you
think he would disagree.
The instructions for the third condition were
identical with those for the second condition, except
that "Protestant Churches" was substituted for "Catholic
Church", and "Protestant" substituted for "Catholic".
37
The instructions for the fourth condition were likewise
identical with those for the second, except that "Jewish
Congregation" was substituted for "Catholic Church",
and "Jew" substituted for "Catholic".
The birth control attitude scale was scored in
the manner recommended in the Instruction Sheet
edited by Thurstone, with, however,
this procedure*
one departure from
The score consisted in the median
scale value of all items which were either single­
checked or double-checked, rather than being based on
the median value of double-checked statements, when any
occurred, and on the single-checked statements,
were double-ehecked.
if none
A high score indicated favorable
attitude toward birth control, and would usually be
interpreted as liberal.
n
The Abbreviated C-R Opinionaire
Since the issues of war and communism, previously
used, were considered to have become so emotionally
toned in recent months, it was doubted that, even under
conditions of anonymity, respondents would express their
frank opinions on these issues*
Moreover, the issue of
birth control was considered to be subject to certain
social taboos which would make it difficult to secure
co-operation in obtaining a large sample of respondents.
A questionnaire to measure conservatism-radicalism which
would contain items on various issues was, therefore,
sought.
The C-R Opinionaire, constructed by T. JT. i*entz
and associates, was considered to be adaptable for the
present study.
Though the C-R Opinionaire was too long
to be administered under four different conditions in
the time allotted for the purpose,
the Manual (/V*) for
this Opinionaire provided item evaluations which made
it possible to choose a limited number of items which
were most representative of the scale as a whole.
The Abbreviated C-R Opinionaire is reproduced in
Appendix D.
This was made up of 20 items from the C-it.
Opinionaire, constructed by T. F. Lentz and associates.
The 20 items from the C-R Opinionaire were selected on
the basis of U-L values reported in the Manual for the
C-R Opinionaire ( i f ) .
The U-L value of an item was
used by Lentz as a measure of its validity for discrim­
inating between the most conservative one-third and
least conservative one-third of subjects, selected on
the basis of their total scores on the C-R Opinions!re.
The 10 radical items with the highest among radical
U-L values, and the 10 conservative items with highest
among conservative U-L values, were chosen from i’orm *>
of the C-R Opinionaire for use in the Abbreviated 0-h
Opinionaire.
In order to randomize the arrangement of
the items, the original numbers corresponding to the
items were written on slips of paper, placed in a box,
and shuffled thoroughly.
Then the slips were drawn,
one by one, and given serial orders corresponding to the
order in which they were drawn, for use in the Abbreviated
C-R Opinionaire.
The first four items in the Abbreviated C-R
Opinionaire were*
1. We should celebrate Pasteur's birthday rather
than Washington's, as he has done the world a
greater service.
2 . Conservative people are usually more intelligent
than radical people.
3. Since the theory of evolution has been accepted
by most scientists, it should be taught in our
schools.
4. At the age of 21, people should have the privilege
of changing their given names.
30
The Abbreviated C-R Opinionaire was presented
in identical form on four separate pages, under each of
four different instructions, requiring respondents to
mark statements (a) according to their own attitudes,
(b) as they believed most Catholics would mark them,
(c) as they believed most Jews would mark them, and (d)
as they believed most Protestants would mark them.
The
questionnaire was stapled together so that the latter
three conditions were presented in each of the N(N-lj
orders possible, and each of the six resulting arrange­
ments were so ordered in the total bundle of question­
naires that an approximately equal number of respondents
would receive the questionnaire in each arrangement.
The instruction to respond according to the respondent's
own attitude, however, was presented first in all
questionnaires.
The instructions for the first condition were
worded as follows I
Below are 22 statements listed to see what
people think about many questions.
If you agree more than you disagree with a
statement, mark it plus (+).
If you disagree more than you agree with a
statement, mark it zero (o).
Be sure to place either a plus or a zero to
the left of each number.
3!
This is not an examination.
People differ
about what is right and wrong in these issues.
Remember that nobody will know what you write, since
your name is not on the paper. Express your own
opinion, even if you think it differs from that
of your friends.
Please indicate your own attitude
by a plus when you agree and by a zero when you
disagree.
The instructions requiring respondents to mark
statements as they believed most Catholics would mark
them were worded as follows*
Below are the same 22 statements listed to
see what people think about many questions.
This
time, please mark them the way you think most
Catholics would mark them.
If you think most Catholics would agree more
than they would disagree with a statement mark it
plus (+).
If you think most Catholics would disagree
more than they would agree with a statement, mark
it zero (o).
Be sure to place either a plus or a zero to
the left of each number.
This is not an examination.
People differ
in their opinions about what a -Catholic would think
is right and wrong in these issues.
Please mark
the statements as you sincerely think most uatkolics
would mark them.
Indicate your own idea of what
most Catholics would sincerely think, oy a ilus if
you think they would agree, and oy a zero if you
think they would disagree.
The instructions for the conaition under ..kich
respondents were to mark statements as they believed
Jews would mark them were identical with those for the
above condition, except that "Jews" was substituted for
"Catholics".
Likewise, instructions for the condition
33
requiring responses in terms of how Protestants would
mark statements were identical with the above, except
that "Protestants" was substituted for "Catholics".
The Abbreviated C-R Opinionaire was scored
according to scoring instructions in the Manual
for the C-R Opinionaire.
That is, the score consisted
in the total number of responses agreeing with the key,
plus one-half the number of items omitted.
The key was
conservatively oriented, plus-responses to conservative
items and zero-responses to radical items being credited
toward the score, so that the higher scores indicated
higher degrees of conservatism.
In addition to the
total score, two types of partial scores were derived*
(a) scores on the first 10 and the second 10 items?1 and
(b) scores on items with key responses of plus and key
responses of zero.
The sets of partial scores were used
for comouting split-half reliabilities.
*
While there were a total of 20 items in the
Abbreviated C-R Opinionaire, there were 22 statements
in the entire questionnaire, as indicated by the fore­
going specimen instructions.
Two items from other scales
irere included, but are r.ot discussed in the present
report.
33
Personal Data Blank, Newspaper Checklist
and Menace Checklist
Attached to each of the questionnaires described
above was a front page devoted to personal data and to
two unstandardized checklists, one relating to news­
papers read by the respondent as frequently as once a
week,
the other relating to groups considered "a menace
to the best interests of the United States".
Personal data asked of respondents in all
samples included*
age at last birthday, date of birth,
preferred religion, respondent*^ birthplace, birthplace
of respondent’s father, and birthplace of respondent*s
mother.
the
This data was asked for purposes of describing
samples.
The exact wording ofthe personal data
questions appears in Appendices
B,
C, and D*
The Newspaper Checklist was worded as follows*
Put a circle around the names of the news­
papers which you read as often as once a week.
If you read any newspapers not listed here, write
their names in the blanks.
Daily News
Eerald-Tribune
New York Times
World-Telegram
Daily Mirror
Evening Journal
PM
Brooklyn Eagle
Bronx Home Nev/s New York Sun
Daily Worker
New York Post
Christian Science
Monitor
The
Chief
____________________
____________________
This wording was the same in all question­
naires, except that the orders in which the different
newspapers appeared in the checklists were different.
PM did not appear in the Communism-’Jar questionnaire,
since this newspaper did not then (1939) exist.
The Newspaper Checklist was used in order to
determine whether there was a relation between the
reading of particular newspapers and the attitudes
expressed toward social issues.
The Menace Checklist was used in the various
questionnaires with the curnose of determining whether
there was any relation between measured attitudes and
hostility toward the aims and practices of generally
well-known groups engaged in political and economic
activities.
The Menace Checklist, as it appeared in con­
junction with the Abbreviated C-R Opinionaire, was
worded as follows*
Put a circle around the names of all the
following groups which you think are a menace to
the best interests of the United States.
labor Unions
Socialist Party
Ku Klux Klan
Coughlinites
C.I.O.
Protestants
labor Spies
Dies Committee
New Dealers
W.P. A.
Tammany Hall
A.P.L.
Brain Trusters
Communist Party
Catholics
Republican Party
Jews
Associated Willkie Clubs
Non-Sectarian Anti-Nazi League
Democratic Party
German-American Bunds
American Civil liberties Union
American Liberty league
N a t ’l. Assn. of Manufacturers
Townsendites
Aliens (in general)
3/
This chedklist was generally similar when
^resented in conjunction with the other questionnaires,
though a few of the items were slightly different in
wording.
Also, the item "aliens (in general)'1 did not
appear with either the Communism-War or the Birth Control
questionnaires, while the item "Associated Willkie
Cluhs* did not aopear with the Communism-War question­
naire.
No method of scoring was used for either the
Newspaper Checklist or the Menace Checklist.
Checking
or not checking a given item constituted the data from
these checklists.
3i
General Problems Relating to
Administration of the Questionnaires
The questionnaires were designed for group
administration and were administered to assembled groups.
The Communism-War questionnaire was presented under only
two conditions of instruction, requiring respondents to
mark statements according to (a) their own attitudes,
and (To) their ideas of how Jews would mark them (form
for Catholic respondents), or how Catholics would mark
them (form for Jewish respondents).
In order to assure
that the respondents in each religious group would
obtain the proper form, it was necessary for either the
Catholics or the Jews to identify themselves by a public
show of hands.
This difficulty was overcome in the use
of the two later questionnaires, the Birth Control
questionnaire and the Abbreviated C-R Opinionaire,
since
both required the respondents to mark statements in
four different ways*
(a) giving their own attitudes,
and giving their opinions of (b) Catholic attitudes,
(c) Jewish attitudes, and (d) Protestant attitudes.
These latter questionnaires were each identical for all
respondents, regardless of their religious affiliation,
so that a single form could be administered to every­
body in the assembled group.
31
The time required for administering the Com­
munism-War and Birth Control questionnaires was approxi­
mately 20 minutes each.
Because of severe limitations
on testing time, the C-R Opinionaire was abbreviated,
as described earlier in this report, rather than being
used in its much longer f orm (60 items).
The Abbreviated
C-R Opinionaire required approximately 25 minutes for
administration.
Respondents were asked not to write their
names on the questionnaires, since it was thought that,
under the condition of anonymity, respondents would be
less likely to give insincere responses.
Corey ( 5 )
found, by the use of a 50-item scale to measure attitude
toward cheating on examination#, that his respondents
did not give reliably different results on signed and
unsigned questionnaires.
However, while these results
were obtained on one scale with one group of subjects,
no evidence has been obtained to prove that such results
would hold generally for all questionnaires and all
groups of respondents.
It was thought advisable to
employ the more cautious procedure recommended by Thur­
stone and Chave (a?), who wrote:
“We shall assume that
an attitude scale is used only in those situations in
which one may reasonably expect people to tell the truth
about their convictions and ooinionsw.
3%
Oral instructions of a general nature pre­
ceded the administering ox each of the questionnaires.
These instructions emphasized the research intention of
the questionnaires, and stressed the frankness of response
made possible by the anonymous conditions.
The subjects of this study may be described
in terms of certain specific data.
obtained through two sources:
These data v/ere
(a) the circumstances
under which respondents were assembled for administering
the questionnaires, and (b) responses to the questions
on the Personal Data Blank.
Educational Status.
The respondents provid­
ing data for this research were all students, or candidates
for admission as students, in liberal arts or engineering
courses in a large metropolitan university.
One sample,
those who responded to the Communism-'ifar questionnaire,
composed of
w a s ^ h i g h school graduates applying for admission to the
university.
These respondents reported for entrance
examinations on August 12, August 26, and September 9,
1939, and responded to the questionnaire during the
period from 9t30 to 10:00 A.M., immediately preceding
the examinations.
A second sample, respondents to the
composed of
Birth Control questionnaire, w a s ^ j u n i o r s and seniors
in engineering who were enrolled in two sections of an
industrial psychology class.
The Birth Control questionr
naire was administered to the respective sections at
8 tOO and at 9*00 A.M., on December 10, 1940.
The third
and largest sample, those to whom the Abbreviated C-h
>tf)
Opinionaire was administered,
composed of
w a s f r e s h m e n in liberal
arts and engineering courses who were completing their
first semester in the university.
This sample responded
to the Abbreviated C-R Opinionaire during a freshman
chapel, which met at 12:00, noon, on January 14, 1941.
Sex.
All respondents were male students, since
only male students are admitted to the two colleges of
the university at which the samples were obtained.
Religious Preferences.
The religious preference
of each respondent was obtained from his Personal Data
Blank,
and provided the basis for classification into
religious groups, as referred to throughout this report.
The distribution of religious preferences in each of the
three samples is shown in Table 1.
The Communism-War
and C-R Opinionaire samples, with respect to percentage
distribution of religious preferences, were quite similar.
Approximately one-fourth were Catholics (25.2$ and 23.4$),
approximately one-half were Jewish (50.4$ and 56.2$), and
approximately one-fourth were Protestants
(24.4$ and 20.4$/.
The distribution of religious preferences in the Birth
Control sample differed from that of the other two
samples, the percentages of Catholics, Jews, and Protestants
being 29.3$, 36.6$, and 34.1$, respectively.
Nativity.
Over 75$ of all respondents were
born in the United States (Column I, Table 2), and in
only three groups, Jewish B-C and Protestant B-C and
C-R, were the percentages less than 90$ (Column I,
•rows E, H, and i).
For each religious group, the
Table 1
Comparative Distributions of Religious
Preferences in Different Samples
Religious
Preference
Catholic
Jewish
Protestant
Total
Comm .-War
P
%
Samples
Birth Control
P
$
C-R i
O pinionaire
P
25.2
50.4
24.4
12
15
14
29.3
36.6
34.1
63
151
55
23.4
56.2
20.4
127 100.0
41
100.0
269
100.0
32
64
31
percentages of respondents of U. S. nativity were quite
similar in the C-W and C-R samples (Column I, rows
A and C, D and F, G and I), while these percentages
differed from those for the B-C sample of Jews and
Protestants (Column I, rows E and H).
The percentage
of respondents who had fathers born in the U. S. was
approximately the same in the C-W and C-R samples, for
each religious group (Column IV, rows A and C, D and F,
G and I).
For Catholics the percentages were 41,7 and
49,2, for Jews 37,5 and 32.4, and for Protestants 65,7
and 63.6.
Only 25^ of Catholics in the B-C sample
(Coluimn IV, row B) had fathers born in the U. S.
With
Table
7L
Percentage Distributions of Birthplaoes of Respondents,
their Fathers, and their Mothers, for each Sample
Group
N
Cath.
A. C-W
B. B-C
C. C-R
Respondent’s
Father's
Mother's
Birthplace
Birthplace
Birthplaoe
U.S. Eur. Etc. U.S. Eur. Etc. V .S .
Etc.
1
11 " H I ” TIT ' V
11
VI "VTY ' VIII
36 97.2
12 100.0
63 96.8
2.8 0.0
0.0 0.0
3.2 0.0
41.7 52.7
25.0 75.0
49.2 49.2
5.6
0.0
1.6
50.0 47.2
25.0 66.7
55.6 44.4
2.8
8.3
0.0
1.6
0.0
4.0
42.1 54.7
53.3 46.7
41.7 56.3
3.2
0.0
2.0
Jew.
D. C-W
E • B —C
F. C-R
64
15
151
96.9 3.1 0.0
80.0 20.0 0.0
92.7 5.3 2.0
37.5 60.9
40.0 60.0
32.4 63.6
Prot.
G. C-W
H. B-C
I. C-R
32
14
55
96.9 0.0 3.1
78.6 14.3 7.1
89.1 7.3 3.6
65.7 28.1 6.2
57.2 35.7
7.1
63.6 25.5 10.9
C-W:
B-C:
C-R:
78.2 15.6 6.2
57.2 35.7 7.1
63.6 25.5 10.9
Communism-War Sample
Birth Control Sample
Abbreviated C-R Opinionaire Sample
+3
respect to birthplaces of the respondents * mothers
(Column VII ), again the percentages for those born in
the U. S. were, in general, more similar between the
C-W and C-R samples than between either of these and the
B-C sample.
The resoondents in each religious group,
then, were, on the whole, more similar in the C-W and
C-R samples than in either of these and the B-C sample,
with respect to nativity and nativity of parents.
From
78# to 100# of all respondents were native-born, while
from 25# to 66# of their fathers and from 25# to 78#
of their mothers were born in the United States.
Ages of Respondents.
The data of Table 3,
comparing the samples with respect to ages of respondents,
reveal close comparability between the C-Vf and C-R
samples (Columns I and III, IV and VI, VII and IX).
The modal ages were the same in these two samples, for
Catholics, Jews, and Protestants.
The modal ages in the
B-C sample (Columns II, V, and VTIl) were 2 to 3 years
higher than for the other two samples.
This was an
obvious finding, in view of the fact that the B-C sample
consisted of upperclassmen, while the other samples were
composed of freshmen and candidates for admission to the
university.
The modal ages of Catholics and Protestants
were a year higher than for Jewish respondents in the
C-V7 and C-R samples.
Table
3
Percentage Distributions of Ages of Respondents in each
Sample (Age Stated as Age at Last Birthday)
Ages
C-W
1
23
22
21
20
19
18
17
16
2.8
0.0
0.0
5.5
16.7
36.1
30.6
5.5
15
2.8
Catholics
B-C
C-R
11
111
8.3
58.4
8.3
16.7
8.3
Jews
d-W
B-C
t
4.8
0.0
1.6
3.2
25.4
49.2
9.5
6.3
3.1
7.8
23.5
46.9
15.6
3.1
63
64
13.3
0.0
40.0
20.0
13.3
13.4
14
N
36
12
C-W:
B-C:
C-R:
15
Protestants
B-C
C-R
TIT
Vlll
It
d-W
ii
2.0
6.0
35.8
43.7
11.3
0.6
0.6
151
3.1
0.0
6.3
9.4
40.6
40.6
7.1
28.6
50.0
14.3
32
14
Communism-War Sample
Birth Control Sample
Abbreviated C-R Opinionaire Sample
3.6
0.0
32.8
40.0
23.6
55
Discussion.
On a •priori grounds, one would
expect that the Birth Control sample was the most
selected of the samples! this sample was composed of
upperclassmen in only one college (engineering), so
that it was pre-selected by all factors which influence
academic mortality, as well as by factors which influaice
choices between an engineering and a liberal arts educa­
tion.
One would expect that the Communism-tfar sample
was least selected, since it was composed of applicants
for admission to both colleges,
and had not yet been
subjected to the factors which would influence selection
or non-selection for college admission.
The C-R
Opinionaire sample consisted of individuals from both
liberal arts and engineering colleges, who had already
been admitted as students, but who had not yet been
subjected to the "weeding-out" process arising from
failure to do satisfactory college work, except, perhaps,
for a few who had dropped out voluntarily.
The comparisons presented in this section have,
in general, indicated, according to expectations, a
fairly close similarity between the Communism-7/ar and
Abbreviated C-R Opinionaire samples, and a fairly general
dissimilarity between either of these and the Birth
Control sample.
Definition of Scores on the
Attitude Questionnaires
Cominunism-War Questionnaire.
The method of
scoring the scales to measure attitude toward communism
and toward war has been described in an earlier section.
Primary scores of the following kind were obtained for
attitude toward each issue*
1. Self-scores for Catholic, Jewish, and Protestant
respondents.
2. Catholic-scores for Jewish and Protestant
respondents.
3. Jewish-scores for Catholic and Protestant
respondents.
Birth Control Questionnaire.
The method of
scoring the scale to measure attitude toward birth control
has been described earlier.
Primary scores of the
following kind were obtained for attitude toward this
issue-*
1.
2.
3.
4.
Self-scores for all respondents.
Catholic-scores for all respondents.
Jewish-scores for all respondents.
Protestant-scores for all respondents.
Abbreviated C-R Opinionaire.
The method of
scoring the Abbreviated C-R Opinionaire has been de­
scribed earlier.
Primary scores of the following kind
were obtained from responses to the Abbreviated C-R
Opinionaire *
1.
2.
3.
4.
Self-scores for all respondents.
Catholic-scores for all respondents.
Jewish-scores for all respondents.
Protestant-scores for all respondents.
Partial scores based on (a) the first 10 and the second
10 items, and (b) the 10 radical items and the 10 con­
servative items of the Abbreviated C-R Opinionaire were
obtained as follows, for all respondents!
1.
2.
3.
4.
Partial
Partial
Partial
Partial
self-scores of both types.
Catholic-scores of both types.
Jewish-scores of both types.
Protestant-scores of both types.
Secondary, or arithmetically derived scores, as follows
were also obtained:
1. Opinion-Distance scores.
2. Misconception scores.
Self-Scores.
Scores based on the responses
under the condition of instruction requiring that the
respondent indicate his own attitude or opinion toward
the issues are referred to as self-scores.
Catholic-Scores.
Scores based on responses
according to the respondent *s idea of how Catholics
would respond are referred to as Catholic-scores.
Jewish-Scores.
Scores based on responses
according to the resoondent's idea of how Jews would
respond are referred to as Jewish-scores.
Protestant-Scores.
Scores based on responses
according to the respondent's idea ox how Protestants
would respond are referred to as Protestant-scores.
For each respondent, then, there were four
scores (except on the Communism-V/ar questionnaire),
namely:
(b)
(a) his self-score, indicating his own attitude
his Catholic-score,
indicating what he thought would
be a typical Catholic's attitude;
(c) his Jewish-score,
indicating what he thought would be a typical Jew's
attitude; and (d) his Protestant-score, indicating what
he thought would be a typical Protestant's attitude.
Opinion-Distance Scores.
The term "opinion-
distance", analogous to the social-distance concept in
sociological terminology, was coined as a term to apply
to differences between scores of the four kinds defined
immediately above.
There were twelve "opinion-distance"
scores for each respondent:
six algebraic differences,
and six absolute differences.
Thus there were opinion-
distance scores consisting in:
1.
2.
3.
4.
5.
6.
Self-score minus Catholic-score.
Self-score minus Jewish-score.
Self-score minus Protestant-score.
Catholic-score minus Jewish-score.
Catholic-score minus Protestant-score.
Jewish-score minus Protestant-score.
Likewise,
there were six opinion-distance scores con­
sisting in absolute differences between:
1.
2.
3.
4.
5.
6.
Self-score and Catholic-score.
Self-score and Jewish-score.
Self-score and Protestant-score.
Catholic-score and Jewish-score.
Catholic-score and Protestant-score.
Jev/ish-score and Protestant-score.
Opinion-distance scores were, then, measures of the
amount by which a respondent's conservatism exceeded
■f?
(or differed from) the conservatism he attributed to
Catholics, Jews, or Protestants, and the extent to which
his opinions of the conservatism of a given group exceeded
(or differed from) his opinion of the conservatism of
another group.
Misconception-Scores.
The term "misconception-
score" refers to the subject’s score when responding
as he believed members of a given group would respond
minus the mean self-score for the given group.
Thus,
a misconception-score consists in the divergence of a
respondentrs opinion of a group's attitudes from the
expressed opinion of the group in question.
Thus, for
each respondent, there were three misconception-scores*
1. A Catholic misconception-score*
the con­
servatism which he attributed to Catholics
minus the conservatism of the Catholic group
of respondents.
2. A Jewish misconception-score*
the conservatism
he attributed to Jews minus the conservatism of
the Jewish group of respondents.
3. A Protestant misconception-score*
the con­
servatism which he attributed to Protestants
minus the conservatism of the Protestant group
of respondents.
Actually,
the term "misconception-score" is a misnomer,
since there was no way of determining that the respondents
had in mind Catholic students, for example, when respond­
ing as they thought Catholics would respond.
words,
In other
the mean self-scores of respondent groups in the
present study could not validly be used as criteria
of the accuracy of the respondent's judgment of how
"most" members of a given group would respond.
term, however,
The
is used for convenience of discussion,
and is operationally interpreted.
3!
Statistical Treatments
One of the major statistical treatments in
the present research consisted in the statistical
evaluation of differences between the means of scores
for different groups, and for different conditions of
instruction.
Corresponding to the comparisons between
mean scores were statistical evaluations of differences
between standard deviations of distributions for different
groups, and for different conditions of instruction.
Thus, there were mean comparisons and variability comrarisons of the types described below.
Inter-Condition Comparisons.
The comparisons
between means and between standard deviations, within'
groups, of scores under the four different conditions
of instruction have been termed "Inter-Condition"
comparisons,
(e. g., comparison between self-scores
and Catholic-scores for Jewish group).
Inter-Group Comparisons.
The cor.oarisons
between means and between standard deviations of scores
for different groups, under a given condition of instruc­
tion, have been termed "Inter-Group" comparisons (.e. g.,
comparison between the self-scores of Jews and Protestants..
Also, comparisons were made between the means
or between standard deviations of self-scores for a
5 2
given group and the means or standard deviations of
scores for the same or another group when responding
according to the subject’s idea of how the given group
would respond (e. g., comparison between self-scores
of Catholics and the Catholic-scores of Protestants*
between the self-scores of Jews and the Jewish-scores
of Catholics,
etc.).
53
Formulae.
In all comparisons of means, the
formula used for the standard error of the mean was
=
-SBiaLL
ITT
in accordance with Fisher’s recommendation(
f.bX )•
This is a more conservative formula than that having
7^ N
as its denominator, since it results in a larger
standard error.
For all inter-condition comparisons, where the
inter-correlations between scores could be computed, the
formula for the standard error of the difference was
ai):Lff#
] / aMl +
“ 2 r12 aMl aM2
*
For inter-group comparisons, where inter-correlations
between scores were not possible, the formula was, simply
Instead of using the criterion requiring that
a difference be 3.00 (or more) times the standard error
of the difference in order to be a reliable difference,
Student's t-distribution was utilized for evaluating
5¥
the significance of all differences.
Differences were
evaluated in terms of two levels of significance, the
.05 level and the .01 level.*
The degrees of freedom
(n) employed for entering the table of t-distribution
was 1 less than the N of the smaller group for inter­
group comparisons,** and 1 less than N for inter­
condition comparisons.
For computing the standard errors of standard
deviations, the formula used was
cDist.
Og =
For computing the standard errors of differ­
ences between standard deviations, the formula for
inter-condition comparisons was
*
A difference significant at the .05 level would not
occur in more than
of similar samples if the null
hypothesis (that the true difference is zero) were true.
A difference wignificant at the .01 level would not occur
in more than 1% of similar samples if the null hypothesis
were true (/&; f.tf,),
**
This was a very conservative procedure, since it
waB equivalent to assuming that the larger group had the
same number of cases as the smaller group.
For inter-group comparisons,
"
y
aal
the formula was
+' h
Student’s t-distribution was used for evaluating differ­
ences.
For evaluating differences between standard
deviations for the Birth Control sample, for which the
IT’s were all less than 20, the variance ratio was
computed.
This involved the use of (N-lJ rather than
N in the denominator of the standard deviation formula.
The formula used for variance ratio was
The larger of the two variances
(standard deviations
squared) was always used in the numerator,
1'he
variance ratio was evaluated by reference to Snedecor's
tables of F ( /
~(ofo).
Another method of statistical treatment consisted
in the correlating of pairs of continuous variables or
scores.
Pearson product-moment coefficients of correlation
were used for this purpose in all samples, except the
Birth Control sample, in which the N's were small, m o t the
jjfc
latter sample, the rank-difference correlation procedure
was used.
In order to evaluate the reliability of
Pearson r fs, the standard error formula recommended by
Pisher
333) was used*
0 r = ~r --------1 - *z
? N - 1
Biserial correlation coefficients were used
for determining relationships between variables such as
self-scores on the Abbreviated C-R Opinionaire and:
responses to particular items,
ia,
(b) the checking or non­
checking of given items in the Menace Checklist, and
(c)
the particular items checked in the Newspaper Check­
list.
The method described by Dunlap ( 7 ) was used for
computing biserial r's.
Reliability of biserial r's
was computed by means of the approximation formula (1, f. 3S0, .
1
a
bis.r
' p<i
,
a, _______ z
~ rbis.
___
y
i
Tetrachoric inter-ccrrelation coefficients
among all items of the Abbreviated C-R Opinionaire, and
among all items of the Menace Checklist were computed.
All tetrachoric r's were computed by means of the com­
puting diagrams of Chesire, Saffir, and Thurstone
Reliability statistics for tetrachoric r's were not
computed.
5*
Chapter IV
Co-operation of Respondents
Before any of the data had been collected for
the present research, there was some doubt about the
practicability of a technique which required that re­
spondents answer a questionnaire according to their
ideas of how members of their own or other groups would
respond.
It was thought that possibly 50# or more of
the respondents would refuse to co-operate when given
instructions to respond in this unaccustomed manner.
Fortunately, as the following analysis indicates, co­
operation of respondents exceeded such a conservative
expectation.
Results with respect to refusals in the
Communism-War sample are shown in Table 4.
No Catholics
or Jews refused to respond under the self-condition,
and only 3.1# of Protestants refused under this con­
dition.
Approximately 10# (actually 11 .1$, 9.4#, and
12.5#) in each group refused to respond under some one
condition.
This left about 90# of respondents in each
group who completed the scales under both conditions.
In the Birth Control sample, all respondents
completed the questionnaire for all conditions.
Perhaps
the fact that the questionnaire was given during a
5?
regiilar class hour would account for the completeness
of co-operation in this instance.
Table 4
Number and Percentage* of Respondents
of each Religion in the Communism-War
Sample who Refused to Respond under Given Conditions
Condition
Catholic
Self-Condition
Catholic-Condition
Jewish-Condition
Any Condition
N =
0 (o.c)
4 (11.1)
4 (11.1)
36
Group
Jewish
0 (0 .0 )
6 (9.4)
6 (9.4)
64
Protestant
1
1
2
4
(3.1)
(4.5)**
(20 .0 )***
(12.5)
32
*
Percentages (in parentheses) computed on basis
of total in group subjected to given condition.
**
N = 22
** *
N = 10
The results for the large C-R Opinionaire
sample provided more possibilities for analysis of the
problem under discussion.
Of a total of 286 respondents,
there were 12 (or 4,2%) who failed to state any religious
preference.
Since the statement of religious preference
was essential for dividing the sample into religious
groups,
there was therefore a loss of 4.2$ of respondents
owing to absence of co-operation in this resoect.
There
were 5 additional respondents (1.7$ of the sample) who
stated preferences (such as Hindu, Buddhist,
etc.) which
could not be classified into the groups originally
kO
planned.
However,
this loss of 1.7$ could not he at­
tributed to lack of co-operation.
Table 5 gives the number and percentage of the
remaining 269 respondents who refused to respond under
given conditions.
Of the Catholic respondents 14.3#
refused to resoond under one or more conditions, while
the percentages were 17.2$ and 7.3$ for Jewish and
Protestant respondents, respectively.
Catholics,
There were 5
7 Jews, and 1 Protestant (7.9$, 4 .6$, and
1 .8$ of the respective groups) who refused under only
one condition.
There were 2 Catholics, 9 Jews, and 1
Protestant (3.2$, 6 .0$, and 1 .8$, respectively) who
refused under two conditions! and 2 Catholics (3.2$),
10 Jews (6 .6$), and 2 Protestants (3.6$) who refused
to respond under three conditions.
Hone refused to
resoond under the self-condition.
For all groups combined, 39 respondents
(14.5$) refused to resoond under one or more conditions.
If to the 39 who failed to respond under one or more
conditions are added the 12 who refused to state a
religious preference, there were a total of 51 (.17,8$)
potential respondents out of the original 286, who
failed to supply responses under all conditions.
Ex­
cluding the 5 who could not be classified, this left
230 (80.5$ of all potential respondents) who responded
under all conditions.
Table
JT
Number and Percentage* of Respondents, of Each Religion,
in the Abbreviated C-R Opinionaire Sample who Refused
to Respond under Given Conditions
Condition
I
_______Group_______
Catholic
Jewish
Protestant
N s 63
N = 151
N « 55
II
III
IV
0 (0.0)
0 (0.0
Cath.-Condition only
1 (1.6)
1 (0.7)
0 (0.0
Jewish-Condition only 2 (3.2)
4 (2.6)
0 (0.0
Prot.-Condition only
2 (3.2)
2 (1.3)
1 (1.8
Cath. & Jewish
0 (0.0)
3 (2.0)
0 (0.0
0 (0.0)
6 (4.0)
0 (0.0
Jewish & Prot.
2 (3.2)
0 (0.0)
1 (1.8
Cath., Jewish,
& Prot.
2
10 (6.6)
2 (3.6
26 (17.2)
4 (7.3
&
Cath.
.
Total
Prot.
fcQ
0 (0.0)
CM
Self-Condition
9 (14.3)
* Percentages shown in parentheses.
If a respondent provided scores under as many
as two conditions
(self-condition and another condition),
however, his results were useful for at least one intercondition comparison or inter-condition correlation.
This left only 14 (4.9/£) of the potential respondents
whose results were completely useless for inter-condition
comparisons.
I’rom the standpoint of co-operation of re­
spondents, then,
the evidence has clearly shown the
technique to be practicable.
!>3
Chapter V
Comparative Attitudes of Religious Groups
In the introductory chapter, a number of
researches were reviewed which seemed fairly consistent
in their findings.
The tentative generalization seemed
justified that Jews were most liberal or radical, Cath­
olics most conservative, and Protestants intermediate
in attitudes toward the issues which had been investigated.
The present chapter is devoted to the analysis of resinlts
under the self-condition, and is therefore a discussion
of what the different religious groups in the several
samples indicated that they themselves believed,
since
the results immediately to be discussed will consist
of attitude-score statistics as traditionally discussed,
some additional evidence will be provided in answer to
the questions*
Which religious group (Catholic, Jewish,
or Protestant) is the most conservative?
most liberal?
Which is the
Can the religious groups be arranged in
a hierarchy with respect to the liberalism-conservatism
continuum?
In Table 6 are comparisons of mean attitude
scores of the three religious groups, as represented in
the three samples, with respect to the issues of communism, war, birth control, and conservatism.
The mean
(o
Table
Self-Scoress Inter-Group Comparisons
among Mean Attitude Scores
Attitude Measure
Group
I
C-R
V
Communism
II
War
III
Birth Control
IV
Catholic
Jewish
Diff.
t
4.104 (36)
4.784 (64)
0.680
3.236**
3.584 (36)
3.418 (64)
0.166
1.221
5.429 (1 2 )
7.590 (15)
2.161
3.950**
13.056 (63)
9.937 (151
3.119
5.387**
Catholic
Protestant
Diff.
t
4.104 (36)
4. 547 (30*
0.443
1.814
3.584 (36)
3.797 (3ft)
0.213
1.401
5.429 (1 2 )
6.650 (14)
1.221
2.052
13.056 (63)
10.664 (55)
2.392
3.630**
Jewish
Protestant
Diff.
t
4.784 (64)
4.547 (30)
0.237
1.040
3.418 (64)
3.797 (31)
0.379
2.984**
7.590 (15)
6.650 (14)
0.940
3.440**
9.937 (151
10.664 (55)
0.727
1.398
** Significant at the .01 level.
N's are placed within parentheses.
(>?
scores of all groups on attitude toward communism
(shown in Column I I ) fell within the range described as
"slightly opposed to communism"
(At).
The mean score
of Catholics was lowest--i. e., most opposed to com­
munism.
The mean score of Jewish respondents was high­
est, while that for Protestants was intermediately placed
with respect to the means for the other two groups.
The results, therefore, corresponded with the findings
reported in the literature.
However,
the differences
between group means were all quite small, and statistically
unreliable, with the exception of that between Catholics
and Jews, a difference significant at the .01 level.*
Column III of Table 6 provides similar inter­
group comparisons with respect to attitude toward war.
The mean scores of all groups fell within the range
described as "moderately opposed to war"
(36).
Prot*^
estants were most favorable to war, Jews least favor­
able, and Catholics intermediate in attitude toward
war.
However, again, the differences were small and
statistically unreliable.
Column IV of Table 6 provides an inter-group
comparison with respect to attitude toward birth control.
*
That is, a difference as large would not occur
in more than 1% of similar samples if the null hypothesis
(that the true difference is zero) were true.
The mean scores for both Jews and Protestants fell
within the range described as "in favor of birth oontrol",
while that for Catholics was within the range described
as "neutral or indifferent"
(Mj.
Without regard to
the reliability or unreliability of differences, Jews
were most in favor of birth control, Catholics least
in favor, and Protestants intermediate.
The differences
between means for Catholics and Protestants, and for
Jews and Protestants, were unreliable.
However, the
difference between the means for Jews and Catholics was
significant at the .01 level.
In Column V of Table 6 inter-group comparisons
of mean scores on the Abbreviated C-R Opinionaire are
presented.
The mean score was highest(most conservative)
for Catholics, lowest (most liberal or radical) for
Jews, and intermediate for Protestants.
The only inter­
group difference which was unreliable was that between
Jewish and Protestant respondents.
The differences
between mean scores for Catholics and Jews and for
Catholics and Protestants were significant at the .01
level.
Table 7 provides inter-group comparisons of
the variability in attitude scores about the group
means.
values.
Entries are in terms of standard deviation
Column II presents comparisons of variability
67
Table
7
Self-Scores* Inter-Group Comparisons of Variability in
Attitude Scores (in terms of Standard Deviation Values)
Attitude Measure
Group
I
Catholic
Jewish
Diff.
P
Communism
II
War
III
,951 (36)
(64)
0 , ,120
0.670 (36)
0.607 (64)
0.063
0,
1, ,071
----
1.945 (12)
0.390 (l5)
---------
- -
24 .49 **
———————
0 ,,811
0.650
Catholic
Protestant
Diff.
*
t
0.951 (36)
0.990 (30)
0.039
@ ---0.228
0.670 (36i
0.558 (31)
0.112
1.155
—
Jewish
Protestant
Diff.
P
t
1.071 (64)
0.990 (30)
0.081
0.607 (64)
0.558 (31)
0.049
0.390 (15)
0.990 (14)
t
0. 506
1.945 (1 2 )
0.990 (14)
------ ----
0.551
3.934 (63)
3.593 (151)
0.341
e
-----
0.836
3.934 (63)
3.154 (55 )
0.780
-----
3.86*
----------
C-R
V
Birth Control
IV
—
--------------
1 .681
3.593 (151)
3.154 (55)
0.439
6.45**
------- --
— — — — — —
1.199
* Significant at the .05 level.
** Significant at the .01 level.
N's are placed within parentheses.
(>%
in scores on attitude toward communism.
The scores of
Jews were most variable, and those of Catholics least
variable.
However,
the differences were small and
statistically unreliable.
Column III of Table 7 presents inter-group
comparisons of variability in war attitude scores.
Here again, differences between groups were small and
statistically unreliable.
On attitude toward birth control the inter­
group comparisons of variability shown in Column IV
indicated greatest variability in the scores of
Catholics, least variability in the scores of Jews,
and an intermediate degree of variability in the scores
of Protestants.
All differences were statistically
reliable, that between Catholics and Jews and that
between Jews and Protestants being significant at the
.01 level, and that between Catholics and Protestants
being significant at the .05 level.*
*
These differences were evaluated by means of
Snedecor’s tables of F, reproduced in Lindquist's
Statistical Analysis in Educational Research ( 14).
F is the symbol used to denote ^variance ratio", which
consists of the square of the larger a over the square
of the smaller a.
This method requires that the de­
nominator in the formula for standard deviation be (H-l)
rather than N (/£ , pp. 60-66).
A difference significant
at the .05 level would not occur in more than 5^ of
similar samples if the null hypothesis (that the true
difference is zero) were true.
On the Abbreviated C-R Opinionaire the inter­
group differences in variability were all unreliable,
as shown in Column V of Table 7.
'Jith respect to variability in attitude scores,
then, Catholics were most variable, Jews least variable,
and Protestants intermediate in variability on attitude
toward birth control.
All other differences in
variability of scores were unreliable (Table 7).
10
Summary
The present chapter has discussed results
hearing upon the expressed attitudes or opinions of the
Catholic, Jewish, and Protestant respondents.
Inter-
groxip comoarisons with respect to mean scores on the
issues of communism, war, and birth control, and on the
Abbreviated C-R Opinionaire, gave confirmatory evidence
for the tentative conclusion from a review of the
literature that Jews are, in general, most liberal or
radical, Catholics most conservative, and Protestants
intermediate in their expressed attitudes.
Differences
between mean scores of Jews and Catholics were highly
reliable on all attitudes excent attitude toward war.
The difference between mean scores of Catholics and
Protestants on the Abbreviated C-R Opinionaire was also
highly reliable.
All differences in mean scores of Jews
and Protestants were unreliable.
The religious groups
did not differ significantly with respect to the
variability in attitude scores, except on attitude toward
birth control.
On this attitude measure, the scores of
Catholics were most variable,
intermediate,
variable.
the scores of Protestants
and the scores of Jewish respondents least
Chapter VI
Differences between Groups
in their Judgments of the Attitudes of
Typical Catholics. Jews. and Protestants
The present chapter is a discussion of the
results bearing upon questions as to differences between
groups in their conceptions of the attitudes of typical
Catholics, typical Jews, a nd ty deal Protestants.
In
the first section inter-group comparisons between the
means of Catholic-scores, Jewish-scores, and Protestantscores are discussed.
These comparisons are interpreted
in terms of the comparative estimates by the different
groups of the degree of favorableness or unfavorableness
in the attitudes of typical Catholics, Jews, and
Protestants,
toward the several issues studied. Follow­
ing this is a discussion of differences between the
group estimates of Catholic, Jewish, and Protestant
attitudes and the actual attitudes expressed by Catholic
Jewish, and Protestant respondents.
Next is a discussion
of the comparative variability in group opinions regard­
ing the attitudes of Catholics, Jews, and Protestants.
Finally, the variability in group opinions concerning
the attitudes of Catholics, Jews, and Protestants will
be compared with the variability in the expressed
attitudes of Catholis, Jewish, and Protestant respondents
IX
Extent of Agreement among Catholic. Jewish, and Protestant
Respondent Croups Regarding the Attitude Positions
~
of Typical Catholics. Jews. and Protestants
The data of this section provide evidence on
the extent of agreement among the different respondent
groups in their opinions about what typical Catholics,
Jews, and Protestants think regarding the issues of
communism, war, and birth control, and in their estimates
of the degree of conservatism in the attitudes of typi. cal
Catholics, Jews, and Protestants.
As indicated by the differences in mean
estimates by each group of respondents (Table 8 , Columns
II and III), the Jews thought Catholics to be more in
favor of communism and war than did the Protestants, but
the differences were small and unreliable, and might
have been due to sampling errors.
The Catholic
respondents judged "tyoical Catnolics" to be sligntly
more in favor of birth control than did the Jewish or
the Protestant respondents,
and the Jewish respondents
judged "typical Catholics" to be slightly less in favor
of birth control than did the Protestants.
However,
none of these differences was statistically reliable,
and hence may be attributed to sampling errors (Table
8 , Column IV).
With respect to general conservatism,
as measured by the Abbreviated C-R Opinionaire, one
significant difference was found (Table 8 , Column V).
7
3
Catholic-Scores* Inter-Group Comparisons
among Mean Attitude Scores
Group
I
Attitude Measure
_________________________________________________
Communism
far
Birth Control
C-B
II
III
IV
V
Catholic
Jewish
Diff.
t
Catholic
Protestant
Diff.
t
Jewish
Protestant
Diff.
t
3.858 (12)
3.380 (15)
0.478
0.736
-
.......
3.766 (58)
3.718 (8)
0.048
0.131
-
3.817 (S8)
3.363 (8)
0.454
0.676
14.633 *60j
13.515 *131)
1.118
2.720**
3.858 (12)
3.646 (14)
0.212
0.277
14.633 v60,
13.821 *53j
00.812
1.580
3.380 (15)
3.646 (14i
0.266
0.383
13.515 *131*
13.821 ^53,
00.306
0.668
** Significant at the .01 level.
N*s are placed within parentheses.
14
Catholics attributed to "most Catholics" a significantly
more conservative attitude than aid Jews♦
of Catholic conservatism
made by
fell mid ay hetJeer,the Catholic
Evaluation
Protestant respondents
and Jewish
evaluations
and was not significantly different f ror* either.
In
general, it may be said that, with one exception,
Catholics, Jews, and Protestants (respondents in this
study) gave about the same estimates when asked to
judge what "typical Catholics" would think about the
issues here studied.
The one exception was that
Catholic and Jewish respondents differed in their opinions
about the conservatism of the "typical Catholic", as
measured by the Abbreviated C-R Opinionaire.
Jewish
respondents thought
him more liberal than the Catholic
respondents thought
him to be.
As shown in Table 9, Column II, Catholics and
Protestants did not differ significantly in their
estimates of the attitude of Jews toward communism.
However,
the Catholic respondents thought Jews to be
significantly more in favor of war than did the
Protestants
(Table 9, Column III).
birth control,
On the issue of
there were no significant inter-group
differences in opinion as to what a "typical Jew"
would believe
(Column IV).
In their estimates of the
general conservatism of Jews (Table 9, Column V), both
nf
Table _2_
Jewish-Scores* Inter-Group Comparisons
among Mean Attitude Scores
Attitude Measure
Group
I
Catholic
Jewish
Diff.
t
Catholic
Protestant
Diff.
t
Jewish
Protestant
Diff.
t
Communism
II
War
III
Birth Control
IV
6.675 (12)
5.647 (15)
1.028
1.576
-
5.906 (32)
5.140 (21)
0.766
1.375
........ -
5.037
(32)
3.926 (2l)
1.111
2.520*
—
6.675 (12)
5.732 (l5)
B.943
1.511
5.647 (15)
5.732 (l4)
0.085
0.127
* Significant at the .05 level.
** Significant at the .01 level.
N fs are placed within parentheses.
C-R
V
11.009 (57)
12.254 (134)
1.245
2.306*
11.009
10.538
0.471
0.719
(57)
(52)
12.254 (134)
10.538 (52)
1.716
3.226**
the Catholic and the Protestant respondents indicated
that they thought "most Jews" would he significantly
more radical than the Jewish group thought "most Jews"
would be.
The estimates of Jewish conservatism by both
Catholics and Protestants were about equal, and the
differences were statistically unreliable.
In general,
then, estimates by the different groups of respondents
respecting the attitudes of "Jews" were about equal,
except on attitude toward war and on general conservatism.
The exceptions, for which statistically reliable differenc
occurred, were thatt
Catholics thought Jews more in
favor of war than did Protestants, and both Catholics
and Protestants attributed more liberalism to "most
Jews" than was attributed to them by the Jewish
respondents.
The data of Table 10 provide a comparison of
the different groups in their conceptions of the
attitudes of Protestants.
Catholics and Protestants
thought "typical Protestants" to be slightly more in
favor of birth control than did Jews, but all the
differences between groups regarding opinions about
Protestant attitudes on this issue were statistically
unreliable, and could be attributed to sampling errors
7 7
Table
m
Protestant-Scores t Inter-Group Comparisons
among Mean Attitude Scores
Attitude Measure
Group
I
Catholic
Jewish
Diff.
t
Catholic
Protestant
Diff.
t
Communism
II
....... -
War
III
.............
.......................
-
.......................
....... -............. —
Jewish....... .................. .....
Protestant
-...............
Diff.
...... -................
t
-......
Birth Control
IV
6.375 (12)
6.113 (15)
0.262
0.457
6.375 (12)
6.379 (l4)
0.004
0.008
6.113
6.379
0.266
0.465
C-R
V
13.202 (57)
12.356 (132)
0.846
2.079*
13.202
12.127
1.075
1.940
(15)
(14)
* Significant at the .05 level.
N's are placed within parentheses.
(57)
(5l)
12.356 (132)
12.127 (51)
0.229
0.482
(Column IV).
Catholics attributed significantly more
conservative attitudes to Protestants than did Jews
(Table 10, Column V), but the differences betv.een Cat hoi
and Protestant respondents, arid betv.een Protestant and
Jewish respondents,
vith respect to t’ e conservatism of
"r.ost Protestants",
v.ere small and statistically un­
reliable.
With but one exception, then, all groups
agreed in their estimates of Protestant attitudes.
The
one exception was that Catholics thought Protestants
significantly more conservative than did Jewish
respondents.
To summarize from the mean comparisons in
Tables 8 , 9, and 10, all groups, with but few exceptions
were about equal in their estimates of the attitudes
of tyoical Catholics, Jews, and Protestants.
ceptions were as follows*
The ex­
Catholic respondents thought
"typical Catholics"
to be more conservative than did
Jewish respondents.
Catholics estimated Jews to be more
in favor of war than Protestants estimated them to be,
while both Catholics and Protestants attributed more
liberalism to "most Jews" than was attributed to them
by Jewish respondents.
And, finally, Catholics
attributed more conservatism to Protestants than was
attributed to them by Jewish respondents.
In all
comparisons, Jews and Protestants agreed in their
estimates of the attitudes of Catholics and Protestants*
Catholics and Protestants agreed in their estimates of
the attitudes of Catholics an d Protestants# and dis­
agreed only with respect to the war attitudes of Jews.
Catholics and Jews had significantly different opinions
about the conservatism of all groups.
These different es
between the estimates of Jews and Catholics accounted
for 3 of the 5 instances in which significant differences
occurred.
Comparisons of Group Estimates of Attitudes with
Expressed Attitudes of Catholic. Jewish.
and Protestant Respondents
The question raised here iss
How do the
group estimates, compared wit h one another in the
previous section, compare with the expressed attitudes
of the Catholic, Jewish, and Protestant respondents?
If the attitudes of the respondent groups could be con­
sidered criteria of the correctness or incorrectness of
the group estimates of the attitudes of typical
Catholics, Jews, and Protestants, the question could oe
answered as to how nearly correct the group estimates
were, or as to which group most seriously misjudged
the attitudes of Catholics, Jews, or Protestants.
ever,
How­
there is no justification for assuming that the
respondent groups were typical of Catholics,
Jews, or
Protestants in general, or that students were thought
of by the respondents as being typical.
Nevertheless,
the data in this section will be of value for indicating
whether respondents in each group considered themselves
typical of individuals with their own religious back­
ground, and for indicating whether attitudes attributed
to typical Catholics, Jews, or Protestants were similar
to the expressed attitudes of Catholic, Jewish, or
Protestant respondents.
In Table 11, Catholic self-scores are com-
%\
Table
//
Comparisons of Mean Catholic Self-Scores with Mean
Catholic-Scores of All Groups
Scores
Attitude Measure
C-R
V
Compared
I
Communism
II
Cath. Self
Jew.
Cath
Diff.
t
4.104 (36)
3.766 (58)
0.338
1.495
3.584 (36)
3.817 (58)
0.233
0.615
Cath. Self
Prot. Cath
Diff.
t
4.104 (36)
3.718 (8)
6.386
1.047
3.584 (36)
3.363 (8)
0.221
0.325
5.429 (12)
3.646 (14)
1.783
2.283*
13.056 (63>
13.821 (53)
0.765
1.207
Cath. Self
Cath. Cath
Diff.
t
------------------..........
..........
........ .
......... ..........
..........
5.429 (12)
3.858 (12)
1.571
4.689**
13.056 (63)
14.633 (60)
1.577
3.828**
War
III
Birth Control
IV
5.429 (12)
3.380 (15)
2.049
3.058*
13.056 (63 )
13.515 (131
0.459
0.827
(Jew. Cath = Catholic-Scores of Jewish Subjects.)
* Significant at the .05 level.
** Significant at the .01 level.
N ’s are placed within parentheses.
pared with Catholic-scores of each respondent group.
Both Jews and Protestants thought "typical Catholics"
to he less in favor of communism than the Catholic group
of respondents actually was (Table 11, Column II).
These differences were small and unreliable, however.
Jews thought typical Catholics more in favor and
Protestants considered them less in favor of war than
the Catholic respondents actually were, but again the
differences were unreliable and attributable to sampling
errors (Table 11, Column III).
All respondent groups,
including Catholics, considered tyoical Catholics to be
significantly less in favor of birth control than the
Catholic respondents really were (Table 11, Column IV).
All groups estimated "most Catholics" to be more con­
servative than the Catholic respondent group was (Table
11 , Column V).
The differences were unreliable for the
Jewish and Protestant estimates, but between the con­
servatism of the Catholic respondents and their own
estimate of the conservatism of "most Catholics", the
difference was highly reliable.
Summarizing,
the
estimates of all groups with respect to the favorableness
of "typical Catholics" toward birth control were less in
favor than was the actual attitude of Catholic respondents.
All other group estimates of Catholic attitudes, with
one exception, were not reliably different from the
23
expressed attitudes of the Catholic respondent group.
This exception was that there was a highly significant
difference between the Catholic estimate of Catholic
conservatism and the conservatism of the Catholic re­
spondents.
With respect to both birth control and
general conservatism, the Catholic respondents considered
themselves significantly less conservative than “typical
Catholics1*.
In Table 12, Jewish self-scores are compared
with the Jewish-scores of each respondent group.
The
Catholic respondents estimated “typical Jews" to be
significantly more in favor of both communism and war
(Columns II and III) than the Jewish respondent group
actually was.
Protestants also thought Jews to be more
in favor of communism and less in favor of war than the
Jewish respondents actually were, but the differences
between these estimates and the attitudes of Jewish
respondents were statistically unreliable.
All groups,
including Jewish respondents, thought typical Jews to
be less in favor of birth control (Table 12 , Column
IV) than the Jewish respondents actually were.
The
differences between the attitudes of Jewish respondents
and the estimates of Jewish attitude toward this issue
by both Jews and Protestants were highly reliable.
Table
/ 2.
Comparisons of Mean Jewish Self-Score with
Mean Jewish-Scores of All Groups
Scores
Compared
I
Jew. Self
Cath. Jew
Diff.
t
Attitude Measure
________________________________________________
Communism
War
Birth Control
C-R
II
III
IV
V
4.784 (64)
5.906 (32)
1.122
2.762**
3.418 (64)
5.037 (32)
1.619
4.522**
3.418
3.296
0.508
1.827
7.590 (15)
6.675 (12)
0,915
2.075
9.937 (151)
11.009 (57)
1.072
1.942
(64) 7.590 (15)
(2l) 5.732 (14)
1.858
4.022**
9.937 (151)
10.538 (52)
0.601
1.105
Jew. Self
Prot. Jew
Diff.
t
4.784 (64)
5.140 (21)
0.356
0.835
Jew. Self
Jew. Jew
Diff.
t
.....-.................
.......................
.......................
.......................
7.590 (15)
5.647 (15)
1.943
4.152**
9.937 (l5l)
12.254 (134)
2.317
9.902**
** Significant at the .01 level.
(Cath. Jew = Jewish-Scores of Catholics.)
F's are placed within parentheses.
%s
All groups, again, judged Jewish conservatism to "be
higher than the expressed conservatism of the Jewish
respondents (Table 12, Column V), but only the difference
for Jewish respondents was statistically significant.
In brief, then, Catholics believed Jews to be more in
favor of communism and war than were the Jewish
respondents in this study# Protestants and Jews estimated
typical Jews to be less in favor of birth control than
was the Jewish respondent groups and Jewish respondents
judged "most Jews* to be more conservative than they
themselves were.
Just as Catholics were found to
estimate the attitudes of "most Catholics" more conserv­
ative than their own attitudes,
so also the Jewish
respondents had attitudes less conservative than those
they estimated for "most Jews*.. Both groups of re­
spondents (Catholic ana Jewish) were found to consider
themselves atypical of people in general having the
same religious backgrounds.
Toward the issue of birth control, all groups
estimated the attitudes of Protestants to be slightly
less in favor than the attitudes of the Protestant
respondent group actually were (Table 13, Column IV).
All differences, however, were statistically unreliable
and attributable to sampling errors.
Again, all groups
judged most Protestants to be more conservative than
14
Table
13
Comparisons
sons of Mean Protestant Self-Score
Selfwith
Mean Protestant-Scores of All
Groups
0
Scores
Compared
I
Prot.
Cath.
Self
Prot
Diff.
t
Attitude Measure
________________________________________________
Communism
War
Birth Control
C-R
II
III
IV
V
........
........
6.650 (14)
6.375 (12 J
0.275
0.854
10.664 (55)
12.356 (S3).
1.692
3.363**
Prot. Self
Jew. Prot
Diff.
t
---.....................
...
........
........... ...........
6.650 (14)
6.113 (l5)
0.537
1.015
10.664 (55)
13.202 4l32>
2.538
4.391**
Prot. Self
Prot. Prot
Diff.
t
---...............
.......................
6.650 (14)
6.379 (l4)
0.271
0.667
10.664 (55;
12.127 (51)
1.463
4.447**
** Significant at the .01 level.
(Cath. Prot = Protestant-Scores of Catholics.)
U's are placed within parentheses.
$7
were the expressed attitudes of the Protestant re­
spondents (Column V).
These differences were highly
reliable, being significant at t h e '.01 level.
The
Protestant group, as was found to be true for Catholics
and Jews, were less conservative than they estimated
"most Protestants" to be, and therefore considered themselves atypical of people in general having similar
religious backgrounds.
Summarizing, the data in Tables 11 , 12 , and
13, comparing the mean self-scores of Catholic, Jewish,
and Protestant respondents with the estimated attitudes
of typical Catholics, Jews, and Protestants, have shown
that all groups of respondents considered themselves
more liberal in attitudes than the general population
of individuals having the same religious affiliation.
The only exception to this generalization was in the
case of the birth control attitudes of Protestants#
while Protestant respondents were slightly more in
favor of birth control than they thought "typical
Protestants" to be, this difference was not statistically
reliable.
All groups estimated Protestant conservatism
to be higher than it was for Protestant respondents,
and all groups, likewise, judged Catholic attitudes
toward birth control to be less favorable tthan was the
attitude of the Catholic group of respondents,
only the
Catholics, however, judged Jewish attitudes toward w a r
and cornnunism to "be more favorable than were the expressed
attitudes of Jewish respondents.
Protestants, as well
as Jews, thought tyoical Jews less favorable to birth
control tlan the Jewish respondents actually were.
tt
Comparative Variability of Estimates of Catholic,
Jewish, and Protestant Attitudes by the
Different Respondent Groups
By comparisons among the different respondent
groups with resoect to the variability in their opinions
about the attitudes of typical Catholics, Jews, and
Protestants, some evidence can be obtained respecting
the relative homogeneity of opinions about the attitudes
of different groups.
If the opinions of a given group
of respondents are more homogeneous (less variable)
than those of another group regarding what typical
Catholics think, for example,
then this would be
interpretable as evidence that a cultural stereotype
respecting what Catholics think is more definite in the
former than in the latter group.
Inter-group com­
parisons of standard deviations of Catholic-scores,
Jewish-scores, and Protestant-scores should,
vide an answer to the questions*
then, pro­
Which group of
respondents is most homogeneous with regard to opinions
about what Catholics, Jews, or Protestants think?
Therefore, in which group do there seem to be the most
definite stereotypes?
Is the stereotype about peonle
of one's own religion more, or less, definite?
The data of Table 14 provide comparisons with
respect to the variability of opinions about what
Catholics think.
Jews were slightly more variable in
Table
Jj£
Catholic-Scores* Inter-Group Comparisons of Variability in
Attitude Scores (in Terms of Standard Deviation Values)
Attitude Measure
Group
I
Communism
II
War
III
Catholic
Jewish
Diff.
F
t
..........
---------.......
----- -----
..........
----------
1.845 (12)
1.604 (l5)
----------
1.32
Catholic
Protestant
Diff.
F
t
.................. .....
1.845 (12)
.....
-....... —
2.190 (l4)
......... .......................
1.41
- ............
Jewish
Protestant
Diff.
F
t
Birth Control
IV
1.199 (58)
0.300 (58)
1.604 (15)
0.875 (8)
1.775 (8)
2.190 (l4)
...... -........................
1.88
35.10**
1.86
C-R
V
2.567 (60)
2.741 (l31)
0.174
.....
0.600
2.567 (60)
2.815 (53 )
0.248
0.685
2.741 (l3l)
2.815 (53)
0.074
----0.220
** Significant at the .01 level.
IT's are placed within parentheses.
7/
opinions about the Catholic attitude toward communism
than were Protestants, but the difference in variability
was not statistically reliable (Column II).
Jews were
significantly less variable than Protestants in opinions
about the war attitudes of Catholics (Column III).
Jews were least variable, Protestants most variable, and
Catholics intermediate in variability of estimates
of the typical Catholicrs attitude toward birth control
(Table 14, Column IV).
None of these differences was
statistically reliable.
Protestants were unreliably
more variable in judgments of Catholic conservatism
(Column V) than \7 *re Catholics or Jews, and Jews were
unreliably more variable than were Catholics.
The or.ly
significant difference in variability of estimates of
Catholic attitudes,
then, was that for Jews and
Protestants in their estimates of the war attitudes of
Catholics.
In this one exception, Jewish respondents
were more homogeneous in their estimates than were
Protestants.
The generally unreliable differences msy
be interpreted to indicate that, if a stereotype
existed regarding the attitudes of Catholics, it did
not exist among one group any more definitely than
among others.
The data of Table 15 provide comparisons
among the respondent groups with respect to the
variability of their estimates of Jewish attitudes.
n
J£
Table
Jewish-Scores* Inter-Group Comparisons of Variability in
Attitude Scores (in Terms of Standard Deviation Values)
Attitude Measure
Group
I
Communism
II
Catholic
Jewish
Diff.
F
t
Catholic
Protestant
Diff.
F
t
Jewish
Protestant
Diff.
F
t
War
III
Birth Control
IV
.......................
............... -......
.......................
2.133 (32)
1.808 (21)
0.325
1.949 (32)
1.199 (21)
0.750
0.834
2.424*
.....
1.750 (l4)
.................
-
1.555 (12)
1.962 (l5)
-.........
1.59
S.504 (57)
3.115 (134J
0.389
.
1.021
1.555 (12)
1.750 (l4)
...........
1.26
-----
3.504 (57)
3.270 (52)
0.234
----0.582
1.962 (15)
3.270 (52)
1.26
* Significant at the .05 level
t'B
C-R
V
are places within parentheses.
3.115 (134)
0.115
-----0.518
While Protestants were less variable in judging the
attitudes of Jews toward communism and war (Columns
II and III) than were the Catholics, they were reliably
so only on the issue of war.
In their estimates of
Jewish attitude toward birth control (Column IV),
Protestant respondents were slightly less variable than
Jews and slightly more variable than Catholics, while
Jews were also somewhat less variable than Catholics,
but these differences were all statistically unreliable.
With respect to estimates of Jewish conservatism,
Protestants were slightly more variable than Jews an d
slightly less variable than Catholics, and Jews were
also slightly less variable than Catholics.
The
generally inconsistent direction of differences, and
their general unreliability,
indicated approximately
equal homogeneity for different respondent groupswith
respect to estimates of Jewish attitudes.
The variability among Protestant respondents
in their judgment of Protestant attitudes toward birth
control (Table 16, Column IV) was slightly less than
that among Jewish respondents, and slightly greater
than that among Catholic respondents, while the
variability among Jews was somewhat greater than that
among Catholics.
All differences were, however, un-
n
Table
/&
Protestant-Scoress Inter-Group Comparisons of Variability
in Attitude Scores (in Terms of Standard Deviation Values)
Attitude Measures
Group
I
Communism
II
War
III
Birth Control
IV
Catholic
1.215 (12)
Jewish
1.862 (15)
....................................
Diff.
F.......
-.....
2.35
t
Catholic..... .......................
1.215 (12)
Protestant
------------------1.290 (l4)
Diff.
....................................
F
t
.....................
1.12
-
.....................
2.901 (57)
3.000 (l32)
0.099
----0.300
2.901
2.801
0.100
(57)
(51)
.....
0.256
Jewish....... .......................
1.862 (15)
Protestant
------------------1.290 (14)
Diff.
....................................
F
C-R
V
2.10
t
3.000 (132)
2.801 (51)
0.199
.....
0.594
IT’s are placed within parentheses.
reliable.
Protestants were slightly less variable in
estimates of Protestant conservatism than were Jews
or Catholics (Column V), but Catholics were slightly
less variable than Jews in this res sect.
Again, the
differences were unreliable.
The data of Tables 14, 15, and 16, comparing
respondent groups with respect to the variability in
their estimates of Catholic, Jewish, and Protestant
attitudes, have, by the general unreliability of
differences,
indicated that, on the whole, there were
no group differences in the variability of estimates.
If cultural stereotypes existed regarding the attitudes
of different religious groups, the above results could
be interpreted as evidence that these stereotypes were
equally definite in all respondent groups, at least in
so far as they were identifiable in terms of the
homogeneity of estimates.
u
Comparisons of Variability in Group Estimates of Catholic,
Jewish, and Protestant Attitudes with the Variability
in Expressed Attitudes of Catholic. Jewish,
and Protestant Resoondent Groups
The results of the preceding section, while
implying that stereotypes regarding Catholic, Jewish,
and Protestant attitudes, if they existed, were equally
definite in all groups, did not give any evidence that
such stereotypes did or did not exist.
Some evidence
regarding the possible existence of such stereotypes
should be provided by a comparison of the variability
in estimates with the variability in expressed attitudes
of the respondent groups.
If the variability in
estimates of attitudes of Catholics, for example, were
found to be generally less for tl.e different respondent
groups than that in the actual attitudes of the
Catholic respondent group, this would be evidence of a
cultural stereotype regarding Catholic attitudes.
Such
a finding would mean that given respondent groups differed
less in their opinions about what Catholics thinh than
did Catholics in their actual attitudes.
With respect to attitude toward communism the
variability in the Protestants' estimates of Catholic
attitude was slightly less, and that of Jews slightly
more, than the variability in the expressed attitudes
of Catholic respondents toward this issue (Table 17,
?7
Table j 1
Comparisons of Variability in Catholic Self-Scores with
Variability in Catholic-Scores of All Groups
(in Terms of Standard Deviation Values)
Distributions
Compared
I
Attitude Measure
Communism
II
War
III
C-R
V
Birth Control
IV
3.934 (63;
2.741 (131)
1.193
0.951 (36)
1.199 (58)
0.248
0.670 (36)
0.300 (58)
0.370
1.945 (12)
1.604 (15)
-----
-----
1.559
4.353**
1.46
— -—
Cath. Self
Prot. Cath
Diff.
F
t
0.951 (36)
0.875 (8)
0.076
0.670 (36)
1.775 (8)
1.105
1.945 (12)
2.190 (14)
----------
1.27
0.300
2.376*
- - - - -
2.503*
Cath. Self
Cath. Cath
Diff.
F
------------------..........
..........
.........
.........
1.945 (12)
1.845 (12)
3.934 (63)
2.567 (60)
1.367
.........
1.11
------
Cath. Self
Jew.
Cath
Diff.
F
t
U
----
------------ -------
— — — — — —— —— —
---- -
3.051**
3.934 (63)
2.815 (53;
1.119
3.873**
* Significant at the .05 level.
** Significant at the .01 level.
(Jew. Cath = Catholic-Scores of Jewish Subjects.J
IT's are placed within parentheses.
Column I I ), but these differences were unreliable.
The variability in Jewish estimates of Catholic
attitudes toward war was reliably smaller than that for
the expressed attitudes of Catholic respondents
(Column III), while the variability in Protestant
estimates was reliably greater.
This could be inter­
preted to imply that a stereotype regarding Catholic
attitudes toward war probably existed among Jews but
not among Protestants.
In their estimates of Catholic
attitude toward birth control (Column IV), all groups
exhibited about the same degree of variability as that
in the expressed attitudes of Catholic respondents.
On the other hand, all groups were less variable in
their opinions about the degree of conservatism of
Catholics than were the Catholic respondents in their
expressions of attitude (Column V).
These differences
were highly reliable.
As indicated by the fact of greater homogeneity
in estimates of Catholic conservatism than in expressed
conservatism of Catholic respondents, then, it w.uld
appear that a stereotype probably existed regarding the
conservatism of Catholics.
The evidence was less clear
with respect to a possible stereotype regarding Catholic
war attitudes, and there appeared to be no evidence
for a stereotype regarding Catholic attitudes toward
7?
Table
/^
Comparisons of Variability in Jewish. Self-Scores with
Variability in Jewish-Scores of all Groups
(in Terms of Standard Deviation Values}
Distributions
Compared
I
Attitude Measure
Communism
II
(64)
(32)
War
III
0.607 (64)
1.949 (32)
1.342
Birth Control
IV
C-R
V
Jew.
Self
Cath. Jew
Diff.
F
t
1.071
2.133
1.062
3.738**
5.325**
0.390 (15)
1.555 (l2)
----------16.05**
.......
3.593 (151)
3.504 (57)
0.089
.....
0.229
Jew. Self
Prot. Jew
Diff.
F
t
1.071 (64)
1.808 (2l)
0.737
.....
2.468*
0.607 (64)
1.119 (2l)
0.512
.....
2.626*
0.390 (15)
3.593 (l5l)
1.750 (l4)
3.270 (52)
-.......... 0.323
20.23**----- -----0.843
Jew. Self
Jew. Jew
Diff.
F
t
------------------0.390 (15)
...
- -.........
1.962 (15)
............................-...... .................. .....
25.3**
3.593 (l5l)
3.115 (134J
0.478
2.265*
* Significant at the .05 level.
** Significant at the .01 level.
(Cath. Jew = Jewish-Scores of Catholics.)
H's are placed within parentheses.
(00
communism or birth control, in so far as the evidence
from homogeneity of estimates may be considered a
criterion.
In Table 18 are compared the variability
measures of the expressed attitudes of Jewish re­
spondents with the variability measures of estimates
of Jewish attitudes.
Both Catholics and Protestants
were more variable in their judgments of Jewish
attitudes toward communism and war (Columns II and III)
than were the Jewish respondents in their expressed
attitudes toward these issues.
Likewise, all groups,
including Jewish respondents, exhibited significantly
greater variability in their estimates of the attitudes
of typical Jews toward birth control than existed in
the expressed attitudes of the Jewish respondent group
(Column IV).
On the other hand, the variability within
all groups respecting Jewish conservatism (Column V)
was less than the variability in the expressed con­
servatism of Jewish respondents, as measured by the
Abbreviated C-R Opinionaire.
However, only in the case
of the estimates by Jewish respondents was the variability
reliably less than that in Jewish attitudes.
In all
comparisons, then, in which reliable differences in
variability occurred between expressed attitudes of
Jewish respondents and estimates of Jewish attitudes,
fOl
Table
I?
Comparisons of Variability in Protestant Self-Scores with
Variability in Protestant-Scores of All Groups
Distributions
Prot. Self
Cath. Prot
Diff.
F
t
Cormunism
II
.
Tar
III
Birth Control
IV
.....
-
1.51
.
.....
0.990 (14)
3.154 (55)
Prot. Self
Jew. Prot
1.862 (15) 2.901
....................................
liff.
F .......
-........
3.55*
t.......
-............
Prot. Self
Prot. Prot
Diff.
F
t
C-R
V
0.990 (14) 3.154 (55)
1.215 (12) 3.000 (57)
-
Compared
I
Attitude Measure
......................
......... ......... -
0.154
......
0.402
(132)
0.253
......
0.622
0.990 (14) 3.154 (55)
1.750(l4fl2.801(5l)
-.- 0.353
1.69 ......
1.176
* Significant at the .05 level.
(Cath. Prot ® Protestant-Scores of Catholics/
N*s are placed within parentheses.
I OX
there was but one exception to the general finding that
the Jewish attitudes were less variable than were the
estimates of Jewish attitudes by the different respondent
groups.
The one exception was that estimates of Jewish
conservatism by Jewish respondents were less variable
than their expressed attitudes.
There was, then, no
clear evidence that a stereotype existed regarding
Jewish attitudes.
The variability in Protestant attitudes is
compared with the variability in estimates of
Protestant attitudes in Table 19.
All groups were
slightly more variable in their judgments of Protestant
attitudes toward birth control than were the Protestant
respondents in their attitu.de expressions toward this
issue (Column IV),
However,
the difference was reliable
only in the case of the estimates by Jewish respondents.
Again, all groups were slightly less variable in their
estimates of Protestant conservatism than were Protestant
respondents in their expressed attituues,
but all the
differences were unreliable.
Summarizing the results in Tables 17, 18, and
19, in which are compared the variability in expressed
attitudes of Catholic, Jewish, and Protestant respondents
with the variability in estimates of Catholic, Jewish,
and Protestant attitudes, it can be said that the
103
variability in estimates was, in general, either reliably
greater than, or not reliably different from, the
variability in expressed attitudes of the respondent
groups.
There were 5 comparisons out of a total of
26 in which an exception to this generalization occurred.
Four of these 5 exceptions occurred with respect to
estimates of Catholic attitudes! all respondent groups
were less variable in their opinions regarding Catholic
conservatism than were Catholic respondents in their
expressed conservatism (Table 17, Column V), and Jewish
respondents were less variable in their estimates of
Catholic war attitudes than were the Catholics in their
expressed attitudes toward this issue (Table 17, Column
III).
The remaining exception was that Jewish re­
spondents were less variable in estimates of Jewish
conservatism than in their own expressed conservatism
(Table 18, Column V).
The evidence,
then, favors the
interpretation that a stereotype probably existed *ith
respect to Catholic conservatism! but there was no
clear evidence that stereotypes existed regarding the
attitudes of Jews or Protestants.
/of
Summary
The present chapter has been a discussion of
points relating to the comparative estimates of respondent
groups with respect to attitudes of typical Catholics,
Jews, and Protestants.
The first section discussed the comparisons
among groups with respect to their estimates of the
favorableress or unfavorableness, or the conservatism
or liberalism, of Catholic,
Jewish, and Protestant
attitudes toward different issues.
’7ith few exceptions,
all respondent grouos agreed in their estimates of the
attitudes of Catholics,
Jews, and Protestants.
Dif­
ferences between Jews and Catholics in their estimates
of the conservatism of all grouos accounted for 3 of
the 5 significant differences wr j.ch occurred in a total
of 22 different comparisons.
Comparisons between estimated attitudes of
Catholics, Jews, and Protestants and the expressed
attitudes of Catholic,
Jewish, and Protestant re­
spondents indicated that all respondent groups con­
sidered their ®wn attitudes atypical of the attitudes
of people in general wit', the sure religious affiliat i on s .
Comparisons of variability in the estii? a tea
by the different respondent groups in dice ted that, on
\0S
the wlole,
there were no significant group differences
in variability of estimates of Catholic, Jewish, or
Protestant attitudes.
If relative homogeneity of
opinions regarding attitudes of a particular group
he considered a criterion of the presence of a
cultural stereotype regarding the attitudes of that
particular group, the evidence indicated that such a
stereotype did not ooerate more strongly in any one
group of respondents than in any other group of
respondents.
Comparisons of the variability in estimates
of Catholic, Jewish, and Protestant attitudes, with the
variability in expressed attitudes of respondent groups,
indicated that the variability of estimates was, in
general, either greater than, or not reliably different
from, the variability in expressed attitudes of the
respondent groups.
There were 5 comparisons out of a
total of 26 in which exceptions to this generalization
occurred.
Pour of the 5 exceptions occurred with respect
to estimates of Catholic attitudes.
The evidence
favored the interpretation that a stereotype regarding
Catholic conservatism probably existed, but there was
no clear evidence of stereotypes regarding Jewish or
Protestant attitudes.
/Ok
Chapter VII
Inter-Cord.ition Comparisons and Relationships between
Self-Attitudes and Attitudes Attributed
to Different Groups
The present chapter deals with inter-condition
comparisons and inter-condition relationships.
That
is, for a given respondent group, comparisons are made
between self-scores and scores based on opinions of how
the different groups would respond to the same attitude
statements, as '.veil as comparisons among a given group's
estimates of typical Catholics, Jews and Protestants.
These comparisons are presented both in terms of mean
scores and in terms of variability.
While the preceding chapter emphasized inter­
group comparisons for each given condition,
the present
chapter emphasizes inter-condition comparisons for each
given group.
The chapter is concluded w i t h a discussion of
product-moment correlations between scores under different
conditions, and with correlations between certain scores
and the difference between pairs of scores.
107
Inter-Condition Comparisons of Mean Scores * Comparisons
among the Attitudes and Estimates of the Attitudes
of Typical Catholics. Jews. and Protestants.
for each Respondent Group
This section is concerned with the comparison
of the favorableness or unfavorableness, and the con­
servatism or radicalism,
in attitudes of a given re­
spondent group with the same group's estimates of the
attitudes of typical Catholics, Jews, and Protestants,
and with comparisons among the attitude positions
attributed by a given group to typical Catholics, Jews,
and Protestants.
By means of such comparisons,
it should
be possible to determine whether members of a given
respondent group consider their own attitudes typical
of the attitudes of a more general group with which
they have religious identification,
or typical of the
attitudes of a different group of different religious
affiliation.
Such comparisons would provide evidence
as to whether a given respondent group identifies itself
in attitudes with any typical population of a given
religious background, and whether any given typical
population is identified in attitudes with some other
typical population of different religious background.
In Table 20 are the inter-condition comparisons
of mean attitude scores for the Catholic group.
The
Catholic respondents were significantly less in favor
10%
2.0
Table
Catholics*
Inter-Condition Comparisons
among Mean Attitude Scores
Attitude Measure
Condition
I
Communism
II
Self
Cath
Biff.
t
.......................
..... ..................
........ -......... ....
................... ....
5.429 (12)
3.858 (12)
1.571
4.689**
13.056 ^63,
14.633 (60)
1.577
3.828**
Biff.
t
4.104 (36)
5.906 (32)
1.802
4.040**
5.429 (12)
6.675 (12)
1.246
1.510
13.056 (63)
11.009 (57)
2.047
3.702**
5.429 (12)
6.375 (12)
0.946
1.166
13.056 (63)
13.202 (57)
0.146
0.320
Biff.
t
.......................
....... .
..........
...............-.......
.......................
3.858 (12)
6.675 (12)
2.817
3.552**
14.633 (60)
11.009 (57)
3.624
6.601**
Biff.
t
.......................
-.................... .
.......................
..................... .
3.858 (12)
6.375 (12)
2.517
3.674**
14.633 (60)
13.202 157)
1.431
4i310**
---------...............
6.675 (12)
6.375 (12)
0.300
0.875
11.009 (57/
13.202 (57;
2.193
4.146**
Self
Jew
War
III
3.584 (36)
5.037 (32)
1.453
4.025**
Self
Prot
Biff.
t
Cath
Jew
Cath
Prot
...................
Jew
Prot
Biff.
t
Birth Control
IV
** Significant at the .01 level.
IT's are placed within parentheses.
C-ft
V
109
of communism and war than they thought Jews to he
(Columns II and III).
Toward hirth control, the
Catholic respondents were significantly more favorable
than they thought a typical Catholic would be, but their
own attitude position on this issue was not significantly
different from that which they attributed to Jews and
Protestants (Column IV).
On the Abbreviated C-R Opinion-
aire (Column V), the Catholics were significantly more
liberal than they thought most Catholics to be, and
significantly ; ore conservative than hJ ey thou,-:] t most
Jews to be.
Between their own conservatism and their
orinions of Protestant conservatism there was no
significant difference.
The Catholic respondents,
then,
did not identify themselves in attitudes with tyoical
Catholics.
They identified themselves with typical
Jews only in their attitude toward birth control.
With
resnect to attitude toward birth control and general
conservatism, as measured by the Abbreviated C-R Opinionaire, Catholics identified themselves with typical
Protestants.
The Catholic group also thought that on the
birth control issue (Column IV, Table 20) Jews and
Protestants would be significantly more favorable than
Catholics would be, and the difference between their
1(0
opinions on what Jews and Protestants would think on
this issue was statistically insignificant.
With respect
to issues in the Abbreviated C-R Opinionaire (Column V),
the Catholics thought that Catholics would be most con­
servative,
Jews most radical, and Protestants intermedi­
ate in attitudes.
These differences were all significant
at the .01 level.
The Catholic respondents, then,
thought that tyoical Catholics, Jews, and Protestants
were significantly different in their attitudes, with
the one exception that they believed Jews and Protestants
were equally in favor of birth control.
In Columns II and III of Table 21, the com­
parisons indicate that the Jewish respondents were
significantly more in favor of communism and less in
favor of war than they considered Catholics to be.
Toward birth control (Column IV ), the Jewish group was
sigr.ificartly more favorable than the Jewish respondents
considered the tvoical Catholic, Jew, or Protestant to
be.
Likewise,
the Jewish respondents were significantly
more liberal or radical, as indicated by mean self-score
on the Abbreviated C-R Opinionaire,
than they considered
most Catholics, Jews, or Protestants to be (Column V).
With respect to the opinions of the Jewish
respondents regarding the attitudes of the different
religious groups toward birth control (Column IV), their
HI
Table <% /
Jews*
Inter-Condition Comparisons
among Mean Attitude Scores
Attitude Measure
Condition
i
Communism
ii
Self
Cath
4.784 (64)
3.766 (58)
1.018
5.185**
Diff.
t
War
hi
Birth Control
rv
3.418 (64) 7.590 (15)
3.817 (58) 3.380 (15)
0.399
4.210
5.051**
10.095**
Self
Jew
Diff.
t
Self
Prot
...............
- 4.152**
Diff.
t
.......................
.......................
.......................
.......................
Diff.
t
-.................
-................
........................
—
Cath
Jew
9.937 (151*
13.515 (131*
3.578
10.555**
7.590(15)9.937 (151*
5.647(15)12.254 (134*
1.943 2.317
9.902**
7.590 (15)
6.113 (15)
1.477
3.155**
.....-..................
Jew
Prot
Diff.
t
..................
9.937 (151/
12.356 (132/
2.419
7.803**
3.380 (15)
13.515 (131,
5.647 (15) 12.254 (134)
2.267
1.261
3.338** 4.602**
Cath
Prot
Diff.
t
C-R
v
2.733
5.647 (l5)
6.113 (l5)
0.466
1.148
** Significant at the .01 level.
N ’s are placed within parentheses.
3.380 (15)
13.515 (l3l)
6.113 (15)
12.356 (l32)
1.159
4.358**
5.221**
12.254
12.356
0.102
0.442
(134)
(132)
responses under the different conditions indicated that
they thought the tyoical Jew and the tippical Protestant
to "be significantly more in favor of birth control than
they thought the typical Catholic to be.
But the
difference between their estimates of attitudes of
typical Jews and Proteitants was unreliable.
Likewise
(Column V), the Jewish respondents thought most Jews
and most Protestants to be reliably less conservative
than most Catholics, as indicated by their scores on
the Abbreviated C-R Opinionaire under the respective
conditions.
However, again, their estimates of Jewish
and Protestant attitudes were not reliably different.
The Jewish respondents did not identify their own
attitudes,
then, with the attitudes of any typical
religious group, not even with the attitudes of typical
Jews.
They estimated Catholics to be more conservative
than either Jews or Protestants, but estimated the
attitudes of Jews and Protestants to be the same.
As shown in Column II of Table 22,
the
Protestant respondents were significantly more in favor
of communism than they thought Catholics to be, and
they believed Jews to be significantly more in favor
of communism than they believed Catholics to be.
Their
own attitude toward communism did not differ significantly
from the attitude which they attributed to Jews.
On
U3
Table
^
Protestants*
Inter-Condition Comparisons
among Mean Attitude Scores
Attitude Measure
Condition
I
Communism
II
War
III
Birth Control
IV
C-R
V
Diff.
t
4.547 (30)
(30)
4.547
3.718 (8)
(8)
3.718
0.829
0.829
2.763*
2.763*
3.797
3.797 (3l)
(31)
3.363
3.363 (8)
(8)
0.434
0.434
0.669
0.669
6.650
6.650 (14)
(14)
3.646
3.646 (l4)
(14)
3.004
3.004
5.678**
5.678**
10.664
10.664 (55)
(55)
13.821
13.821 (53)
(53)
3.157
3.157
6.731**
6.731**
Diff.
t
4.547
4.547 (30}
(30)
(21)
5.140 (21)
5.140
0.593
1.273
1.273
3.797
3.797 (31)
(31)
3.926 (21)
(21)
3.926
0.129
0.500
0.500
6.650 (14)
6.650
(14)
5.732 (14)
(14)
5.732
0.918
1.712
1.712
10.664 (55)
10.664
10.538
10.538 (52)
0 .126
0.230
-.....
.......................
6.650 (14)
6.379 (l4)
0.271
0.667
10.664 (55)
12.127 (5l)
1.463
4.447**
Diff.
t
3.718 (8)
5.140 (21)
1.422
2.722*
3.363 (8)
3.926 (2l)
0.563
0.778
3.646 (14)
5.732 (14)
2.086
3.506**
13.821 (53)
10.538 (52)
3.283
7.582**
Diff.
t
.......................
.......................
.......................
....... ...
..........
3.646 (14)
6.379 (14)
2.733
5.157**
13.821 (53j
12.127 (51 >
1.694
3.683**
5.732 (14)
6.379 (14)
0.647
1.123
10.538 (52;
12.127 (51;
1.589
2.873**
Self
Cath
Self
Jew
Self
Prot
Diff.
t
Cath
Jew
Cath
Prot
Jew
Prot
Diff.
t
.......................
* Significant at the .05 level.
** Significant at the .01 level.
N ’s are placed within parenthesis.
attitude toward war (Column III), the Protestant group
of respondents did not differ significantly from their
estimates of Catholic or Jewish attitudes, and their
estimates of Jewish and Protestant attitudes were not
reliably different.
In their attitude toward birth control,
the
Protestant respondents were significantly more favor­
able than they considered a typical Catholic to be
(Column IV).
However,
their attitudes toward this
issue did not differ significantly from their estimates
of Jewish or Protestant attitudes.
The Protestants,
by their responses on the Abbreviated C-R Opinionaire
(Column V), indicated that they were significantly less
conservative than they considered most Catholics or
most Protestants to be, but not significantly more or
less conservative than they considered most Jews to be.
Protestant respondents thought a typical
Catholic to be significantly less in favor of birth
control (Column IV) than they thought either a typical
Jew or a typical Protestant to be, but they did not
think typical Jews differed reliably from typical
Protestants on this issue.
The Protestants thought
most Jews and most Protestants to be reliably less con­
servative than most Catholics on the Abbreviated C-R
Opinionaire (Column V).
They also thought most Jews
IIS
to be significantly less conservative than most
Protestants,
The Protestants identified their own
attitudes on all issues with their estimates of typical
Jewish attitudes.
They identified their own attitudes
toward war with the attitudes of Jews and Catholics,
and identified in their estimates the war attitudes of
Catholics and Jews,
They identified their own attitudes
toward birth control with their estimates of the birth
control attitudes of Jeww and Protestants, but not with
their estimates of Catholic attitudes on this issue.
In general, the comparisons in Tables 20,
21, and 22 have shown that respondents in all groups
were more likely to identify their own attitudes with
their estimates of some other group's attitudes, rather
than with their estimates of the attitudes of typical
individuals with their own religious affiliations.
Catholic respondents identified their own attitudes with
their estimates of Protestant attitudes, while Protestants
identified their own attitudes with their estimates
of typical Jewish attitudes.
The Jewish respondents
thought all "typical" groups less liberal than themselves.
Typical Jews and Protestants were, in general, thought
by all groups to have similar attitudes, except that both
Catholic and Protestant respondents thought Jews to be
more liberal than Protestants on the issues in the
Abbreviated C-R Opinionaire,
//*
Inter-Condition Comparisons of Variability measurest
Variability Comparisons among Attitudes and
Estimates of Attitudes of Typical
Catholics. Jews. and Protestants
The comparisons among attitudes and estimates
of attitudes wit h respect to their variability, for
given respondent groups, should provide further evidence
for or against the existence of stereotypes regarding
the attitudes of Catholics, Jews, and Protestants.
If the respondents of a given group are more homogeneous
with respect to estimates of the attitudes of typical
Catholics, Jews, or Protestants than with respect to the
expression of their own attitudes,
this evidence would
be interpretable as favoring the existence of stereotypes.
If respondents are significantly more homogeneous in
their estimates of the attitudes of one typical group
than in their estimates of another typical group, this
homogeneity would be interpretable as evidence favoring
the existence of a stronger stereotype regarding the
attitudes of the former than regarding the attitudes
of the latter.
In Table 23 are comparisons of variability
in attitudes and estimates for the Catholic respondents.
The Catholic respondents to the Communism-War question­
naire were significantly less homogeneous with respect
to their opinions about how Jews would respond than
117
Table
^
3
Catholicss Inter-Condition Comparisons of Variability in
Attitude Scores (in Terms of Standard Deviation Values).
Attitude Measure
Condition
I
Communism
II
Self
Cath
...... .................
-.........
-------------------
Diff.
War
III
Birth Control
IV
8
t
Self
Jew
Diff.
P
t
Self
Prot
Diff.
8
t
Cath
Jew
1.11
....................................
0.951 (36)
2.133 (32)
1.182
0.670
1.949
1.2.79
(36)
(3 2 )
4.130**
4.957**
.......................
....... .................. .....
-................................
Diff.
1.945 (12)
1.845 (l2)
----------
----------
----------
J
-
t
................. ......
C-a-l
V
3.934 (63)
2.567 (60)
1.367
--------3.873**
1.945 (12)
1.555 (l2)
....... ...
1.50
... .......
3.934 (63)
3.504 (57)
0.430
..........
0.951
1.945 (12)
1.215 (l2)
3.934 (63)
2.901 (57)
2.54
1.033
....... .
2.662**
1.845 (12)
1.555 (12)
----------
2.567 (60)
3.504 (57)
0.937
1*40
- ............... ........
..........
2.325*
-
1.845 (12)
1.215 (12 )
-.....
2.29
- ........
2.567 (60/
2.901 (57/
0.334
1.140
Jew..........
- ............. Prot
Diff.
---------P
t
1.555 (12)
1.215 (12)
3.504 (57)
2.901 (57)
0.603
---------1.453
Cath
Prot
Diff.
S
t
........
-
1.63
* Significant at the .05 level.
** Significant at the .01 level,
ITrs are placed within parentheses.
nt
they were in their own attitudes on these issues (Columns
II and III).
On the issue of birth control, there were
no significant differences in the variability of scores
under different conditions (Column IV).
The Catholic
respondents to the Abbreviated C-R Opinionaire (Column
V), however, were significantly less variable in opinions
about the attitude position of Catnolics and Protestants
than in their own attitudes expressed under the self­
condition.
There was no significant difference in
variability between their own attitudes and their
estimates of Jewish attitudes, though the variability
in the latter was somewhat smaller.
The variability
in their estimates of Catholic attitudes was significantly
smaller than that in their estimates of Jewish attitudes,
indicating that,
if homogeneity of opinion be taken as
evidence of a cultural stereotype, the Catholic-stereo­
type was the more definite among Catholic resoondents.
The evidence from the variability comparisons between
the attitudes of Catholics and their estimates of
attitudes, may be interpreted to mean that Catholics
were influenced by a Catholic-stereotype and a
Protestant-stereotype regarding attitudes toward issues
in the Abbreviated C-R Opinionaire, since the homogeneity
of opinions regarding the conservatism of Catholics and
Protestants was significantly greater than the homo-
n?
geneity of expressed attitudes by Catholic respondents.
In Table 24 are presented similar inter-condition comparisons with respect to the variability in
scores of Jewish respondents under the different in­
structional conditions.
Their own scores on attitude
toward communism (Column II) did not differ significantly
in variability from the scores which they attributed
to Catholics.
However, the variability in their opinions
on how Catholics would respond on the issue of war
(Column III) was significantly less than the variability
in their self-attitudes on this issue.
On the issue
of birth control (Column IV), the Jewish respondents
were significantly more variable in their opinions about
what the tyoical Catholic, Jew, or Protestant would
think than they were in their own attitudes on this
issue.
On the other hand, the Jewish respondents were
significantly less variable in opinions on how most
Catholics, Jews, or Protestants would respond to the
Abbreviated C-R Opinionaire (Column V) than they were
with respect to their own attitudes.
There were no significant differences in the
variability of their estimates between the attitudes of
Catholics and Jews, Catholics and Protestants, or Jews
and Protestants, on the issues of birth control or on
the issues in the Abbreviated C-R Opinionaire (Columns
!%0
Jews*
Inter-Condition Comparisons of Variability in
Attitude Scores (in Terms of Standard Deviation Values)
Attitude Measure
Condition
I
Communism
II
Self
Cath
1.071 (64)
1.199 (58)
0.128
0.877
Birth Control
IV
C-R
V
0.607 (64) 0.390
(15)
0.300 (58) 1.604
(15)
0.307
— .
0.852
---------16.9**
5.117**
---------3.264**
- ....... —
- .........
-.......
0.390 (15)
1.962 (15)
3.593 v15l)
3.115 U 3 4 ,
0.478
---------2.265*
-.......... ------------------Diff.
..........
F------- ------------------t
—
0.390 (15)
1.862 (15)
Diff.
F
t
Self
Jew
—
Diff
F
t
Self
Prot
War
III
Cath
Jew
Diff.
F
t
Cath
Prot
Diff.
F
t
Jew
Prot
Diff.
F
t
.....
-
-------------- ----— - ---------------------------- -----
-...............
......... -
-------------------
25.3**
-.........
22.9**
3.593 U5l)
2.741 vl31,
3.593 (l5l)
3.000 (l32)
0.593
-----------2.045*
1.604 (15)
1.962 (15)
1.49
2.741 (131)
3.115 (134)
0.374
--1.612
1.604 (15)
1.862 (15)
2.741 (131)
3.000 (132)
0.259
...........
1.112
1.35
...... ....
1.962 (15)
---------1*862 (15)
-..............
---------1.11
*Significant at the .05 level.
**Significant at the .01 level.
IPs are placed within parentheses.
3.115 (134)
3.000 (l32)
0.115
...........
0.481
IV and V).
The variability in the attitudes of Jewish
respondents on the Abbreviated C-R Opinionaire was
greater than the variability in their estimates of
Catholic, Jewish, and Protestant conservatism.
This
finding would seem to be evidence for the existence
of a stereotype among Jewish respondents regarding the
conservatism of typical members of all the groups con­
sidered.
There did not appear to be a stereotype re­
garding attitudes of either Catholics, Jews, or
Protestants, toward the birth control issue# but Jew­
ish respondents were more homogeneous in estimates of
Catholic war attitudes than in their own attitudes
toward war.
In Table 25 are inter-condition comparisons
of variability in scores for Protestant respondents.
On attitude toward communism (Column II), the Protestant
respondents were significantly less variable in their
attitudes than in their judgment of Jewish attitudes,
and on attitude toward war (Column III) they were
significantly less variable in their own attitudes than
in their estimates of either Catholic or Jewish attitudes.
They were also reliably less variable in their own
attitudes toward birth control (Column IV) than in
their estimates of either Catholic or Jewish attituaes.
On the Abbreviated C-R Opinionaire (Column V), there
m
Table
Protestantss
Inter-Condition Comparisons of Variability in
Attitude Scores (in Terms of Standard Deviation Values)
Attitude Measures
Condition
I
Communism
II
Self
Cath
0.990 (31)
0.875 (8)
0.115
0.558 (3l)
1.775 (8)
1.217
0.483
2.753*
Diff.
F
t
Self
Jew
Diff.
F
t
Self
Prot
Diff.
F
t
Cath
Jew
Diff.
F
t
Cath
Prot
Diff.
F
t
Jew
Prot
0.990
1.808
0.818
War
III
(31) 0.558 (31)
(21) 1.199 (2l)
0.641
2.656*
3.287**
Birth Control
IV
0.990 (l4)
2.190 (14)
...........
4.89**
.......
1.775 (8)
1.119 (21)
0.576
2.584*
1.164
3.154 (55)
2.815 (53)
0.339
.....
0.885
0.990
(l4) 3.154 (55)
1.750
(14) 3.270 (52)
..........0.116
3.13*
.....
.....
0.270
.......................
0.990 (14)
.......................
1.290 (14)
....................................
........ ...............
1.69
0.875 (8)
1.808 (21)
0.933
C-R
V
2.190 (14)
1.750 (14)
...........
1.57
-----
2.190 (14)
1.290 (14 )
....................................
2.89*
3.154 (55;
2.801 (ttly
0.353
......
1.176
2.815 (53y
3.270 (52)
0.455
.....
1.137
2.815 (53)
2.801 (51)
0.014
......
0.037
.......................
1.750 (14) 3.270 (52)
........................ 1.290 (l4) 2.801 (5l)
Diff.
........
0.469
F
.......................
1.84
......
t.......
-......................
1.116
* Significant at the .05 level.
** Significant at the .01 level.
Ur*s are placed within parentheses.
123
were no significant differences in the variability of
their scores between any of the possible pairs of con­
ditions.
The Protestant respondents were significantly
less variable in their opinions concerning what Catholics
think on the issue of communism (Column II) than they
were concerning what Jews think on this issue, and re­
liably less variable in their opinions regarding attitudes
of Protestants toward birth control (Column IV) than
in their opinions about attitudes of Catholics on this
issue.
The evidence did not indicate the existence of
a stereotype among Protestants.
Summarizing, the evidence in Tables 23, 24,
and 25, consisting of variability comparisons among
attitudes and estimates of attitudes, indicated the
probable existence among Catholics of a stereotype con­
cerning Catholic and Protestant conservatism! the
stereotype concerning Catholics was evidently neither
more nor less definite than the stereotype concerning
Protestants, since the variability difference in the
resnective estimates was not reliable.
Jewish re­
spondents appeared to have a stereotype concerning the
conservatism of all groups, which was apparently equally
definite with resnect to all groups, since there were no
significant differences in the variability for the
respective estimates.
There was no evidence indicating
the existence of a stereotype among Protestants.
These
statements are not meant to imply that no stereotypes
could have existed in cases where the evidence did not
clearly point to their existence.
While the presence
of significantly less variability in extimates than in
the expressions of attitudes may be interpreted, with
some confidence,
to be evidence of a stereotype, the
absence of such a difference could not be interpreted
as proving the non-existence of a stereotype.
ns
Correlations between Inter-Condition Seta of Scores*
Relations between Attitudes of Respondents and
Attitudes Attributed to Typical Catholics.
Jews, and Protestants! Opinion-Distances
as Related to Conservatism
The present section deals with the correla­
tions between scores under different conditions and
with the relation of self-scores on the Abbreviated
C-R Opinionaire to opinion-distances.
In Table 26 are presented the inter-condition
correlation coefficients for the three groups of re­
spondents on the Abbreviated C-R Opinionaire.
The
correlations between self-scores for each individual
group of respondents and the scores based on responses
under conditions requiring that respondents answer as
members of their own religion would answer were
relatively high (+.575, +.658, and +.685 for the
Catholic, Jewish, and Protestant respondents).
interpretation is that the more conservative
The
(or radical;
a respondent was, the more conservative (or radical)
he thought most members of his own religion would be.
The attitudes of Catholics (Row A) were more
like what they thought the attitudes of Protestants would
be (+.497) than what they thought the attitudes of Jews
would be (+.349).
Likewise,
the Catholics thought the
attitudes of most Catholics (Row B) were more similar
to those of Protestants (+.586) than to those of Jews
(+.094).
Their opinions of what Protestants would think
Table
2.(d
Group Comparisons with Respect to Inter-Condition
Correlation Coefficients** Abbreviated C-R
Opinionaire Scores
(N'b in parentheses)
Condition
Group
I
Condition
II
Catholic --A* Self
Catholic
III
Jewish
IV
Protestant
V
+.575 * .087
(60)
+.349 ± .117
(57)
+.497 ± .101
(57)
+.094 ± .134
(56)
+.586 ± .089
(56)
B.Cath.
C. Jew.
Jewish
D. Self
+.246 ± .129
(54 J
+.204 ± .084
(131)
E. Cath.
+.658 ± .049
(133)
+.381 ± .075
(132)
+.427 ± .073
(127)
+.612 + .055
(129)
P. Jew.
Protestant G. Self
+.622 ± .055
(126)
+.348 ± .121
(53)
H, Cath.
I. Jew.
* Pearson r's
+.236 ± .132
(52)
+.685 ± .075
(51)
+.325 ±. 125
(52)
+.315 ± .127
(51)
+.169 ± .137
(51)
±
ar
m
were more similar to their opinions of what Catholics
would think (+.586) than to their opinions of what Jews
would think (+.246) (Column V, Rows B and C).
The attitudes of Jewish respondents (Row D)
were somewhat more similar to their opinions of what
Protestants would think (+.381) than to their opinions
of what Catholics would think (+.204).
The Jewish
respondents thought the attitudes of Catholics (Row E)
to be more similar to the attitudes of Protestants
(+.612) than to the attitudes of Jews (+.427), and the
attitudes of most Jews (+.622) and of most Catholics
(+.612) to be more similar to the attitudes of most
Protestants (Column V) than were their own attitudes
(+.381).
The attitudes of Protestants (Row G) were
more closely associated with their opinions regarding
Catholic attitudes (+.348) than with their opinions
regarding Jewish attitudes (+.236).
Likewise, the
Protestant respondents thought the attitudes of most
Catholics to be more similar to the attitudes of most
Protestants (+.315) than they thought the attitudes of
most Jews (+.169) to be (Column V).
All groups, then, thought their own attitudes
to be most siniilar to the attitudes of members of their
own religion (Table 26).
Catholics thought their own
attitudes and the attitudes of Catholics to be more
/a?
similar to Protestant than to Jewish attitudes.
Jews
thought their own attitudes and the attitudes of Jews
to he more similar to Protestant than to Catholic
attitudes.
Protestants thought their own attitudes
and the attitudes of Protestants to he more similar
to Catholic than to Jewish attitudes.
In Table 27 are presented correlations
between algebraic opinion-distances and the scores
between which the differences existed.
The question
on which the data in Table 27 were meant to provide
evidence was*
Which is the more important determinant
in the difference between a pair of inter-condition
scores?
Is this difference to be attributed more to the
variance in self-attitudes or to the variance in opinions
on what another group would think?
If the former
answer were given, this would mean that there was a
relationship between conservatism and the extent to which
the individual thought his attitudes differed from those
of members of a given religion.
In Rows A, E, and I are presented, for the
three groups, the correlations between self-scores on
the Abbreviated C-R Opinionaire and the self-minusCatholic, self-minus-Jewish, and self-mirus-Protestant
opinion-distances.
In each group the correlation between
self-score and self-minus-Catholic opinion-distance is
larger than for the correlations involving other opiriiondistances*
Taking this fact in isolation, it could be
concluded that the more conservative an individual was,
the more his self-score exceeded his Catholic-score.
However, each of the correlations between self-score
and self-minus-Catholic opinion-distance would not have
accounted for more than 50# of the variance in the
opinion-distances.
If there were no relation whatever
between the trait of conservatism and the extent to which
an individual would think his degree of conservatism
exceeded that of most Catholics, the correlations should
be sufficiently high to account for 50# of the variance,
since, in such a case, the opinion-distance would be
equally accounted for by each of the scores between
which it was a difference.
Another characteristic of the correlations
presented in Table 27 is that, in all instances,
the
correlation between self-score and a given opiniondistance is larger than that between the given opiniondistance and the other score from which it was derived.
If the total variance accounted for by each pair of
correlations equalled unity, it would be valid to con­
clude that the trait of conservatism (as measured by
the Opinionaire) was positively related to opinion-
130
Table
7
Group Comparisons of Correlations* between Algebraic
Opinion-Distances and the Scores from which the
Differences were Derived* Abbreviated C-R Opinionaire
Group Related
Variable
Scores consisting in Self'-Sc. minus*
Catholic-Score
I
Cath.
A. Self-Sc.
+.757 ± .056
(60)
B. Cath-Sc.
-.097 ± .128
(60)
C. Jew-Sc.
Jewish-Score Prot-Score
II
III
+.652 ± .076
(57)
-.462 ±. 105
(57)
-.253 ± .125
(57)
D. Prot-Sc.
Jew.
E. Self-Sc.
+.739 ± .040
(131)
P. Cath-3c,
-.478 ±. 068
(131)
G. Jew.-Sc.
+.524 ± .063
(134)
+.649 ± .051
(132)
-.394 ± .073
(134)
H. Prot-Sc,
Prot.
+.712 ± .066
(57)
-.451 ± .070
(132)
I. Self-Sc.
+.632 ± .083
(53)
J. Cath-Sc.
-.497 ± .104
(53)
K, Jew. Sc.
+.569 ± .095
(52)
+.495 ±. 106
(51)
-.378 ± .120
(52)
L. Prot-Sc.
♦Pearson r's ± o r.
N's are placed within parentheses.
-.406 ± .118
(51)
131
distance.
However, the total variance did not equal
unity for any pair of correlations, owing, among other
factors, to unreliability of the instrument under dif­
ferent conditions.
Such attenuation could be corrected
for if any method were known by which to determine the
reliability of the opinion-distances.
The interpreta­
tion of the correlations in Table 27, then, cannot
validly be made in terms of the existence or non­
existence of a relationship between the trait of con­
servatism and the extent to which individuals think
themselves different in attitudes from a given religious
group.
Such difficulties of interpretation, however,
are partially removed when the correlations in Table
28 are considered.
In this table are presented cor­
relations between self-scores of the different groups
and opinion-distance scores not directly dependent on
the self-scores for their derivation.
That is, the
opinion-distances consist in differences between
Catholic-scores and Jewish-scores, between Catholicscores and Protestant-scores, and between Jewishscores and Protestant-scores.
The self-scores of
Catholics (Row A) were not significantly correlated
with any opinion-distance scores not dependent on self­
scores for their derivation.
However, the self-scores
Table
%
Group Comparisons of Correlations* between Self-Scores on
Abbreviated C- R Opinionaire and Algebraic Opinion-Distances
(not involving Self-Scores)
Opinion-Distance
Cath. minus
Jew.
II
Cath. minus
Prot.
Ill
Jew. minus
Prot.
IV
A. Catholic
+.079 ± .133
(57)
+.039 ± .135
(55)
-.060 ± .135
(55)
B, Jewish
-.456 ± .071
(127)
-.231 ± .084
(129)
+.344 ± .079
(126)
C. Protestant
+.023 ± .140
(52)
-.305 ± .128
(51)
-.159 ± .137
(51)
Group
I
* Pearson r ’s ± or ,
N fs are placed within parentheses.
IS3
of Jewish respondents
(Row B) were negatively correlated
with the Catholic-minus-Jewish opinion-distances, and
positively correlated with the Jewish-minus-Protestant
opinion-distances.
That is, the wore conservative the
Jewish respondents were, the less likely they were to
think most Catholics to he more conservative than most
Jews.
Also,
the more conservative the Jewish respondents
were, the more likely they were to think most Jews more
conservative than most Protestants.
The self-scores of the Protestant group (Row
C) were correlated significantly above zero only with
the Catholic-minus-Protestant opinion-distance.
Since
this was a negative correlation (-.305), it could be
interpreted to mean that the more conservative Protest­
ants were, the less likely they were to consider most
Catholics to be more conservative than most Protestants.
These results raised the question as to
whether there would be similar relationships between
self-scores and absolute ooirion-distances— that is,
absolute differences between pairs of scores based on
responses under different conditions, without regard
to which of the scores was greater.
In Table 29 are
presented the correlations between self-scores and
absolute opinion-distances,
so defined.
Correlations
in Rows A, D, and G should be disregarded,
since they
Table
29
Group Comparisons of Correlations* between Self-Scores
on Abbreviated C-R Opinionaire and
Absolute Opinion-Bistances**
Difference
between
Group
I
II
Cath. A. Self-Sc.
AMD
Cath-Score
III
Jewish-Score
IV
Prot-Score
V
-.689 ± .068
(60)
+.278 * .123
(57)
-.368 ± .116
(57)
-.046 ± .133
(57)
-.103 ± .134
(55)
B. Cath-Sc.
-.014 ± .136
(55)
C. Jew.-Sc.
Jewish
D. Self-Sc.
-.143 ± .086
(131)
E. Cath-Sc.
-.449 ± .069
(134)
-.480 ± .067
(132)
-.248 ± .084
(127)
-.184 ± .084
(129)
P. Jew.-Sc.
Protestant
G. Self-Sc.
-.099 ± .089
(126)
-.611 ± .087
(53)
H. CathrSc.
+.203 ± .134
(52)
-.397 ± .119
(51)
+.008 ± .140
(52)
-.247 ±.132
(51)
I. Jew.-Sc.
+.093 ± .140
(51)
* Pearson r ’s ± ar .
** Absolute opinion-distances are between the scores in
the rows and columns which locate the correlation
entries.
IPs are placed within parentheses.
were correlations between self-scores and opiniondistance scores consisting of the differences between
self-scores a n d other scores.
These would be of un­
certain interpretation, as pointed out above; a further
consideration to clarify this point arises from the
fact that, even if all Catholic-scores, for instance,
were constant, the correlation between self-scores an d
the difference between self-scores a n d Catholic-scores
would be unity, since it would be the correlation between
the magnitude of self-scores and this magnitude
diminished by a constant quantity.
The remaining
correlations in Table 29 were not significantly above
zero in magnitude,
so that there appears to have been
no significant relationship between self-scores and
absolute opinion-distance scores not directly derived
from the self-scores.
Summary of Inter-Condition Findings
The present chapter has dealt with inter­
condition comparisons of mean scores and of standard
deviations, with inter-condition correlations between
scores, and, finally, with the correlations between
self-scores and opinion-distance scores.
Catholic respondents were shown to be less in
favor of commuiiism, less in favor of war, and more con­
servative than they thought Jews to be.
They were more
in favor of birth control and less conservative than
they thought most Catholics to be.
They thought most
Catholics to be less in favor of birth control and more
conservative than they thought either Jews or Protestants
to be, and thought most Protestants were more conservative
than most Jews.
The Jewish respondents were more in favor of
communism, less in favor of war, more in favor of birth
control, and less conservative than they thought Catholics
to be.
They were likewise more in favor of birth control
and less conservative than they thought most Jews or
Protestants to be.
They thought most Jews and Protestants
to be more in favor of birth control and less conservative
than most Catholics.
The Protestants were more in favor of com­
munism and birth control and less conservative than
137
they thought Catholics to he.
They were likewise less
conservative than they thought most Protestants to he.
Protestants thought Jews to he more in favor of com­
munism and birth control, and less conservative,
they thought Catholics to he.
than
In their group opinion,
the Protestants thought Protestants to be more in favor
of birth control and less conservative than Catholics.
They alBo thought Jews to he less conservative than
Protestants.
In general, respondents in all groups were
more likely to identify their own attitudes with their
estimates of the attitudes of some other group than
with their estimates of the attitudes of typical
individuals with their own religious affiliation.
Catholic respondents identified their own attitudes with
their estimates of Protestant attitudes, while Protestant
respondents identified their own attitudes with their
estimates of typical Jewish attitudes.
The Jewish re­
spondents thought typical members of all groups less
liberal than themselves.
Catholics were less variable in attitudes
toward communism and war than in their estimates of
Jewish attitudes toward these issues, but more variable
in conservatism than in their estimates of the conservat­
ism of most Catholics and most Protestants.
They were
m
also less variable in their estimates of conservatism
of Catholics than in their estimates of the conservatism
of Jews.
Jewish respondents were more variable in
attitudes toward war and in conservatism than in their
opinions on what Catholics would think regarding these
issues, but less variable in attitudes toward birth
control than in their opinions about Catholic attitudes
toward this issue.
They were less variable in attitudes
toward birth control, but more variable in responses
toward the issues in the Abbreviated C-R Opinionaire,
than in their oninions about the attitudes of most Jews
and most Protestants.
Protestant respondents were less variable in
attitudes toward war and toward birth control than in
their opinions of Catholic attitudes on these issues.
They were likewise less variable in attitudes toward
communism, war, and birth control than in their ideas
of Jewish responses toward these issues.
In opinions
regarding Catholic attitudes toward communism,
they were
less variable than in their coinions on Jewish attitudes.
They were also less variable in opinions about responses
of Protestants toward the birth control issue than in
opinions about Catholic responses toward this issue.
In general,
if significantly less variability
/3?
in estimates of attitudes than in the expression of the
respondents' own attitudes be interpreted as evidence
of the existence of a stereotype, the findings in this
respect indicated the existence among Catholics of a
stereotype concerning Catholic and Protestant con­
servatism, and indicated the existence among the Jews of
a stereotype concerning the conservatism of all groups.
Inter-condition correlations of scores reveal ed
that respondents in each group thought members of their
own religion would give responses more similar to their
own responses on the Abbreviated C-R Opinionaire than
would members of other religions.
Catholics thought
their own attitudes and attitudes of Catholics to be
more similar to those of Protestants than to those of
Jews.
Protestants thought their own attitudes and
attitudes of most Protestants to be more similar to
those of Catholics than to those of Jews,
Jewish
respondents thought their own attitudes and the atti­
tudes of most Jews to be more closely related to those
of Protestants than to those of Catholics.
Correlations of self-scores w i t h ooiniondistance scores seemed to indicate that the more con­
servative the Jewish respondents were, the less likely
they were to think most Catholics to be xaore conservative
than most Jews, and. the more likely they were to think
most Jews more conservative
than most Protestants.
Also, the interpretation seemed justified that the more
conservative the Protestants were, the less likely they
were to consider most Catholics to be more conservative
than most Protestants.
/¥/
Chapter VIII
Reliability Studies and Evaluation of Items
in the Abbreviated C-R Opinionaire
The present chapter deals with data relating
to the reliability of the Abbreviated C-R Opinionaire
under the several conditions of instruction, with the
equivalence of sets of partial scores based on two
methods of halving the Abbreviated C-R Opinionaire, with
biserial correlations between total scores a n d responses
to particular items, and with inter-item tetrachoric
correlations*
In Table 30 are presented reliability
estimates for the Abbreviated C-R Opinionaire,
puted by three methods*
com­
(a) Spearman-Brown corrected
correlations between the scores on the first end second
halves of the Opinionaire,
(b) Spearman-Brown corrected
correlations between scores based on radical statements
and on conservative statements, and (c) reliability
computed by means of "Formula 20" published by nichardson and Kuder
(an.
n
rtt “ -----n - 1
This latter formula is as follows*
of - Lpq
•--- 1— ~ - = ---of
in which n = the number of items,
of = variance of scores on total test
(i.e., standard deviation squared),
p = proportion of correct responses to
each item*
q = 1 - p,
= reliability of total test*
Table
30
Reliability Estimates for the Abbreviated C-R Opinionaire
under Different Conditions, Computed by
Three Methods: (l) 1st vs. 2nd Half,
Corrected for Whole Test by S-B Formula!
(2) Conservative vs. Radical Items,
Corrected by S-B Formula! and
(3 ) Richardson and Kuder’s "Formula 20"
Condition
Croup
Jewish
V
Protestant
VI
1st vs. 2nd •6254.094 .3884157
C vs R
•7434,065 .6294.096
.2114.209
.3844.164
.4384.149
.4364.149
Formula 20 .758
.484
.653
. 523
60
57
57
Method
II
I
Catholic
1
IV
Self
III
Cath.
N
63
Jewish
1st vs 2nd .6994.058 .4644.080 .5704.073
C vs R
. 730±055 .3074,079
*4524.079
Formula 20 .726
H
151
.5074.077
.2964.085
.492
.595
.540
131
134
132
Protestant
1st ]rs. 2nd .6744.081 .6014,109
C vs. R
•5624.118 .4254.158
•821&.050
.4724.147
.4634.151
.2834.201
.540
•596
.486
53
52
51
Total Sample
Formula 20 .738
.509
.621
.528
269
244
243
240
Formula 20 .596
55
a. O
T T
=
—
2jl
y r -
- rij; )
(1 . r1I)g
iH3
This formula is considered to yield a slight under­
estimate of reliability as compared with reliability
computed by the usual split-test methods.
However,
since this formula was based on a rational procedure
for establishing equivalence of variability in the two
halves of a test, it is here interpreted as the most
nearly correct of the three estimates of reliability in
Table 30.
As there were no "correct" responses to the
Abbreviated C-R Opinionaire, p was taken as the pro­
portion of responses of agreement with given items.
In general, as shown in Table 30, the
reliability under the self-condition was much higher
than that under other conditions of instruction.
The
reliability was approximately the same (Column III)
for the Catholic and Jewish groups under the self-condi­
tion, but higher for these groups than for the Protestand group.
Reliability under the Jewish-condition was
uniformly higher than under other conditions, excepting
the self-condition.
The different types of reliability
coefficients were different for a given group under a
given condition (excluding the self-condition).
However,
the magnitudes of standard errors were such that the
reliability coefficients computed by the split-test
methods might be considered equivalent, within the
ranges of their respective chance errors, to those com­
puted by Formula 20.
The fact that reliability was
m
lower under other conditions than uhder the self-condition would he expected, since, on a priori grounds,
it would seem reasonable that each individual would he
more consistent in expressing his own opinions than in
making a "best guess" ahout how members of some given
group would respond.
Whistler and Remmers (30) found
similarly low reliabilities for their Scale to Measure
Individual and Group Morale when their subjects were
asked to compare aspects of "Present Day American" life
with those for "Pioneer Americans", etc.
These authors
suggested that "the degree of organization of attitude
patterns affects the reliability of measurements of
them".
In Table 31 are presented comparisons between
mean scores on the first 10 and on the second 10 items
of the Abbreviated C-R Opinionaire under the different
conditions of instruction.
Each group, under three out
of four conditions, yielded equivalent scores on the
first and second halves of the Opinionaire, as indicated
by the unreliability of differences between means.
The
Catholic and Jewish groups had mean Catholic-scores
significantly greater on the first than on the second
half, while the Protestant group yielded scores on the
first half which were significantly larger than on the
second half under the self-condition.
Table
31
Comparisons between Mean Scores on First and Second
Halves of Abbreviated C-R Opinionaire
Catholic
II
Group
Jewish
III
6.802
6.754
0.048
0.157
IE!
5.142 (151) 5.900
5.288 (151) 5.264
0.636
0.146
2.585*
0.880
Cath* First Half
Second Half
Diff.
t
7.833
7.283
0.550
2.099*
IE!
7.263 (131)
6.752 (l3l)
0.511
2.903**
Jew.* First Half
Second Half
Diff.
t
5.395 (57)
6.114 (57)
0.719
1.959
6.440 (134) 5.557 (52)
6.306 (134) 5.481 (52)
0.076
0.134
0.757
0.390
Prot.* First Half
Second Half
Diff.
t
6.640
7.061
0.421
1.447
6.553 (132) 6.480 (51)
6.295 (132) 6.147 (51)
0.258
0.333
1.099
1.410
Conditions
I
Self* First Half
Second Half
Diff.
t
IE!
* Significant at the .05 level.
** Significant at the .01 level.
N's are placed within parentheses.
Protestant
IV
7.406
6.915
0.491
1.980
IE!
IE!
/
In Table 32 are similar comparisons between
mean scores on the 10 items to which zero-responses
were conservative and mean scores on the 10 items to
which plus-responses were conservative.
The comparisons
indicate that, in general, scores on the two halves so
determined were significantly different, or non­
equivalent.
The Catholic group yielded equivalent scores
under the self-condition/ the Jewish group had equivalent
scores under the Jewish-conditionf and the Protestant
group had equivalent scores under the Protestantcondition.
Among the remaining comparisons, all of
which showed reliable differences, a strikingly con­
sistent tendency appeared*
Under the self-condition
the scores were higher for the items with which dis­
agreement was the conservative response than for items
with which agreement was the conservative response;
under all other conditions, the exact reverse occurred-that is, scores were higher for items to which responses
of agreement constituted conservative responses.
other words,
In
total self-scores were more determined by
disagreement with radical statements than by agreement
with conservative statements, while scores under other
conditions were more determined by agreement with
conservative statements than by disagreement with radical
statements.
Thus, it would appear that respondents
*0?
lH-7
Table
32s
Comparisons between Mean Scores on (l) Radical Items and
(2) Conservative Items in Abbreviated C-R Opinionaire
Group
Catholic
II
Jewish
III
Protestant
IV
Self* Radical Items
Conservative Items
Diff.
t
6.817 (63)
6.738 (63)
0.079
0.312
5.732 (151)
4.669 (151)
1.033
6.006**
5.882 (55)
5.245 (55)
0.637
2.235*
Cath* Radical Items
Conserv. Items
Diff.
t
7.117 (60)
8.000 (60)
0.883
4.107**
6.584 (131)
7.431 (I3l)
0.847
4.235**
6.557 (53)
7.764 (53)
1.207
4.064**
Jew. Radical Items
Conserv. Items
Diff.
t
5.272 (57)
6.237 (57)
0.965
2.681**
6.284 (134) 4.923 (52)
6.463 (134) 6.096 (52)
0.179
1.173
0.895
3.460**
Prot* Radiaal Items
Conserv. Items
Diff.
t
6.184 (57)
7.518 (57)
1.334
4. 584*“*
5.947 (132)
6.894 (132)
0.947
3.865**
-Conditions
I
* Significant at the .05 level.
** Significant at the .01 level.
IT’s are placed within parentheses.
6.108 (51)
6.520 (51)
0.412
1.230
were somewhat more likely to disagree than to agree with
statements, while they believed tyoical members of given
religious groups more likely to agree than to disagree.
In Table 33 are presented the biserial cor­
relations between total self-scores on the Abbreviated
C-R Opinionaire and the responses to particular items
under the self-condition.
The correlations for radical
items (Column V) were generally higher than those for
conservative items (Column II ).
This finding was quite
consistent with that in Table 32, which indicated that
self-scores were more determined by disagreement with
radical items than by agreement with conservative items.
The fact, however,
that even the lowest correlation for
conservative items (Column II) was significantly positive
(more than 3 cr-bis r) justifies the interpretation that
the least discriminating conservative item was still
more likely to be agreed with by those respondents having
the higher conservatism scores than by those having the
lower scores.
The three least discriminating conservative
items were*
10. Three meals a day will always be the best
general rule.
17. Democracy as practiced in the United States is
the best of all modern governments because it
is the most suited to the needs of modern times.
8. One is never justified in taking another’s life,
even when it would be a merciful act.
Ift
Table
33
Contribution of Abbreviated C-R Opinionaire Items to Total
Score!
Items arranged in Order of Magnitudes of
Biserial r rs with Total Score, and Compared
with U-L Values of Lentz
(N = 269)
Radical Items
Conservative Items
Item
No.
I
II
U-L
Value
III
22
+ .573 ± .060
56
3
16
+ .564 ± .056
59
21
+ .561 db .058
6
11
rbis
0bis
Item
No.
IV
rfris °bis
V
U-L
Value
VI
-.684 ± .050
46
48
1
-.681 ± .067
47
+ .527 ± .060
55
15
-.662 ± .050
51
+ .421 ± •066
58
18
-.649
±
.053
48
+ .389 ± .070
49
5
-.631
±
.053
43
.338 ± .072
47
4
-.576 ± .057
41
8
+ .329 ± .071
48
9
-.576 ± .086
51
17
+ .313 ± .081
50
13
-.531 ± .059
48
1 0
+ .278 ± .072
60
2 0
-.478
44
2
19
+
CM
12
1
47
c•
db .049
db
.062
t50
The three most discriminating conservative
items were*
22. It is more important to believe in God than
to be unselfish.
16. Any science which conflicts with religious
beliefs should be taught cautiously, if at
all, in our schools.
21. People who are religious will be happier in
the future life than will others.
The three least discriminating radical items
were *
20. It is to be hoped that men will improve the
comfort of their dress by abandoning or re­
placing the present necktie and collar.
13. Criminals should be treated like sick persons.
9. Socially-minded experts, rather than voters,
should decide the policies of government.
The three most discriminating radical items
were *
3. Since the theory of evolution has been accepted
by most scientists, it should be taught in our
schools.
12. The metric system of weights and measures should
be adopted instead of our present system.
1. We should celebrate Pasteur’s birthday rather
than Washington’s, as he has done the world a
greater service.
In Columns III and VI of Table 33 are presented
the U-L values of items, as reported by Lentz
The
rank orders of the magnitudes of biserial correlations
were quite dissimilar to the rank orders of U-L values.
/57
However, when the mean of U-L values for the 5 items
having highest biserial correlations was
compared with
the mean for items having the lowest biserial correla­
tions, the former was higher, so that, despite the
small range of the U-L values, there was nonetheless
some relation between the two measures of item validity.
Tables of inter-item tetrachoric correlations,
one for each group of respondents, are presented in
the appendix.
Condensed tables, based on these inter­
item correlations, will be discussed here.
If certain
items are validly to be described as conservative while
other items are described as radical, the resoonses to
such items should be positively and negatively correlated,
respectively, with total scores, as was found to be the
case (Table 33).
Likewise, responses to any two items
termed conservative, or to any two items termed radical,
should be oositively correlated, 'while responses to any
two items, one of which is termed radical and the other
of which is termed conservative, should be negatively
correlated.
Table 34 shows the distribution of inter-
item tetrachoric correlations whose signs were the
reverse of what would be expected in terms of the above
proposition.
S’or the Catholic group (Column II) only
26 sign reversals occurred (that is, 13.7^ of all cor­
relations for the group).
Of these, 15 were less than
i5X
Table
2j£
Inter-Item Tetrachoric Correlation Coefficients!
Frequency of Sign Reversals (Conservative vs.
Radical Items Positively Correlated^
C vs« C or R vs. R Negatively Correlated)
Magnitude of
Correlations
I
.So
Groups
Catholic
II
Jewish
III
Protestant
IV
- .59
1
.40 - .49
1
2
5
- .29
4
12
.10 - .19
5
6
14
- .09
15
10
22
Total Sign
Reversals
26
16
55
•
CO
o
.30 - .39
o
•
o
Range of N's
60-63
Median rtet#
.087
144-151
.080
48-55
.139
IS3
Table
3S"
Inter-Item Tetrachoric Correlation Coefficientsi
Frequency of Correlations of Various
Magnitudes (Without Sign Reversals)
Magnitude of
Correlations
Groups
Catholic
Jewish
rrotestant
.70 - .79
5
.60 - .69
4
4
2
.50 - .59
12
4
5
.40 - .49
23
22
16
.30 - .39
29
36
21
.20 - .29
38
39
24
.10 - .19
20
35
28
.00 - .09
32
30
31
163
170
128
Range of H's
60-63
144-151
48-55
Median r^et.
.278
.251
.220
Total without
sign reversals
1
ISt
.10, 5 were between .10 and *19, and 6 were above .20
but less than .40.
For the Jewish group, 16 correlations
occurred with signs the reverse of that expected.
Of
these, 10 were below .10, and 6 were between .10 and.19.
For the Protestant group, 55 (or 20% of all correlations)
occurred with signs the reverse of that expected.
these, all but 7 were below .30.
Of
13/hen the distributions
of correlations with sign reversals (Table 34) are com­
pared with the distributions of correlations without
sign reversals
(Table 35), the probability becomes
apparent that the majority of correlations with sign
reversals were not sigrificantly above zero, and there­
fore had wrong signs owing to chance errors.
Disre­
garding this interpretation, however, the fact that a
very large proportion of the correlations was positive
or negative in accordance with expectations argues for
the validity of the terms "conservative" and "radical"
used in describing the items.
With regard to the magnitudes of the inter­
item correlations, Tables 36, 37, and 38 indicate that
pairs of items highly inter-correlated for a given group
of respondents were not so highly inter-correlated for
other groups.
Table 36 presents the 9 pairs of items
most highly inter-correlated for the Catholic group
(Columns I and II).
Columns III and IV show that between
these same pairs of items the correlations were not
{5S
Table
36?
C-R Opinionaire*
Group Comparisons of Inter-Item
Tetrachoric Correlation Coefficients, involving
Item Pairs with Correlations of .60
or Above for Catholic Group
Items
Inter-Correlated
I
3
5
5
6
9
12
12
12
15
&
&
&
&
&
&
&
&
&
16
12
21
22
22
15
17
18
17
Group
Catholic
(K’s=60-63)
II
Jewish
(IP 8=144-151)
III
-.63
+ .68
-.70
+.70
+.60
+.78
-.72
+.60
-.72
-.33
+.32
-.21
+.28
---- *
+.27
-.30
+.31
-.38
Protestant
(R»s=48-55)
IV
-.30
+.32
-.30
+.28
+.15
+.25
-.08
+.40
-.40
* Correlation not computed because less than .05 in
row or column total.
Tabl£ 3 1
C-R Opinionaire*
Group Comparisons of Inter-Item
Tetrachoric Correlation Coefficients, involving
Item Pairs with Correlations of .50
or Above for Jewish Group
Items
Inter'-Correlated
I
1
2
2
9
9
12
13
15
&
&
&
8c
&
&
&
&
12
21
22
17
18
16
18
21
Group
Jawiih(IP8=144-151)
II
+ .60
+.50
+ .50
-.68
+. 66
-.50
+ .60
-.52
Catholic
(BP s=60-63)
III
-.04
$. 33
-.42
+.35
-.48
+.32
-.23
Protestant
(IP8=48-55;
IV
- -
—
+.60
+ .28
-.22
+.35
-.30
+.23
+.17
Table
3$
C-R Opinionaire* Group Comparisons of Inter-Item
Tetrachoric Correlation Coefficients, involving
Item Pairs with Correlations of .50
or Above for Protestant Group
Items
Inter-Correlated.
I
2
3
4
4
5
5
9
15
16
&
&
&
&
&
&
&
&
&
21
22
8
21
8
9
15
18
19
Group
Protestant
(H* s=48-55)
II
+ .60
-.70
-.57
-.57
-.60
-.50
+.52
+.55
+ .53
Catholic
(¥’8=60-63)
III
+.17
-.31
-.06
-.29
-.40
+.20
+.27
+.38
+ .05
Jewish
(¥*8=144-151)
IT
+.50
-.40
-.11
-.36
-.08
+.38
+.36
+.38
+.30
nearly so high for the Jewish and Protestant groups.
Similarly, Table 37 presents
the 8 pairs of items most
highly inter-correlated for the Jewish group.
Again
the correlations between these same pairs of items were
much lower for Catholics and Protestants,
likewise,
Table 38 shows that item pairs most highly correlated
for Protestants were generally less highly correlated
for the Catholic and Jewish groups.
15%
Summary
The present chapter has dealt with data relating
to the reliability of the Abbreviated C-R Opinionaire,
with equivalence of partial scores, with biserial cor­
relations between total self-scores and the responses
to particular items, and with inter-item tetrachoric
correlations.
Reliability of scores was greater for the
self-condition than for other conditions.
Reliabilities
under other conditions than the self-condition were
interpreted to be lower because of the reasonableness
of the hypothesis that each individual would be more
consistent in expressing his own opinions than in
deciding what members of given groups would think.
Comparison of mean scores on halves of the
Opinionaire Bhowed that the method of division based
on the first ten and the second ten items yielded more
cenerally equivalent partial scores than did the method
which made use of scores from conservative items and
from radical items.
Comparisons in the latter ca^e led
to the interpretation that under the self-condition
scores were more determined by disagreement with radical
statements than by agreement with conservative state­
ments, while the reverse was true with resoect to scores
under other conditions.
159
Biserial correlations between total self­
scores and responses to particular items were larger
for radical than for conservative statements*
Items
most and least highly correlated with total scores were
listed.
Distributions of magnitudes of inter-item
tetrachoric correlations with sign reversals, together
with the great proportion of correlations with signs
according to expectation, was interpreted as evidence
that the terms "conservative” and "radical”, as
applied to given items, were correctly so apolied.
Pairs of items highly inter-correlated for
a given group were much less highly correlated for other
groups.
This finding could be interpreted as evidence
that the organization of attitudes was different f or tthe
different respondent groups.
\k>0
Chapter IX
Studies of Factors Related to Croups Thought to be
^Menaces". and to Newspapers which the
Respondents Claimed to Read
The present chapter deals with findings
based on responses to the two unstandardized check­
lists, relating to groups thought to be "a menace to
the best interests of the United States” and relating
to newspapers which were checked as being read as often
as once a week by the respondents.
The analysis of
results from these checklists has been based on responses
of the C-R Opinionaire sample, the data from all re­
spondents having been treated as data from a single
group.
The number of cases varied from 267 to 269.
In Table 39 are presented the items from the
menace Checklist,
the checking of which was positively
correlated with self-scores on the Abbreviated C-R
Opinionaire.
That is, the items in Column I of Table
39 are names of groups which were considered "menaces"
by conservative students more frequently than by radical
or liberal students.
Items whose biserial correlations
with conservatism were more than 3 times their standard
errors were "Socialist Party” and "Aliens (in general)".
"When a less rigid criterion of reliability, consisting
of the requirement that the correlations be only as
/£/
Table
Biserial r ’s*
Scores on Abbreviated C-R Opinionaire versus
Particular Menaces Checked
iltems Possibly Conservative)
(R = 267)
^Menaces*
Biserial r ’s
II
I
1.
2.
5.
9.
13.
14.
17.
18.
19.
26.
Labor Unions ......
Socialist Party ....
C. I. 0. ........... ....
Hew Dealers ........
Brain Trusters
Communist Party ....
Jews ...............
Assoc. Willkie Clubs
Ron-Sect. Anti-Razi League
Aliens (in general) ....
Table
Biserial r ’s*
+.213
0bis r
III
±
±
±
db
±
db
+.082
+.080
+.354
±
±
±
±
.106
.073
.089
.109
.106
.083
.125
.119
.112
.097
^0
Scores on Abbreviated C-R.Opinionaire versus
Particular Menaces Checked
(Items Possibly Liberal)
(R = 267)
^MenacesH
I
3.
4.
7.
8s
11.
12.
21.
22.
23.
25.
3j_
Biserial r ’s
II
Ku Klux Klan ...........
Coughlinites ...........
Labor Spies ............
Dies Committee .........
Tammany Hall ...........
A. }j. L. ...............
Uerman-American Bund ...
Amer. Civil Liberties Union
American Liberty League .
Townsendites ...........
-.343
-.431
-.174
-.428
-.305
-.051
-.392
-.037
-.282
-.126
0Bis~r
III
±
±
±
±
±
±
±
±
±
.085
.068
.075
.107
.074
.119
.098
.114
.095
.108
large as 2 times their standard errors, was employed,
"C.I.O” , “Hew Dealers”, and "Communist Party” were
added to the list of groups which were likely to he
thought "menaces” more frequently by the more con­
servative individuals.
Since all the groups mentioned
above are groups oopularly considered “radical” , it
is not surprising that their being considered “menaces"
was positively correlated with conservatism.
Table 40 lists the items from the Menace
Checklist which were negatively correlated with self­
scores on the Abbreviated C-R Opinionaire.
Items whose
negative correlations with conservatism scores were as
large as 3 times their standard errors were "Ku Klux
Klan",
"Coughlinites” , “Dies Committee”, “Tammany
Hall", and “German-American Bund”.
The correlation for
"American Liberty League" fell just short of being 3
times its standard error.
A complete table of inter-item tetrachoric
correlation coefficients between responses to items of
the Menace Checklist is included in the appendix.
The
inter-correlations between certain sets of the items
are included in Tables 41 and 42.
The question upon
which these tables present some evidence might be
stated as follows*
Yfere items which were significantly
correlated, either negatively or positively, with self-
H>3
j£L
Table
Tetrachoric Correlation Coefficients between "Menaces"
having Positive Biserial r's (equal to 2
r)
with Scores on Abbreviated C-R Ouinionaire
(N = 267)
Menace
2
Menace*
5
9
14
26
+.39
+.40
+ . 55
+ .25
+ .52
+.45
+ .42
+ .20
+ .35
5
9
+ .58
14
* Particular ‘menaces' may be identified by reference to
Table 39.
Table
L\~'X
Tetrachoric Correlation Coefficients between "Menaces"
having Significantly Negative Biserial r ’s with
Scores on Abbreviated C-B Opinionaire
(N = 267)
Menace*
Menace
4
3
4
8
11
+ .55
8
11
21
+.03
+.26
+ .32
+ .13
+ .02
-.02
+.20
-.37
-.02
* Particular 'menaces’ may be identified by reference to
Table 40.
scores on the Abbreviated C-R Opinionaire, themselves
highly intercorrelated?
The inter-item tetrachoric
correlations in Table 41 were between items having
positive biserial correlations as large as twice their
standard errors with conservatism scores.
These items
can be identified by reference to Table 39.
All the
correlations in Table 41 were positive, and, though
not high, fairly uniform in magnitude (+.20 to +.58J.
Compared to the inter-item correlations in Table 42
(between items having significantly negative biserial
correlations with scores on conservatism), these cor­
relations (Table 41) were strikingly uniform.
Some
negative correlations and correlations of approximately
zero appear in Table 42.
In other words, the groups
more likely to be thought "menaces" by conservative
individuals were more uniformly related in the
opinions of the resoondents than were groups more likely
to be thought "menaces" by radical individuals.
Tables 43 and 44 present the biserial cor­
relations between the total number of groups which were
considered to be "menaces" and responses to each
particular group in terms of whether or not it was a
"menace".
Since the checking of each item contributed
the same as the checking of any other item to the
total number of items checked, the use of biserial
/65-
V3
Table
Biserial r's*
Total Number of Menaces Checked versus
Particular Menaces Checked (Significant Correlations)
(N = 269)
"Menace”
I
1.
2.
6a
7.
9.
10 •
11 .
12a
13.
17.
21.
22.
23.
25.
26.
Biserial r
II
Labor Unions ...........
Socialist Party .......
Ca la 0# ........a......
Labor Spies ............
New Dealers ..........
^ a
P a A* a a a a a a a a a a a a a a a
Tammany Hall ............
A . Pa L. a a a a a a a a a a a a a a a
Brain Trusters .........
Jews ...................
German-American Bund ...
Amer. Civil Liberties Union
American Liberty League
Townsendites ...........
Aliens (in general) ....
+.668
+.421
+.679
+.398
+.770
+.457
+.416
+.741
+.391
+.786
+.642
+.568
+.505
+ .733
+ .374
®bis r
III
±
±
±
±
±
±
±
±
±
±
±
±
±
±
±
.081
.069
.063
.067
.076
.092
.069
.118
.098
.229
.080
.094
.084
.076
.096
Table jt!t
Biserial r's* Total Number of Menaces Checked versus
Particular Menaces Checked (NOn-Significant Correlations )
(N = 269)
"Menace"
I
3.
4.
8.
14.
18.
19.
24.
Ku Klux Klan ...........
Coughlinites ...........
Dies Committee .........
Communist Party ........
Assoc. Willkie Clubs ...
Non-Sect. Anti-Nazi League
N a t ’l. Assn. of Manufact.
Biserial r
II
-.110
-.025
+.129
+.109
+.172
+.172
+.201
^bis r
III
±
±
±
±
±
±
±
.090
.079
.117
.084
.118
.110
.135
correlations here would he analogous to the use of bi­
serial correlations between the items and the total
score on any test.
However, within limits, the “total
score1* here could be considered a measure of the
relative lack of discrimination with which the respondents
responded to the items of the Menace Checklist.
Thus,
if total number of items checked as “menaces" were
accepted as such a measure (of discrimination in check­
ing), then those items highly correlated with total
number of items checked could be interpreted to be the
items more likely to be checked by persons wfith rel­
atively less discrimination.
Table 43 presents the
items for which correlations with total number of items
checked were significantly oositive.
Table 44 presents
items for which the correlations were not significantly
above zero.
In Table 43, the five highest correlations
were for “Jews", “Hew Healers",
and "C.I.Q".
"A.P.L",
"Townsendites",
Of these, only two had biserial correla­
tions with conservatism scores as large as twice their
standard errors, and none had such correlations as large
as 3 times their standard errors.
In Table 44, the
five lowest correlations were for “Eu Elux Elan",
"Coughlinites", “Dies Committee",
and “Associated Willkie Clubs“*
“Communist Party",
Of these items, three
had biserial correlations with conservatism scores at
least 3 times their standard errors, one had a correla­
tion twice its standard error, and only one item was
correlated to the extent of less than twice its standard
error.
The inference would seem to be reasonable that,
whatever "total number of menaces checked” might be a
measure of, it would not be a measure related to con­
servatism.
This inference,
tested by a product-
moment correlation between self-score on the Abbreviated
C-R Opinionaire and total items checked as "menaces*1,
was fcuna to be true, since the resulting correlation
was only +.081, with a standard error of .061.
A further indication of the possible meaning
of "total menaces checked" was sought in the correlation
of this score ("total menaces checked") with the total
number of newspapers checked.
If the total number of
items checked in the Menace Checklist were partially
to be accounted for by what might be termed a low
"checking threshold", or an indiscriminate tendency to
check items, then there should be a significantly positive
correlation between the number of "menaces" checked and
the number of newspapers checked.
The correlation was
+.257, with a standard error of .057.
This was a
significantly positive correlation, though the per­
centage of variance accounted for by this correlation
was only between 6% and 1%,
From this indication, then,
n>t
it would seem that, while a tendency indiscriminately
to check items might possibly have accounted in some
degree for the number of "menaces” checked, this
tendency made only a very small contrioution to the
total number of "menaces" checked.
An alternative
interpretation, equally reasonable, would be that there
was a real relation between the number of newspapers
read by a respondent and the number of groups he con­
sidered to be "menaces".
Considering the reasonableness of the latter
interpretation, an analysis was made to determine whether
given newspapers reported to have been read by the re­
spondents might possibly have influenced the number of
groups the respondents considered to be "menaces".
Table 45 presents the biserial correlations between
the total number of "menaces" checked and the checking
or not checking of particular newspapers.
The only
correlations equal to as much as three times their
standard errors were for the Daily New s . the New fork
Times, and the Wo rid-Tele/gram.
These correlations were
negative, meaning that the fewer "menaces" a respondent
checked, the more likely he was to check the given
newspapers; or, vice versa, the more "menaces" he
checked,
the less likely he was to check the given
newspapers.
Whether this relationship should be
lb?
Table
4-5*
Biserial r ’st Relation between Number of "Menaces'* Checked
and Particular Newspapers Reported as Read Once a Week
(N = 269)
Newspapers
Biserial r*s
I
0bis r
III
II
±
±
±
±
±
±
±
±
±
±
Daily Kdswss*'.........
New York Times .....
Daily Mirror ........
Herald-Tribune .....
World-Telegram .....
Evening Journal ....
E M ..................
Bronx Home News ....
New York Sun ........
New York Post .......
Table
.073
.073
.087
.083
.074
.087
.079
.096
.083
.077
^(a
Biserial r'si Relation between Scores on Abbreviated
C-R Opinionaire and Particular Newspapers Reported
as Read Once a Week
(N = 269)
Newspapers
I
Daily News .......
New York Times ...
Daily Mirror ....
Herald-Tribune ...
World-Telegram ...
Evening Journal ..
P M ...............
Bronx Home News ..
New York Sun .....
New York Post ....
Biserial r ’s
II
ffbis r
III
±
±
±
±
±
±
±
±
±
±
.078
.075
.086
.083
.077
.084
.073
.096
.082
.073
no
interpreted in terms of newspapers' influence, in terms
of a more basic characteristic of individuals which
results in their reading particular newspapers and also
in their considering few groups to be "menaces", or
in terms of a tendency to check the more noticeable
items, cannot be determined on the basis of the present
data.
Table 46 presents evidence on the relation
between conservatism and the reading of particular
newspapers.
The only biserial correlations between
scores on the Abbreviated C-R Opinionaire and the
checking of particular newspapers, which were as large
as three times their standard errors, were for the
Hew York Times. P M . and the Hew York Post.
Since
these correlations were negative, they would mean
that the more liberal or radical a respondent was, the
more likely he would be to read the given newspapers,
assuming validity of the respondents' reports regarding
newspapers which were read.
Table 47 presents biserial correlations
between the total number of newspapers checked a n d the
checking or not checking of particular newspapers.
If
the assumption is made that, on the average, an
individual is unlikely actually to read more than two
or three different newspapers with any regularity,
these
in
Table
jjj
Biserial r's* Relation between Total Number of Newspapers
Reported as Read at Least Once a Week
and Particular Newspapers Checked
In = 269)
Newspapers
I
Biserial r ’s
II
Daily News ........
New York Times
Daily M i r r o r ..............
Herald-Tribune ............
World-Telegram ............
E v e n i n g - J o u r n a l ...........
P M .........................
Bronx Home News
New York S u n
New York P o s t
+.051
+.010
+.462
+.045
+.090
+.364
+.218
+.320
+.290
-.090
ffbis r
III
±
±
±
±
*
±
±
±
±
±
.078
.078
.074
.083
.077
.079
.076
.091
,078
.078
correlations could be interpreted in terms of the validity
of the checking of different newspapers.
sumption,
On this as­
that the individuals who checked relatively
fewer newspapers were making the more valid responses,
the lowest correlations would indicate which news­
papers were checked with the greatest relative validity.
The five lowest correlations were for the Daily hews,
the New York Times. the Herald-Tribune. the World-Telegram.
and the New York Post.
The validity of the above assump­
tion becomes questionable, however, when the relative
positions of these particular newspapers in the check­
list are considered.
Each of the first four newspapers
mentioned immediately above occupied either the first
or the second position in their respective columns,
while the New York Post occupied the last position in
the column in which it appeared.
The interpretation
would possibly be the reverse of that stated above*
That those who checked the fewest newsoapers were, in
general, the ones who checked only the most noticeable
newspapers--i. e., those appearing at the top or bottom
of columns.
173
Summary
The present chapter has discussed results
based on the unstandardized checklists relating to
"menaces” and to newspapers.
Certain groups were found
to be more likely to be considered "menaces" by the
more conservative individuals, while others were more
likely to be considered "menaces" by radical individuals.
Responses to the former were more uniformly and positively
intercorrelated than were responses to the latter.
"Jews", "New Dealers",
"A.i’.L.", "Townsendites", and
"C.I.O." were found to be checked relatively more
frequently by individuals checking many "menaces",
while "Ku Klux Klan",
"Coughlinites", "Dies Committee",
"Communist Party", and "Associated V/illkie Clubs" were
relatively more likely to be checked by persons who
checked few "menaces".
were,
Responses to the former list
in general, not significantly correlated with
conservatism, while responses to the latter list were,
in general, sigrificantly correlated, positively or
negatively, with conservatism.
No significant correla­
tion was found between conservatism and total number of
"menaces" checked; but there was a significant, though
low, positive correlation between the total number of
"menaces" checked and the total number of newspapers
checked.
Total number of "menaces" checked was negatively
correlated with checking the particular newspapers
New York Times, Daily Ne ws . and World-Telegram.
News­
papers which were found to he read relatively more
frequently by liberal or radical students than by con­
servative students were the New York Times. P L . and the
New York Post. Newspapers relatively more frequently
checked by those who checked few newspapers were the
Daily News. the New York Times. the Kerald-Tribune. the
World-Telegram. and the New York Pos t, all of which
occupied the more noticeable positions in their
respective columns.
Chapter X
Summary and Conclusions
The present research was meant to answer
certain hypothetical questions relating to the attitudes
of Catholic, Jewish, and Protestant students and the
estimates by these students of "typical" Catholic,
Jewish, and Protestant attitudes.
The research was based on a methodology
relatively new in the field of attitude investigation.
This methodology required that a respondent mark
statements in attitude questionnaires in accordance
with four different instructions:
in terms of his own attitudes,
(a) to give responses
(b) to give responses
in terms of his idea of how most Catholics would
respond,
(c) to give responses in terms of his idea
of how most Jews would respond, and
(d) to give re­
sponses in terms of his idea of how most Protestants
would respond.
The respondents in this study were in three
samples, each of which v/as subdivided into a Catholic
group, a Jev/ish group, and a Protestant group, on the
basis of stated religious preferences.
Respondents in
all samples were either students, or candidates for
admission as students, in a large metropolitan
university, which admits only male students.
The first
sample, respondents to the Communism-War questionnaire,
consisted of 36 Catholics, 64 Jews, and 32 Protestants,
who were candidates for admission.
The second sample,
respondents to the Birth Control questionnaire, was
composed of 12 Catholics,
15 Jews, and 14 Protestants,
who were junior and senior engineering students.
The
third and largest sample, respondents to the Abbreviated
C-R Opinionaire, was composed of 63 Catholics, 151 Jews,
and 55 Protestants, who were completing their first
semester in liberal arts or engineering courses.
The attitude measures employed were;
the
scale to measure Attitude toward Communism, Form A,
prepared by L. L. Thurstone; the scale to measure
Attitude toward War, Form A, prepared by R. C. Peterson;
the scale to measure Attitude toward Birth Control,
Form A, prepared by K. A. Wang and L. L. Thurstone;
and the Abbreviated C-R Opinionaire, adapted from the
C-R Opinionaire, Form J, prepared by T. F. Lentz and
associates.
With each questionnaire was included a
Personal Data Blank, a Hewspaper Checklist, and a
Menace Checklist, prepared by the writer.
Results were analyzed in terms of:
117
1. Co-operation of respondents.
2. Comparative attitude positions and attitude
variabilities for respondent groups.
3. Comparative estimates of attitudes of typical
Catholics, Jews, and Protestants, between
respondent groups.
4. Comparative estimates of attitudes of typical
Catholics, Jews, and Protestants, within
respondent groups.
5. Reliability of the Abbreviated C-R Opinionaire,
and relations obtaining among items and between
items and total scores.
6. Relations obtaining among items of the news­
paper Checklist and Menace Checklist and
between these items and certain available scores.
The chief results of the present study may
be summarized as follov/s:
1. The general methodology of this study was
found to be practicable from the standpoint of the
oo-operation of respondents.
Only 17.8$ of re­
spondents to the Abbreviated C-R Opinionaire failed
to supply complete responses for all four conditions
of instruction.
Only about 5$ failed to give
complete responses for at least two conditions.
(Pages
2. In general, Catholics were most conservative
with respect to the issues studied.
Jews were
generally most liberal or radical, and Protestants
were intermediate in their attitudes.
Differences
between the attitude positions of Catholics and
n%
Jews were highly reliable on all issues except war.
However, all differences between the attitude
positions of Jews end Protestants were unreliable,
and differences between Catholics end Protestants
were also unreliable except on general conservatism.
(Pages £ 3 “£
)
3. The variability in attitudes was about
equal for all respondent groups, except in attitudes
toward birth control.
On this issue, the greatest
individual differences were among Catholics, the
smallest individual differences among Jews, and the
individual differences among Protestants were
intermediate in extent.
(Pages ££"£? )
4. With few exceptions, all respondent groups
agreed among one another in their estimates of the
attitudes of typical Catholics, Jews, and Protestants.
The exceptions were:
Catholic respondents thought
typical Catholics to be more conservative than did
Jewish respondents.
Catholics estimated Jews to
be more in favor of war than Protestants estimated
them to be, while both Catholics and Protestants
attributed more liberalism to most Jews than was
attributed to them by Jewish respondents.
Finally,
Catholics attributed more conservatism to Protestants
than was attributed to them by Jews.
Differences
179
between Jews and Catholics in their estimates of
the conservatism of all groups accounted for 3 of
the 5 instances, out of a possible 22, in which
significant differences between group estimates
occurred.
(Pages 7 2 “ 7? )
5. All respondent groups thought their own
attitudes to be more liberal than those of typical
individuals with the same religious affiliation.
All groups estimated Protestant conservatism to
be higher than it was among Protestant respondents,
and Catholic attitude toward birth control to be
less favorable than it was among Catholic re­
spondents.
Protestants and Jews thought typical
Jews less favorable to birth control than were
the Jewish respondents.
Protestants,
Catholics, out not
estimated Jewish attitudes toward war
and toward communism to be more favorable than
were the attitudes expressed by Jev/ish respondents.
(Pages. $0-?$)
6. In general, the extent of individual dif­
ferences in estimates of typical Catholic, Jev/ish,
and Protestant attitudes was about e^ual for the
three respondent groups.
If homogeneity of estimates
be considered a criterion of the existence of
stereotypes regarding attitudes possessed by ad­
herents to particular religions, the stereotypes
no
were equally influential with all respondent groups*
(Pages
)
7* Individual differences in the estimates
of typical Catholic, Jewish, and Protestant attitudes
were, in general, either reliably more extensive
than, or not reliably different from, the individual
differences in the expressed attitudes of the
respondent groups*
In only 5 comparisons in a
total of 26 did an exception to this generaliza­
tion occur*
Four of the five exceptions occurred
with respect to estimates of Catholic attitudes.
The evidence favored the interpretation that a
stereotype probably existed concerning the
relative conservatism of Catholics.
(Pages
>03
)
8. Catholic respondents identified their own
attitudes with their estimates of Protestant
attitudes, while Protestant respondents identified
their own attitudes with their estimates of typical
Jewish attitudes.
The Jewish respondents thought
typical members of all religions to be less liberal
than they themselves were*
Typical Jews and
Protestants were, in general, thought by all groups
to have equally conservative attitudes.
(Pages
lOl-Hit)
9. Variability comparisons among the attitudes
and attitude estimates, for the separate respondent
groups, favored the existenoe among Catholio students
IV
of a stereotype concerning Catholic and Protestant
conservatism.
These stereotypes concerning Catholics
and Protestants were apparently equally definite.
Jewish respondents appeared to have stereotypes
concerning the conservatism of all groups, which
stereotypes were all equally definite.
There was
no evidence indicating the existence of a stereo­
type among Protestants.
However, no criterion of
the absence of a stereotype was available in the
data.
(Pages II& ' I M
10.
)
The conservatism of individual respondents
in all groups was more closely related to their
own estimates of the conservatism of typical members
of their own religion than to their own estimates
of the conservatism of a typical member of any
other religion.
The attitudes of individual
Catholics and their estimates of the attitudes of
typical Catholics were more closely related to
their estimates of Protestant attitudes than to
their estimates of Jewish attitudes.
Attitudes of
Protestants and their estimates of the attitudes
of Protestants were more closely related to their
estimates of Catholic attitudes than to their
estimates of Jewish attitudes.
The attitudes of
the individual Jewish respondents and their estimates
Ill
of Jewish attitudes were more closely related to
their estimates of Protestant attitudes than to
their estimates of Catholic attitudes.
(Pages
l2SmIA9)
11. The more conservative Jewish respondents
were, the less likely they were to think Catholios
to be more conservative than Jews, and the more
likely they were to think most Jews to be more
conservative than most Protestants.
The more
conservative the Protestants were, the less likely
they were to consider that most Catholics were
more conservative than most Protestants.
(Pages
12. Scores on the Abbreviated C-R Opinion­
aire were more determined by disagreement with
radical items than by agreement with conservative
items when respondents marked statements according
to their own attitudes, but were more determined
by agreement with conservative items than by dis­
agreement with radical items when respondents
marked the statements according to their ideas of
how typioal members of their own or of some other
group would respond.
(Pages
IW-IW)
13. Reliability of the Abbreviated C-R Opinionaire
was* in general, higher when respondents expressed
their own attitudes than when they made estimates
H3
of the attitudes of typical members of their own
or of some other group*
(Pages
lH-1-IW)
14. Biserial correlations between total
scores on the Abbreviated C-R Opinionaire and
responses to particular items were higher for
radical than for conservative items.
This finding
was in agreement with the previous finding that
total scores were more determined by disagreement
with radical items than by agreement with conservative
items.
(Pages
15. All conservative items were positively
and all radical items negatively correlated with
conservatism soores.
Likewise, most conservative
items were positively inter-correlated, most
radical items positively inter-correlated, and
most conservative-radical pairs of items negatively
inter-correlated.
These findings indicated that
the terms "radical" and "conservative" applied to
describe items were properly so applied.
(Pages
1^1-15+)
16. As indicated by the fact that pairs of
items highly inter-oorrelated for a given respondent
group were much less highly inter-correlated for
other respondent groups, the organization of con­
servative and radical attitudes was apparently
n y
different for the different (Catholic, Jewish, and
Protestant) respondent groups.
(Pages/5V-/57 )
17. Certain groups in the Menaoe Checklist
were more likely to be considered menaces to the
best interests of the United States by conservative
than by radical students— notably,
"Socialist
Party” , "Aliens (in general)", "C.I.O.", "New
Dealers", and "Communist Party".
notably,
"Ku Klux Klan",
Other groups—
"Coughlinites",
"Dies
Committee", and "German-American Bund"— were more
likely to be considered menaces by radical students
than by conservative students.
(Pages
I^O-I^X)
18. Items of the Menace Checklist whioh
were positively related to conservatism were more
uniformly and positively inter-correlated than were
items which were negatively correlated with con­
servatism.
This fact oould be interpreted as
evidence that groups considered to be menaces by
conservative students were more closely associated
in the ideas of students than were groups considered
to be menaces by the more radical students.
(Pages /4JHW-)
19. The total number of items in the Menace
Checklist which were checked by respondents was not
significantly correlated with conservatism, but was
correlated positively with total number of newspapers
checked.
However, the latter correlation was low,
and would not Justify the conclusion that the
checking of items in both checklists was merely
due to a tendency indiscriminately to check many
or few items.
20.
(Pages / 6 7 * )
Biserial correlations indicated that the
New York Times, P M , and the New York Post were
more likely to be read by radical than by con­
servative students.
(Pages IW-/7 0)
f U
Implications of the Researoh Findings
Certain sooial psychologists--notably, La Piere
( I Jl ), and la Piere and Farnsworth ( 1 3 )- -have criticized
attitude research because it does not neoessarily afford
predictions of overt behavior.
Thurstone and Chave
( 13 )
(25),
However, Lundberg (
likert
(IS ),
1 7 ),
and Murphy
have discussed such criticisms and have assigned
to attitude research a significance in its own right*
The present writer believes that the motivation at the
basis of most attitude investigations must consist in
some such view as that expressed by Murphy
( 19 ),
who
writes:
After all, a m a n fs categorical agreement or
disagreement with a rather strongly stated opinion
about Chinese, or Jews, or Communists, or Rotarians.
is in everyday life regarded (if the man be sincere)
as a significant part of his behavior. There seems
no reason why this behavior should suddenly become
non-significant when it is made the subject of
careful inquiry.
Since studies of expressed attitudes, or of
expressions of what the respondent "wants people to think
he thinks"
(17 ),
are considered to yield important
data for sooial psychology, it should be apparent that
an equally important source of data is the study of
estimated attitudes, or the study of "what respondents
say they think other people think".
The method employed
in the present investigation has yielded data
i%7
on attitude estimates as well as on expressed attitudes.
One of the outstanding results of this study
was that students were more liberal than they thought
most people of the same religion to be.
Yet the attitudes
of these students were more olosely related to those
they estimated for people of the same religious affilia­
tion than to those they estimated for people of some
other religious background.
Each individual of a given
religion, then, estimated most people of the same religion
to be somewhat more conservative than himself, regardless
of how conservative he himself was.
Thus, since
Catholic respondents were, in general, more conservative
than respondents of the other religions, they estimated
most Catholics to be more conservative than other groups
estimated Catholics to be.
Likewise, the more conservative
the Jewish respondents were, the more conservative they
thought most Jews to be, and therefore the less lihely
they were to think Catholics or Protestants more con­
servative than Jews.
The fact that Catholic respondents thought
Catholics to be more conservative than Protestants,
and also more conservative than they themselves actually
were, resulted in an identification of their own attitudes
with those of Protestants.
Likewise,
the faot that the
Protestants estimated Jews to be less conservative than
lit
Protestants, and the fact that they were themselves
less conservative than they estimated Protestants to be,
led to an identification of their own attitudes with
those of Jews*
Similarly, Jewish respondents were more
liberal than their estimates of the attitudes of most
Jews, and they estimated Jewish attitudes to be more
liberal than Catholic or Protestant attitudes, so that
they themselves were more liberal than their estimates
for any of the groups*
Stereotypes concerning Catholio, Jewish, and
Protestant attitudes were evident.
Two separate find­
ings supported the interpretation that stereotypes
existed:
(a) all groups agreed that Catholics were
most conservative, and all but Jewish respondents agreed
that Jews were most radical and that Protestants were
intermediate, and (b) individual differences in attitude
estimates within groups were frequently less extensive
than individual differences in the expression of attitudes.
The relative ranks assigned members of the
three religions with respect to conservatism exactly
paralleled the relative ranks of the three respondent
groups in their expressed conservatism.
The difference
between the two sets of ranks was simply that the attitude
positions represented by the latter were more liberal
tnan those represented by the former.
m
Recommendations for Further Research
Inter-condition analyses similar to that in
the present study could he carried out by use of a single
group of respondents*
Significant interpretations
needed in social psychology might be based on the study
of the attitudes of:
1. Employees toward unions, and their estimates of
the attitudes of foremen, superintendents, and
company officials.
2* Democrats toward the constitution, and their
estimates of the attitudes of Republicans,
Socialists, Communists, etc.
3. Children toward various issues, and their
estimates of their parents’ attitudes.
A large variety of fruitful hypotheses might be tested
by means of this general technique applied to given
groups, with the use of various attitude measures.
A number of methodological problems appear
to justify investigation, among which the following may
be listed:
1.
Could a rating-scale technique be sub­
stituted for the method used in this study?
That
is, would similar results be obtained if respondents
were asfced merely to rate typical Catholic,
Jewish,
and Protestant attitudes toward a given issue
on, say, a five-point scale ranging from strong
approval to strong disapproval?
An experimental
test involving the use of both methods, with the
190
same respondent group, would seem necessary to
answer this question.
E. The Abbreviated C-R Opinionsire used in this
study did not have the desired reliability under
some conditions.
Verification of the present
results should involve the use of a longer, more
reliable measure of conservatism.
Retest reliability
of the measure, under the several conditions of
instruction, probably should be determined.
3.
The finding that under the self-condition,
scores were more determined by disagreement with
radical items than by agreement with conservative
items and that under other conditions the reverse
was true should be subjected to further investiga­
tion.
If this finding were verified generally,
it would have important implications for question­
naire construction.
For example, this finding
would imply that on scales constructed by the
Thurstone method, there might be a constant error
in the self-scores in the direction of conservatism,
and a constant error in the opposite direction
in the other types of scores.
If such were the
case, scores on the Thurstone scales for a study
like the present.one should probably consist in the
mean of the median scale value of checked items
/?/
and the median scale value of crossed items.
4. The present study provided no evidence
regarding the correctness or incorrectness of the
estimates of the respondents, since the mean attitude
positions of student groups could not he considered
criteria of the attitude positions of typical
members of the different religions#
Studies in
which such criteria can successfully be applied
seem desirable.
Studies of this kind would require
either that a representative sample of all Catholics,
Jews, or Protestants be tested, or that the in­
structions define more completely the criterion
group whose attitudes are to be estimated.
5. The Menace Checklist showed promise as a
possible measure of conservatism.
The items of this
checklist might be scored in terms of the validity
with which they individually discriminated between
conservatives and radicals, and administered, in
conjunction with an accepted measure of conservatism,
to additional groups of respondents.
A defect in
the checklist as administered in this study was
that the instructions did not provide for indications
by the respondents that they did not consider a
particular group to be a menace.
Provisions for
both negative and affirmative responses to the
items would be an improvement.
I
Appendix A
Bibliography
1* Allport, G. W . , "Attitudes", in Murchison, C., (ed.),
Handbook of Soolal Psychology. Worcester; Clark
University Press, 19^5 •
2. Allport, G. 7/., "The Composition of Political Attitudes",
American Journal of Sociology. 1929, 3 5 , 220-238.
3. Carlson, H. B., "Attitudes of Undergraduate Students",
Journal of Social Psychology. 1934, J5, 202-213.
4. Chesire, L., Saffir, M . , and Thurstone, L. L.,
Computing diagrams for the Tetraohorio Correlation
Coefficient. Chicago: University of Chicago Bookstore,
1933.
5. Corey, S. M., "Signed vs. Unsigned Attitude Question­
naires". Journal of Educational Psychology. 1937.
28, 144-1457------------------------ *----- ^
6. Droba, D. D., "A Soale of Militarism-Pacifism",
Journal of Educational Psychology. 1931, 22, 96-111.
7. Dunlap, J. W . , "Note on Computation of Biserial
Correlations in Item Evaluation", Psychometrika,
1936, 1, 51-58.
— ----------8. Guilford, J. P., Psychometric Methods, New York;
MoGraw Hill, 1936.
9. Harris, A. J., Remmers, H. H . , and Ellison, C. E.,
"The Relation between Liberal and Conservative Attitudes
in College Students, and Other Factors", Journal of
Social Psychology, 1932, _3, 320-336.
10. House, F. N., "Measurement in Sociology", American
Journal of Sociology, 1934, 40, 1-11.
11. Jones, V., "Attitudes of College Students and the
Changes in Such Attitudes during Four Years of College",
Journal of Educational Psychology. 1938, 29, 14-25.
12. La Piere, R. T., "The Sociological Significance of
Measurable Attitudes", American Sociological Review,
1938, 3, 175-182.
13. La Piere, R. T., and Farnsworth, P. R., Social
Psychology. New York; MoGraw Hill, 1936.
2
14. Lentz, T. F., jet al», Manual for C-R Opinionaire,
St. Louis:
Charaoter Research Institute, Washington
University, 1935.
15. Lilcert, R., "A Technique for the Measurement of
Attitudes” , Arohives of Psychology, Ro. 140. Pp. 55.
----1932.
16. Lindquist, E. F., Statistical Analysis in Educational
Research. Hew York!
Houghton Mifflin, TU4T5!
17. Lundberg, G. A., Social Research. Rew York:
Green, 1929.
Longmans,
18. Merton, R. K., ’’Fact and Factitiousness in Ethnic
Opinionnaires". American Sociological Review. 1940.
5, 13-28.
-----------19. Murphy, G., and Murphy, L. B., Experimental Social
Psychology. Rew York:
Harper and Bros., 1931!
20. Relson, E., ”Radicalism-Conservatism in Student
Attitudes", Psychological Monographs. 1938, 50,
Ro. 4, Pp. 32.
21. Richardson, M. W., and Kuder, G. F., "The Calculation
of Test Reliability Coefficients Based on the Method
of Rational Equivalence", Journal of Educational
Psychology. 1939, 30, 681-687.
22. Spencer, D., Fulcra of Conflict: A Rew Approach to
Personality Measurement. Yonkers: World Book Co., 1939.
23. Sukov, M . , and Williamson, E. G. , "Personality Traits
and Attitudes of Jewish and Ron-Jewish Students",
Journal of Applied Psychology, 1938, 22, 487-492.
24. Thurstone, L. L . , Instructions for Using the Scale
Attitude toward Communism, Chicago; University of
Chicago Press, 1931.
25. Thurstone, L.
and Chave, E. J., The Measurement of
Attitude, Chicago: University of Chic'ago Press, 192U7
26. Thurstone, L. L., and Peterson, R. C., Instructions
for Using the Sc -le Attitude toward War! Chicago:
University of Chicago Press, l93l.
3
27. Vetter, G. B., "The Measurement of Social and Political
Attitudes and the Related Personality Factors",
Journal of Abnormal and Social Psychology, 1930, 25,
14$-189.
28. 7/ang, K. A., and Thurstone, I. L . , Instructions for
Using the Scale Attitude toward Birth Control/"Chicago:
University of Chicago Press, 193(37
29. Whistler, 1., and Remraers, H. H . , "A Scale for
Measuring Individual and Group Morale", Journal of
Psychology. 1937, 4, 161-165.
30. Whistler, L., and Remmers, H. H . , "liberalism,
Optimism, and Group Morale;
A Study of Student Attitudes",
Journal of Social Psychology, 1938, j); 451-467.
31. Winslow, C. N., "A Study of the Extent of Agreement
between Friends’ Opinions and their Ability to Estimate
the Opinions of Each Other", Journal of Social Psychology,
1937, 8, 433-442.
H
Appendix B
The Communism-War Questionnaire
(Form for Catholio Respondents ) +
Do not write your name on this paper* Answer the following
questions ar accurately as you can.
1. On your last birthday, what was your age?___________
2. What was the date of your birth?
Month
Kay
Year
3. Where were you born?
City
State or Country
4# Where was your father born?
City
State or Country
5. Where was your mother born?
_______
City
State or Country
6. What is your religious affiliation?____________________
7. Put a circle around the names of the newspapers which you
read as often as once a week.
If you read other newspapers,
not listed here, as often as once a week, write their
names on the blank lines.
Daily News
New York Times
Daily Mirror
Herald-Tribune
World-Telegram
Evening-Journal
Bronx Home News
Daily Worker
New York Sun
New York Post
The Chief
Brookljm Eagle
Christian Science Monitor
_____________________ _
8. Put a circle around the names of all the following groups
which you think are a menace to the best interests of the
United States.
Labor Unions
Socialist Party
K u Klux Elan
Coughlinites
C.I.O.
Labor Spies
Dies Committee
New Dealers
W.P.A.
Tammany Hall
A.F»L»
'Brain Trusters'
Communist Party
Catholio Church
Republican Party
The Jews
Non-Sectarian Anti-Nazi
League
Demooratio Party
German-Amerioan Bunds
American Civil Liberties
Union
American Liberty League
Labor's,Non-Partisan League
Amer. Assn. of Manufacturers
Townsendites
*
The form for Jewish respondents is not reproduced
here, since it differed only in that "Catholio" was sub­
stituted for "Jew" under the second instruction.
The
titles of the questionnaires did not appear on any forms
presented to the respondents.
s
Do not write your name on this paper.
This is a study of attitudes toward war and communism. Below
you will find 40 statements of different attitudes toward
war and oommunism.
Put a oheok mark
with the statement.
Put a oross (X) if you disagree with the statement.
If you cannot decide about a statement, you may mark it
with a question mark.
This is not an examination.
People differ about what is right
and wrong in these issues. Remember that nobody will know
what you write, since your name is not on the paper. Ex­
press your own attitude, even if you think it differs from that
of your friends” Please indioate your own attitude by a
check mark when you agree and by a cross when you disagree.
)
1. Under some conditions, war is necessary to maintain
justice.
) 2. Both the evils and the benefits of communism are
greatly exaggerated.
) 3. The benefits of war rarely pay for its losses even
for the viotor.
) 4. It is only the ignorant and incompetent who want
communism.
) 5. War brings out the best qualities in men.
j 6. The whole world must be converted to communism,
j 7. War has some benefits, but it*s a big price to pay
for them.
} 8. Workers can hardly be blamed for advocating communism,
j 9. There is no conceivable Justification for war.
) 10. Communism is a much more radical change than we
should undertake.
) 11. War is often the only means of preserving national honor.
) 12. Police are justified in shooting down the communists.
) 13. War is a ghastly mess.
j 14. Give Russia another twenty years or so and you*ll see
that communism can be made to work.
) 15. I never think about war and it doesn*t interest me.
) 16. If Russia today is a sample of how communism works,
we don*t want it.
) 17. War is a futile struggle resulting in self-destruction.
) 18. Communism should be established by force if necessary.
) 19. The desirable results of war have not received the
attention they deserve.
) 20. The communist may be rough but he has the right idea,
j 21. Pacifists have the right attitude, but some pacifists
go too far.
) 22. I am not worrying, for I don*t think there*s the
slightest ohanoe that communism will be adopted here.
) 23. The evils of war are greater than any possible benefits,
j 24. The communists are the most destructive element in
our civilization.
) 25* Although war is terrible it has some value.
) 26. Communism is the solution to our present economic problems.
) 27. International disputes should be settled without war.
) 28. We should be open-minded about communism,
j 29. War is glorious.
) 30. The ideals of communism are worth working for.
6
Do not write your name on this paper.
Continued—
(
(
(
(
(
(
(
(
{
{
) 31. Defensive war is justified hut other wars are not.
) 32. I am not sure that communism solves the problems of
capital and labor.
) 33. War breeds disrespeot for human life.
) 34. The whole communistic scheme is unsound.
) 35. There can be no progress without war.
) 36. Communists should set up a dictatorship of the
workers in the United States.
) 37. It is good judgment to sacrifice certain rights in
order to prevent war.
) 38. Vfe should not reject communism until it has been
given a longer trial.
) 39. War is the only way to right tremendous wrongs.
) 40. I£ a man has the vision and the ability to acquire
property, he ought to be allowed to enjoy it
himself.
Please read carefully
This is another study of attitudes toward war and communism.
On the next page you will find 40 more statements expressing
different attitudes toward war and communism.
This time
please mark them the way you think a member of the Jewish
congregation would mark them, if he were expressing his
sincere opinion about the statements.
Put a check mark ( ) if you
with the statement.
Put a cross (X) if you think
with the statement.
If you cannot deoide about a
statement, you may mark
think a Jew would agree
a Jew would disagree
J e w ’s answer to the
it with a question mark.
This is not an examination. People differ in their opinions
about what a Jew would think is right and wrong in these
issues. Please mark the statements as you sincerely believe
a Jew would mark them.
Indioate your own idea of what a Jew would think by a check
mark if you think he would agree, and by a cross if you
think he would disagree.
Do not write your name on this paper.
y
Please answer according to instructions on preceding page;
(
(
{
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
{
(
)
1. Under some conditions, war is neoessary to maintain
Justice.
) 2. Both the evils and the benefits of communism are
greatly exaggerated.
) 3. The benefits of war rarely pay for its losses even
for the viotor.
) 4. It is only the ignorant and incompetent who want
oommunism.
)
5. War brings out the best qualities in men.
) 6. The whole world must be converted to communism.
)
7. War has some benefits, but it's a big price to pay
for them.
) 8. Workers can hardly be blamed for advocating oommunism.
]
9. There is no oonoeivable Justification for war.
) 10. Communism is a muoh more radical change than we
should undertake.
) 11. War is often the only means of preserving national honor.
) 12. Police are Justified in shooting down the communists.
j 13. War is a ghastly mess.
) 14. Give Russia another twenty years or so and you'll see
that oommunism can be made to work.
) 15. I never think about war and it doesn't interest me.
) 16. If Russia today is a sample of how oommunism works,
we don't want it.
) 17. War is a futile struggle resulting in self-destruction.
) 18. Communism should be established by force if neoessary.
) 19. The desirable results of war have not received the
attention they deserve.
) 20. The communist may be rough but he has the right idea.
) 21. Pacifists have the right attitude, but some pacifists
go too far.
) 22. I am not worrying, for I don't think there's the slight­
est chance that oommunism will be adopted here.
) 23. The evils of war are greater than any possible benefits.
) 24. The communists are the most destructive element in
our civilization.
) 25. Although war is terrible, it has some value.
j 26. Communism is the solution to our present economic problems.
) 27. International disputes should be settled without war.
) 28. We should be open-minded about oommunism.
) 29. War is glorious.
) 30. The ideals of communism are worth working for.
} 31. Defensive war is Justified but other wars are not.
) 32. I am not sure that oommunism solves the problems of
capital and labor.
) 33. War breeds disrespect for human life.
) 34. The whole communistic scheme is unsound.
) 35. There can be no progress without war.
) 36. Communists should set up a dictatorship of the
workers in the United States.
) 37. It is good Judgment to sacrifice certain rights in
order to prevent war.
) 38. We should not reject communism utotil it has been given
a longer trial.
) 39. War is the only way to right tremendous wrongs.
) 40. If a man has the vision and the ability to acquire
property, he ought to be allowed to enjoy it
himself.
%
Appendix £
The Birth Control Questionnalre
Do not write your name on this paper.
questions as accurately as you can.
Answer all the following
1. On your last birthday, what was your age? ____
years.
2. What was the date of your birth?
________
Month
Day
Year
3. What is your preferred religion? (Be sure to check one
preference.)
Catholic________
Jewish__________
Protestant
4. Where were you born?____
CJity
State or country
5. Where was your father born?
State or country
UTIy
6. Where was your mother born?
State or country
City
7. Put a circle around the names of the newspapers which you
read as often as once a week.
If you read any newspapers
not listed here, write their names in the blanks.
Daily Hews
Hew York Times
Daily Mirror
Herald-Tribune
World-Telegram
Evening Journal
PM
Bronx Home News
Daily Worker
New York Sun
New York Post
The Chief
Brooklyn Eagle
Christian Science Monitor
8. Put a circle around the names of all the following groups
which you think are a menaoe to the best interests of the
United States.
labor Unions
Socialist Party
K u Klux Klan
Coughlinites
C.I.O.
Protestant Churches
Labor Spies
Dies Committee
New Dealers
W.P.A.
Tammany Hall
A.F.L.
Brain Trusters
Communist Party
Catholic Church
Republican Party
The Jews
Associated Willkie
Clubs
Non-Sectarian Anti-Nazi
League
Democratic Party
German-American Bunds
American Civil Liberties
Union
American Liberty League
Nat'l. A s s ’n. of
Manufacturers
Townsendites
(Turn to next page.)
Do not write your name on this paper*
This is a study of attitudes toward birth control.
Below you will
find 20 statements expressing different attitudes toward birth
control.
Put a check mark (
if you agree with the statement.
Put a double check ( * V j if you strongly agree with the
statement.
Put a cross (X) if you disagree with the statement.
If you cannot decide about a statement, you may mark it
with a question mark.
This is not an examination. People differ about what is right
and wrong in this issue. Remember that nobody will know what you
write, since your name is not on the paper. Express your own
attitude, even if you think it differs from that of your friends.
Please indicate your own attitude by a check mark when you agree,
by a double check when you agree emphatically, and by a cross
when you disagree.
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
)
)
1. Birth control is a legitimate health measure.
2. Birth control is necessary for women who must help
earn a living.
) 3. The practice of birth control may be injurious
physically, mentally, or morally.
) 4. We simply must have birth control.
j 5. The practice of birth control is equivalent to murder.
) 6. Birth control has both advantages and disadvantages.
) 7. Only a fool can oppose birth control.
j 8. Birth control increases the happiness of married life.
) 9. Deoenoy forbids the use of birth control.
) 10. Birth control should be absolutely prohibited.
) 11. Birth control is the only solution to many of our
social problems.
} 12. Birth control reduces the marital relation to the
level of vice.
) 13. Birth
control has nothing to do with morality.
) 14. Birth
control information should be available to everybody
j 15. Birth
control is morally wrong in spite of its
possible benefits.
) 16. Uncontrolled reproduction leads to overpopulation,
social unrest, and war.
) 17. Birth
control is race suicide.
) 18. People
should be free to do whatever they wish about
birth control.
) 19. The practice of birth control evades man's duty to
propagate the race.
) 20. The slight benefits of birth control hardly justify it.
(Turn to next page.)
Do not write your name on this paper*
fC
This is another study of attitudes toward birth control. Below
you will find 20 statements expressing different attitudes toward
birth control.
This time please mark them the way you think a
typical member of the Catholic church would mark them, if he
were expressing his sincere opinion about the statements.
Put a check mark ( >/) if you think a typical Catholio would
agree with the statement.
Put a double check
if you think a typical Catholic
would strongly agree with the statement.
Put a cross (X) if you think a typical Catholic would
disagree with the statement.
If you cannot decide about a typical Catholic's answer to
the statement, you may mark it with a question mark.
This is not an examination. People differ in their opinions
about what a typical Catholic would think is right and wrong in
this issue. Please mark the statements as you sincerely
believe a typical Catholic would mark them.
Indicate your own idea of what a typical Catholic would think by
a check mark if you think he would agree, by a double check if
you think he would agree emphatically, and by a cross if you
think he would disagree.
( )
( )
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
1. Birth control is a legitimate health measure.
2. Birth control is necessary for women who must help
earn a living.
) 3. The practice of birth control may be injurious
physically, mentally, or morally.
)
4. Y/e simply must have birth control.
)
5. The practice of birth control is equivalent to murder.
)
6. Birth control has both advantages and disadvantages.
)
7. Only a fool oan oppose birth control.
)
8. Birth control increases the happiness of married life.
)
9. Decency forbids the use off birth control.
) 10. Birth control should be absolutely prohibited.
) 11. Birth control is the only solution to many of our
social problems.
) 12, Birth control reduces the marital relation to the
level of vice.
) 13. Birth control has nothing to do with morality.
) 14. Birth control information should be available to everybody.
) 15. Birth control is morally wrong in spite of its possible
benefits.
) 16. Uncontrolled reproduction leads to overpopulation,
social unrest, and war.
) 17. Birth control is race suicide.
) 18. People should be free to do whatever they wish about
birth control.
) 19. The practice of birth control evades man's duty to
propagate the race.
) 20. The slight benefits of birth control hardly justify it.
If. B . : Two additional pages were in the original questionnaire,
with instructions to mark statements according to respondents'
idea of how (a) Protestants, and (b) Jews would respond.
In­
structions were identical to those above exoept for substitution
of "Protestant" and "Jew" instead of "Catholio".
Appendix D
The Abbreviated C-R Oplnlonalre
Do not write your name on this paper* Answer the following
questions as accurately as you can*
1* On your last birthday, what was your age?
years.
2. What was the date of your birth
___________
Month
Day
Year
3. What is your preferred religion? (Be sure to check one
preference)
Catholic_______________
Jewish
_____________
Pro t estant
4. Where were you born?
City
State or Country
5. Where was your father born?
City
(
State or Country
6 . Where was your mother born?
State or Country
ETEy"
Put a circle around the names of the newspapers which you read
as often as once a week.
If you read any newspapers not
listed here, write their names in the blanks.
Daily Hews
New York Times
Daily Mirror
PM
Bronx Home News
Daily Worker
Herald-Tribune
World-Telegram
Evening Journal
Brooklyn Eagle
New York Sun
New York Post
Christian Science Monitor
The Chief
8 . Put a circle around the names of all the following groups
which you think are a menace to the' best interests of the
United States.
labor Unions
Sooialist Party
Ku Klux Klan
Coughlinites
C.I.O.
Protestants
labor Spies
Dies Committee
New Dealers
W.P.A.
Tammany Hall
A . F .1.
Brain Trusters
Communist Party
Catholics
Republican Party
Jews
Associated Willkie
Clubs
Non-Sectarian Anti-Nazi
league
Democratic Party
German-American Bunds
American Civil liberties
Union
American liberty league
Nat'l. Ass'n. of
Manufacturers
Townsendites
Aliens (in general)
(Turn to next page, please.)
Do not write your name on this paper*
u
Below are 22 statements listed to see what people think about
many question*1
.
If you agree more than you disagree with a statement, mark
it plus ( + )•
If you disagree more than you agree with a statement, mark
it zero (0)•
Be sure to place either a plus or a zero to the left of each
number.
This is not an examination. People differ about what is right
and wrong in these issues. Remember that nobody will know what
you write, since your name is not on the paper. Express your
own opinion, even if you think it differs from that of your
friends. Please indicate your own attitude by a plus when you
agree and by a zero when you disagree.
1. We should celebrate Pasteur's birthday rather than
Washington's, as he has done the world a greater service.
2. Conservative people are usually more intelligent than
radical people.
3. Since the theory of evolution has been accepted by most
scientists, it should be taught in our schools.
4. At the age of 21, people should have the privilege of
changing their given names.
5. Cremation is the best method of burial.
6 . The Star Spangled Banner is the most stirring in theme
and noble in sentiment of national anthems.
8 . One is never justified in taking another's life, even
when it would be a merciful act.
9. Socially-minded experts, rather than voters, should
decide the polioies of government.
10. Three meals a day will always be the best general rule.
11. Evan in an ideal world there should be protective tariffs.
12. The metric system of weights and measures should be
adopted instead of our present system.
13. Criminals should be treated like sick persons.
15. Armistice Day should be celebrated with less martial
spirit.
16. Any science which conflicts with religious beliefs
should be taught cautiously, if at all, in our schools*
17. Demooracy as practiced in the United States is the best
of all modern governments because it is the most
suited to the needs of modern times.
18. Our oourts should be in the hands of sociologists rather
than lawyers.
19. Trial by Jury has been, and always will be, the most
effective way of securing justice.
20. It is to be hoped that men will improve the comfort of
their dress by abandoning or replacing the present
necktie and collar.
21. People who are religious will be happier in the future
life than will others.
22. It is more important to believe in God than to be
unselfish.
(Turn to next page, please)
BTSTj Two items relating to war and communism were numbered
7 and 14. These items were not scored.
Do not write your name on this paper.
Below are the same 22 statements listed to see what people think
about many questions. This time, please mark them the way you
think most Catholics would mark them.
If you think most Catholics would agree more than they would
disagree with a statement, mark it plus (+* ).
If you think most Catholics would disagree more than they would
agree with a statement, mark it zero (0).
Be sure to place either a plus or a zero to the left of each number.
This is not an examination. People differ in their opinions about
what a Catholic would think is right and wrong in these issues.
Please mark the statements as you sincerely think most Catholics
would mark them. Indicate your own idea of what most Catholics
would sincerely think, by a plus if you think they would agree,
and by a zero if you think they would disagree.
( )
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
(
1. We should oelebrate Pasteur's birthday rather than
Washington's, as he has done the world a greater service.
)
2. Conservative people are usually more intelligent than
radical people.
)
3. Since the theory of evolution has been accepted by most
scientists, it should be taught in our sohools.
)
4. At the age of 21, people should have the privilege of
ohanging their given names.
)
5. Cremation is the best method of burial.
)
6. The Star Spangled Banner is the most stirring in theme
and noble in sentiment of national anthems.
) 8. One is never justified in taking another's life, even
when it would be a merciful act.
) 9. Socially-minded experts, rather than voters, should
decide the policies of government.
) 10. Three meals a day will always be the best general rule.
) 11. Even in an ideal world there should be protective tariffs.
) 12. The metric system of weights and measures should be
adopted instead of our present system.
) 13. Criminals should be treated like sick persons.
) 15. Armistioe Day should be celebrated with less martial spirit
) 16. Any science which oonfliots with religious beliefs should
be taught cautiously, if at all, in our schools.
) 17. Demooraoy as practiced in the United States is the best of
all modern governments because it is the most suited to
the needs of modern times.
) 18. Our courts should be in the hands of sociologists rather
than lawyers.
) 19. Trial by jury has been, and always will be, the most
effective way of securing justice.
) 20. It is to be hoped that men will improve the comfort of
their dress by abandoning or replacing the present
neoktie and collar.
) 21. People who are religious will be happier in the future
life than will others.
) 22. It is more important to believe in God than to be unselfish
N. B.: Two additional pages were in the original questionnaire,
with instructions to mark statements according to respondents'
idea of how (a) Protestants, and (b) Jews would respond, in­
structions were identical to those above except for substitution
of "Protestant" and "Jew" instead of "Catholic".
II
APPENDIX E
Scores of Catholics, Jews, and Protestants on "Attitude
toward Communism" and "Attitude toward War"
Self-Scores and Jew-Scores of 36 Catholics
Subject
Communism________________ War_______________
___________Self-Score
2
3
7
10
11
12
13
15
16
19
21
45
46
47
48
49
50
51
52
53
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
4.55
5.05
4.50
5.05
5.40
4.70
3.00
4.05
4.70
3.80
3.00
2.20
3.00
3.55
3.30
4.70
4.25
3.30
4.05
3.55
4.25
.2.70
3.80
5.05
4.70
5.40
3.80
5.05
4.30
4.50
5.05
5.05
4.05
3.30
2.70
3.00
Jew-Score
6.80
7.80
7.40
7.80
3.80
7.80
8.30
4.05
7.40
5.90
7.20
2.20
8.20
4.30
4.70
3.55
3.00
7.00
8.20
8.20
3.30
8.20
2.70
- - —
- - —
6.80
-------
5.05
3.80
3.00
2.70
-------
4.05
7.20
8.20
8.30
Self-Score
2.80
4.20
3.60
3.60
4.20
5.60
4.20
3.20
3.70
3.20
3.35
2.95
3.50
3.70
5.10
3.60
3.50
3.20
3.50
3.60
3.20
3.20
3.60
3.20
3.50
3.50
3.50
3.45
2.80
4.70
2.40
3.70
2.80
2.40
4.20
2.40
Jew-Soore
3.55
6.65
3.60
6.50
3.35
5.10
3.35
2.80
6.50
3.35
7.90
2.95
4.20
4.70
4.70
2.80
2.10
8.75
2.65
7.90
7.15
7.15
7.90
-------------
3.50
-------
3.60
3.50
3.20
2.80
3.20
6.80
3.35
2.80
APPENDIX E (Continued)
Self-Scores and Catholic-Scores of 64 Jews
Subject
Communism
Self-Score
6
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
63
64
65
66
67
68
69
70
71
72
73
74
75
102
103
104
105
106
107
6»40
3.55
6.40
5.05
5.05
4.25
5.05
3.80
4.70
3.80
5.90
4.05
4.05
5.05
4.70
3.00
5.05
4.25
4.25
4.30
4.70
5.05
5.40
5.40
5.40
7.20
4.70
3.55
3.55
2.70
4.70
4.50
4.50
5.40
4.70
3.30
3.30
4.70
3.30
Cath.-Score
•
m m m
2.70
3.80
3.30
3.30
3.30
------------------
4.05
------------------
3.30
4.30
8.30
4.30
3.30
2.20
3.55
4.70
2.20
------------------
----
3.55
5.40
3.80
5.90
2.45
2.70
2.70
3.00
2.70
----
2.70
3.25
3.30
3.30
3.00
3.30
3.55
2.20
3.30
War
Self-Score Cath.-S<
2.95
3.70
3.35
3.60
3.60
3.20
3.50
3.60
2.80
3.50
2.40
3.50
3.45
3.60
2.40
2.40
2.85
3.35
3.35
2.85
2.95
3.35
3.60
3.20
3.35
3.20
2.95
3.35
3.50
3.35
3.35
3.35
3.20
3.35
3.35
4.70
4.20
3.50
3.35
3.20
3.70
3.60
3.60
3.35
------------------
3.35
w am ^
3.66
3.50
8.50
3.20
4.20
5.75
3.20
2.95
3.50
.
.
.
.
2.40
3.50
4.20
3.20
5.60
6.00
3.60
3.20
3.50
----
3.20
2.80
2.80
3.35
4.20
4.70
2.95
6.50
3.50
\k>
APPENDIX E (Continued)
Self-Scores and Catholic-Scores of 64 Jews
(Continued)
Subject
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
Communism
Self-Score Cath.-Score
4.35
5+05
3.80
5.90
4.25
4.70
3.30
4.30
5.90
7.00
3.80
7.00
4.30
7.80
5.05
4.70
3.80
4.50
4.70
3.80
6.40
5.55
5.05
4.50
5.65
3.00
5*05
2.45
5.90
3.80
3.30
3.30
4.50
2.45
4.30
3.30
2.70
4.05
3.30
5.40
3.30
3.55
4.70
4.05
3.80
5.05
7.80
4.25
3.55
3.25
War
Self-Score Cath.-Score
3.20
2.80
3.50
3.70
3.70
2.80
3.70
3.20
2.80
3.20
3.50
3.50
3.60
3.60
3.60
3.50
3.50
2.10
3.70
3.50
3.35
3.35
4.20
6.80
3.50
3.50
3.20
2.80
3.70
3.35
3.35
3.35
3.70
4.70
2.40
3.50
2.40
3.60
3.70
3.60
3.50
6.50
2.80
4.70
3.60
3.60
4.70
3.50
3.35
3.60
17
APPEKDIX E (Continued)
Self-Scores and Catholic-Scores
of 10 Protestants
Subject
44
54
55
56
57
58
59
60
61
62
Communism_________________________War___________
Self-Score Cath.-Score Self-Score Cath.-Score
------------------
------------------
6.40
4.85
3.00
4.30
4.30
3.80
------------------
3.80
7.40
3.80
5.05
3.55
3.55
. . . . .
2.70
5.40
2.70
3.55
3.50
4.70
3.35
4.70
2.80
4.20
3.50
------------------
3.20
5.50
3.50
3.60
3.60
3.60
------------------
2.40
3.35
6.65
3.50
Self-Scores and Jew^Scores
of 22 Protestants
Communism
Self-Score
X
4
5
8
9
14
17
18
20
22
23
24
92
93
94
95
96
97
98
99
100
101
4.50
4.60
4.70
3.55
5.05
4.30
5.05
5.40
4.35
4.50
3.80
6.80
3.30
4.70
4.25
3.55
3.30
4.25
4.05
4.50
4.30
4.30
war
Jew-Score
Self-Score
Jew-Score
3.60
3.70
2.40
3.60
3.35
3.20
3.60
3.70
3.50
6.65
3.60
2.80
4.20
3.70
5.75
3.50
3.20
3.35
4.70
7.15
----
3.50
3.50
3.70
4.20
3.60
2.80
4.70
3.35
3.50
4.20
3.70
3.50
5.25
3.70
3.60
3.60
3.50
3.60
3.35
3.70
3.60
2.70
3.60
3.20
5.40
5.90
6.80
3.80
5.40
5.90
3.55
4.25
4.25
7.40
2.70
5.40
4.05
3.80
7.00
7.40
8.20
3.80
8.20
2.05
It
APPENDIX F
Self-Scores, Catholic-Scores, Jew-Scores, and ProtestantScores on "Attitude toward Birth-Control"
Scores of 12 Catholics
Subject Self-Score
1
2
3
4
5
6
7
8
9
10
11
12
5..40
2.80
4.95
7.00
5.80
6.60
6.70
7.40
5.80
7.60
2.30
2.80
Cath.-Score
Jew-Score
2.55
3.30
2.80
3.45
3.20
4.50
6. 60
7.40
2.30
5.60
2.30
2.30
Prot.-Score
2.80
6.60
7.95
6.60
6.60
5.80
7.40
7.40
7.80
5.40
8.15
7.60
6.50
5.80
7.50
4.95
4.50
5.80
6.60
7.50
5.80
5.60
7.80
8.15
Jew-Score
Prot.-Score
Scores of 15 Jews
Subject Self-Score
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
6.60
7.80
7.80
7.60
7.60
7.60
8.00
7.50
7.10
7.50
8.00
8.15
7.60
7.50
7.50
Cath.-Score
2.80
3.20
2.30
3.05
2.00
2.55
2.55
3.30
2.55
6.60
6.50
2.80
2.55
5.40
----
2.80
3.60
7.50
7.40
4.50
7.40
7.55
3.45
7.10
6.60
7.40
3.05
3.60
7.00
5.80
3.30
3.20
7.50
7.00
5.40
7.60
6.60
3.05
7.40
7.50
7.50
6.60
4.05
7.40
7.60
19
APPENDIX
F
(Continued)
Scores of 14 Protestants
Subject Self-Score
13
14
15
16
17
18
19
20
21
22
23
24
25
26
5.80
7.50
5.80
6.60
6.20
7.60
7.50
7.50
7.00
6.60
7.50
7.40
4.50
5.60
Cath.-Score
2.80
3.05
3.45
2.55
5.60
3.05
2.80
2.80
5.80
5.80
6.50
2.55
2.30
2.00
Jew-Score
7.50
2.55
6.20
5.80
8.00
7.40
7.60
3.30
5.60
6.20
5.80
3.30
5.60
5.40
Prot.-Score
5.60
7.80
7.10
7.40
8,15
5.60
4.50
4.50
6.70
6.60
7.40
7.40
4.95
5.60
20
APPENDIX G
Self-Scores, Catholic-Scores, Jew-Scores, and. ProtestantScores on Abbreviated C-R Opinionnaire
Scores of 63 Catholics
Subject Self-Score
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
91
92
93
94
95
96
97
16.0
16.0
13.0
7.0
11.0
15.0
15.0
20.0
13.0
13.0
18.0
9.0
11.0
8.0
4.5
8.0
9.0
6.0
15.0
17.0
14.0
7.0
9.0
15.0
9.0
14.0
4.0
7.0
17.0
15.0
7.0
12.0
18.0
12.0
18.0
12.0
14.0
10.0
14.5
7.0
Cath.-Score
Jew-Score
Prot.-Score
12.0
14.0
16.0
12.0
13.0
14.0
13.0
18.0
14.0
13.0
15.0
11.5
11.0
13.0
11.0
9.0
12.0
17.0
17.0
17.0
14.0
10.0
6.0
17.0
13.0
14.0
14.5
10.0
17.0
14.0
12.0
11.5
17.0
18.0
17.0
16.0
13.0
16.5
15.0
15.0
13.0
10.5
12.0
19.0
10.0
12.0
15.0
17.0
15.0
13.0
12.0
8.0
8.0
7.0
9.0
11.0
15.0
9.5
5.0
7.0
14.0
11.0
9.0
10.0
8.0
*1.0
6.0
4.0
17.0
15.0
12.0
10.5
9.0
11.0
11.0
13.0
13.0
8.0
11.0
6.0
13.0
14.0
10.0
13.0
10.0
12.0
15.0
16.0
15.0
13.0
12.0
11.0
13.0
12.0
9.0
11.0
12.0
12.0
15.0
14.0
16.0
9.0
8.0
14.0
16.0
13.0
14.0
9.0
14.0
14.0
10.0
12.0
19.0
12.0
15.0
12.0
9,0
14.5
10.0
19.0
21
APPENDIX G (Continued)
Scores of 63 Catholics
(Continued)
Subject Self-Score
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
17.0
6.0
7.5
14.0
16.0
15.0
16.0
11.0
18.0
10.0
14.0
18.0
14.0
13.0
13.5
15.0
18.5
15.0
10.5
14.0
15.0
8.0
15.0
Cath.-Score
18.0
14.0
14.5
13.0
16.0
15.0
17.0
13.0
18.0
10.0
14.0
18.0
16.0
13.0
13.5
17.0
---IS. 0
--- -
17.0
13.0
----
13.0
Jew-Score
Prot.-^core
14.0
7.0
7.0
11.0
17.0
6.0
17.0
9.0
6.0
7.0
8.0
8.0
9.0
13.0
17.0
5.0
12.0
11.0
14.0
9.0
16.0
8.0
18.0
8.0
14.0
16.0
16.0
13.0
----
------ -
-------------
9.0
----
4.0
11.0
---15.0
-------
11.0
10.0
----
APPEKDIX G (Continued)
Scores of 151 Jews
Subject Self-Score
53
90
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
8.0
12.0
2.0
14.0
8.0
3.0
5.5
12.0
11.0
7.0
14.0
13.0
11.0
13.0
4.5
5.0
5.0
4.0
6.0
10.0
11.0
10.0
13.0
9.0
7.0
7.0
9.0
3.0
5.0
12.0
9.5
7.5
6.0
6.0
11.0
7.0
6.0
9.0
10.0
17.0
16.0
11.0
8.0
1.0
Cath.-Score
---13.0
16.5
11.0
16.0
15.0
8.0
14.5
15.0
12.0
13.0
14.0
12.0
11.0
11.0
15.0
11.5
10.0
18.0
13.0
16.0
12.0
13.0
15.0
11.0
15.0
9.0
18.0
7.0
13.0
16.0
16.5
11.0
15.0
16.0
9.0
12.0
14.0
15.0
9.0
18.0
12.0
11.0
17.0
Jew-Score
---13.0
17.0
15.0
11.0
10.0
4.0
12.0
14.0
7.0
13.0
14.0
14.0
13.0
11.0
9.0
12.0
9.0
14.0
11.0
10.0
13.0
13.0
15.0
11.0
6.0
9.0
11.0
5.0
12.0
11.5
10.5
5.0
10.0
12.0
9.0
11.0
14.0
15.0
12.0
16.0
12.0
7.0
3.0
Prot.-Score
---13.0
16.0
12.0
11.0
9.0
7.0
14.0
16.0
8.0
13.0
12.0
14.0
11.0
10.0
17.0
11.0
9.0
18.0
13.0
14.0
10.0
13.0
15.0
11.0
10.0
7.0
15.0
7.0
12.0
12.0
10.5
7.0
9.0
15.0
9.0
15.0
15.0
14.0
9.0
16.0
13.0
8.0
14.0
13
APPENDIX
G (Continued)
Scores of 151 Jews
(Continued)
Subject Self-Score
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
5.0
5.0
5.0
14.0
10.5
16.0
15.0
9.0
14.0
5.0
11.0
15.0
15.0
14.0
16.0
9.0
12.0
11.0
13.0
6.0
11.0
12.0
14.0
7.5
4.5
13.0
9.0
6.0
8.0
10.0
9.5
5.0
14.0
12.0
6.0
10.0
9.0
10.0
7.0
9.0
8.0
4.0
6.0
12.0
Cath.-Score
8.0
10.0
9.0
13.5
14.0
15.5
15.0
14.0
14.0
7.0
17.5
18.0
14.0
15.0
6.0
12.0
17.0
18.0
13.0
10.5
12.0
16.0
15.0
13.0
12.5
15.0
14.0
14.0
9.0
10.5
9.5
11.0
10.0
12.0
12.0
13.0
9.0
13.0
6.0
13.0
13.0
16.0
11.0
12.0
Jew-Score
8.0
6.0
9.0
15.0
12.0
17.0
17.0
10.0
14.0
5.0
16.0
15.0
15.0
14.5
16.0
15.0
17.0
16.0
13.0
11.0
13.0
13.0
15.0
8.0
8.0
12.0
12.0
6.0
8.0
10.0
9.5
6.0
14.0
12.0
10.0
13.0
9.0
13.0
7.0
15.0
12.5
14.0
12.0
14.0
Prot.-t>core
11.0
6.0
7.0
14.0
10.0
14.0
13.0
11.0
11.0
5.0
17.0
15.0
15.0
13.0
10.0
12.0
16.0
18.0
13.0
9.5
14.0
16.0
13.0
12.0
9.5
11.0
9.0
12.0
10.0
10.0
9.5
6.0
11.0
12.0
10.0
10.0
9.0
13.0
9.0
13.0
13.0
12.0
11.0
15.0
ACTEKPIX G (Continued)
Scores of 151 Jews
(Continued)
Subject Self-Score
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
9.0
10.0
12.0
11.0
12.0
10.0
12.0
10.0
8.0
7.0
9.0
12.0
12.0
13.0
13.0
8.0
11.0
13.0
8.0
16.0
3.0
11.0
9.0
6.0
8.0
9.0
11.0
8.0
10.0
17.0
5.0
7.0
16.0
9.0
5.0
7.0
5.0
10.5
14.0
5.0
9.0
13.5
Cath.-Score
16.0
13.0
9.0
14.0
16.0
15.0
14.0
14.0
15.0
12.0
14.0
12.0
16.0
12.0
14.0
7.0
14.0
15.0
18.0
14.0
13.0
10.0
14.0
14.0
10.0
15.0
13.0
15.0
14.0
17.0
14.0
12.0
18.0
10.0
9.0
13.0
14.0
---14.0
---11.0
----
Jew-Score
8.0
10.0
14.0
11.0
11.5
11.0
15.0
12.0
13.0
9.0
15.0
12.0
14.0
15.0
15.0
12.0
13.0
16.0
17.0
16.0
9.0
11.0
12.0
13.0
13.0
9.0
11.0
11.0
10.0
17.0
15.0
11.0
16.0
10.0
9.0
9.0
10.0
---14. 0
------17.5
Prot.-Score
8.0
7.0
9.0
12.0
10.0
14.0
18.0
12.0
8.0
11.0
12.0
12.0
16.0
7.0
16.0
14.0
11.0
14.0
16.0
17.0
10.0
10.0
11.0
11.0
12.0
14.0
16.0
13.0
11.0
17.0
16.5
6.0
16.0
11.0
9.0
13.0
14.5
12.0
17.0
---9.0
----
A*
APPENDIX G (Continued.)
Scores of 151 Jews
(Continued)
Self-Score
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
4.5
2.0
Cath.-Score
Jew-Score
5.5
---7.0
13.0
13.0
13.0
----
-------
10.0
12^0
12.5
12.0
16.0
14.5--------- ---5.0
14.0
7.5
---8.5
7.0
---11.0
9.0
10.0
11.5
9.0
14.0
14.5
----
Prot.-Score
----
------9.0
---12.0
---------8.0
----
12.0
---
13.5
13.0
15.5
9.0
------9.0
14.5
-------—
-------
---9.0
---13.5
13.0
u
APPENDIX G (Continued)
Scores of 55 Protestants
Subject Self-Score
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
11.0
11.0
6.0
6.0
5.5
11.0
9.0
10.0
10.5
9.0
11.0
13.5
10.0
11.0
11.0
5.0
5.0
13.0
17.0
10.0
13.0
16.0
13.0
14.5
16.0
12.0
14.0
8.0
11.0
9.0
5.0
11.0
9.0
6.0
12.0
12.5
10.0
8.0
12.0
8.0
7.5
8.0
6.0
Cath*“Score
12.0
11.0
15.0
13.0
16.0
9.0
16.5
16.0
14.0
10.0
11.0
15.0
15.0
14.0
12.0
13.0
11.0
11.0
17.0
15.0
18.0
14.0
10.0
15.0
17.0
14.0
13.0
14.0
10.0
17.0
6.0
14.0
13.5
6.0
15.0
13.0
8.0
13.0
13.0
19.0
14.0
13.0
14.0
Jew-Score
11.0
10.0
14.0
12.5
5.0
8.0
10.5
15.0
8.0
10.0
12.0
14.0
12. 0
11.0
14.0
8.0
4.0
9.0
11.0
8.0
4.0
11.0
5.0
10.0
12.0
10.0
6.0
10.5
6.0
16.0
4.0
11.0
12.0
6.0
13.0
13.0
11.0
9.0
13.0
8.0
10.0
6.0
15.0
Prot.-Score
12.0
9.0
8.0
10.0
16.0
11.0
13.5
11.0
10.5
10.0
11.0
12.5
11.0
11.0
11.0
13.0
8.0
12.0
16.0
14.0
15.0
14.5
15.0
14.5
17.0
11.0
12.0
8.0
15.0
13.0
6.0
16.0
10.0
5.5
12.0
12.0
13.0
7.0
13.0
15.0
12.0
10.0
12.0
27
APPENDIX G (Continued)
Scores of 55 Protestants
(Continued)
Subject Self-Score
44
45
46
47
48
49
50
51
52
54
55
56
8.0
11.0
9.0
13.0
5.5
8.0
7.0
11.0
10.0
18.0
15.0
10.0
Cath.-Score
14.0
17.0
13.0
16.0
11.0
9.0
12.0
14.0
13.0
18.0
-------
Jew-Score
13.0
7.0
15.0
15.0
6.0
10.0
5.0
14.0
10.0
-------
Prot.-Score
9.0
14.0
14.0
13.0
6.0
10.0
7.5
14.0
-------------
APPENDIX H — DISTRIBUTIONS
Score Distributions of Catholics, Jews, and Protestants
on "Attitude toward Communism" and "Attitude toward mi-ar"
Jewish Subjects
Score
Intervals
8 *4-8*5
8.2-8.3
8.0-8.1
7.8-7.9
7.6-7.7
7.4-7.5
7.2-7.3
7.0-7.1
6.8-6.9
6.6-6.7
6.4-6.5
6.2-6.3
6.0-6.1
5.8-5.9
5.6-5.7
5.4-5.5
5.2-5.3
5.0-5.1
4.8-4.9
4.6-4.7
4.4-4.5
4.2-4.3
4.0-4.1
3.8-3.9
3.6—3.7
3.4—3.5
3.2—3.3
3.0-3.1
2.8-2.9
2.6-2.7
2.4-2.5
2.2-2.3
2.0-2.1
If
Communism
War
Self-Score Cath.-Score Self-Score Cath.-Score
1
1
1
1
1
2
1
2
3
1
3
1
5
2
9
2
10
4
8
2
6
2
1
4
3
4
3
4
1
5
16
3
1
6
3
3
64
2
2
58
1
4
2
3
13
13
21
13
9
14
9
6
3
3
1
64
58
*7
APPENDIX H--DISTRIBUTIONS (Continued)
Catholic Subjects
Score
Intervals
8.6-8.7
8.4 —8.5
8.2-8.3
8.0-8.1
7.8-7.9
7.6-7.7
7.4-7.5
7.2-7.3
7.0-7.1
6.8-6.9
6.6-6.7
6.4—6.5
6.2-6.3
6.0-6.1
5.8-5.9
5.6-5.7
5.4-5.5
5.2-5.3
5.0-5.1
4.8-4.9
4.6-4.7
4.4-4.5
4.2—4.3
4.0-4.1
3.8-3.9
3.6-3.7
3.4-3.5
3.2—3.3
3.0-3.1
2.8-2.9
2.6-2.7
2.4-2.5
2.2-2.3
2.0—2.1
N
Communism
Self-Score
Jew-Score
War
Self-Score Jew-ocore
1
7
3
3
2
2
1
2
2
1
1
2
1
1
2
6
1
1
1
4
3
3
3
3
1
1
2
1
2
2
4
1
8
7
7
2
2
7
4
5
1
2
3
4
1
1
2
2
2
1
1
3
36
32
36
1
32
30
APPENDIX H — -DISTRIBUTIONS (Continued )
Protestant Subjects
Communism
Score
Self- Cath.- JewIntervals Score Score Score
8*2-8.3
8.0-8.1
7.8-7.9
7.6-7.7
7.4-7.5
7.2-7.3
7.0-7.1
6.8-6.9
6.6-6.7
6•4—6.5
6.2—6.3
6.0-6.1
5.8-5.9
5.6-5.7
5.4—5.5
5.2—5.3
5.0-5.1
4.8-4.9
4.6-4.7
4.4-4.5
4.2-4.3
4.0-4.1
3.8-3.9
3.6-3.7
3.4-3.5
3.2-3.3
3.0-3.1
2.8-2.9
2.6-2.7
2.4-2.5
2.2-2.3
N
War
Cath.
- JewSelfScore Score Score
1
1
1
1
1
1
3
1
3
3
6
1
3
4
2
1
1
1
1
1
2
2
1
1
3
3
5
1
1
2
2
2
1
1
2
1
2
13
8
3
2
1
9
5
2
1
1
1
1
2
1
30
8
21
31
8
21
31
APPENDIX H— DISTRIBUTIONS (Continued)
Attitude toward Birth Control
Catholic Subjects
Score
Intervals
8.0-8*3
7.6-7.9
7.2-7.5
6.8-7.1
6.4-6.7
6.0-6.3
5.6-5.9
5.2-5.5
4.8-5.1
4.4-4.7
4.0-4.3
3.6-3.9
3.2-3.5
2.8-3.1
2.4-2.7
2.0-2.3
N
SelfScores
1
1
1
2
2
1
1
Cath.Scores
Jew- ^
Scores
Prot.Scores
1
1
3
2
1
1
2
1
3
2
1
1
1
4
1
1
1
2
1
12
3
1
1
3
12
1
12
12
Attitude toward Birth Control
Jewish Subjects
Score
Intervals
8.0-8.3
7.6-7.9
7.2-7.5
6.8-7.1
6.4-6.7
6.0-6.3
5.6-5.9
5.2-5.5
4.8-5.1
4.4-4.7
4.0-4.3
3.6-3.9
3.2-3.5
2.8-3.1
2.4-2.7
2.0-2.3
N
SelfScores
3
6
4
1
1
Cath.Scores
2
JewScores
5
2
1
Prot.Scores
2
4
2
1
1
1
1
1
15
2
3
5
2
15
2
1
2
1
2
2
15
15
3X
APPENDIX H--DISTRIBUTIONS (Continued)
Attitude toward Birth-Control
Protestant Subjects
Score
Intervals
8.0-8.3
7.6-7.9
7.2-7.5
6.8-7.1
6.4-6.7
6.0-6.3
5.6-5.9
5.2-5.5
4.8-5.1
4.4-4.7
4.0-4.3
3.6-3.9
3.2-3.5
2.8-3.1
2.4-2.7
2.0-2.3
N
SelfScores
1
5
1
2
1
3
Cath.Scores
JewScores
1
1
2
1
3
2
4
1
1
1
3
1
2
3
1
2
1
14
Prot.Scores
1
5
2
2
14
2
1
14
14
33
APPENDIX H — DISTRIBUTIONS (Continued)
C-R Opinionaire Scores
Catholic Subjects
Score
Intervals
20
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
N
SelfSeore
Cath.Score
JewScore
J^rot. Score
1
1
6
3
4
10
8
5
3
3
3
4
3
6
2
2
63
5
10
5
5
10
12
4
4
3
1
1
60
4
4
2
5
4
8
4
7
6
5
4
1
2
57
2
1
1
6
5'
9
6
9
4
5
5
3
1
57
APPENDIX H — DISTRIBUTIONS (Continued)
C-R Opinionaire Scores
Jewish Subjects
Score
Intervals
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
3
2
1
SelfScore
2
5
4
10
11
16
14
12
19
11
12
9
14
5
3
2
2
151
Gath.Score
7
4
12
17
24
16
15
10
10
9
2
3
2
131
JewScore
7
7
14
13
19
16
15
11
14
5
3
4
4
1
1
134
Prot.ocore
3
5
11
8
14
14
16
16
13
15
5
8
3
1
132
3 5
APPENDIX H--DISTRIBUTIONS (Continued)
C-R Opinionaire Scores
Protestant Subjects
Score
Intervals
19
18
17
16
15
14
13
12
11
10
9
8
7
6
5
4
N
SelfScore
1
1
2
1
2
5
4
10
7
5
6
2
4
5
55
Cath.Score
1
2
4
4
6
10
10
3
5
3
2
1
2
53
JewScore
1
4
4
4
5
6
9
2
5
1
5
3
3
52
Prot.Score
1
3
4
6
6
8
7
6
2
3
2
2
1
51
C o m b in ed
( H * 267 j
3f*
VO
oo co com in h
(M
in in o co cm to i m co cm co h co
O W H H t f O I t o CM H rH t o i n
+ + 1 + + I I 4* W + + + +
in
CM
in CM I O O O C M t O t O O C " 0 > G O r 4
to to I W to r t r l
to lO
i n to
+ + + + + ++ +
4 - 4 - 1 4 *+............................................
to CM CM00 00
? H CM to ^
to
in ^cM co oM oaio coo iO co
C M H C M O O t O t O C M C M W i —I W O
+ + + +
+ + + + + + + +
tO O to -M*
M M H IO
+ +
++
CM
G ro u p s
CM
CM
CM
o>
A ll
H
C o r r e la tio n
A p p e n d ix
I
C o e ffic ie n ts *
CO
Q>
5? ¥ I I + +
r - COCMCM’M ' O C ^ O O C M
O H tO O H tO tO H tO O
> m H
I tO CM
•I +
I
I 4* 4*
t
I
I
t
I
to O to
to CM O
+ + +
H 00
O to
+ I
I
o>
tor-cocM HinoiococoinHco
• O H H C M CM CM H O O N H O
4- +
1 1 4*
I
4* +
tO CM "M*CO i-4 r4 l
'J* CM O O CM tO I
4- +
4- 1 + 4* I
1
t
I
4* +
CMto O to
*• + I +
+ + + +
CM CM in CO H CO
M H O N H O
+ + If + +
I CM CM CM
I H
+
CM O
+ +
to
+
©
r)
O
(4
£
+
I +
+
+
1 +
I i n CO I to
I O CM • CM
o
I—I
+
l l + l l
+ +
irnn o 'f m o io o i o
C Mt n c MO' M' CMi n c M I H
i o
I o
T e tr a c h o r ic
o>
C MrHCM OCM'tfCM'^rH
+ + + + + + + + +
+ I+ + + + +
a> O 00 CO CM CM o
to 'M1 O CM I p H
+ + + + + + ¥
o to to to o
to <p> H O CM
+ I+
I
O CM to CO 00
CM
p CM
+ + + +
In t e r - " M e n a c e *
+ + i+
cm H in
+
to
CM
I
G a>
0)
o
3 rt
A
p ©
+ + + + •+ + + +
OOtDCMCOOOOtO
CO
in
+
•
a> P
o o
(4
Pi 1
(0
o cm in
+ +
00 O 00 o
to to
CM
o m o o w tO M 'o o
n w c M to w o H f-
3
bD
O -H
o to
•M* a> m to
r- to o to
CM
in CM CM CM • o I to to ^
• CM I i n to qm
+ I+ I+ I+ + +
0)
¥
+ + + I+ + I + I + I
to CO CM CMo o> t
to iH H CMto H I
rH
1 +
o
•rt
+>
a)
i—
I
a) •
p G
V o
> (4
u ©
0)
n t5
G©
o E
S
o o
1
—
1
OH
+> o
<H
P
©m
XJ P
p G
o •H
O
g Pi
•H
pH
•G c4
a) a
p •H
p O
•H ©
a xj
o
tH
© •H
P
0) ©
* ©
tO
i
—I
+
G
p
¥
+ +
tOtOOCM OOQOtO'+C-OCOOO
( O W O H M O H M O H t f O N
09
to
a>
•rl
P
P
G
(0
i+
in t>
*t ¥
co
+
HC M tO '^’ m t- C O O O r H C M t O 'M 'in C 'C n H C M t O in
r4t-4rH«HrHrHHiHCMCMCMCM
m tJ
g ©
•H P
O
Pi ©
l—
1
© x*
a r-i
•H P
Oo
a
w
&
A
to
37
CM
CM O
rM
CM O
a>
£
0
•H
o>
H O
OOOOCMa>0»QOCMtOC^OC“ COtOinoCM
PM
O
CO
COOO>OCMOC'*iOCMiOOCMcOGOC>rWOtOCMintOtOtOOOtOtOOtOCMCM
+
+ + + 1 1 + 1 + + + I + I 1
1
o
d
a>
m
to
p
at
•rl
t>
a)
rt
i
pm
E-»
o
o
rt
0
at
Ph
p
v
Em
a
<u
OCMCOtOCOHWC'-OmcOOrM
CMOOOCMtOOCMOCMC-tOtp
+ 1 1 + 1
1
+
H
o
u
in
rt
H«
Pm
o
•H
O O lO M D rl'O M O lO aJO iO C O
OJQJtOOtOQCMCMtOCM^'CM'sfO
+ + 1 I 1 + + I + 1 I I + I
rt
o
•H
p
at
a)
to
rM O
ik
H *
a
M
M’ C D N O O O O i W C ' t O N O ^ W i O
OHCMCMCM'sfO'ti} H O r - O H C - H
1 + + 1 1 + 1 1 + + 1
1 1 1
h*
4i
co
rt
rt
rH
HO
O
HO
M
o>r t
CO O
tO O
-p
M
1
Pi
CD
-P
c
M
Ph
S
o
Pm
C!>
o
I—i
to
«
+
++
+ +
CMOCMCMHOtOOOlOinin
^H^OHOtOM'MHpW
1 + 1 1 +
rt
p
at
o
+ 1 + 1 + 1
ooCMoinoomotoco
WHHOWrlCMIOHHlO
•11+1+111+1 +
Hfinooooiohooo
OtONOtOtOrMH + O
1 1 + + 1 1+ 1
intocotoooinm o o
mwHtootoKinm
1 + 1 1 1+ + »+
cocooe-ococoH
«+OHtftOCOCMCM''3<
1+ 1 + 1 + + 1
CMCM O i n O O to
•IO t o O CM CM to
+ 1+1+1 1
C
MO CMt o O O
Hj'M'M'OM'N
1+111 +
H CMCM^ in
nootfo
11111
o totoin
O CM to
^ CM rM
^ rt +
1+
COo
<+ CM
tort + 1
in
H
u
CM
+ 1
'r t
o
a
a
a>
p
M
d
rM
£
o
rt
ID
C
D
C
D
•rM
cH U
h
u
51
+
O _
CM r t
rt
g
+
ocMtocMtnco'M, r - o o m c M C M r - o t o a > o
£
•H
w
1 MOO>000>COOO«i,rM|flCMW inQ!QOO
I H W C M t-W ptO H S loS tO C M 'd'litflW H
1 + 1
1 1 + + 1
+ 1 + 1 1
+ 1 + 1
u
•H
at
Ml
co to H totno to o in o oc'-tn o cM 'tftoc-ocO t O t O ^ t O C - H t O e i t O i n o O t O C M t p ’^ ' r M i n M '
+ + i i i + + i + + i t » i + + i + i +
o«rtrtoortoortrt$rt o o rto rto
rt c M t o # n t n a 3 0 5 0 H C M t o ^ t n t o t ^ r o t t O r H
H
u
-p
£
rt
£
o
•H
-p
a)
a> •
PM £
Ph tsD
o •H
o CD
rM
rM
rM
1
P
m
*
o
+
a>
a rt
at o
rt at
O a>
a) d
> a)
P
m
£
a> o
09 rM
£ rM
•
(L>
rM
at
o
C
D
o o
C
D
d
O P
p
p
•rH
p
p
Ot«i
£
•H
o
a) rt
p
p l“t
•H <rt
a
a
o •rM
o
(D o
d
pH d
at
at
C
D
p tp
£
£
at
Pm
t*
C
D
rt
p
a
o
Ph
<H
-H
C
D
•H
rt at
at
£
Tt
at a)
a at
CD
o
C
O
rM
•rM
o
Ph
a>
a
p
•H
A rt
C
D
•rM
—
EM
n
rt
•
•
rt
i*
CM
CM O
rH
CM O
a>
0
•rl
ca
G
h
■H
£
1
o
co
H
a
Xt
a>
p
at
•rt
l>
©
G
+>
H
G
a
toH ocM HC -cM ooocM coointnom ^'
H H H Q H tO O M 'lO O lM 'lO tO N tO K lM 1
I + + I 1 + + I + +
t 1 t l + + i
Pi
w a o o t Q r i o t o m o r i o i n a o o o
M 'H H H C M H H iO O C M lO lO W K lM 'W
+ I +
I + 1 1 + I 1
+ + + + I I
SI
P
a)
H
i
—I
0
G
G
o
o
o
p
a>
H
a
a>
p
H
1
o
•H
p
at
H Q
to
H O
p t o t o H oo t o cm i i n m p cm p t o
tO tO tO CM CM 'J* CM t p t o i n CM CM P
1 + 1 I 1 + + I + + 1 1 + I
to
H
Pi
pcMiniotoincototoooc-oop
^•tOtOQCMCMPtOtOtOCMCMin
+ 1 + + + 1 + + 1 I + + +
o •H
o 02
rH 1
I
i n cm to p m o o i c m w p h
PpCMW CMOH I p H H H
+ + + 1 1
1 1 + 1 1 +
to
H PCi
m inooin^pcBCM M toto
W P P P H C M rl'tO tflO
+ + )+ + 1 + + 1 1 +
CM _
H PCi
pm intocM C -cM topto
tOtOPtOtO^rHCOCMCO
+ 1 + + + 1 + + 1 1
H
H O
P CM O O CM P to 1 CM
tO p tO rlp tfO 1 P
1 + 1 + + + 1 1 1
O
rH O
CM m cn CM
to O CM
HPO tfCM CM OrH
1 1 1 1 1 + 1 1
M
ai •
G G
G tiO
G
o
•+
a>
O .G
C
Da
P t aJ
a 0
0 xt
>
oo a
to pci
G
p
o
0 0
G xt
ctl
Pt at
rH Xl
C
ri at
0
•rH G
in oo
&
o
G
P
WPS
o
0 0
a +>
?•?
in
?
M U
.G
M
•
02
•rt
£
£
pq
(XiOpcipcittOOpgOUpcilKil « O U 0* O « O
1
at
0
p <H
G •H
•rt
o 07
in cm p
H P P
+ »
pci
G
G
•rl
X} o
0 Pt
p
p 1
—
1
•H at
a fi
o •rt
a)
p
rH CM tO*«#intOCOPOHCMtO'l,inPC'COO>OH
h h h h h h h h h
a>
—i
15 i
o
a)
02 rH
o
—1 to
G I
o o
o «H a)
xt
op
G
0
M t f HCMH
1 + 1 1 1
to o
0
G
CM 94 I P 00 P o
rH 1 rH to O H
+ 1 1 1 +
1
to oo p h oo in
P H O H O p
I t
11 +
CM P P CM O
Oi Pci
at
G
G
IB
G
A0
0
+30
-10
+05
+23
P
•H
G
p
rH
Items
«
H
i
0
G
O
02
CMCMCMCMCMt OCMCOCMOO' f l OCOCO
CMtOtOHCM'tfptOpCMtOCMtOtOCM
I + I I I + + 1 + + 1 1 I I
+
14
$i
Hi
o>
o
G
* *
G
IflOOtOrlCOOM'rlOCOiO'J'CylM'COBCMiq
rH
a>
•H
O
CM «
at
xi
in co o t-'^to oo cM o r-cocM in cM o o io
tO p H H P tO H H C M H tO P O H P H H C M
+ + + + 1 1 1 + 1 1 + + 14- 1 1 + 1
G
•H
m o o cm o oo m i p t n p i n o H c o c M t o p c M o o
CM in H1 O r t N H I C M O ^ C M t O ' M ' C M r t ' M ' H O W
1 +
1 1 1+
1 1 + + 1 1 1 1 + + 1 + 1 +
h cm cm
•
p
•rt
P
P
at
G
at
m
0)
P
a
0
G
01
at
pf
a
0)
p
•rl
ID
•H
£
*
♦
3f
1 cooo'M'toointooooNtflincoioMOto
OI
NO
^ O
t O O N O O i O N ' t H O ' d ' O 1 C- lO N 0» C- O
n t o ^ in 10 o n n to w n H 1 h w h o o o
IP l i I P P P 1 1 P P 1+ + + + 1
IQ
a)
•H
G
N Pi
tQ10 (Ji O O O IO 1 OOlQO^’t f Oi f l i OON
N O O O N O O 1N H -^O H O H O ^O
P p P
1
P l I l P l P
P P P I
o»
HO
oi - pcoc~oi noONNNt oi >oi ot ooi o
N W H O O H O 'fO H 'f^ O N W H O
I P P I
P l l P P I H f r P l P I I
©
G
•H
rt
G
0
•H
G
•H
£
Pi
o
00
rHPS
©
CHO
CO<ONC*H<#iONonoOO»lQOO
NHNON^HCMHNOO'-JWCM
I I P I l P l l l l l l P l P
IO
HO
" P O O O t O NNO O Oi O OO NO
HtOtOHCMCMCMHOOtOtOHH
I P I l l P P l P l l P l I
I
a
G
o
M
M
s
3
(QnocooxsoiococootQ^ionto
1
•P
rt
•H
t>
a>
G
fit
Ml
C
O
I N C -O O N H H O H tO H ^H O O H H H
t P l l l P p p i l l l P I P I I l i P
•H
•P
rt
rl
©
G
G
O
O
G
O
.G
0
rt
G
P>
a>
EH
a
a
lO
to
H Pi
Eh C
M_
rHPS
H
HO
M O
HO
1 O 'sf N N m «£> I N to
1 to
C
Mto C
MO 1 O O
1 1P P P 1 1 1 1P
WOtOWtOt-N^O
HHOQtOHO -^O
1P P P 1 P 1 •
1 (J> C
O•t* IP'd* O KJ
1N N O 0 N H O
1 1 p 1 I 1P 1
<
J>PS
C
OCJ
10 N
e- 0 N
OH C
Min to N
p p 1 1 1p
10 O
h rt w 0 to
O O to H to
p 1 1 1 1
10 P
S
in n in co
H H H to
p 1p p
©
■P
G
OlOOCJii OC' OOOt'O
O r l W O O N WHOrllQ
1 1p p 1 1 p p 1 IP
co m 0 0 0 to 0
©
1
G
I O nO HW O iO H^N lO O
i i i P P i p i p
OOlCOlOOWM'C-WlOMn
lO W O lO O H M ntfN H p
1p p 1
p p 1 1 1p p
w
P>
M
1 > ^ o n n ^ w o )io o tO N
1 1 P
to tOC
M
Pi
G
o
G
O
•<*PS
? ? ?
was
1N
1 'Si'
1 1
■P
G
a)
-P
(0
0)
p
o
111
C
O
NO
ID
a
a>
p
M
M<
1
G
P
©
•
G G
O bO
•H •rl
-P to
rt
1—I 1
©
G G
G O
O
O P
H A'
1—1 O
<! rt
©
• 'rt
d) ©
0
£
rt 0
PiH
m H
O
d> <G •
©
t*
G -P H
d) G rt
to •ri 0
G O ©
O Pi
©
O
1—1 ■
rt
O rt G
P> a pi
•rl •rl
'rt O P»
rt © •P
■p ■rt rt
■p
•H rt G
a
.rt
B
•H
rt
©
G to .G
rt rt -P
to t J a
-p rt 0
G © G
•H G ti-i
O
Pi © to
X> rt
1—1
£
rt *rt
ah a
•H P ©
t> 0 -p
,G •H
A
*#
*
m
•
Jz.
to
to
•H
£
*
*
♦
PSOPStfPtfOOpSOOpSPS* ps 0 0 PS0 PS0
HWl O^l OI OCOCi OHNt O' f i ni OMDCi OH
HHHHHHHHHHNN
L ibrA*T
n . y.
U httJ
Документ
Категория
Без категории
Просмотров
0
Размер файла
8 285 Кб
Теги
sdewsdweddes
1/--страниц
Пожаловаться на содержимое документа