|
The Newspaper and Its Public
A Standardized Test
to Measure the Public's Attitude
Toward a Newspaper
JAMES E. BRINTON /CHILTON R. BUSH
Stanford University
THOMAS M. NEWELL
Alfred Politz Research, Inc.
Institute for Communication Research
Department of Communication and Journalism
Stanford University
Price: $3.75
Publication and distribution of
this study has been made possible
through the generosity of
MERRITT C. SPEIDEL
to whom seven communities owe
thanks for their civic-oriented
newspapers.
-XI5
PREFACE
This study is presented in two parts. The first part reports the
findings and is written in nontechnical language for readers who have
not had training in statistical research.
The second part, a technical appendix, is directed to those research-
trained readers who may wish to know more about the method and who may be
asked to interpret the findings for the first kind of reader. Readers
who are not interested in the method may skip the second part.
The study was financed primarily from the budget of the Institute
for Communication Research, which, in part, is supported by individual
gifts. One gift that we wish to acknowledge especially is from the
estate of Miles W. Kresge, Jr.
Although most of the work was done by the authors, we wish to
acknowledge the assistance of Jack Lyle, who did the statistical work
for Studies 6 and 7. Merrill B. Samuelson supervised the administration
of the test in the seventh community. Dan B. Clark II & Associates
administered the test in three communities.
The person to whom we feel most indebted is Wilbur Schramm who was
partly a collaborator and partly an author. His contribution can be
best described as recovering the fumble on the ten-yard line and carry-
ing the ball into the end zone.
J.E.B.
C.R.B.
T.M.N.
CONTENTS
Page
Preface . . . . . . . . . . . . v
I. The Test . . . . . . . . . . . 1
II. Why This Test Was Developed . . . . . . . 6
III. How This Test Was Developed . . . . . . . 8
IV. How the Test Is Used . . . . . . . . 23
V. Two General Findings from the Test -- Intensity of
Attitudes toward the Press, and Factors
in Attitudes toward the Press . . . 28
VI. Readers Evaluate Their Newspaper -- Results of the Test
in One Community -- Appendix to Chapter VI 34
VII. Other Findings from the Test -- Results in Different
Communities, Results When There are
Competing Newspapers, Attitudes of Women,
Attitudes of Newspaper BEmployees . . . 74
VIII. Toward an Improved Understanding and Opinion of the
Press . . . . . . . . . 85
TECHNICAL APPENDIX
A. Dimensionality and Reliability of the Attitude Scales 89
B. How the Intensity Plots Were Made . . . . . 104
C. The Factor Analysis . . . . . . . . . 107
D. The Involvement Scale . . . . . . . . 112
E. The "Liberalism/Conservatism" Scale . . . . . 115
F. The Sample . . . . . . . . . . . 116
G. Questions Tested and Eliminated . . . . . . 118
H. Bibliography . . . . . . . . . . 126
LIST OF FIGURES
Figure Number Page
1 Political and Economic Fairness . . . . 29
2 Confidence in Political Leadership . . . 29
3 General Satisfaction . . . . . . . 37
4 Racial and Religious Fairness . . . . . 42
5 Adequacy of News Content . . . . . . 42
6 Check upon Government . . . . . . . 47
7 Morally Objectionable Content . . . . . 47
8 Responsibility for Advertising Content . . 51
9 Authoritarianism . . . . . . . . 51
10 Representativeness . . . . . . . 54
11 Independence from Pressure . . . . . 54
12 Responsibility for Accuracy . . . . . 59
13 Human Dignity and Worth . . . . . . 59
LIST OF TABLES
Table Number
1. General Satisfaction, and Religious Faith
or Preference . . . . . . . .
2. Adequacy of News Content, and Evaluation of
Newspapers in General . . . . . .
3. Adequacy of News Content, and Time Spent
Reading . . . . . . . . . .
4. Responsibility for Advertising Content, and
Evaluation of Newspapers in General . . .
5. Authoritarianism, and Political Party
Affiliation . . . . . . . . .
6. Authoritarianism, and Number of Years a Reader.
7. Independence from Pressure, and Membership
in a Labor Union . . . . . . .
8. Independence from Pressure, and Income . .
9. Responsibility for Accuracy, and Political
Party Affiliation . . . . . . .
10. Human Worth and Dignity, and Age . . . .
11. Political and Economic Fairness, and Political
Party Affiliation . . . . . . .
12. Political and Economic Fairness, and Labor
Union Membership . . . . . . .
13. Political and Economic Fairness, and Length
of Residence in Fifth Study . . . . .
14. Confidence in Political Leadership, and
Political Party Affiliation . . . . .
15. Confidence in Political Leadership, and Opinion
About Whether Newspapers Should Offer Advice.
16. Comparative Scores for Two Newspapers . . .
Page
17. Readers' Evaluation of Competing Newspapers . 78
18. Combined Favorable and Very Favorable Attitudes
of Men and Women in Study 4 . . . . 80
19. Attitudes of Editorial Employees in Study 5 . 81
20. Attitudes of Editorial, Advertising, and
Circulation Employees, and Predictions by Key
Editorial Personnel in Study 4 . . . 83
A. Arrangement of Data for Scale Analysis . . 90
B. Individuals Ranked by New Scores . . . 92
C. Final Arrangement of Data in Scale Pattern . 93
D. Coefficients of Reproducibility and Scalability 98
E. Reproducibility, Scalability, and Reliability
of Scales . . . . . . . . 103
P. Arithmetical Basis of an Intensity Plot . . 106
G. Factors of the Test . . . . . . . 107
H. Original Correlation Matrix . . . . . 109
I. The Factors in Study 6 . . . . . . 111
J. General Satisfaction and Involvement . . 114
THE TEST
Copyright 1957 by Chilton R. Bush
1. How accurate is the
in its local news stories?
2. In your experience, do
headlines in the
give you an accurate idea
of what really happened?
3. If you heard a news item.
over the radio and then read
a conflicting version of the
same story in the _,
which one would you believe?
4. How often does the __
in its news columns and head-
lines, try to make a hapoen-
ing sound more exciting than
it really is?
., Do you think the
usually gives as much space
to speeches by a candidate
the paper is against as it
does to speeches by a candi-
date it is for?
6. Does the usually
present both sides of
important political issues?
7. If the was
against a man who was run-
ning for public office,
would it be fair to him or
not?
8. Which, if either, has the
better chance of getting
his side of a story about
a local strike into the
: the head of
the business or a labor
union official?
9. If a Negro got in a seri-
ous fight with a white man
in this area, how fair
would the be
toward the Negro?
10. Does the print
both sides of issues that
involve different races?
11. Does the seem
fair to all religious
groups?
12. Do you think the
really cares about the
poor people in this town?
13. Does a wealthy man get
better treatment in the
than a poor
man?
14. How true is this state-
ment: The names of some
local people are in the
very often.
while interesting news
about many other local
people hardly ever gets
in the paper.
15. If a good friend of the
owners of the
got arrested for drunken
driving, would the
print the
story?
16. Would it be easier for a
personal friend of the
owners of the
to get a story in the
paper than for a person
who didn't know the
owners at all?
17. Does the amount of money
a man has make a differ-
ence in the treatment he
gets in the ?
18. If a big local advertiser
called up the publisher
of the and
asked him not to print a
certain story, would the
print the story
anyway or leave it out?
19. If an important wealthy
man got into trouble with
the law and asked the
publisher of the
not to publish
a news story about the
matter, would the
print the story
anyway or leave it out?
20. Some people say that most
newspapers won't print
anything tnat might make
them lose advertising.
Do you think this is true
of the 7
21. How well does the
keep its readers up to dat,
on national and world
affairs?
22. How much of the interest-
ing local news do you
feel the gives?
23. What do you think of this
statement: "The
keeps the people of this
area well informed."
24. When you finish reading
the do you
feel it has given you a
clear idea of what has
happened during the day?
25. What about this state-
ment: I can be as well
informed by reading the
as by reading
any other paper I can
get.
26. Do you think the
pays enough attention to
what goes on in the state
government?
27. What kind of a watch does
the keep for
graft in the city and
county government?
28. If a contractor did a bad
paving job for the city,
do you think the
would try to find out
about it and print the
facts?
29. Do you think the
keeps its readers well
informed about the way the
local government conducts
its business?
30. Does the help to
see that laws are enforced
in this area?
31. If the people who run the
had to choose
between their own personal
interests and the best
interests of the whole
area, which would they
choose?
32. Do you think the
tries to run this area to
suit itself?
33. When someone in this area
does something that helps
the community, does the
usually give
him credit for it?
34. What about this statement:
The not only
advocates strongly the
public improvements it
itself wants, but cam-
paigns just as hard for
other improvements that
are equally desirable.
3$. How many people in this
area pay attention to the
's advice on hoW
to vote for President of
the U.S.?
36. If you were not sure about
how to vote on a local
bond issue, would you take
the advice of the
on how to vote?
37. If you did not understand
one of the measures on a
state election ballot,
would you take the advice
of the on how
to vote?
38. When the prints
a'n editorial, do you
usually feel it has made
a complete study of the
subject being discussed?
39. Do you think the people
who run the _____ i
believe what is said in
the advertisements the
paper prints?
40. How much of the adver-
tising in the
do you feel you can
believe?
41. Some people say some news-
papers don't care what
kind of ads they print so
long as they make money.
Do you think this is true
of the ?_
42. Does the allow
local stores to make false
claims in their advertise-
ments in the paper?
43. How often
something
that gave
something
feel that
the world
and hate?
have you read
in the
you a "lift"--
that made you
goodness rules
more than greed
14. Does it seem to you that
the would rather
print bad things about
people than good things?
45. Do the news items in the
suggest that
there are more people in
the world with weak
character than with
strong character?
46. How often does the
print some-
thing that makes you
feel there are a lot of
good people in the
world and not Just a lot
of bad people?
47. How often does the
report events
in which the people
involved show qualities
of courage or sacrifice?
48. How often does the
print pictures
that overemphasize sex?
49. How often does the
print news
about crimes committed
in such communities as
New York, Hollywood,
and Chicago?
50. How often have you read
something in the
that you think
children or teen-agers -
should not read?
51. On the whole, what sort of
job do you think the
is doing?
52. If a friend of yours moved
to this area, would you
advise him to subscribe to
the 7
53. How much do you feel you
really need the 7
54. If the presses at the
broke down and
the paper couldn't be
Printed again for two or
three days, how much would
you miss the paper?
55. When the
arrives, how much do
you usually want to
read it?
56. Everything considered,
what do you think of the
as a
newspaper?
II
WHY THIS TEST WAS DEVELOPED
We began to develop this test, ten years ago, because we believed that the
people of a democracy must have a reasonable amount of confidence in their
newspapers if the political system is to survive.
At about that time two Presidents of the United States had been denouncing
the "one-party press,* some labor union leaders had been telling their members
not to believe the newspapers, and numerous high school teachers had been
parroting Upton Sinclair and George Seldes.
At about that same time, the trend in consolidations, mergers, and sales
of newspapers had reached a point where nine daily newspaper towns in ten had
single ownership of their papers. Critics were quick to point out that this
placed a special obligation of public service on publishers, and if this
obligation were not met the whole basis of our "free market place of ideas"
would be threatened.
It was not hard at that time to envision a future crisis in which our
people might have to choose between believing the facts reported in their news-
papers and the facts presented by a demagogue of the stripe of Huey Long or
Joe McCarthy.
That situation the newspapers themselves might be able to prevent if they
had a reliable instrument for ascertaining how they stood with their readers.
We set out, therefore, to construct an instrument by which the individual
newspaper could not only measure its esteem in a general way, but could also
learn in what respects it was highly esteemed, and in what respects not; and
how highly it was valued by different sections of the public. After such
self-analysis, we felt, any newspaper could make its own contribution to im-
proving democratic processes.
Whether or not such a test now is as essential as we thought it was ten
years ago, the following prediction, made recently by a distinguished former
editor, is worth quoting:
There will be no strong and insistent public demand /Tor free-
dom of expression/ unless freedom of expression is obviously
valuable to the public. A freedom that is enjoyed by nobody in
town except one newspaper publisher is of small value to anybody
else and will not be vigorously defended by the public. It is my
belief that unless a great majority of those publishers who have
attained a monopoly accept the responsibility that goes with
monopoly there will be some form of political interference with
freedom of the press within the next ten years.1
1Gerald W. Johnson, Peril and Promise: An Inquiry Into Freedom of the
Press. (New York: Harper and Brothers, 1957), p. 52.
III
HOW THIS TEST WAS DEVELOPED
Our goal has been to develop a measure of public attitudes toward specific
newspapers--not toward "the press" in general, but toward whatever newspaper an
individual in a particular community habitually reads. Thus it would be a test
which any publisher might use to find out what his readers think of his paper.
In order to make it easy for a publisher to use in a local community, we wanted
to make the test as short as it could be, and still be reliable.
In the process of making the test there were four steps, which overlapped
in time. These were
(1) Determining the content of attitudes toward the press. Before we
could measure attitudes, we had to find out how people think about the press.
What yardsticks do they use to evaluate it? From our own knowledge as former
newspaper men, from the literature on the press, and by interviewing editors
and publishers, we were able to isolate 12 attitudinal dimensions--accuracy,
adequacy, fairness, et cetera--which appear to enter into people's attitudes
toward the press. After testing these in two communities, we sent the list
and the questions representing each dimension to a selected group of editors
in both large and small communities, requesting them to supply any additional
dimension or question that they thought should be added to the test. They
were unable to add to our list, and after trying out the test in five
additional communities we are satisfied that we have an adequate sample of the
content of attitudes toward newspapers.
(2) The second problem was to construct an attitude scale to represent
each of these dimensions. We began by developing more than two hundred
questions which were pretested on a large class of journalism students at the
University of Oregon. Their responses indicated that many of these questions
could be eliminated because they were merely different ways of asking the same
thing. Then the questions were combined into Guttman scales. The making of
these scales is a statistical process which cannot be explained in detail here.
It amounts to finding a group of questions which fit together like the rungs on
a ladder. Just as anyone on a given rung of a ladder has passed all the lower
rungs, so will anyone who says yes to a question which is high on a Guttman
scale also says yes to all the questions which are lower on the scale. A scale
of this kind has three great advantages: it makes it possible to do more with
shorter tests, it gives more reliable results than can be obtained from single
questions, and it makes sure that all the questions in a given scale are
measuring the same attitude--they are all rungs in the same ladder. By test-
ing, revising, and retesting, we were able to construct for each of the
dimensions scales which have very high reliability, although it was not until
the seventh time the test was administered that we got the last scale to meet
our standard of reliability. Technically trained readers will find the criti-
cal coefficients of all these scales in Technical Appendix A. The list of 132
questions which were found not to meet the criteria of the test is in Appendix
G.
(3) The third step was to try out our scales in communities of different
kinds, until the test had met our standards and could be released for general
use. We gave the test in seven communities in four states--Washington, Oregon,
California, and Texas. Circulations of the newspapers studied were approxi-
mately 215,000, 205,000, 170,000, 110,000, 33,000, and 29,000. In all these
communities we measured the attitudes of men toward the newspaper, and in
three communities we measured the attitudes of women as well. The scale
analysis, however, was based on men because it was our observation that men
have more structured attitudes about the political and economic performance of
the newspaper. In chapter VII we shall report some of the attitudinal differ-
ences between men and women. In each community we used a probability sample of
the readers of the newspaper. A probability sample is one in which every
reader has an even chance of being selected to be interviewed. It is the kind
of sample which makes it possible for the researcher to calculate the likeli-
hood that his sample really represents the entire readership. After the
questionnaire was administered in each community, it was coded, punched on IBM
cards, tabulated, and analyzed. The scaling analysis was repeated after every
administration of the test, and such revisions were made each time as seemed
indicated. It was only after the test was given in the seventh community that
it met every requirement, and no further revision seemed to be indicated.
(4) Finally, correlations, factor analyses, and other statistical oper-
ations were performed on the data in order to see, at greater depth, what was
being measured and what general findings were being made. Two of these general
findings are reported in the next chapter.
To make clearer what we have been talking about, let us now list the atti-
tudinal dimensions which have gone into the test, and the questions used under
each of these dimensions to form an attitude scale.
The Attitudinal Dimensions
An attitudinal dimension is a component of a person's total attitude
toward something. Although the reader will have a general attitude of favor-
ableness or unfavorableness toward the newspaper, he will also have specific
attitudes, some of which will be more favorable than others. Por example, he
may be highly favorable to the newspaper so far as adequacy of news content is
concerned, but less favorable toward it on the grounds of political and eco-
nomic fairness. These are the attitudinal dimensions which we decided to
build into the tests
-11-
Adequacy of News Content By this we mean completeness of coverage of
local, regional, national, and world news as measured by the reader's satis-
faction or dissatisfaction over the way he is informed by the paper. The
following questions make up this scale on the tests
21. How well does the keep
its readers up to date on national
and world affairs?
22. How much of the interesting local
news do you feel the
gives?
23. What do you think of this statement:
"The keeps the people of
this area well informed?"
24. When you finish reading the _
do you feel it has given you a clear
idea of what has happened during the
day?
25. What about this statement: I can be
as well informed by reading the
as by reading any other
paper I can get.
When the test is administered in a nonmetropolitan community, question 25
is especially useful in putting the scores for this dimension into perspective.
A second dimension is:
Responsibility for Accuracy. By this is meant performance of the news-
paper in carrying out its alleged covenant with the reader to present the news
truthfully and accurately, and not sensationally. These questions make up
this scale:
1. How accurate is the
in its local news stories?
2. In your experience, do headlines in
the give you an accurate
idea of what really happened?
3. If you ieard a news item over the
radio and then read a conflicting
version of the same story in the
which one would you
believe?
4. How often does the in
its news columns and headlines,
try to make a happening sound more
exciting than it really is?
Fairness and Representativeness. This means representative and impartial
treatment, in the news and editorial columns, of all citizens and segments of
society, and especially of candidates, political issues, labor unions, busi-
nesses, religions, races, and citizens of low social status.
It is possible to make separate scales out of five segments of this
dimension. However, we have combined some of them so as to form three scales,
under the general heading of Fairness and Representativeness.
Political and Economic Fairness:
5. Do you think the
usually gives as much space to
speeches by a candidate the
paper is against as it does to
speeches by a candidate it is
for?
6. Does the usually
present both sides of important
political issues?
7. If the was against a
man who was running for public
office, would it be fair to him
or not?
8. Which, if either, has the better
chance of getting his side of a
story about a local strike into
the : the head of the
business or a labor union
official?
Racial and Religious Fairness:
9. If a Negro got in a serious fight
with a white man in this area,
how fair would the be
toward the Negro?
10. Does the print both
sides of issues that involve
different races?
11. Does the seem fair to
all religious groups?
Representativeness:
12. Do you think the really
cares about the poor people in
this town?
13. Does a wealthy man get better
treatment in the
than a poor man?
14. How true is this statement: The
names of some local people are in
the very often, while
interesting news about many other
local people hardly ever gets in
the paper.
15. If a good friend of the owners of
the got arrested for
drunken driving, would the
print the story?
-14-
16. Would it be easier for a personal
friend of the owners of the
to get a story in the
paper than for a person who
didn't know the owners at all?
17. Does the amount of money a man
has make a difference in the
treatment he gets in the
Independence from Pressure. This implies resistance to pressure for
partial treatment of advertisers, religious sects, persons atop the pyramid of
the power structure, and other interests. This is favoritism due to fear of a
sanction, and is, therefore, to be distinguished from fairness and representa-
tiveness, which is favoritism due to preference and prejudice.
18. If a big local advertiser called up
the publisher of the and
asked him not to print a certain
story, would the print
the story anyway or leave it out?
19. If an important wealthy man got
into trouble with the law and
asked the publisher of the
__ not to publish a news
story about the matter, would
the _______ print the story
anyway or leave it out?
20. Some people say that most newspapers
won't print anything that might make
them lose advertising. Do you think
this is true of the ?
Check upon Government. This refers to the newspaper's performance in
reporting malfeasance, misfeasance, and nonfeasance of elected and appointed
government officials. Historically, of course, the newspaper has always
claimed to be a check upon government, and this is the implication of the First
Amendment as well as of the names selected by many newspapers -- for example,
Advocate, Defender, Examiner, Guard, Guardian, Monitor, Sentinel, and Tribune.
-15-
How many readers still think their newspaper should be a check on govern-
ment? In two of the communities where we administered this test, only 9 and
11 per cent, respectively, of the readers in the sample disagreed with the
statement that "newspapers should be a check on governments." In two other
communities, however, 25 per cent disagreed with the statement that "one of the
main functions of a newspaper is to be a check upon governments; that is, to
report how well and how honestly public officials carry out their duties."
The following questions make up this scale:
26. Do you think the pays
enough attention to what goes on
in the state government?
27. What kind of a watch does the
keep for graft in the
city and county governments?
28. If a contractor did a bad paving
Job for the city, do you think the
would try to find out
about it and print the facts?
29. Do you think the keeps
its readers well informed about
the way the local government
conducts its business?
30. Does the help to see
that laws are enforced in this
area?
Confidence in Leadership on Political Issues and Candidates. This has to
do with readers' confidence in the newspaper's capability and sincerity in its
recommendations to voters concerning candidates and measures.
There appears to be considerable difference among readers in what they
think newspapers should do in this respect. For example, about one half of the
readers in every community in which this question was asked said "No" to: "Do
you think the should give advice to its readers on whom and what to
vote for?"
These questions make up the scale:
35. How many people in this area pay
attention to the 's
advice on how to vote for
President of the U.S.?
36. If you were not sure about how to
vote on a local bond issue, would
you take the advice of the
on how to vote?
37. If you did not understand one of
the measures on a state election
ballot, would you take the advice
of the on how to vote?
38. When the prints an
editorial, do you usually feel it
has made a complete study of the
subject being discussed?
Authoritarianism. This means the tolerance or intolerance of the news-
paper for others' objectives and points of view as evidenced in connection with
community decisions. The motivation for intolerance is self-interest or desire
for dominance. The motivation for tolerance is the newspaper's high regard for
the public interest as distinguished from its own interest and its own vanity.
This dimension is distinguished from "Fairness and Representativeness" because
it involves the newspaper's self-interest or ego, not just its preference or
prejudice.
These questions are in the scale:
31. If the people who run the
had to choose between their own
personal interests and the best
interests of the whole area, which
would they choose?
-17-
32. Do you think the tries
to run this area to suit itself?
33. When someone in this area does
something that helps the
community, does the ____
usually give him credit for it?
34. What about this statement: The
not only advocates
strongly the public improvements
it itself wants, but campaigns
just as hard for other improve-
ments that are equally desirable.
Human Worth and Dignity. To paraphrase William Faulkner, writing in a
different context, this dimension refers to the performance of the newspaper
in presenting man as an animal with spirit -- by reminding readers of the
presence in society of people who exhibit courage, sacrifice, and compassion,
and by suggesting to readers that man's spirit will endure and prevail over
the baser animal drives.
43. How often have you read something
in the that gave you a
"lift"--something that made you
feel that goodness rules the world
more than greed and hate?
44. Does it seem to you that the
would rather print
bad things about people than
good things?
45. Do the news items in the
suggest that there are more people
in the world with weak character
than with strong character?
46. How often does the print
something that makes you feel there
are a lot of good people in the
world and not just a lot of bad
people?
-18-
47. How often does the report
events in which the people involved
show qualities of courage or
sacrifice?
Morally Objectionable Content. This refers to the newspaper's responsi-
bility for upholding the established social and moral values. This is not a
measure of what the newspaper ought or ought not to print, but of what its
readers think it ought or ought not to print.
48. How often does the print
pictures that overemphasize sex?
49. How often does the print
news about crimes committed in such
communities as New York, Hollywood,
and Chicago?
50. How often have you read something
in the that you think
children or teen-agers should not
read?
Responsibility for Advertising Content. This is the responsibility of the
newspaper to protect its readers from exploitation by commercial interests who
buy advertising space.
39. Do you think the people who run
the believe what is
said in the advertisements the
paper prints?
40. How much of the advertising in the
?_______ do you feel you can
believe?
41. Some people say some newspapers
don't care what kind of ads they
print so long as they make money.
Do you think this is true of the
______?
-19-
42. Does the allow local
stores to make false claims in
their advertisements in the paper?
These twelve dimensions (counting Fairness and Representativeness as
three dimensions) make up the specific attitude measures of the test. How-
ever, we have usually added two or three other sets of questions, which
follow.
General Satisfaction. Although these questions form a scale, they do not
constitute a dimension, because they are answered in a total frame of refer-
ence. They are included so that we may know how each of the specific
dimensions relates to the reader's general satisfaction with the newspaper.
In that sense, their total score supplies something of a criterion for each of
the dimensions; and the extent to which they correlate with each dimension
tells us something of how much each dimension contributes to the general or
total evaluation of the newspaper.
51. On the whole, what sort of job do
you think the _______ is doing?
52. If a friend of yours moved to this
area, would you advise him to
subscribe to the ?
53. How much do you feel you really
need the ?
54. If the presses at the
broke down and the paper couldn't
be printed again for two or three
days, how much would you miss the
paper?
55. When the arrives, how
much do you usually want to read
it?
56. Everything considered, what do you
think of the as a
newspaper?
Utility. The following questions, which are not strictly a part of our
test but are available for use when needed, do not represent an attitudinal
dimension in the sense that the earlier dimensions do. Rather they represent
a generalized judgment of how useful the newspaper is to the individuals being
tested. But they are clearly related to the reader's general evaluation of the
paper.
Does the way the writes the news make it
easy for you to understand what happened?
How much entertainment does the give you
with its columns, pictures, comic strips, articles,
and news stories
In your work, business, or daily living, do you find
that the is helpful?
How satisfied are you with the delivery service of
the 7
On the whole, would you say that the is
an interesting newspaper?
Are the things you read in the easy or
hard to find?
How much do you enjoy reading the ?
How often has the ____ helped you with some
practical advice or information that was useful?
What about this statement: Newspapers, including
the print so much about war, conflict,
and people's misfortunes, they ought to have more
humor to balance their content.
Expectations. It will help to interpret some of these scores if we have
some benchmarks. What does the reader really expect of the newspaper?
-21-
Questions like the following help us put the other answers into context:
Somebody has said that one of the main functions of a
newspaper is to be a check upon government; that is,
to report how well and how honestly public officials
carry out their duties. Do you agree or disagree?
Regardless of what you think of the ____, how
would you rate American newspapers in general?
Do you think the should give advice to its
readers on whom and what to vote for?
Demography
In addition to these attitude questions, we have always asked respondents
for certain demographic information, so that we can tell what kinds and what
groups of people hold a given attitude. Under this heading, we usually ask for
age, sex, education (last grade completed in school), occupation, political
party affiliation, number of years in the community, number of years a reader
of the paper, and such other information as may be helpful in understanding the
local situation.
Are There Other Important Dimensions?
The question may legitimately be asked, whether the dimensions we have in
the test are the only ones which enter into a reader's attitudes toward his
newspaper. Undoubtedly they are not. Some critics of the press have sug-
gested, for example, "invasion of privacy"; others have suggested that "over-
emphasis of sports or other entertainment content" might be an area that would
contribute one or more dimensions. The dimensions now in the test, however,
are the only ones where we have found attitudes sufficiently structured to
justify this level of testing.
Reliability of the Test
In Appendix A, which will be of interest to research trained readers and
can be skipped by others, will be found reliability figures for the scales that
make up the test. In nontechnical terms, we can say that the instrument, as
-22-
here presented, has met extremely high requirements for reliability. The
results are well above those usually demanded of such a test, and we feel
that a publisher can use the test with confidence.
IV
HOW THE TEST IS USED
Permission to Use the Test. We will grant permission to any publisher to
use the copyrighted test without charge provided that these conditions are
complied with by the publisher who intends to make the findings public:
1. That we approve the sample. It should be a probability sample of the
newspaper's readers.
2. That we approve the research agency which administers the test.
3. That we approve the questionnaire, if any changes are to be made in
the form presented here.
4. That the findings be made available to us (on a confidential basis if
desired).
5. That the findings with regard to a competitor not be publicized
without that competitor's consent.
For the publisher who is distant from a competent research agency, we are
willing to design the sample, develop the questionnaire, and do the scoring and
analysis on a cost basis.
For any publisher who wants to use the test, we stand ready to give what-
ever advice and direction we can.
Administering the Test. The drawing of a probability sample from readers
will present no problem to a competent research man. If a publisher does not
have access to research help, we are willing, as we have said, to supervise the
drawing of his sample.
Interviewing may present a special problem because the instrument is
rather long to maintain in every case good rapport between interviewer and
-23-
-24-
respondent if the test is administered by personal interview. The best method
we have found is to combine self-administration with personal interviewing.
We recommend that the test be administered over a weekend and be carried on in
the following way:
1. Call on the persons selected for interview on Saturday and
get the responses to the first four or five questions by personal
interview.
2. Leave the unfinished questionnaire with the respondent to
be completed within twenty-four hours.
3. Call back on Sunday to pick up the questionnaire. The
interviewer uses this opportunity to check the responses to make
sure that all questions have been answered.
For the handling of the few special cases in which this pro-
cedure does not operate satisfactorily, we can supply detailed
information.
Handling the questionnaires, coding, tabulating, and analyzing will pre-
sent no problem for a trained survey research man. IBM machines, of course,
will speed up the work. For the publisher who does not have access to such
help, we stand by to assist as best we can. Inquiries should be addressed to
the holder of the copyright.
Uses of the Test: The main use of the test is to tell the publisher how
high or low the newspaper is valued on the particular dimensions. The test can
also be used to measure the attitudes of types of persons: for example, the
newspaper's employees, teachers of the social sciences in high schools, and
persons who have a relationship with the newspaper other than as readers (e.g.,
advertisers, school administrators, and other sources of news). One warning:
When the test is administered to such persons the number is often small. That
means that small observed differences are not statistically significant. Tests
must be applied by a trained research worker.
It is possible to administer the test to scattered members of a group who
do not read the same newspaper by having them respond with respect to the
particular newspaper they read. In that instance, the object measured is more
like newspapers in general than like a particular newspaper.
When supplemented with questions that are relevant in certain problem
areas, the test can be used to develop certain explanations. For example, in
the area of employee attitudes the test results could be a significant variable
when related to job satisfaction and similar variables.
Studies can be made by communication specialists to compare readers'
attitudes in single ownership vs. competing ownership situations; to relate the
findings to the life history of certain readers who have had good or bad ex-
periences with the newspaper; and to relate the findings to a content analysis
of editorial and news content about community conflicts over a specified
period.
By research of a more sophisticated kind, additional explanations might be
developed by factor analyzing the scores of readers with high education and of
low education; that is, to test whether there is a different patterning of
attitudes in different education groups. Research-trained readers will think
of other studies that might be made.
Interpreting the Results of the Test. Two sets of scores will be readily
available to the user of the test. These are the scores on individual
questions, and the total scores on the separate attitudinal dimensions. A
third set of scores can be obtained by figuring the intercorrelation of the
dimension scores, and, if desired, by factor analyzing these correlations and
thus revealing the clustering or patterning of attitudes in the community.
Publishers will certainly want the first two kinds of scores, but only a few
-26-
will feel the need, or have the research resources at hand, to go into the cor-
relations and factor analysis.
When a publisher has the scores for his readers, how does he interpret
them?
To date, there is no norm, no standard, to which he can compare the scores
to see whether his readers are giving him better or worse than average grades.
As the test is used in more and more communities, and as these results are sent
in to us, we may be able to establish such norms and make them available to
future users of the test. We have not been able to state such averages or norms
for the seven communities where the test has so far been given, because the
instrument was then in a state of development and was not given in exactly the
same form in any two communities. In Chapter VI, however, we present the
results of the test when given in its present form in a community where the
newspaper is, by general agreement, public-spirited and efficient. Publishers
who want to use the test themselves will find these Chapter VI results very
interesting.
Even if we had a national norm, however, the test should not be inter-
preted solely with reference to that standard. There will almost certainly be
local circumstances to be taken into account. The publisher will understand
these circumstances better than anyone else, and ultimately must make his own
decisions as to what the test means in terms of community criticism and
endorsement of his policy and practice. The political tensions, the history of
conflict within the community, the community economic structure, the religious
groupings--these and other elements must be taken into account in passing
judgment on the test results.
As more and more test results come into us, we shall be able to give
publishers guidelines in interpreting the scores. Even without such additional
guides, however, any editor and publisher will find the answers to the
-27-
questions most illuminating and helpful. This has certainly been the experi-
ence wherever the test has been given thus far.
In the next chapter will be found a number of suggestions and results
which will be of interest to the publisher who wants to interpret his own test
results.
V
TWO GENERAL FINDINGS FROM THE TEST -- INTENSITY OF ATTITUDES
TOWARD THE PRESS, AND FACTORS IN ATTITUDES TOWARD THE PRESS
It may be helpful to users of the test if we describe two general findings
which have emerged from the uses so far made of the test.
Intensity of Attitudes
In the first place, only a relatively small percentage of readers seem to
feel intensely about these attitudinal dimensions.
Look at Figures 1 and 2. Technically-trained readers will recognize them
as plots of the intensity components of Guttman scales, but it is not necessary
to know anything about how they are made in order to understand what they mean.
These are charts of how intensely people felt -- in the community where
the test was most recently given -- about two of the scales. Notice that the
charts are about the same shape: a mountain peak at either end, and a broad
flat valley in the middle. The higher the line rises, the more strongly
readers felt about their answer. The unfavorable answers are on the left;
favorable ones on the right. The whole width of the chart represents all the
readers who were interviewed; you will see that the base line is marked out in
ten per cent intervals.
What Figure 1 means, then, is that about 11 per cent of the readers felt
strongly that their newspaper was not fair, and about 22 per cent felt strongly
that it was fair. Those in the middle -- about 62 per cent -- didn't feel very
strongly that it was either fair or unfair.
Figure 2 is the same kind of picture. This represents how strongly people
felt about the political leadership which the newspaper sought to take. Here
-28-
FIGURE 1
Political and
Economic
Fairness
Intensity
Curve
FIGURE 2
Confidence in
Political
Leadership
Intensity
Curve
10 20 30 o0 So 60 7"0 8"0 90
percentiles
CO INTENT
p U*.
"le
Nr
T c60
Ee
N n 50"
St
I i o0*
T1
Ye
s 30'
20.
10.
10 20 30 1o 50 60 70
percentiles
CONTENT
80 90
about 36 per cent felt that the paper was trying to take too much leadership --
perhaps shoving people around a bit. On the other hand, about 18 per cent felt
strongly that the paper was taking only a proper amount of political leader-
ship. But 46 per cent had no strong opinion either way.
In the next chapter you will see a number of these intensity curves, and
you will observe that many of them have about the same shape. The significant
part of the chart is the broad, low valley in the middle, representing the
people who have no strong attitude either way. This might be called a valley
of ignorance, because it suggests a rather disturbing ignorance of, or interest
in, the job newspapers do.
Patterns of Attitude
In the second place, it is clear that readers, unless they are reminded,
do not typically think of newspapers in terms of all these specific dimensions.
They have much more generalized attitudes toward newspapers.
The evidence for this comes from factor analysis. Technically-trained
readers will find the essential data on the most recent factor analysis we have
performed with this test data in Appendix C. Without getting into details of
how factor analyses are made, or the mathematical details of results, however,
we can say quite simply what the method does and what the results mean.
Factor analysis is a mathematical means of finding the clusters or group-
ings within a large number of related measures. For example, at a certain
stage of psychologists' study of human intelligence, a large number of intelli-
gence measures were obtained from the same persons, and the results were then
correlated and factor analyzed to discover the large factors that constitute
the abilities we call intelligence. Similarly, in our own department at
Stanford, two research workers have recently made a factor analysis to deter-
mine the elements that make for stylistic readability.
-31-
We made factor analyses after different administrations of the test. The
one we present in detail in Appendix C was based on the seventh community trial
of the test. This, it will be remembered, was the first occasion when the test
was used in complete and final form. From this analysis, we found that the
twelve attitudinal dimensions fitted into three larger factors, or clusters of
attitudes.
The first of these is a general evaluative factor. In other words, more
often than any other way, the reader typically thinks of his paper simply as a
good or a bad paper, apparently without defining "good" or "bad" very sharply.
This turned out to be a very big factor. All except one of the dimensions were
closely related to it. Readers who wish to read Appendix C will recognize that
about 60 per cent of all the variance we were able to get out of the factor
analysis belonged to this factor. A similar large cluster appeared in all the
factor analyses we performed on studies using this test. Therefore, we can say
with some confidence that readers seem to think of their newspaper chiefly in
the broad terms of it being a good or a bad paper.
The second factor, or cluster, which appeared we can call a bias factor.
This was a more specific factor than the first one we have been describing.
The attitudinal dimensions which contributed to this second factor were Inde-
pendence from Pressure, Representativeness, Authoritarianism, Political and
Economic Fairness, and Confioence in Political Leadership, in that order. This
factor, or one very much like it, has also appeared in the other analyses we
have done. It seems likely, therefore, that many readers do tend to think of
their papers quite specifically in terms of whether it slants the news toward
its own viewpoint or carries editorials that reflect its publisher's or
editor's own prejudices, self-interest, or desire for dominance.
The second factor was a little less than half as large (technically, the
loading was less than half as great) as the first one; and the third factor was
a little less than half as big as the second. This third factor seemed to be
somewhat like the second, but drew its strength mostly from but two attitudinal
dimensions Authoritarianism and Political Leadership. It seems, therefore,
to be a kind of political dictatorship factor, and indicates that a consider-
able number of readers do think of their paper specifically in terms of how
much it tries to direct their political opinions.
It is not surprising that bias and political dictatorship should emerge as
general attitude factors toward newspapers. For one thing, of course, the
newspaper has always been closely related to the political process of a com-
munity. Since the population of a community usually varies in its political
and economic loyalties and prejudices, a part of the population will invariably
have an unfavorable attitude toward the newspaper when the paper habitually
takes a definite stand or advocates a point of view about which there is not
general community agreement. Nevertheless, the fact that bias and political
dictatorship are salient in readers' minds should mean something to publishers,
especially in single ownership situations.
Two dimensions that a newspaper man might expect to emerge as basic
factors are Accuracy and Adequacy of News Content. They did not emerge in the
seventh study except to contribute to the general evaluative factor. In the
sixth study, however, they made up the third basic factor, replacing political
dictatorship.
This sixth study was one in which we did not have complete confidence in
the sample because it appeared to contain too many individuals from the higher
education and income levels. The particular newspaper had a circulation of
more than 200,000, was Democratic in a heavily Democratic community, and had a
strong competitor which also was Democratic. We have reason to believe that
only a small proportion of its readers were in disagreement with its political
and economic policy. Therefore, more readers tended to evaluate the paper for
-33-
Adequacy of Content than for Political. Dictatorship. (Table I in Technical
Appendix C)
Until we have done more studies in more kinds of communities, we are not
prepared to say that the basic underlying attitude clusters in regard to the
press are exactly the same in all situations. We suspect that the good-bad
factor is always present, but that the others sometimes change. Yet it seems
clear that the average reader will organize most of his attitudes toward his
newspaper in only a few significant measures. He will have a general opinion
as to whether it is a good or a bad newspaper. He will have an opinion as to
whether it is or is not a biased newspaper. In different places, in con-
nection with different newspapers, he will have a definite opinion as to
whether the paper's news content is accurate and adequate, and whether it
tries to impose its own political leanings on its readers.
VI
READERS EVALUATE THEIR NEWSPAPER --
RESULTS OF THE TEST IN ONE COMMUNITY
In the seventh community in which we administered the test we found that
the individual questions met all of the criteria we had established. So now we
have scores that are unidimensional and reliable and can report them for that
particular newspaper.
The circulation of this newspaper--the only one published in that com-
munity--is approximately 33,000. This community has a somewhat larger
proportion of college educated readers than the typical community and has grown
considerably in recent years. The newspaper is independent, but generally
supports the Republican party and its candidates. Republican voters constitute
a small majority of the voting population (48 per cent Republican, 43 per cent
Democratic, 6 per cent independent, and 3 per cent refused to state--in our
sample). The newspaper traditionally has had a strong editor and publisher and
has participated in all important community decisions. It has been a better
watchdog over public officials than has the usual newspaper.
The findings reported in this chapter refer only to this particular news-
paper. In Chapter VII we report some findings for other newspapers, and for
women, for employees, and for readers who are subscribers to both of the com-
peting newspapers in one community.
General Satisfaction
As something of a criterion, we asked six questions of a general nature
about the newspaper. We expected the relation of these answers to the scores
on attitudinal dimensions, and to the demographic characteristics (age,
-35-
education, etc.) of the readers, to give us a valuable base line for the rest
of the test, and tell us what kinds of readers are most and least satisfied
generally with their newspaper.
The responses to these questions on General Satisfaction follow. The
number in parentheses after the question is the average score for all respond-
ents. The least favorable answer is 1, "don't know" is 3, and the most
favorable answer, 5; so 3.9, on the first question, indicates a very favorable
average response. The percentages below the question indicate the percentage
of readers who checked each of the responses. For example, for Question 51
exactly 18.4% said that they thought the paper was doing a very good job, 62.8%
said they thought it was doing a good job, et cetera.
51. On the whole, what sort of job do you think the
is doing? (3.9)
18.4% A very good job
62.8 A good job
11.7 A poor job
0.0 A very poor job
4.6 Don't know
2.5 Not answered
52. If a friend of yours moved to this area, would you
advise him to subscribe to the ? (4.2)
1.4% No, certainly not
7.4 No, probably not
51.1 Yes, probably
36.2 Yes, certainly
1.4 Don't know
2.5 Not answered
53. How much do you feel you really need the ? (4.1)
2.8% Not at all
11.7 Not much
44.7 Some
37.9 Very much
0.4 Don't know
2.5 Not answered
-36-
54. If the presses at the broke down and the
paper couldn't be printed again for two or three
days, how much would you miss the paper? (3.4)
18.1% A very great deal
43.2 Quite a lot
29.1 A little
6.4 Not at all
0.7 Don't know
2.5 Not answered
55. When the arrives, how much do you usually
want to read it? (4.0)
24.1% Very much
58.8 Pretty much
13.1 Not very much
0.4 Not at all
1.1 Don't know
2.5 Not answered
56. Everything considered, what do you think of the
? (3.4)
10.7% An excellent newspaper
54.6 A good newspaper
28.0 A passable newspaper
2.8 Not much of a newspaper
1.4 Don't know
2.5 Not answered
These answers should encourage a publisher. Looking at the intensity
chart (Figure 3) he can see that about 62 per cent of his readers are generally
well satisfied with the paper, only about 9 per cent are definitely unfavor-
able, and 29 per cent do not feel strongly either way.
Furthermore, his research adviser, looking at the correlations (Table H,
Appendix C) can tell him that General Satisfaction is closely related to every
attitudinal dimension except Morally Objectionable Content. In other words,
the high score on General Satisfaction reflects a broad satisfaction with the
paper.
The publisher's research adviser can tell him also that General Satis-
faction with his paper is not closely related to age, income, education,
political party affiliation, labor union membership, or number of years a
-37-
FIGURE 3
General Satisfaction Intensity Curve
90 *
80 *
P 70 *
60
r 60 *
c
20 .
n0 *
i 40 *
1
s 30 "
20 *
10"
10 20 30
IO 50 60 70 80 90
percentiles
CONTENT
reader. This is not always the case. In our Study Number 5, for example,
satisfaction with the paper depended on political party membership. In the
same study, age was related positively to satisfaction, and education was
on the borderline of a significant relationship. The publisher of the news-
paper in Study 7, however, can reflect with some pleasure that both young
and old are generally satisfied with his paper, as are Democrats and Re-
publicans, rich and poor, long-time readers and new readers, and all the
other demographic groups we have mentioned.
There is one respect, however, in which General Satisfaction in this
seventh study (and in the third study as well) seems to depend partly on the
group membership of a reader. That is religion. If one looks at Table 1
(which, like the other tables, has been put in the appendix to this chapter, so
as not to impede the progress of readers who don't like to read tables), one
sees that a larger proportion of Catholics than of Protestants are dissatisfied
with the paper (page 66).
Utility
In Test Number 7, we used nine questions on Utility. As previously sug-
gested, we think of these as independent single questions, rather than as a
dimension. The total score on these Utility questions, however, relates more
highly than any of the dimensions to General Satisfaction. These are the
responses on Utility:
Does the way the writes the news make it easy
for you to understand what happened? (4.2)
32.3% Very easy
56.7 Fairly easy
6.7 Not very easy
0.4 Not at all easy
0.4 Don't know
3.5 Not answered
How much entertainment does the give you (with
its columns, pictures, comic strips, articles, and news
stories)? (4.1)
0.4% None at all
9.6 Not much
54.6 Some
31.2 A great deal
0.7 Don't know
3.5 Not answered
-39-
In your work, business, or daily living, do you find that
the is helpful? (3.8)
27.0% Yes, it helps quite a lot
48.6 Yes, it helps some
17.7 No, not much
2.1 No, not at all
1.1 Don't know
3.5 Not answered
How satisfied are you with the delivery service of
the ? (4.3)
56.7% Very satisfied
28.4 Fairly satisfied
8.2 Not very satisfied
3.2 Not at all satisfied
3.5 Not answered
On the whole, would you say the is an interesting
newspaper? (3.9)
0.3% No, not at all interesting
12.8 No, not very interesting
66.7 Yes, fairly interesting
16.3 Yes, very interesting
0.4 Don't know
3.5 Not answered
Are the things you read in the easy or hard
to find? (4.2)
33.3% Very easy to find
55.0 Fairly easy to find
7.1 Fairly hard to find
0.7 Extremely hard to find
0.4 Don't know
3.5 Not answered
How much do you enjoy reading the ? (4.3)
0.0% Not at all
7.4 Not much
43.3 Some
45.8 Very much
3.5 Not answered
How often has the helped you with some practical
advice or information that was useful? (3.5)
5.7% Never
20.5 Seldom
55.0 A few times
14.9 Many times
0.4 Don't know
3.5 Not answered
-40-
What about this statement: Newspapers, including the
print so much about war, conflict, and
people's misfortunes, they ought to have more humor
to balance their content. (2.9)
6.6% Strongly agree
44.6 Agree
41.4 Disagree
2.8 Strongly disagree
1.1 Don't know
3.5 Not answered
A question about type faces was eliminated because it did not meet the
criteria of scaling analysis. It is reproduced in Appendix F.
Average Scores on the Dimensions
Here are the average scores for each attitudinal dimension in the test,
and for Utility and General Satisfaction. Remember that a score of 1 is the
worst possible, and a score of 5 is the best possible.
4.40 Racial and Religious Fairness
3.74 Adequacy of News Content
3.68 Check on Government
3.67 Morally Objectionable Content
3.62 Responsibility for Advertising Content
3.60 Authoritarianism
3.56 Representativeness
3.54 Responsibility for Accuracy
3.54 Human Worth and Dignity
3.53 Average Score for All Dimensions This Newspaper
3.00 Political and Economic Fairness
2.98 Confidence in Political Leadership
2.97 Independence from Pressure
3.91 Utility
3.83 General Satisfaction
Looking at these average scores, it is clear that the newspaper is highly
valued for its utility. The only attitudinal dimension which has a higher
score is Racial and Religious Fairness. Likewise, the General Satisfaction
score is considerably higher than the average score for all dimensions and
higher than any dimension except Racial and Religious Fairness; this must be
because the Utility aspect of the paper makes some contribution to General
Satisfaction. We have found it to be true of nearly every paper we have
studied, that the General Satisfaction score is higher than the average score
for the separate dimensions.
Noteworthy in these scores is the high valuation placed on the newspaper's
Racial and Religious Fairness. Less pleasing to the publisher must be the
lower than average ratings given three political dimensions, Political and
Economic Fairness, Confidence in Political Leadership, and Independence from
Pressure. Apparently, the readers think more highly of the paper's adequacy of
content and fairness to minority groups than of its political fairness and
independence.
Let us now look at the responses to questions in each of these attitudinal
dimensions.
Racial and Religious Fairness
Most pleasing to the publisher of this paper must be the intensity plot
for this attitudinal dimension (Figure 4). This indicates that about 81 per
cent of his readers have a favorable opinion of the paper's racial and re-
ligious fairness, only 19 per cent are unfavorable or neutral, and no one feels
very strongly unfavorable. Here are the answers to the questions which repre-
sent Racial and Religious Fairness:
9. If a Negro got in a serious fight with a white man
in this area, how fair would the be
toward the Negro? (4.4)
0.4% Not at all fair
3.2 Not very fair
39.0 Pretty fair
53.5 Very fair
2.5 Don't know
1.4 Not answered
-42-
90*
80.
P 70"
le
Nr
T c60
Ee
N n 50'
St
I i40-
T1
Ye
s 30.
20.
10.
10 2*0 3 0 10 60 70 8 90
percentiles
CONTENT
FIGURE 5
Adequacy of
News Content
80.
P 70-
le
Nr
T c 60*
Ee
N m 50'
St
I i
T 1 0"
Ye
s 30-
10 20 30 41 50 60 70 80 90
percentiles
CONTENT
FIGURE I4
Racial and
Religious
Fairness
Intensity
Qirve
-43-
10. Does the print both sides of issues that
involve different races? (4.2)
29.1% Yes, always
61.7 Yes, usually
3.9 No, not very often
0.7 No, almost never
3.2 Don't know
1.4 Not answered
11. Does the seem fair to all religious groups? (4.6)
61.7% Yes, very fair
34.4 Yes, pretty fair
2.1 No, not very fair
0.0 No, not at all fair
0.4 Don't know
1.4 Not answered
These scores are not related to the religious faith of readers. For
example, there is no significant difference between the responses of Negroes
and the responses of whites. Nor is it related to any of the other demographic
characteristics of the audience. Across the board, the reading audience gives
this paper a vote of confidence and approval on its racial and religious
fairness.
Adequacy of News Content
The first thing the publisher will look at, when he studies this atti-
tudinal dimension, will again be the intensity plot. This is in Figure 5.
Here he will see (or his research adviser will calculate for him) that about
52 per cent of his readers are strongly favorable to his news content, so far
as adequacy is concerned. About 31 per cent are neutral, and only about 17 per
cent are definitely unfavorable. Furthermore, those who are unfavorable do not
feel so strongly about it as do the favorable readers. This, like the response
on Racial and Religious Fairness, should please the publisher.
-44-
Here are the responses on the questions which deal with adequacy of
content:
21. How well does the keep its readers
up to date on national and world affairs? (4.0)
23.4% Bxtremely well
62.8 Fairly well
9.9 Not very well
2.1 Not at all well
0.4 Don't know
1.4 Not answered
22. How much of the interesting local news do you feel
the gives? (4.1)
29.1% Nearly all of it
61.0 Most of it
8.1 Very little of it
0.0 Almost none
0.4 Don't know
1.4 Not answered
23. What do you think of this statement: "The
keeps the people of this area well informed?" (3.7)
12.1% Strongly agree
67.0 Agree
15.2 Disagree
3.9 Strongly disagree
0.4 Don't know
1.4 Not answered
24. When yo. finish reading the do you feel it
has given you a clear idea of what has happened during
the day? (3.8)
13.8% Yes, very clear
69.2 Yes, fairly clear
11.0 No, not very clear
2.5 No, not at all
2.1 Don't know
1.4 Not answered
25. What about this statement: "I can be as well informed
by reading the as by reading any other paper
I can get." (3.1)
11.0% Strongly agree
41.5 Agree
36.5 Disagree
7.8 Strongly disagree
1.8 Don't know
1.4 Not answered
-45-
It is generally assumed by publishers that the main reason that most
people buy a newspaper is for its news content, and it is also widely assumed
that more people prefer the newspaper, in a competitive situation, that pre-
sents the most news. We are not able, with these studies, to test the second
hypothesis. Our studies, however, appear to confirm the first one. For
example, we can say from the results on this test that
(a) Those readers who are most satisfied with the amount of news in their
own newspapers are the ones who value newspapers most highly (see Table 2 in
the appendix to this chapter).
(b) Those who spend the most time reading their newspaper tend to be the
ones most satisfied with the amount of news they find (see Table 3).
It is worth noting that readers' evaluation of adequacy of news content
appears to be unrelated to education, age, or any of the demographic variables.
The satisfaction with this paper's news coverage is distributed through the
population.
Therefore this publisher has every right to feel that his readers approve
his news coverage, and that he has good reason to maintain it.
Check on Government
A glance at Figure 6 will show that about 38 per cent are definitely
favorable to this newspaper's performance as a check on government, about
27 per cent definitely unfavorable, and about 35 per cent neutral. In other
words, readers seem to divide into three fairly even groups, although the
unfavorable one is the smallest.
-46-
Here are the responses:
26. Do you think the pays enough attention to what
goes on in the state government? (3.6)
1.1% No, it never does
21.3 No, sometimes it doesn't
64.2 Yes, it usually does
9.2 Yes, it always does
1.4 Don't know
2.8 Not answered
27. What kind of watch does the keep for graft
in the city and county government? (3.6)
4.3% It doesn't watch at all
18.8 Not a very close watch
59.2 A fairly close watch
11.0 A very close watch
3.9 Don't know
2.8 Not answered
28. If a contractor did a bad paving job for the city, do you
think the would try to find out about it and
print the facts? (3.7)
25.2% Yes, certainly
44.3 Yes, probably
23.0 No, probably not
3.9 No, certainly not
0.8 Don't know
2.8 Not answered
29. Do you think the keeps its readers well informed
about the way the local government conducts its business?
(3.9)
3.5% Very poorly informed
11.3 Not so well informed
57.9 Fairly well informed
24.1 Very well informed
0.4 Don't know
2.8 Not answered
30. Does the help to see that laws are enforced in
this area? (3.6)
13.1% Yes, it always does
58.6 Yes, it usually does
20.2 No, sometimes it doesn't
2.5 No, it never does
2.8 Don't know
2.8 Not answered
This dimension is not related to any of the demographic characteristics.
FIGURE 6
Check on
Government
Intensity
Curve
FIGURE 7
Morally
Objectionable
Content
Intensity
Curve
90 *
80 *
70 *
60 *
50 *
40 *
30 *
20 *
10 *
10 20 30 4*0 5o 60 70 80 90
percentiles
CONTENT
90
80 *
P 70 *
e
r
60"
e
n 50 *
t
i
1 40
e
s 30 *
20 *
10 *
10 20o 3'0 L'o 60 70 8"0 9 0
percentiles
CONTENT
Morally Objectionable Content
The first thing the publisher will notice about this dimension of the test
(Figure 7) is the large number of readers in the neutral group. Very few are
really unfavorable (6 per cent), and a large group is favorable (41 per cent),
but the majority simply have no strong opinion one way or the other (53 per
cent). This may be because the newspaper here being studied contains compara-
tively little of this kind of news. We might get less neutrality in responses
if we administered the test in a community where the newspaper carried a good
deal of objectionable material.
Strangely enough, this dimension of attitude does not correlate with any
of the other dimensions, nor with General Satisfaction, nor with religion or
any other demographic characteristics. It is a lone wolf, and may be measuring
something that is apart from a general attitude toward the newspaper. It is
also a very difficult dimension to measure, because almost everyone says that a
newspaper should carry little about crime and little about sex scandals, and
yet it is well known that many people like to read such news. It was not until
the seventh study that we were able to develop a set of questions for this
dimension that would meet the criteria of scale analysis. This we finally did
by substituting a concrete question for one that had referred to crime news
generally.
These were the answers to questions which lie in this dimension:
48. How often does the print pictures that
overemphasize sex? (4.2)
26.6% Never
67.7 Not very often
3.9 Often
0.0 Very often
0.4 Don't know
1.4 Not answered
-49-
49. How often does the print news about crimes
committed in such communities as New York, Hollywood,
and Chicago? (2.8)
0.0% Very often
47.2 Often
35.8 Not very often
14.2 Never
1.4 Don't know
1.4 Not answered
50. How often have you read something in the that
you think children or teen-agers should not read? (4.0)
2.1% Very often
6.4 Often
70.2 Not very often
19.2 Never
0.7 Don't know
1.4 Not answered
Responsibility for Advertising Content
There are no neutrals when it comes to expressing opinions on this news-
>aper's advertising stewardship. As Figure 8 shows, 52 per cent are definitely
Favorable and 49 per cent definitely unfavorable. Feeling is less strong on
the negative end, but this is a subject on which everyone seems to have a
lefinite feeling.
These are the responses to questions in the scale:
39. Do you think the people who run the believe
what is said in the advertisements the paper prints?
(3.5)
6.0% All of it
65.2 Most of it
23.8 Not much of it
0.0 None of it
3.2 Don't know
1.8 Not answered
40. How much of the advertising in the do you
feel you can believe? (3.7)
0.0% None of it
18.4 Not much of it
74.5 Most of it
4.2 All of it
1.1 Don't know
1.8 Not answered
41. Some people say some newspapers don't care what kind of
ads they print so long as they make money. Do you think
this is true of the ? (3.5)
3.5% Yes, certainly
25.2 Yes, probably
53.5 No, probably not
15.3 No, certainly not
0.7 Don't know
1.8 Not answered
42. Does the allow local stores to make false
claims in their advertisements in the paper? (3.8)
22.6% No, never
40.7 No, not very often
8.2 Yes, fairly often
1.8 Yes, very often
24.8 Don't know
1.8 Not answered
This is a rather puzzling dimension in certain ways. For example, more
people seem to believe newspaper advertising (Question 40) than think the
publisher believes it (Question 39)! Although the responses in Study Number 7
are not related to demography, there is some evidence in our earlier studies anc
in other newspaper research done here, that younger people tend to have less
favorable attitudes toward advertising policy than do older people. But there
is one thing certain about it: what a reader thinks of the advertising policy
of a newspaper is a very important part of his general attitude toward the
paper. Readers of this report who are not allergic to reading tables should
look at Table 4. It shows a very close relationship between evaluation of news-
papers in general and the paper's responsibility for its advertising content.
Authoritarianism
Figure 9 shows that about 34 per cent of the readers are favorable on this
dimension, about 16 per cent unfavorable, and about 50 per cent with little or
no feeling about it. A more conservative interpretation might divide the
readers 55 per cent favorable and 45 per cent unfavorable.
-51-
FIGURE 8
Responsibility
For Advertising
Content
Intensity
Curve
90 *
60 *
70 *
60 *
50 *
40 *
30 *
20 *
10 *
10 20 30 1O0 50 60 70 80 90
percentiles
CONTENT
FIGURE 9
Authoritarianim
Intensity
Curve
90 *
80 *
p 70 *
le
Nr
Ee
N n 0
St
I 1 40
T1
Ye
s 30 *
20
10
10 20 30
ho 50 60 70 80 90
percentiles
CONTENT
The most interesting thing about the results on this dimension is that
they are different by party affiliation. Table 5 shows that Democrats are much
more likely than Republicans to think that the paper--which is Republican in
policy--is trying to impose its own political will on the community. We found
an even stronger adverse attitude on the part of Democratic readers of a Re-
publican newspaper in a metropolitan community which was predominantly
Democratic.
Attitude on the newspaper's authoritarianism is also related to the number
of years a respondent has read the paper (Table 6). The long-time readers
tend to think the paper more authoritarian. Perhaps the newer readers (who are
also newer residents) have participated in fewer community decisioLs and are
unaware of former community conflicts or political campaigns. (The chronologi-
cal age of the respondent, however, is not related to authoritarianism.)
Here are the responses to individual questions:
31. If the people who run the had
between their own personal interests and
interests of the whole area, which would
choose? (3.2)
8.2% The area's interests, always
43.6 The area's interests, usually
33.7 Their own interests, usually
4.6 Their own interests, always
6.7 Don't know
3.2 Not answered
32. Do you think the tries to run
suit itself? (3.2)
14.5% No, certainly not
40.4 No, probably not
26.6 Yes, probably
12.8 Yes, certainly
2.5 Don't know
3.2 Not answered
to choose
the best
they
this area to
33. When some one in this area does something that helps
the community, does the usually give him
credit for it? (4.1)
0.4% No, it never does
2.8 No, it hardly ever does
72.7 Yes, it usually does
20.5 Yes, it always does
0.4 Don't know
3.2 Not answered
34. What about this statements The not only
advocates strongly the public improvements it itself
wants, but campaigns just as hard for other improve-
ments that are equally desirable. (3.5)
9.6% Strongly agree
59.9 Agree
22.3 Disagree
2.1 Strongly disagree
2.9 Don't know
3.2 Not answered
Representativeness
A conservative interpretation of the intensity plot for this dimension
(Figure 10) would indicate that about 42 per cent of readers are favorable,
about 32 per cent unfavorable, and the other 26 per cent neutral. The answers
to specific questions are:
12. Do you think the really cares about the
poor people in this town? (3.9)
28.4% Yes, very much
49.6 Yes, some
17.0 No, not very much
2.5 No, not at all
0.7 Don't know
1.8 Not answered
13. Does a wealthy man get better treatment in the
than a poor man? (2.8)
11.3% No, not at all
14.3 No, very little better
33.3 Yes, some better
12.4 Yes, a whole lot
26.9 Don't know
1.8 Not answered
FIGURE 10
Representativeness
Intensity
Curve
FIGURE 11
Independence
from Pressure
Intensity
Curve
p
90 *
80 *
70
60 *
50 .
40 *
30 *
20 *
10 *
90 *
80 *
70 *
I e
Nr
T c 60 *
Ee
N n 50 -
St
i '40 *
TI
Ye
s 30 *
20 *
10 *
10 20 30 hO 50 60 70 80 90
percentiles
CO NTE NT
10 20 30 5 6 5o 60 70 o0 90
percentiles
CONTENT
-55-
14. How true is this statement: The names of some local
people are in the very often, while inter-
esting news about many other local people hardly every
gets in the paper. (2.5)
23.4% There's a lot of truth in it
40.8 There's some truth in it
26.2 There is very little truth in it
5.7 It's not true at all
1.8 Don't know
2.1 Not answered
This dimension does not appear to be related to any of the demographic
characteristics, although in a metropolitan community we found a strong
relationship between political party affiliation and Question 12.
Independence from Pressure
On this dimension (see Figure 11) fully 70 per cent of the readers had no
strong attitudes whatsoever. However, among those who did feel strongly, those
unfavorable overbalanced those favorable, 22 per cent to 8 per cent.
Labor union members, as Table 7 shows, thought the paper less free from
pressure than did nonunion men. Readers with high income were more likely to
think of it as free from pressure than were those of low income (Table 8). It
is therefore the union workers, the lower income groups, who are the pub-
lisher's chief critics in this attitudinal area and their suspicion of him
probably has to do with his supposed relations to business, industry and
finance. Responses on this dimension, however, are not related to our
"Liberalism/Conservatism" scale. In one metropolitan community which was pre-
dominantly Democratic and the newspaper we measured was Republican, we found
Independence from Pressure to be highly related to political party affiliation.
In this particular community there was no relationship to political affili-
ation, but we did find a strong correlation with party affiliation in a
metropolitan community in which we gave the test.
These are the figures on the individual questions:
15. If a good friend of the owners of the _______ got
arrested for drunken driving, would the
orint the story? (3.1)
7.4% No, certainly not
30.9 No, probably not
20.6 Yes, probably
7.4 Yes, certainly
29.4 Don't know
4.3 Not answered
16. Would it be easier for a personal friend of the owners
of the to get a story in the paper than for
a person who didn't know the owners at all? (2.4)
4.9% No, certainly not
10.6 No, probably not
42.9 Yes, probably
16.0 Yes, certainly
21.3 Don't know
4.3 Not answered
17. Does the amount of money a man has make a difference
in the treatment he gets in the ? (2.9)
9.2% No, none at all
18.8 No, very little
31.5 Yes, some difference
7.1 Yes, a lot of difference
29.1 Don't know
4.3 Not answered
18. If a big local advertiser called up the publisher of
the and asked him not to print a certain
story, would the print the story anyway or
leave it out? (3.0)
5.3% Would print it anyway, certainly
40.0 Would print it anyway, probably
39.4 Would leave it out, probably
2.8 Would leave it out, certainly
8.2 Don't know
4.3 Not answered
-57-
19. If an important wealthy man got into trouble with the
law and asked the publisher of the not to
publish a news story about the matter, would the
print the story anyway or leave it out? (3.3)
3.5% Would leave it out, certainly
26.6 Would leave it out, probably
52.5 Would print it anyway, probably
7.4 Would print it anyway, certainly
5.7 Don't know
4.3 Not answered
20. Some people say that most newspapers won't print
anything that might make them lose advertising. Do
you think this is true of the ? (3.1)
8.8% No, certainly not
40.0 No, probably not
39.4 Yes, probably
5.7 Yes, certainly
1.8 Don't know
4.3 Not answered
Responsibility for Accuracy
There is no great amount of strong feeling in this area. About 6 per cent
of the readers are definitely unfavorable (Figure 12), about 24 per cent are
definitely favorable, and the other 70 per cent are in the neutral band.
There is a hint in the results (see Table 9) that Republicans may be more
likely than Democrats or Independents in this community to consider the paper
accurate, but the difference is not enough to be clearly significant.
These are the results:
1. How accurate is the in its local news
stories? (4.0)
19.0% Very accurate
74.6 Fairly accurate
4.6 Notso accurate
1.4 Not at all accurate
0.4 Not answered
2. In your experience, do headlines in the give
you an accurate idea of what really happened? (3.7)
13.9% Yes, always
64.8 Yes, usually
15.6 No, sometimes they don't
5.3 No, often they don't
0.4 Not answered
3. If you heard a news item over the radio and then read
a conflicting version of the same story in the _
which one would you believe? (3.3)
6.8% The radio, certainly
56.6 The radio, probable
28.5 The probably
3.5 The certainly
4.2 Don't know
0.4 Not answered
4. How often does the in its news columns and
headlines, try to make a happening sound more exciting
than it really is? (2.8)
4.2% Very often
36.3 Once in a while
45.9 Hardly ever
11.4 Never
1.8 Don't know
0.4 Not answered
Human Worth and Dignity
The publisher of this paper will be pleased to find a dimension in which
there are no really unfavorable attitudes. On this one (see Figure 13) there
are about 35 per cent with no strong feelings pro or con, and 65 per cent who
are definitely favorable to what the paper does.
Answers on this dimension vary somewhat by age. Older people are some-
what more likely to be favorable than young people (Table 10). However the
opinion is so preponderantly favorable that this is not an important
consideration.
-59-
FIGURE 12
Responsibility
For Accuracy
Intensity
Carve
FIGURE 13
Human Worth
And Dignity
Intensity
Ourve
90 *
80 *
P 70 *
I
To
Ee
N n 50 *
St
l i 0
T1
s 30 *
20 *
10
10 20 30 60 $0 60 70 80 90
percentiles
CONTENT
90 *
80 *
70 *
60 .
50 *
h0 *
30 *
20 *
10 *
10 200 0 5o0 o60 70 0o 90
percentiles
CONTENT
These are the responses, by percentage:
43. How often have you read something in the that
gave you a "lift"--something that made you feel that
goodness rules the world more than greed? (2.8)
6.7% Never
51.8 A few times
33.3 Often
4.6 Very often
3.6 Not answered
44. Does it seem to you that the would rather print
bad things about people than good things? (4.1)
31.6% No, absolutely not
53.9 No, not entirely
9.2 Yes, pretty much so
0.3 Yes, to a great extent
1.4 Don't know
3.6 Not answered
45. Do the news items in the suggest that there
are more people in the world with weak character than
with strong character? (3.8)
21.3% Never
52.1 A few times
15.6 Often
2.1 Very often
5.3 Don't know
3.6 Not answered
46. How often does the print something that makes
you feel there are a lot of good people in the world and
not Just a lot of bad people? (3.4)
9.6% Never
51.4 A few times
31.5 Often
2.1 Very often
1.8 Don't know
3.6 Not answered
47. How often does the report events in which the
people involved show qualities of courage or sacrifice? (3.6)
13.4% Very often
56.0 Often
25.9 A few times
0.0 Never
1.1 Don't know
3.6 Not answered
-61-
Political and BEconomic Fairness
In this dimension, as in some others, a very large proportion of the read-
ing audience appears to have no strong feelings one way or the other. Figure 1
(referred to earlier) shows that about 27 per cent are favorable, about 11 per
cent are unfavorable, and about 62 per cent are neutral.
However, the unfavorable readers tend to be Democrats and labor union
members; the favorable ones to be Republicans and nonunion men (Tables 11 and
12). In an earlier study, we found that the liberal readers, separated out by
a liberalism-conservatism scale, tended to evaluate the paper's fairness less
favorably than did the conservative readers. In still another study (see
Table 13), we found a definite relationship between this dimension and length
of residence in the community. Trying to explain this, we discovered that
single ownership of the newspapers had come fairly recently to the community,
and that some years ago there had been in the community another newspaper of
liberal tendencies. Furthermore, one of the newspapers now under single owner-
ship had an editorial writer who was extremely caustic in his references to the
party not represented by the present newspapers. Apparently, therefore, some
of the older residents of the community were looking back to the old days, and
to them the present papers seemed "unfair."
These were the responses, by percentages
5. Do you think the gives as much space to
speeches by a candidate the paper is against as it
does to speeches by a candidate it is for? (3.0)
14.6% No, certainly not
31.0 No, probably not
41.3 Yes, probably
12.4 Yes, certainly
0.3 Don't know
0.4 Not answered
-62-
6. Does the usually present both sides of
important political issues? (3.6)
14.6% Yes, always
60.1 Yes, most of the time
19.6 No, not very often
4.6 No, almost never
0.7 Don't know
0.4 Not answered
7. If the was against a man who was running for
public office, would it be fair to him or not? (3.3)
12.0% Not at all fair
24.5 Not so fair
48.6 Pretty fair
13.4 Very fair
1.1 Don't know
0.4 Not answered
8. Which, if either, has the better chance of getting his
side of a story about a local strike into the _
the head of the business or a labor union official? (3.6)
1.1% The union official, certainly
5.7 The union official, probably
59.2 Equal chance
22.0 The head of the business, probably
10.3 The head of the business, certainly
1.3 Don't know
0.4 Not answered
Confidence in Political Leadership
On this dimension, the newspaper was evaluated lower than on any other
except Independence from Pressure. The intensity plot (Figure 2) shows that
about 18 per cent of readers are favorable, about 42 per cent neutral, and
about 36 per cent unfavorable.
Strangely enough, while Democrats among the readers are clearly less
favorable than Republicans, still the difference is probably not great enough
to be significant (Table 14). Rather, the unfavorable attitude seems to result
from the fact that almost half the readers who have opinions on the subject
seem to feel that a newspaper like this in a one-paper city should not offer
its political advice to the reader. Of the 268 people who answered the
question, "Do you think the should give advice to its readers on who
-63-
and what to vote for?" 132 responded on the negative side, 136 on the positive
side. It was overwhelmingly the negative voters on this question who gave the
newspaper a low evaluation on its Political Leadership. If you want to see
this in more detail, look at Table 15.
Results by percentages:
35. How many people in this area pay attention to the
's advice on how to vote for President
of the U.S.? (2.6)
20.2% Very few people
34.4 A few people
33.7 A lot of people
3.2 Most people
7.1 Don't know
1.4 Not answered
36. If you were not sure about how to vote on a local
bond issue, would you take the advice of the
on how to vote? (2.8)
5.0% Yes, certainly
40.i Yes, probably
36.1 No, probably not
16.7 No, certainly not
0.7 Don't know
1.4 Not answered
37. If you did not understand one of the measures on a
state election ballot, would you take the advice of
the on how to vote? (2.9)
5.0% Yes, certainly
42.9 Yes, probably
34.7 No, probably not
14.9 No, certainly not
1.1 Don't know
1.4 Not answered
38. When the _______ prints an editorial, do you
usually feel it has made a complete study of the
subject being discussed? (3.6)
8.2% Yes, very complete
69.2 Yes, fairly complete
17.0 No, not very complete
2.8 No, not complete at all
1.4 Don't know
1.4 Not answered
-64-
Involvement and Liberalism-Conservatism
We have developed and tested scales to measure a reader's "involvement"
with the paper -- meaning chiefly how much he talked about it -- and to measure
the position of a reader on a base line from most conservative to most liberal.
The latter scale has never proved of much use with this test, although in at
least one community it correlated highly with political party affiliation. The
involvement scale correlates highly with General Satisfaction, Human Worth and
Dignity, and Utility. Both these scales are reprinted in the Appendix.
Summary
Let us now summarize some of the things found out in this test community
about the attitudes different kinds of readers have toward the newspaper.
Age: Older readers saw more Human Worth and Dignity material in the news-
paper than did younger readers.
Religion: Catholics reported less General Satisfaction with the paper
than did Protestants.
Politics: Democrats and sometimes Independents evaluated this independent
Republican newspaper less favorably than did Republicans on Political and
Economic Fairness, and on Authoritarianism.
Labor union membership: Members of a labor union evaluated the newspaper
less favorably than did nonmembers on Independence from Pressure and on
Authoritarianism.
Income: Readers of high income evaluated the newspaper more favorably
than did others on Independence from Pressure.
Number of years a reader: Those who had been reading the newspaper only a
few years evaluated it more highly than did others with regard to Authori-
tarianism. This finding is a little puzzling, and may not be typical of other
newspapers.
-65-
Time spent reading: Those who devoted more time to reading the newspaper
evaluated it more highly on Adequacy of News Content.
In general summary, then, it can be said that a large majority of the
readers of the paper expressed satisfaction with it. They gave it a high grade
on utility, and an extraordinarily high mark on Racial and Religious Fairness.
Most of the attitudinal dimensions were favorably evaluated, although for many
dimensions there was a very large percentage of persons who had no attitude to
express. On three of the dimensions, however, the scores were notably lower
than on the others. These were Political and Economic Fairness, Confidence in
Political Leadership, and Independence from Pressure. We have found out that
it was chiefly the Democrats, the labor union members, and the readers with low
incomes who felt that the newspaper was lacking in these three respects. There
was also a large bloc of readers, by no means all Democrats or union members,
who believed that a newspaper in a one-newspaper town should not urge its
choices for offices and issues on its readers.
Perhaps this amount and kind of reader dissatisfaction is inevitable.
When a partisan newspaper publishes in a one-paper city, it can hardly please
everybody with its political content. The unfavorable attitudes expressed on
these three dimensions may, therefore, be nothing to worry about, particularly
since the General Satisfaction, the Utility, and most of the attitudinal scores
are encouragingly high. But it is at points like this that a thoughtful and
foresighted publisher will begin to review his policy and practice.
APPENDIX TO CHAPTER VI
TABLE 1
General Satisfaction and Religious Faith
or Preference
Score on General Satisfaction
Low Medium High
Number (7-18) (19-25) (26-30) Total
Protestants 195 3% 54% 38% 100%
Catholics 42 21 41 38 100%
Others 17 53 24 23 100%
Total 254
Chi Square for difference between Protestants and Catholics was 38.30,
p=.001, df=4.
Newspapers
in general
TABLE 2
Adequacy of News Content and Evaluation
of Newspapers in General
Score on Adequacy
Low Medium High
Number (7-17) (18-20) (21-25)
50.0% 25.0%
26.0 45.1
17.8 37.0
Not answered 16
df=4.
25.0%
28.9
45.2
Very poor
and poor 20
Good 173
Very good 73
Total 266
Chi Square=13.64, p=.01 with
Total
100%
100%
100%
-67-
Time spent
Less than 19
20-29 mins.
More than 30
Total
Chi Squar
TABLE 3
Adequacy of News Content and Time Spent Reading
Score on Adequacy
Low Medium Higk
Number (7-17) (18-20) (21-2
mins. 57 33.3% 49.1% 17.6
53 32.1 39.6 28.3
ains. 159 20.8 39.6 39.6
269 Not answered 13
e-11.65,
pa.05
with df=4.
TABLE 4
Responsibility for Advertising Content and Evaluation
of Newspapers in General (Question 68)
Scores on Responsibility for Advertising
Low Medium High
Newspapers are: Number (7-12) (13-15) (16-20) Total
Poor, very poor 20 75.0% 15.0% 10.0% 100%
Good 174 24.7 31.6 43.7 100%
Very good 72 13.9 27.8 58.3 100%
Total 266 Not answered 20
Chi Square=33.45, p-.001 with df=4.
!5)
*%
Total
100%
100%
100%
Authoritarianism and
Party
Strong and
Moderate GOP
Strong and
Moderate Demo.
Independents
Total
Chi Square=8.32,
Number
105
95
72
272
p>.02,4.05
Authoritarianism anc
Years a Reader Number
Under 5 years 72
6 to 9 years 44
Over 10 years 150
Total 266
Chi Square=10.21, p=.02 with e
TABLE 5
Political Party Affiliation
Score on Authoritarianism
Low Medium High
(8-15) (16-19) (20-25)
20%
30
24
Not answered
with df=4.
TABLE 6
I Number of Years a Reader
Score on Authoritarianism
Low Medium High
(8-15) (16-19) (20-25)
13% 33% 54%
27 48 25
21 42 37
Total
100%
100%
100%
Total
100%
100%
100%
TABLE 7
Independence from Pressure and Membership in a Labor Union
Score on Independence from Pressure
Low Medium High
nion Number (6-15) (16-19) (20-30)
54 39.0% 37.0% 24.0%
ber 209 24.9 35.4 39.7
l 259 Not answered 22
Chi Square=6.77, pm.05 with
Independence
Income
Under $5,000
$5,000-$6,999
$7,000 and over
Total
Chi Square=16.14,
df =2.
TABLE 8
from Pressure and Income
Score on Independence from Pressure
Low Medium High
Number (6-15) (16-19) (20-30)
86 32.6% 36.0% 31.4%
83 31.3 44.6 24.1
76 22.4 25.0 52.6
245 Not answered 37
p=.01 with df=4.
Total
100%
100%
100%
Labor U
Member
Nonmemb
Tota
Total
100%
100%
TABLE
Responsibility for Accuracy and
9
Political Party Affiliation
Score on Accuracy
Low High
Political Party Number (6-17)
Strong and Moderate
Republicans 105 31.4%
Strong and Moderate
Democrats 95 37.9
Independents 72 48.6
Total 272 Not answered 10
Chi Square=4.83, p=.10 with df=2.
(18-25)
68.6%
62.1
51.4
Human Worth
TABLE 10
and Dignity,
Score on T
Low
Age Number (13-14)
Below 30 97 34.0%
40-49 105 23.8
Over 50 61 13.1
Total 263 Not answered
Chi Square=12.65, p-.05 with df=4.
and Age
o Value the Good
Medium High
(15-20) (21-25)
52.6% 13.4%
50.5 25.7
57.4 29.5
19
Total
100%
100%
100%
Total
100%
100%
100%
-71-
TABLE 11
Political and Economic Fairness and Political Party Affiliation
Score on Political and Economic Fairness
Low Medium High
ical Party Number (4-9) (10-13) (14-20) T(
Strong and Moderate
Republicans 105 5.7
Strong and Moderate
Democrats 95 30.5
Independents 72 15.3
Total 272
Chi Square=25.03, p=.001 with df=4.
Labor
Member
Nonmemi
Toti
31.4% 62.9%
31.6
32.0
Not answered
TABLE 12
Political and Economic Fairness and Labor Union Membership
Score on Political and Economic Fairness
Low Medium High
Union Number (4-9) (10-13) (14-20)
58 44.8% 38.0% 17.2%n
ber 216 28.2 36.1 35.7
al 274 Not answered 8
Chi Square=9.75, p=.01 with df-2.
Polit
total
37.9
52.7
10
100%
100%
100%
Total
100%
100%
%
-72-
Political and
Length of Residence
Less than 5 years
5-9 years
More than 10 years
Total
Chi Squarewl2.36,
TABLE 13
Economic Fairness and Length of Local Residence
in the 5th Study
Score on Political and Economic Fairness
Number Low High
42 44.2% 55.8%
71 60.2 39.8
155 63.5 36.5
268 Not answered
p-.01 with df=2.
Total
100%
100%
100%
18
TABLE 14
Confidence in Political Leadership
and Political Party Affiliation
Score on Confidence in Political Leadership
Low Medium High
Political Party Number (4-9)
Strong and Moderate
Republicans 103 21.4%
Strong and Moderate
Democrats 94 36.0
Independents 71 21.1
Total 268
Chi Square-7.34, p>.10,<.20 with df-4.
(10-13) (14-20) Toi
36.9% 41.7% 100%
30.0 34.0
36.6 42.3
Not answered 14
tal
100%
100%
Should Offer Advice
No
Yes
Total
Chi Square=51.26
TABLE 15
Confidence in Political Leadership
and Opinion About Whether
Newspapers Should Offer
Advice on Candidates
Score on Confidence in Poli
Low Medium
e Number (4-9) (10-13)
132 43.2% 36.4%
136 10.3 31.6
268
), p=.001 with df=2.
Not answer
tical Leadership
High
(14-20) Tol
20.4% 1
58.1 1
red 14
tal
VII
OTHER FINDINGS PROM THE TEST--RBSULTS IN DIFFERENT COMMUNITIES,
RESULTS WHEN THERE ARE COMPETING NEWSPAPERS,
ATTITUDES OF WOMEN, ATTITUDES
OF NEWSPAPER EMPLOYEES
Here we want to present certain other findings from the test. These are
in answer to the questions: How do test results differ from city to city?
What attitudes toward the press does one find in a community where there are
competing newspapers? How do women's attitudes toward the newspaper differ
from men's attitudes? And how do the attitudes of newspaper employees toward
the newspaper differ from attitudes of other readers?
Comparing Newspapers in Different Communities
We have presented at length in this report only findings from our Study
Number 7 because it is the first of the studies in which all of our criteria
for scaling analysis have been met completely. We have presented a few com-
parative findings from the other studies only for illustrative purposes or when
there has been some reason to believe they would add some explanation to the
meaning of the test.
To illustrate how scores might be i-nfluenced by the characteristics of the
community as well as of the newspaper, we have listed in Table 16 a summary of
the scores by dimension for the sixth as well as the seventh study. We have
put this table in the text, rather than in an appendix,because it is easy to
read and because we think even nontechnical readers will be interested in
looking at it.
-74-
-75-
TABLE 16
Comparative Scores for Two Newspapers*
Adequacy of News Content
Morally Objectionable Content
Authoritarianism
Racial and Religious Fairness
Human Worth and Dignity
Responsibility for Advertising
Content
Check Upon Government
Responsibility for Accuracy
Political and Economic Fairr.-as
Confidence in Political Leadership
Independence from Pressure
Average Score
Utility
General Satisfaction
Study No. 6
Mean
Score Rank
4.25 1
4.12 2
3.91 3
3.86 4
3.86 4
3.85
3.83
3.79
3.56
3.45
3.15
3.78
4.24
4.25
Study No. 7
Mean
Score Rank
3.74 2
3.67 4
3.60 6
4.40 1
3.54 7
3.62
3.68
3.54
3.00
2.98
2.97
3.53
3.91
3.83
*Representativeness was combined with
in Study Number 6.
Political and Economic Fairness
Study Number 6 was in a metropolitan community with two competing news-
papers of the same political party preference. The community overwhelmingly
agreed with this party preference. We expected, therefore, that the atti-
tudinal scores in Study 6 would be higher than those in Study 7.
It should be remembered that the scores may not be exactly comparable
because a few of the questions were not exactly the same, and because the
sample in Study 6 was skewed toward the upper income levels. Nevertheless,
the scores do come out as predicted. It will be seen that the average score
for Study 6 is about seven per cent higher than the average score for Study 7.
This seems to us to reflect a greater amount of agreement with the political
position of the paper.
The difference in rank order of the attitudinal dimensions are harder to
explain. It is clear that the newspaper in Study 7 gets distinctly more
approval for its Racial and Religious Fairness and for its activities as a
Check Upon Government. Similarly, the newspaper in Study 6 is more highly
valued for its policy in regard to Morally Objectionable Content, and its lack
of Authoritarianism. Perhaps the most interesting comparison is of the three
lowest dimensions. These are the same for the two papers, and suggest that the
"low grades" we noted for these dimensions in Study 7 may be typical of most
newspapers. This political and economic area is apparently the area in which
the newspaper finds'its readers most suspicious and most alert to real or
fancied failings.
After a considerable number of studies of different newspapers has been
done, we should have something like a norm so that we could compare different
newspapers and different types of newspapers. Even so, we should need to take
into account the political and economic context of each community and the
history of the particular newspaper in that community, before interpreting the
scores of a test in that community.
Competing Newspapers
In one metropolitan community we administered the test to readers of two
competing newspapers, Including a sample of those who read both newspapers.
One was a morning and the other an evening newspaper. Newspaper A had a con-
siderably larger circulation in the city zone (our sample) than did Newspaper B.
We regarded Newspaper A as being superior to Newspaper B in several respects.
Our judgment seemed to be confirmed by the fact that Newspaper A's circulation
in the city zone was considerably larger, apart from the question of readers'
preference for a morning or an afternoon newspaper.
We expected, however, that we would find no differences of attitude among
those who took only one of the newspapers. That is, we assumed that the
reader of one newspaper would evaluate that newspaper at about the same level
that the reader of the other newspaper would evaluate the newspaper of his
choice.
What we found is exhibited in Table 17, which reports percentages of "very
favorable" responses. This table, like the others in this chapter, is easy to
read and therefore is put in the text.
The table shows that on every dimension the reader who read only News-
paper B gave a higher evaluation to that newspaper than the readers who read
only Newspaper A assigned to their newspaper (Columns 1 and 3). The following
differences, however, are not statistically significant: General Satisfaction,
Adequacy, Accuracy, Racial and Religious Fairness, Responsibility for Adver-
tising Content, and Morally Objectionable Content. The other differences are
barely significant statistically.
On the sample of readers who read both papers we had the respondents
answer two questionnaires--one for each newspaper. Thus we had an evaluation
of Newspaper A and of Newspaper B by the same readers who regularly read both.
These differences are presented in Columns 2 and 4 of the table.
-78-
TABLE 17
Readers' Evaluation of Competing Newspapers
(Percentage of "Very Favorable" Responses)
General Satisfaction
Utility
Adequacy of News Content
Responsibility for Accuracy
Independence from Pressure
Confidence in Political Leadership
Political and Economic Fairness*
Racial and Religious Fairness
Check Upon Government
Responsibility for Advertising
Morally Objectionable Content
Authoritarianism
Human Worth and Dignity
Newspaper A
"Only" "Both"
Readers Readers
50% 45%
55 52
33 35
24 20
15 11
10 11
27 22
44 38
21 16
21 20
31 23
38 35
29 26
Newspaper B
"Only" "Both"
Readers Readers
52% 36%
61 51
37 31
25 15
22 10
17 10
35 20
48 35
27 17
24 19
32 19
45 32
36 24
*Includes the Representativeness dimension.
We found that readers of both newspapers tended to evaluate both News-
paper A and Newspaper B at a lower level than did readers who read only one of
those papers. In the case of Newspaper A, the dimensions on which the per-
centages of very favorable responses "went down" were Racial and Religious
Fairness, Morally Objectionable Content, and Authoritarianism. On the other
dimensions the differences are not statistically significant.
The "decline" in evaluation of Newspaper B by readers of both newspapers
was considerably more than for Newspaper A, as the table shows. The largest
-79-
difference was on General Satisfaction--from 52 per cent to 36 per cent. The
"both" percentages were considerably lower than the "only" percentages on
Utility and on all other dimensions except those of Responsibility for Adver-
tising Content and Morally Objectionable Content.
Some of this difference can be accounted for by the differences in the
samples of "only" and "both"readers. The "both" sample includes more readers
of high income and high education. The "both" readers are probably more
knowledgeable and, therefore, more critical. This may be inferred from the
lower percentages for Newspaper A as well as the lower percentages for Newz-
paper B. The lower evaluations by "both" readers of Newspaper B than of
Newspaper A appear to confirm our a priori judgment that Newspaper A is
superior to Newspaper B in several respects.
We summarize the findings as follows:
1. Readers who read only Newspaper B evaluate their newspaper as highly
as, or a little higher than, do readers who read only Newspaper A.
2. Readers of both newspapers evaluate both newspapers lower than do
those who read only one of the newspapers.
3. Readers of both newspapers evaluate Newspaper B considerably lower
than they do Newspaper A.
In a second metropolitan community we tested readers of the two newspapers
separately but asked readers to express their attitudes about just one
newspaper--the one they read. We did not ask the readers' of both newspapers to
answer two questionnaires. Except for a slight difference on Utility, the
differences between the two newspapers were not significant. Both are regarded
as good newspapers and the city zone circulations are about the same; both
support the same political party. We had expected that we would find no
significant differences.
-80-
The Attitudes of Women
In three communities we administered the test to women as well as men.
Women consistently scored higher than men on most dimensions. Table 18 shows
these differences as found in Study Number 4 in which the newspaper had a
circulation of 90,000. The data represent the percentage of respondents who
TABLE 18
Combined Favorable and Very Favorable Attitudes
of Men and Women in Study Number 4
(in percentages)
Men Women
General Satisfaction 77% 86%
Adequacy of News Content 79 84
Responsibility for Accuracy 84 84
Check upon Government 79 82
Independence from Pressure 54 57
Political and Economic Fairness* 69 78
Morally Objectionable Content 88 88
Racial and Religious Fairness 85 86
Responsibility for Advertising Content 75 81
Confidence in Political Leadership 58 63
Authoritarianism 72 77
Human Worth and Dignity 84 90
*Includes the Representativeness dimension.
were "very favorable" or "favorable." The scores for women are considerably
higher on General Satisfaction and on Political and Economic Fairness, and are
somewhat higher on five other dimensions. Many of the small differences may
be due to chance. The sample size, however, was in excess of 400 men and 400
women.
Employees' Attitudes
In the last four studies we administered the test to the employees and in
one study to editorial executives. Table 19 presents the attitudes of editorial
employees, the same editorial employees' predictions of readers' responses, and
the actual responses of men readers. These figures are from Study Number 5.
TABLE 19
Attitudes of Bditorial Bnployees in Study Number 5:
Their Own Attitudes and Their Predictions
of Readers' Attitudes
(in percentages)
Own Predicted
Responses Responses*
General Satisfaction 96% 94%
Utility 89 83
Adequacy of News Content 99 98
Responsibility for Accuracy 98 91
Authoritarianism 89 72
Independence from Pressure 65 33
Political and Economic Fairness** 90 69
Racial and Religious Fairness 98 91
Check upon Government 94 78
Responsibility for Advertising Content 92 78
Confidence in Political Leadership 75 68
Morally Objectionable Content 95 87
Human Worth and Dignity 89 74
Readers'
Responses
95%
93
95
89
85
55
80
97
81
83
61
92
87
*As they think the average reader would respond.
**Includes the Representativeness dimension.
-82-
Editorial employees' own attitudes are consistently higher than the
readers' attitudes, except on utility. The smaller differences, of course,
will not be statistically significant.
Editorial employees underestimated readers' favorable attitudeson seven of
the dimensions and on Utility. They overestimated readers' favorable attitudes
on Confidence in Political Leadership. They matched readers' responses on
General Satisfaction and Adequacy of News Content.
We present no explanation of these findings, for we are not quite certain
what they mean. Under some circumstances, it is well known that respondents
project their own attitudes when they are asked to estimate how other people
feel about a certain matter--for example, contributing to the Community Chest
Fund. In the present situation, the editorial employees were asked to state
their own attitudes on one questionnaire and to estimate readers' attitudes on
a second questionnaire. In another study, in which we asked editorial em-
ployees only to predict readers' attitudes without stating their own attitudes,
their scores were consistently and considerably lower than the actual responses
of readers. Yet other employees in that study virtually duplicated the
attitudes of readers.
Table 20 shows the attitudes of editorial, advertising, and circulation
employees in Study Number 4. Let us look first at the differences among the
different kinds of employees. Editorial employees have somewhat more favorable
attitudes than do advertising employees and about the same attitudes as circu-
lation employees. They are lower than either of the others on Confidence in
Political Leadership but--as would be expected--higher on Accuracy. Circu-
lation employees--as might be expected--are higher than the others on Adequacy
of News Content. Advertising employees are no higher than other employees on
Responsibility for Advertising Content.
-83-
TABLE 20
Attitudes of Editorial, Advertising, and Circulation Employees,
and Predictions of Key Editorial Personnel
in Study Number 4
Key Readers'
Edit. Adv. Circl. Pers.* Responses
General Satisfaction 76% 76% 73% 50% 77%
Adequacy of News Content 81 80 90 43 79
Responsibility for Accuracy 95 85 92 57 84
Authoritarianism 82 72 82 57 72
Independence from Pressure 67 64 61 18 54
Political and Economic Fairness** 69 70 70 16 69
Racial and Religious Fairness 89 82 88 66 85
Check upon Government 85 81 87 51 79
Responsibility for Advertising 90 87 89 44 75
Confidence in Political Leadership 55 68 60 29 58
Morally Objectionable Content 93 87 94 70 88
Human Worth and Dignity 87 84 89 60 84
*As they think the average reader would respond.
**Includes the Representativeness dimension.
Responses of advertising employees just about match those of readers.
Responses of circulation employees are generally higher than readers'.
Responses of editorial employees are higher than readers on Accuracy, Inde-
pendence from Pressure, Check on Government, and Responsibility for Advertising
Content. Small differences should be regarded as not statistically significant,
especially since the number of employees was not large although nearly all
employees received the test.
-84-
We also asked key editorial personnel to predict readers' responses.
Their estimates were 20 to 75 per cent lower than readers' actual responses.
We have no explanation for such large discrepancies, except that (1) the
editorial executives had a self-critical attitude and (2) they were afraid
they might overestimate readers' attitudes. Their responses may indicate a
defensive attitude.
We had few hypotheses about employee attitudes when we administered the
tests. More investigation in this area would be worth while provided the
investigator developed some additional instruments that would relate person-
ality characteristics, social attitudes, and some other variables to expressed,
predicted, and projected attitudes toward the newspaper, the newspaper
profession, and job satisfaction.
VIII
TOWARD AN IMPROVED UNDERSTANDING AND OPINION OF THE PRESS
As some of the findings we have presented show, there is an appalling
ignorance about newspapers -- their objectives, motivations, and procedures.
The intellectuals are apparently as ignorant as the rest of the lay public.
And on the other side of the fence, the key employees of the newspaper seem to
be farther off than other employees in their estimates of what the public
thinks of the press.
It is to the advantage of both press and public to decrease this area of
misunderstanding. It is to the advantage of the newspaper to educate the
public as to its objectives, motivations, and procedures. And it is of advan-
tage to the community for its leaders to know more about the newspaper's
intentions, aspirations, and methods of operation. This is especially
desirable when the newspaper is a vigorous leader and therefore is in disagree-
ment with a segment of the community on almost any important issue.
The responsibility for better understanding rests with the press itself,
with the schools, and with every citizen. Here we shall talk mostly about the
press, because in these pages we have presented a measuring instrument by which
publishers can inform themselves more accurately than before of community
attitudes toward their newspapers.
The question is, what does a publisher do when he has used this Test of
Public Attitudes toward the Press, and discovered not only areas of ignorance
concerning the newspaper, but also attitudes which are not so favorable as he
would like?
-85-
-86-
The test may give him reason to review some of his policies and practices.
It will probably make him more willing to do what he can to reduce public
ignorance and misunderstanding of the press. For example, he may want to do a
better job with groups and organizations that visit his plant, and in supplying
speakers and exhibits to the schools. He may make more use of his own paper to
explain his own paper. And we want to make another suggestion.
The device of Newspaper Day in October is not very effective, although it
has some value. What is needed is a way to bring newspaper executives and
community leaders together to 'discuss certain community problems. One device
that has merit for the nonmetropolitan community is a Citizens' Advisory
Council. This was tried with some success in one community (see Editor and
Publisher, July 28, 1951, p. 36). The publisher selects for a one-year term a
number of citizens representative of the interests in the community and is host
at a quarterly dinner. The group discusses community problems, including the
newspaper's attitude toward some of the problems. The meeting supplies an
opportunity both for the publisher to learn more about the problems under dis-
cussion and for the individual citizen to understand the motivations and the
limiting operating procedures of the newspaper. In this situation--as one of
the present writers observed at such a meeting--the publisher is seldom on the
defensive: certain members of the Council usually come to his defense so that
he is residually the beneficiary of a lot of good advice. For the successful
operation of such a Council, the publisher should (1) change the membership
after one year, and (2) have it understood by the members that he is not
necessarily bound by their advice.
Just as democratic government is more efficient and more rewarding for the
individual constituents than is dictatorship because criticism flows both ways,
so is the newspaper a better social institution when it solicits criticism and
advice from the responsible sections of the community. For, in the long run,
there must be a press in which readers have confidence.
TECHNICAL APPENDIX
Appendix A
DIMENSIONALITT AND RELIABILITY OP THE SCALES
Determination of Dimensionality
The area tests were formed by grouping or writing questions to fit defi-
nitions of the areas. This procedure was described early in Part I. Implicit
in this operation was the hypothesis that questions so grouped formed homo-
geneous tests of single dimensions. The Guttman scale analysis technique (5)*
was the method used to test the hypothesis of unidimensionality for each scale.
The analysis, performed separately for each scale, consists of a graphic
arrangement of the response data, which then must meet certain criteria, by
question and in total, to be judged a unidinensional scale. The process, based
on the Cornell technique developed by Guttman, was followed in analyzing all
scales.
An Example of Scale Analysis
In order to illustrate the process of scale analysis by the Cornell tech-
nique and to demonstrate the application of several criteria of scalability, an
example is presented.
The initial step is scoring of the tests, using a set of arbitrary item
weights** ranging from "4" for the most favorable response category down to "1
for the least favorable category. Each individual is then given a score on the
*Reference numbers refer to appropriate articles or books listed in the Bibli-
ography on page 126.
**The weights used in this example scale differ slightly from those used in the
scales reported in Part I. In those scales the response weights ranged from
1 to 51 the example scale shows only four categories. In most reported
Guttman scales, items are followed by five response categories.
-90-
scale, the score being the sum of the weights for his response to each item.
Thus, in Table A, Case No. 1 had a score of 24 (second column), made up of a
"4", the most favorable response to each of the six items.
The second step is the arrangement of individuals in order of total score,
from high to low. Bach individual's total score and his numerical response to
each item are recorded in tabular form. The individuals form rows and the
items form columns in the table. This arrangement of data is shown in Table A.
TABLE A
Arrangement of Data for Scale Analysis
Case Original New Response to Items
No. Score Score 1 2 3 4 5 6
1 24 (6) (1) 4 (1) 4 (1) 4 (1) 4 (1) 4 (1) 4
2 20 (6) (1) 4 (1) 4 (1) 3 (1) 3 (1) 3 (1) 3
3 19 (5) (1) 4 (1) 3 (0) 2 (1) 3 (1) 3 (1) 4
4 19 (6) (1) 4 (1) 3 (1) 3 (1) 3 (1) 3 (1) 3
5 18 (5) (0) 3 (1) 3 (1) 3 (1) 3 (1) 3 (1) 3
6 18 (4) (0) 3 (1) 3 (0) 2 (1) 3 (1) 4 (1) 3
7 18 (5) (0) 3 (1) 3 (1) 3 (1) 3 (1) 3 (1) 3
8 17 (4) (0) 3 (1) 3 (0) 2 (1) 3 (1) 3 (1) 3
9 17 (4) (1) 4 (1) 3 (0) 2 (1) 3 (1) 3 (0) 2
10 17 (4) (0) 3 (1) 3 (0) 2 (1) 3 (1) 3 (1) 3
11 17 (5) (1) 4 (1) 3 (0) 1 (1) 3 (1) 3 (1) 3
12 16 (3) (0) 3 (0)T (0) 2 (1) 3 (1) 3 (1) 3
13 16 (3) (0) 3 (0) 2 (0) 2 (1) 3 (1) 3 (1) 3
14 15 (2) (1) 4 (0) 2 (0) 1 (0) 2 (1) 3 (0) 2
15 15 (3) (0) 3 (1) 3 (0) 1 (0) 2 (1) 3 (1) 3
16 15 (2) (0) 3 (0) 2 (1) 3 (0) 2 (0) 2 (1) 3
17 14 (2) (0) 2 (0) 2 (0) 2 (1) 3 (1) 3 (0) 2
18 13 (1) (0) 2 (0) 2 (0) 2 (0) 2 (0) 2 (1) 3
19 12 (0) (0) 3 (0) 2 (0) 1 (0) 2 (0) 2 (0) 2
20 10 (0) (0) 2 (0) 2 (0) 2 (0) 1 (0) 1 (0) 2
The system used here departs from the Cornell technique in the method of re-
cording data. In the Cornell technique, each item is represented by four
columns, one for each response category, and responses are recorded by placing
an "X" in the appropriate column. As modified here, each item is represented
by only one column and the item responses are recorded numerically. This
change makes it practical to use IBM equipment in the analysis.
|
