Sources in New York Times' Coverage of China
Before and After June 4th, 1989
A Content Analysis Focusing on Influence of Crisis Situation On Sourcing
Patterns of US Media in World News Coverage
Paper Submitted to AEJMC Newspaper Division for MacDougall Competition 1998
By Guoli Li
E. W. Scripps School of Journalism
Ohio University
Contact: Guoli Li, Scripps Hall, Ohio University, Athens, OH 45701
Phone: (740) 594-5588 Email: [log in to unmask]
Sources in New York Times' Coverage of China
Before and After June 4th, 1989
A Content Analysis Focusing on Influence of Crisis Situation On Sourcing
Patterns of US Media in World News Coverage
Abstract
This is a study of the New York Times' coverage of the students'
movement in China in 1989 and focuses on the influence of this political
incident on the sourcing patterns of the New York Times when covering China. By
comparing the New York Times' coverage of China in four time periods both before
and after the students' demonstration, the author finds that the crisis
situation impacted the sourcing patterns of the New York Times both during and
after the event and has a lingering effect.
Introduction
Unexpected events such as wars, riots, disasters and other drastic social
upheavals have been fodder pursued by hungry mass media in most countries,
wherever the events occurred. Such events also provide communication researchers
with opportunities to study the performance of mass media, because they often
enable researchers to focus studies for various purposes. The students'
demonstration in China in 1989 was one such event. This study attempted to find
out what sources were used by the western press, in this case the New York
Times, during the crisis situation, how they were used and what is the effect on
the sourcing patterns thereafter. It attempts to add to the understanding of the
dynamics of the media over time and in a particular historical context.
Literature Review
Numerous studies have demonstrated that the elites, or the powerful, have
been the dominant sources in the media of both the United States and other
counties, be they totalitarian states or democracies. Elite sources supply most
of the information in both national and state newspapers in the United States
(Brownet). A study of the news source use in the stock crash of 1987 found that,
even during such a crisis situation, news media favor prestige sources (Losorsa
and Reese). Few studies have concentrated on the use of sources during such
unexpected or unusual events as wars, riots, natural disasters or other violent
social upheavals in a place other than the host country of the reporters.
Studies that dealt with such issues failed to look at how such situations
impacted the sourcing patterns of foreign press stationed in the host countries
of the events in the future.
So far, the majority of the studies of the coverage of world events have
focused on the content, which is also a part of this study. A study of the New
York Times, ABC, NBC and CBS showed that only events that are deviant from
American values or of economic and political significance to the US are covered.
"World events presented mostly in the news were deviant events and those with
economic or political significance to the US. The next most prominent events
were deviant, but with low significance" (P. Shoemaker, et al.). Dissidents and
demonstrating students apparently fit into this category. Yet, literature review
shows that few studies have been devoted to the effect of such skewed coverage
on the sourcing patterns of the media themselves. One study found that anonymous
attributions are used in 80 percent of both national and international news
coverage in Time and Newsweek. More international stories than national stories
contain anonymous attribution. The most frequently quoted anonymous source was
some type of official, followed by an aide, diplomat, expert, staffer and
advisor (K. T. Wulfemeyer).
Rationale, Questions and Hypotheses
The author has several reasons for choosing this event as the subject of
study. First, it is an unexpected event with shocking effects and aftermath.
Second, the host country, China, has a drastically different political system
and cultural values than the United States. Third, the special relationships
between the governments and media of the United States and China may enhance the
informativeness of the findings produced from the study.
Both news content and the sourcing pattern were taken into consideration
when the study was designed. What issues received the heaviest coverage? What
changes have taken place in the content of US media's coverage of China after
the Tian'anmen Square crackdown? Who were the most frequently quoted sources
during the movement and how were they treated by the press? What changes have
taken place in the sourcing patterns both during and after the incident? These
are the major questions this study attempted to address.
H1: The coverage of politics and government-related issues and human rights
increased drastically during the students' movement and remained dominant in the
coverage of China thereafter.
H2: The citation of non-government sources saw a sharp increase during the
crisis situation. Sources that rarely appear in the routine coverage entered the
press. Students and dissidents became dominant sources in both 1989 and 1997,
which were treated as unusual years.
H3: The frequency of anonymous sources grew substantially in 1989.
The study is also expected to provide insight into other related fields.
Method
The New York Times' coverage of China in four years, 1985, 1989, 1993 and
1997 were studied. Four weeks were constructed for 1985 and 1993 respectively,
which were treated as normal years. The other two periods were 28 consecutive
days in 1989 with the June 4 Incident being the midpoint and 28 consecutive days
in 1997 with October 28 being the midpoint. Chinese President Jiang Zemin
started a state visit to the United States on October 28,1997. This time period
was chosen as a reference framework in an attempt to show if the visit, which
happened when the memory of the Tian'anmen Square crackdown remained fresh,
might have caused any repercussion of the 1989 movement and thus affect the
source use. Only stories that are datelined in mainland China were coded.
Paul J. Deutschmann's category system was modified to define story types in
this study. Some points that need clarifying: round-ups and news features, in
spite of their content, were categorized as "general human interest" stories.
Education, sports, culture and public welfare were collapsed into one category.
"Human rights" was singled out as one category. Stories about the students'
movement and dissidents were put into this category.
Sources were grouped into either government or non-government sources. A
source was either identified or unidentified. Ambiguous or mixed sources were
coded as non-government sources. Some points that need clarifying: foreign
government refers to any government other than the Chinese government and
international organizations. "Official" refers to any Chinese who happened to be
an official but did not represent the government when appearing in the news.
Journalists were put into this category because of their de facto official
status in China. Teachers, students, scholars and artists were grouped into one
category.
The sampling and coding of the New York Times was done at the microfilm room of
a major midwestern university in February and March 1998. Before the coding work
began, 12 days, or more than ten percent of the sampled days, were randomly
pulled from 1985 and 1997 and 16 valid cases were found. Three graduate students
were trained to code the stories at the same time. The intercoder reliability,
based on percentage of agreement, ranged from 92 percent to 100 percent. The
overall intercoder reliability was 96 percent.
Findings
News content: H1 was confirmed. A total of 129 stories were found in the
sampled issues of the New York Times, more than 50 percent of which appeared in
the four consecutive weeks in 1989, or the crisis situation period. As
hypothesized, the coverage of politics and governments and human rights
dominated the media content. It is not surprising considering the historical
situation and the media performance at that time. Table 1 shows that, of the 129
stories, 42 or 32.6% appeared on the front page. In 1989 nearly half of the
stories appeared on the front page. Politics and government received the
heaviest coverage, constituting 25.6% of all the stories. Human rights received
the second heaviest coverage, constituting 23.3% of the total coverage.
Table 1
year
No of stories
Front page
Economic activity
Politics
Human interest
Education
Crime
Accident
Science
Human rights
Diploma-
tic affairs
Miscella-neous
1985
17
1, 5.8%
1
1989
84
37, 44%
1
14
7
13
1
1
1993
17
2, 11.8%
1
1
1997
11
2, 18%
1
1
Total
129
42
2
15
8
14
2
1
X2 = 7.936; df = 3; p ( .005
Table 2
Value label
Frequency
Valid percent
Economic activity
12
9.3
Politics and government
33
25.6
General human interest
26
20.2
Education, sports, culture
3
2.3
Crime
2
1.6
Accident and disaster
1
3.1
Science and invention
7
.8
Human rights
30
23.3
Diplomatic affairs
10
7.7
Miscellaneous
5
3.87
Total
129
100
The focus on politics and human rights is in conformity with previous studies
of US media's coverage of world events.
Sources: H2 and H3 were mostly confirmed. A total of 383 sources were found
to be used, consisting of 224 non-government sources and 159 government sources.
As Table 3 shows, the percentage of government sources dropped from 84.4% in
1985 to 36.5% in 1989, then climbed from 37.8% in 1993 to 52.3% in 1997. The
changes in the percentage of non-government sources goes in the opposite
direction, increasing sharply from 15.6% in 1985 to 63.5% in 1989, then sliding
to 62.2% in 1993 and further to 47.7% in 1997. It shows that the sourcing
pattern during a crisis situation does have a lingering effect, at least in this
case. The crisis situation brought about an abrupt change in the sourcing
patterns, which remained relatively stable over a period of time and nearly
returned to the original pattern gradually. How strong is the effect, how long
the effect lasts and in what manner it changes needs further study.
Table 3
Year
1985
1989
1993
1997
Total
Government
27, 84.4%
107, 36.5%
14, 37.8%
11, 52.3%
159
NonGovmnt
5, 15.6%
186, 63.5%
23, 62.2%
10, 47.7%
224
Total
32
293
37
21
383
X 2 = 29.43; df = 3; p ( .0000
In terms of the attribution of the sources, nearly half of the sources are
unidentified or unclear. One hundred and eight one sources or 47.3% of the total
are unidentified. Table 4 shows that both the frequency and percentage of
unidentified sources are higher during the crisis situation, comprising 56% of
the sources used in 1989. Further analysis of the data showed that more
non-government sources than government sources are anonymous and more foreign
government sources than Chinese government sources are anonymous. Among the most
cited anonymous sources are western diplomats, other Chinese and unconfirmed
sources or reports. Forty cases, or 30.25%, of the first sources cited in the
stories are anonymous. The high frequency of anonymous sources can be
interpreted in various ways. A New York Times reporter wrote within a week of
the June 4 crackdown in the Tian'anmen Square that "confirming information has
been particularly difficult these days because sources are drying up in a time
of political retrenchment (Nicholas D. Kristof)."
Table 4
Year
1985
1989
1993
1997
Total
Identified
26
128
32
16
202
Unidentified
6
165
5
5
181
Total
32
293
37
21
383
X 2 = 40.58; df = 3; p ( .0000
Interesting changes were found in the use of non-government sources. As
many source studies have found, such elite or empowered people as government
officials and executives dominate the sources. This study partly confirmed this
contention. In this study, Chinese and foreign executives combined comprised
66.7% of the non-governmental sources in 1985 and 47.8% in 1993, which were
treated as normal years. They were the most frequently used sources in the two
years. The percentage was 30% in 1997, also the highest of the non-government
sources. While in 1989, the percentage of executives among the non-government
sources was merely 5.4%, which is stunningly low considering the heavy media
coverage during that period.
Table 5
Year
1985
1989
1993
1997
Total
Executive
3, 66.7 %
10, 5.4 %
11, 47.8 %
3
27
Employee
0
11
1
1
13
Dissident
0
7
0
1
8
Teacher, student,
0
54, 29 %
4, 17 %
1
59
Armyman
0
7
0
0
7
Bureaucrat, journalist
0
11
2
1
14
Other Chinese
1
49, 26.3 %
2
2
54
Other foreigner
1
23
3
2
28
Unclear
0
14
0
0
14
Total
5
186
23
10
224
As hypothesized, teachers, students and scholars were the most frequently
used non-government sources in 1989, jumping from none in the sample of 1985 to
29% of all the sources in 1989. The percentage of the teachers, students and
other scholars of the non-governmental sources in 1993, four years after the
students' demonstration, was 17%, second after executives, but then it dropped
to 10% in 1997. The hypothesis that Chinese President Jiang Zemin's visit to the
United might impact, or increase, the frequency of students, scholars and
dissidents appearing in the US media was not confirmed by this study. A
plausible explanation is that only stories dated in China were coded and
analyzed, while most news coverage of the visit originated in the Untied States.
Ten cases were found during this period. A brief review of the New York Times'
coverage of the visit during the period will demonstrate another picture.
Summary and Discussion
These data suggest that a crisis situation does impact both the media
content and the sourcing patterns concerning related issues. The greater the
impact, the slower it tails off. There are complex factors underlying the
changes.
Obviously, further research of similar events and subjects needs to be done
to substantiate the findings and further studies might bring to light more
enlightening findings as to such questions as how the sources were used or what
they said. There is sound reason to believe that how the sources are used will
provide more insight into the understanding of the media source use than the
mere fact that they are used.
The frequency of other Chinese is the second highest in the non-government
sources. It took up 26.3 percent of the non-government sources in 1989. The
frequency of unidentified sources in this category is very high. One plausible
explanation is the factor of the accessibility of the sources. A further
breakdown of this category might show us something interesting.
How long will be the effect of the crisis situation on the sourcing pattern
remains a largely unanswered question. Although the study shows that there are
obvious changes in the patterns of source use before and after the crisis
situation, more samples need to be studied to see how the changes have taken
place from year to year.
As Guido H. Stempel III and Hugh M. Culbertson stated in a source study
more than a dozen years ago, dominance and prominence might both reflect source
assertiveness, credibility (as viewed by reporters), accessibility, quotability
and other factors. The study of the "other factors" will help us better
understand the use of sources.
References
Dominic L. Losorsa and Stephen D. Reese, "News Source Use in the Crash of
1987: A Study of Four National Media," Journalism Quarterly 67:1: 60-71
Guido H. Stempel III and Hugh M. Culbertson, "The Prominence and Dominance
of News Sources in Newspaper Medical Coverage", Journalism Quarterly Vol. 61:
671-76 (1984)
Jane Delano Brown, et al, "Invisible Power: Newspaper News Sources and the
Limits of Diversity", Journalism Quarterly 64: 45-54 (1987)
K. Tim Wulfemeyer, "How and Why Anonymous Attribution is Used by Time and
Newsweek", Journalism Quarterly Vol. 62: 81-86 (1985)
Nicholas D. Kristof, "Seeing shadows where once there were leaders, facts
and informants," New York Times June 9, 1989 A 11
P. Shoemaker, Lucig H. Danielian and Nancy Brendlinger, "Deviant Acts,
Risky Business and US Interests: The Newsworthiness of World Events", Journalism
Quarterly Vol. 68: 781-95
P. Shoemaker, "Bias and Source Attribution", Newspaper Research Journal
l:5: 25-31 (Fall 1983-Summer 1984)
|
