Air Pollution Coverage, Anti-Chinese Sentiment, and Attitudes Towards Foreign Policy in South Korea

Abstract

Unlabelled:
Air pollutants allegedly originating from China have become a thorny issue in South Korea. Despite a neutral view of the topic on the part of the South Korean government, recent public polls show a high correlation between the air pollution issue and negative sentiment toward China. How has the media reported on China regarding air pollutants in South Korea? What is the effect of media reports on air pollution on anti-Chinese sentiment and foreign policy attitudes? By examining news headlines and Twitter data in 2015 and 2018, this work finds that media reports blaming China for air pollution doubled during the 2015-2018 period. Discourse surrounding air pollution also shifted: negative sentiment directed at both the Chinese government and the Chinese people increased in 2018 compared to 2015. In addition, an original online survey experiment shows that China-blaming articles have a causal effect on increasing related resentment, particularly toward Chinese people, and that this effect is moderated by age group. Such articles have also had negative effects on foreign policy attitudes via increased anti-Chinese sentiment; greater hostility toward the Chinese people is found to have a causal effect on reduced support for strengthening relations with their country.

Supplementary information:
The online version contains supplementary material available at 10.1007/s11366-023-09849-z.

Full text

RESEARCH ARTICLE
Journal of Chinese Political Science
https://doi.org/10.1007/s11366-023-09849-z
Abstract
Air pollutants allegedly originating from China have become a thorny issue in
South Korea. Despite a neutral view of the topic on the part of the South Korean
government, recent public polls show a high correlation between the air pollution
issue and negative sentiment toward China. How has the media reported on China
regarding air pollutants in South Korea? What is the eect of media reports on
air pollution on anti-Chinese sentiment and foreign policy attitudes? By examin-
ing news headlines and Twitter data in 2015 and 2018, this work nds that media
reports blaming China for air pollution doubled during the 2015–2018 period. Dis-
course surrounding air pollution also shifted: negative sentiment directed at both the
Chinese government and the Chinese people increased in 2018 compared to 2015.
In addition, an original online survey experiment shows that China-blaming articles
have a causal eect on increasing related resentment, particularly toward Chinese
people, and that this eect is moderated by age group. Such articles have also had
negative eects on foreign policy attitudes via increased anti-Chinese sentiment;
greater hostility toward the Chinese people is found to have a causal eect on re-
duced support for strengthening relations with their country.
Keywords Sino-Korea relations · foreign policy attitudes, · environmental
pollution · public opinion
Accepted: 24 February 2023
© The Author(s) 2023
Air Pollution Coverage, Anti-Chinese Sentiment, and
Attitudes Towards Foreign Policy in South Korea
Esther E.Song1
Esther E. Song
esther.song@giga.hamburg.de
1 German Institute for Global and Area Studies, Institute for Asian Studies,
Rothenbaumchaussee 32, 20148 Hamburg, Germany
1 3

E. E. Song
Introduction
In 2022, China-South Korea relations entered their 30th anniversary. The two coun-
tries have enjoyed increased economic and cultural exchange since the establishment
of these ties in 1992. Since 2004, China has been South Korea’s biggest trading part-
ner, with a volume thereof reaching USD 0.14 trillion in 2021 [36]. Chinese immigra-
tion to South Korea makes up the largest proportion of such arrivals, accounting for
43.6% [38]. Aside from economic partnership, China also plays an important role in
determining peace in the peninsular region by having the greatest degree of leverage
over North Korea.
This article covers a relatively new but seldom examined issue that has become a
central topic informing anti-Chinese sentiment in South Korea since the mid-2010s:
air pollution. Although it is known that the South Korean public’s attitude toward
China worsened due to the Northeast Project (dongbukgongjeong) in 2004 and the
Terminal High Altitude Missile Defense (THAAD) crisis in 2016, more recently air
pollution has become the China-related issue with the highest negative ratings among
the country’s citizens. For example, according to a survey conducted by SisaIn in
April 2021, the top reason for an unfavorable view of China is air pollution [39]. The
latter has become a key topic impacting South Korea-China relations since the mid-
2010s due to increased controversy over the source of this pollution. While promis-
ing stronger measures against China to the domestic audience, the South Korean
government has taken a mild stance on the issue, merely requesting cooperation from
the Chinese government. For instance President Moon Jae-in, during his presidential
race in 2017, mentioned that he would “discuss the matter of air pollutants in the
China-Korea summit” [30], being later criticized for failing to take a stronger posi-
tion regarding the issue.
A backlash from civil society then ensued: the Air Pollutant Informant (misemeonji
allimi) [1], a Facebook group created in April 2017, has been dedicated to attribut-
ing blame to the Chinese government and criticizing South Korean domestic experts
who advocate for shared responsibilities between the two countries. In March 2018,
an online petition was led on the Blue House website that demanded the South
Korean government seek resolution from its Chinese counterpart, with it receiving
270,000 signatures [3]. In response, both parties agreed to establish the China-Korea
Environmental Cooperation Center [14]. Yet in 2019, Lu Kang, the spokesperson
for the Ministry of Foreign Aairs in China, denied that the latter was the source of
such pollution [68], sparking public protests by conservative groups in front of the
Chinese Embassy in Seoul [35].1
Regarding the increased anti-Chinese sentiment circulating among the South
Korean public, some have raised the question of whether the media has exacerbated
1 These groups included Freedom Union (jayuyeondae), Freedom Korea National Defense Corps (jayu-
daehan hogukdan), and Turn Right. Far-right groups and the Liberty Korea Party (LKP, later merged into
the People’s Power Party) used anti-China slogans as part of protests against the South Korean government
in 2018 and 2019. For example, in a large-scale protest waged by the LKP in 2019, slogans included “Anti-
China & Pro-U.S.,” while the focus of the rally was largely anti-President Moon [40]. Such anti-China
slogans were also part of protests focused on criticizing the Moon administration’s foreign policy, which
leaned toward being pro-China.
1 3

Air Pollution Coverage, Anti-Chinese Sentiment, and Attitudes Towards…
such animosity: in 2019, Li Ganjie, head of the Ministry of Ecology and Environment
in China, told Cho Myung-rae, head of the Ministry of Environment in South Korea,
that the latter’s media exaggerates the role of China in contributing to air pollution in
South Korea [31]. Likewise, previous works have shown that media outlets reporting
on air pollution are biased against China [34, 55]. Inspired by these discussions, this
article examines how the media has reported on China and domestic air pollution in
South Korea, and whether this has contributed to increasing anti-Chinese sentiment
and less benign foreign policy attitudes. Despite the importance of this topic, there
is currently no work that empirically examines the link between reporting patterns
regarding air pollution and anti-Chinese sentiment, or the implications of this for
views on the two countries’ relations. By combining text analyses of social media
data and a survey experiment, the causal link between media reporting on air pollu-
tion, anti-Chinese sentiment, and foreign policy attitudes is thus examined.
The ndings hereof contribute to the growing literature highlighting the intersec-
tion between the media, public opinion, and policymaking [43, 62]. Regarding for-
eign policy, the International Relations scholarship has discussed how public opinion
can aect related decisions: it has been found that in democracies, leaders can be pun-
ished through elections and therefore they care about public opinion when it comes to
foreign policy [45, 57]. Extant works have noted that the media acts as a “conveyor
belt” of information and frames an event in ways consistent with elite rhetoric [7, 70].
The media thus acts as a passive channel transmitting elite views to the general pub-
lic, and so helps harden the latter’s attitudes in line with those of the state. However,
a growing body of literature has also divulged that the media can play a proactive role
in shaping public opinion in ways impacting foreign policy too. Entman showed that
the media is likely to have an independent inuence on foreign policy when leaders
debate the frames for a given course of action [17]. More recently, the proactive role
of the media in inuencing foreign policy has been highlighted in relation to the rise
of new media outlets [6]. By examining how China’s contribution to air pollution in
South Korea has been reported on, and the eects of this on anti-Chinese sentiment,
this article contributes to the burgeoning scholarship emphasizing the proactive role
of the media in shaping foreign policy attitudes.
This is also broadly connected with the increasing work on xenophobia and its
consequences. Recent studies have found that governments, media, and conspiracy
theories can create narratives targeting specic groups of people, inciting xenophobic
sentiment [27, 42, 52]—as also linked to nationalism [22, 67]. Such resentment has
repercussions for various government preferences, including vis-à-vis immigration
[21], foreign policy [41], and trade [37]. In the context of South Korea, anti-Chinese
sentiment has been relatively understudied compared to anti-Japanese or anti-Amer-
ican views, which have already been found to have strong eects on foreign policy
[13, 49, 53]. The scholarship has viewed anti-Chinese sentiment in South Korea as
weaker than its anti-Japanese counterpart due to dierent historical experiences with
the countries involved [15]. Yet due to the aggravation of anti-Chinese sentiment in
recent years, reaching an all-time high in 2022 [2, 60], the phenomenon has begun
to attract greater academic attention—particularly in relation to the 2022 presidential
elections as well to the security implications of such resentment [58, 59, 64].
1 3

E. E. Song
The use of social media data to examine trends in anti-Chinese sentiment in South
Korea has both its advantages and its limitations. For example, regarding the latter,
there is the issue of representativeness: while results vary by country, the consensus
is that Twitter users are younger (for the United Kingdom, see [46]; for the United
States, see [8]; for South Korea, see [25]). Yet, there are also benets to be had in
using online- versus survey data to examine anti-Chinese sentiment: the digital envi-
ronment provides opportunities for freer speech, which makes it easier to detect the
phenomenon in question. Due to the anonymity the Internet aords, people can say
things they might otherwise not do so face-to-face [61]. This tendency has also been
found to be stronger regarding hate speech and racism—anonymity can embolden
people to be more outrageous, obnoxious, and antagonistic in what they say [9, 16,
54]. Based on this, I utilize both social media- and survey data to gauge anti-Chinese
sentiment. First, I use social media data to show descriptively the trends herein asso-
ciated with the air pollution issue; second, I test the causal eect of China-blaming
reporting on anti-Chinese sentiment and less benign foreign policy attitudes.
The article is organized as follows. First, I describe how public sentiment toward
China in South Korea has changed over time since 1992. Feelings had been positive
in the early 1990s but would gradually worsen due to China’s indulging of North
Korea and coercive diplomacy surrounding the THAAD missile base. In more recent
periods, the issue of cross-border air pollution has become a top contributing factor to
high levels of resentment toward China among South Korean citizens. Second, I ana-
lyze media coverage of air pollution in the 2015–2018 period and show how articles
blaming China for it increased year on year. Compared to 2015, such articles had
doubled by 2017. Third, I analyze Twitter data to show how the South Korean public
discourse regarding air pollution and China would evolve: by 2018 hostility directed
at the Chinese government and the Chinese people had emerged, having not featured
in 2015. Lastly, I show results from a unique online survey experiment testing the
causal eect of China-blaming articles on anti-Chinese sentiment, and subsequently
on how the resentment aroused here has aected foreign policy attitudes.
Changing Public Perceptions of China in South Korea
In 1992, when China and South Korea rst established ocial relations, public per-
ceptions in the latter were positive due to increased expectations that the former
would play a positive role in realizing Korean unication. China then ranked as the
second-most favorable neighboring country, surpassing Japan [10, 63]. In 1996, the
percentage of respondents in one study agreeing that China is an important ally to
South Korea was more than double those believing so for the US (47.1 versus 23%)
[19]. This all aligns with the South Korean government’s expectation that diplomatic
ties with China would enable denuclearization and peaceful unication. This posi-
tive outlook was also embedded in ocial Ministry of Foreign Aairs documents
released regarding the rapprochement initiated with China in 1992 to these two ends
[47].
The turning point came in 2004 however, when China launched the Northeast
Project that created controversy among South Korean politicians and was covered
1 3

Air Pollution Coverage, Anti-Chinese Sentiment, and Attitudes Towards…
widely in the domestic media. The Northeast Project was highly contentious due
to its alleged politicization of history and negation of sovereignty. Thus, China was
framed as a hostile superpower having goals of “Chinese expansionism” (junghwa-
paegwonjuui) in the South Korean media. Public perceptions also worsened: between
2004 and 2006, respondents with a favorable view of China decreased from 49 to
32% [30]. The trade dispute arising over South Korea’s restriction of garlic imports
from China and the subsequent retaliation from the latter also contributed to negative
feelings here [29].
Attitudes toward China further worsened throughout the 2010s, due to increased
concerns over the country’s role in failing to contain threats from North Korea—
which became a widely covered topic due to a series of attacks by the latter on South
Korea, including the shelling of Yeonpyeong Island and the sinking of the Cheonam
patrol corvette in 2010. Throughout the second decade of the new century such per-
ceptions continued to decline, reaching their lowest point in 2013 when just 23% of
respondents in one study reported having a favorable view of China—a 26% decrease
therein from 2004 [18]. In 2016, the THAAD crisis—a foreign policy dispute that
emerged over the establishment of a missile basis in Seongju, South Korea—raised
a series of concerns over China’s coercive diplomacy. Negative perceptions of China
were also reected in contemporary polls, with unfavorable views thereof increasing
from 37% to 2015 to 60% in 2018 [60]—representing a 23% increase in just three
years.
Although the number of works on the causes of anti-Chinese sentiment in South
Korea is small compared to those on anti-Japanese views due to South Korea’s dif-
ferent historical experiences with the two countries [15], a growing trend can still be
detected here—with it having a particular focus on nationalism and domestic polar-
ization, and specic attention being given to the Northeast Project and the THAAD
disputes. Some scholars have focused on nationalism: it was found that the historical
dispute over the ancient kingdom of Koguryo has had a causal eect on anti-Chi-
nese sentiment [20]. Domestic polarization and media reporting have also received
scrutiny: for example it has been also found that, compared to the Northeast Project
issue, domestic polarization was higher regarding THAAD, where South Korea’s
right-wing parties and conservative media used rhetoric designed to weaken support
for the incumbent government’s pro-China policy and to strengthen a pro-US stance
instead [28].
Meanwhile, the issue of air pollution has begun to rank higher here than these two
disputes in recent years. According to a survey by SisaIn, the top reason for negative
sentiment toward China was air pollution (89.4%), followed by COVID-19 (86.9%)
[39]. In the same survey, over 94% of respondents answered that China is respon-
sible for South Korea’s domestic air pollution. Although the origins of the latter are
a scientic issue, the question has been raised in both politics and scholarship as to
whether the media is playing a role in the heightened belief that air pollution can be
directly attributed to China. The South Korean government has remained neutral on
the issue: when President Moon met Yang Jiechi, member of the Politburo of the
Chinese Communist Party, during a meeting in March 2018, he mentioned that “there
are domestic causes for air pollution in Korea, but also partially from China”—thus
framing the issue as a joint one [32]. Yet when examining the strong backlash from
1 3

E. E. Song
civil society, as reected in the aforementioned online petition uploaded to the Blue
House website, there are strong indications that the media has indeed played a role
in steering blame toward China. In the petition it says: “The media also reports that
the cause of air pollution originates from China, yet the government is irrespon-
sive.” Media reports are cited as evidence. The petition also asks: “Why do we need
to cooperate? Air pollutants are from China” [3, 5]. Previous works on the South
Korean media’s reporting vis-à-vis air pollution in the country have also found that
responsibility for it is attributed to China alone: blaming the latter has emerged as
one of four frames used, and can be found in both liberal and conservative outlets
alike [55]. Furthermore, most articles that have mentioned the causes of air pollution
have attributed blame to China, as opposed to domestic or individual behavioral fac-
tors [34]. These ndings raise several questions: How exactly has the media directed
blame toward China for air pollution in South Korea? How has the public discourse
regarding Chinese responsibility for such pollution changed over time? Does this
attribution aect anti-Chinese sentiment and negative perceptions of the two coun-
tries’ relations? I turn to these matters in the following.
Media Coverage of Air Pollution and China-Blaming: 2015–2018
To examine the patterns in air pollution reporting and how they might have fueled
anti-Chinese sentiment, headlines from the largest news aggregator in South Korea,
Naver News, were crawled (https://news.naver.com/). I scoured the top-ve search
results for each day from 2015 to 2018 after searching the term “China,” and then
ltered out news articles that mentioned “ne dust air pollutants.” This yielded 145
articles from 75 unique sources.2 Other than collecting data from individual news
sources, reliance is on Naver News for several reasons. The key one, however, is that
it has been found that most South Korean citizens consume such information through
search sites and news aggregators [51]. According to recent research, 72% of respon-
dents said that they use search engines and news aggregators to consume news. This
is much higher than the proportion (5%) stating that they consume news directly from
related websites, and hence my choice here.
Media reports linking China to ne dust air pollutants increased fourfold between
2015 and 2017 (see Fig. 1 below). Not only did the total number of related articles
increase, but the proportion of those blaming China for South Korea’s air quality also
went up twofold—from approximately 36% in 2015 to 75% in 2017. These articles
included a phrase that directly attributes blame to China for this air pollution: “ne
dust air pollutants from China” (junggukbal misemeonji). At the same time, while
accounting for a smaller proportion than the China-blaming articles, there was also
an increase in ones mentioning additional sources of pollution (such as domestic
factors); these went from none in 2016 to approximately 13% doing so in 2018.
2 Articles from liberal outlets accounted for approximately 12% hereof, while those from conservative out-
lets accounted for 3%. Liberal outlets refer to the following: Kyunghyang Shinmun, Nocut News, Newsis,
Pressian, and Hankyoreh. The remaining outlets were categorized as neutral. Conservative outlets were:
TV Chosun, Chosun Biz, Chosun Ilbo, Joongang Ilbo, Channel A, Donga Ilbo, and Hankook Gyeongjae.
1 3

Air Pollution Coverage, Anti-Chinese Sentiment, and Attitudes Towards…
Another category that grew in size was articles reporting on South Korea’s air pollu-
tion issue, with them mushrooming in 2018. The two categories seem to have grown
due to the online petition posted on the Blue House website in 2018 urging China
to take greater responsibility, as well as following South Korean government eorts
to increase cooperative measures with the latter [3, 5]. Lastly, articles discussing the
state of China’s domestic air pollution grew slightly during the 2017–2018 period,
but the proportion remained small. This included not only reports on worsening air
pollution within China but also on the latter’s eorts to tackle it [48, 50]. Table 1
below shows example headlines for each category of report established.
The statistics show that reports on air pollution and those attributing blame to
China would heavily increase in the post-2016 period. Against this backdrop, a key
question arises: Have these articles fueled anti-Chinese sentiment, and, if so, how has
the pattern changed over time? To answer this, I rst examine anti-Chinese sentiment
related to air pollution in Twitter data from 2015 to 2018.
Topic Modeling of Tweets: 2015 and 2018
To answer the question of how public sentiment changed toward China amid increas-
ing media reports attributing blame to the latter for South Korea’s air pollution, I
draw on related tweets from 2015 to 2018 respectively. These data were used to
delineate changes in the public discourse after the increased coverage of China and
more hostile rhetoric seen from 2017 onward. For comparison, I collected all tweets
Fig. 1 Categorization of news on air pollution linked to China (2015–2018)
1 3

E. E. Song
for 2015 and 2018 by searching the keyword “China” and “ne dust pollutants.” I
avoided using tweets from 2016 to 2017 due to the strong salience of the THAAD
dispute in those years, which made it dicult to sample tweets related to China and
air pollution. As media coverage of THAAD dissipated after South Korea and China
resolved to halt further expansion of the missile base in November 2017, I chose the
period after that year as the point of comparison with the earlier one of 2015.
The data were collected using an open-source Python package “snscrape,” which
enables the amassing of historical tweets (archived at: https://github.com/JustAno-
therArchivist/snscrape).3 As a result, the 2015 sample included 401 tweets; for 2018,
the total number collected was 2,951. The research method used was topic modeling.
Specically, I used the Non-Matrix Factorization (NMF) model to detect topics in the
tweet corpus. NMF is an unsupervised method that categorizes text corpora without
labels. In addition, compared to other topic models such as Latent Dirichlet Alloca-
tion (LDA), it has been found that NMF performs better when dealing with shorter
texts. While LDA assumes that each text in the sample discusses several topics, NMF
does not—making the latter more applicable to shorter, single-issue texts [12].
3 As this package collects only publicly available data on Twitter, Institutional Review Board approval was
not obtained for the collection procedure.
Categorization News Headlines
Articles that
blame China for
S.Korea’s air
pollution
“Fine dust air pollutants at high levels across
the country, ‘the cause is China’” (Korea Eco-
nomic Daily, 2015)
“Fine dust air pollutants at high due to smog
from China” (Newsis, 2017)
“Sudden attack by ne dust air pollutants from
China…high levels during this weekend” (SBS,
2017)
“Huge smog from China…ne dust air pollut-
ants make it hard to breathe” (SBS, 2018)
Articles that
mention addi-
tional sources of
pollution other
than China
“The source of horric ne dust air pollutants is
not only China” (OhmyNews, 2017)
“Are the ne-grained air pollutants from China?
Probably not this time” (Hankyoreh, 2017)
“China’s contribution was small for January’s
ne dust air pollutants…around 38∼57%”
(Yeonhap News, 2018)
“Recent ne dust air pollution can be attributed
more to domestic causes” (Yeonhap News,
2018)
Articles on
China’s domestic
air pollution
“China announces red alert…ne dust air
pollutants 20 times more than safety baseline”
Newsway, 2015)
“Liaoning province starts managing air quality
in a serious manner” (Yeonhap News, 2017)
Other reports
on S.Korea’s air
pollution
“POSCO ICT exports dust collector to China”
(Yeonhap News, 2016)
“Ministry of Environment reports that it will
increase cooperation with China in order to
reduce air pollutants” (Yeonhap News, 2018)
Table 1 Categorization of News
on Air Pollution and China
(N = 145)
1 3

Air Pollution Coverage, Anti-Chinese Sentiment, and Attitudes Towards…
Text preprocessing, such as removing punctuation, duplicates, and emojis, was
conducted prior to applying the NMF model. In addition, I removed any tweets
related to weather reports and advertisements so that the sample only includes ones
posted by individuals. I used the KoNLPy (Korean NLP in Python; https://konlpy.
org/en/latest/) open Korean Text module to tokenize tweets to nouns. I chose term
frequency and inverse document frequency (TF-IDF) vectorization. Compared to the
count vectorizer (the bag-of-words approach, which counts the frequency of unique
words that appear in a string), TF-IDF notes how many times a unique word occurs
across all strings—therefore eectively ignoring common words.
Results: Anti-Chinese Sentiment Linked to Air Pollution Increased in
2018
The optimal number of topics for the respective text corpora for 2015 and 2018 was
found by obtaining the coherence score [56]. Developed in studies of natural lan-
guage processing, this score evaluates whether each output topic is coherent on a
scale of 0 to 1. According to the coherence score, the optimal number of topics for
the 2015 corpus was 5; for 2018, meanwhile, it was 15. In both periods, the number
of tweets spiked when pollution levels increased: when examining changes in the
number of tweets in each topic category, the number surged around March and April
in both years, when pollution levels spiked, and then decreased during the summer,
when pollution rates decreased. The number of tweets spiked again in November and
December, when pollution levels increased. Illustrations of these trends are provided
in the Appendix (Figures A1, A2).
There are similarities and dierences between the topics extracted from the 2015
versus 2018 corpora. Among those remaining similar throughout, one comprised a
discussion of Asian dust that usually occurs in spring (Topic 1 in Table 2; Topic 6
in Table 3 below); another was complaints about reworks being the cause of the
air pollution coming from China (Topic 3 in Table 2; Topic 4 in Table 3). At the
same time, there was a sharp dierence between the topics obtained from the two
periods: compared to 2015, tweets conveying anger toward China and the Chinese
people were prevalent throughout 2018. These “venting” tweets often included racist
or otherwise derogatory remarks targeted at the Chinese people (Topics 3 and 10 in
Table 3). The examples here also reveal that the primary cause of the increase in such
racist and derogatory remarks was blaming China for South Korea’s air pollution.
Another dierence is that in 2015 there were more discussions on the actual origins
of ne dust pollutants (Topics 2 and 5 in Table 2), while in the 2018 corpus this
inquiry did not emerge. Further, in 2018 more grievances were expressed toward the
South Korean government for its reluctance to resolve the issue with China, which
seems to have been driven by the controversy arising from the online petition posted
on the Blue House website calling for China to be held responsibility and stronger
government eorts from the latter to tackle the issue (Topic 14 in Table 3).
1 3

E. E. Song
Online Survey Experiment
Data and Hypotheses
Although Twitter data show that there would be an increase in anti-Chinese senti-
ment coinciding with the greater number of articles appearing in the media blaming
China for the air pollution issue, they do not conrm whether this actually aected
Table 2 NMF Topic Analysis Results for 2015 Tweet, Selected Topics Only (N = 401)
Topic
Number
Topic Summary Sample Tweets (English)
1 Asian dust,
mask, wind,
government,
because, re-
ally, MERS,
we, counter-
measures,
regulations
Linking annual dust
to China and the
wind; comments on
how the rst MERS
patient in China
was Korean; asking
government coun-
termeasures and
regulatory eorts
‘China creates dust wind with ne dust air
pollutants’
‘Korea reciprocated by sending MERS patient
and received ne dust air pollutants in ex-
change, looks like fair trade’
‘How is the government regulating companies
to reduce ne dust air pollutants? Since when
has ne dust air pollution become so severe and
how did the government regulate or de-regulate
companies? The government just thought it was
from China.’
2 Korea,
emerged, right
now, extent,
itself, air, con-
clusion, talk,
MERS
Discussion over the
actual origins of
ne dust air pollut-
ants; venting over
anti-Korean senti-
ment in China over
a Korean MERS
patient
‘In fact some of the ne dust air pollutants are
from China but a lot of them are also emitted
domestically a lot of people blame China when
we talk about decreasing air pollution however
a lot of it comes from automobiles and coal fac-
tories however when I say these things facebook
users get angry and tell me to blame China’
‘actual air quality around car road are worse
than publicly announced indicators the cause of
ne dust air pollutants are not coal factories or
China but are automobiles on the street’
‘so China sends over Asian dust and the Chi-
nese people commits all kinds of crime includ-
ing chopping people up or phone scamming and
Chinese people are criticizing Koreans for send-
ing over MERS sounds like they like damaging
people but they hate when others do it to them
let’s just spread MERS to this ch*** country’
3 Today, density,
Seoul, mask,
reworks,
smog, dust,
reason, Chinese
New Year,
yesterday
Discussion over
whether ne dust
air pollutants are
due to reworks in
China during Chi-
nese New Year
‘Is it really true that ne air pollutants are from
reworks used during the Chinese New Year in
China is that a direct cause’
‘I heard today’s air pollution is because of all
the reworks during Chinese New Year’
5 Korea, we,
emerged,
country,
think, people,
environment,
half, continue,
problem
Discussion over
whether the origins
of ne dust air
pollutants are from
Korea, discussions
that both sides share
responsibility
‘ne dust air pollutants…half of it is actually
from Korea so we have equal responsibility
with China’
‘China is about % responsible for the ne dust
air pollutions in Korea over half originates from
within the major causes are gas emissions from
diesel automobiles and smoke from thermal
power stations’
1 3

Air Pollution Coverage, Anti-Chinese Sentiment, and Attitudes Towards…
Table 3 NMF Topic Analysis Results for 2018 Tweet, Selected Topics Only (N = 2,951)
Topic
Number
Topic Summary Sample Tweets (English)
1 Really,
f***ing, weath-
er, please, pain,
smog, protest,
earth, have
done, school
Complaints about
smog and air pollution
and anger venting
towards China; hoping
more days o from
school and companies
due to bad air quality
‘air quality is really bad today so if we can’t
blame China then let’s have some days o
school and work’
‘I want to really get rid of China this ne dust is
so annoying people don’t take it seriously when
we call it dust this is more like cancer-inducing
smog’
3 F***, f***ing,
son of a b****,
county, reason,
bulls***, heat
wave, this,
c****, oneself
Anger over air pol-
lution and attribut-
ing blame to China;
venting anti-Chinese
sentiment
‘air quality is so s***** today China f***ing
a********.’
‘there is so much rain today but tomorrow’s
forecast says air is super bad what the f***
China’
4Because, re-
works, respira-
tory, sound,
thermal power
plant, face,
denitely, sky,
how, Chinese
New Year
Complaints over re-
works during Chinese
New Year in mainland
as cause of air pollu-
tion; complaints and
assertions that air pol-
lution is due to China
‘f*** heard all this ne dust is because re-
works in China’
‘ne dust is denitely because of China is not
because of thermal power plants or automobiles
that’s all bulls****’
‘so everyone knows now that all this ne dust
comes from China they can’t get away with it
give me back my throat and nose’
6 Asian dust,
tomorrow,
one, citizen,
now, smog,
measures, af-
ternoon, really,
about
Complaints over
Asian dust and how
it originates from
China, blaming China;
expressing hatred
towards China by
attributing blame for
Asian dust and smog
‘f*** it is so hard to breathe under this mask
because of all the Chinese ne dust this feels
like s***’
‘in spring it’s the Asian dust in summer there’s
the heat wave in fall there is ne dust and win-
ter there is serious smog China m***********’
10 S***, c****,
Chinese peo-
ple, numbers,
bulls***, heat
wave, problem,
always, see,
commute
Blaming China for
the heat wave and air
pollution; venting ager
towards the Chinese
people for ne dust air
pollutants
‘China f****** gives us ne dust and heat
waves’
‘of course these c**** f***** will not like
hearing it but we are suering for months
because air pollution and it feels like s*** the
only nice Chinese are the dead ones’
14 Domestic,
problem, solve,
element, gov-
ernment, mea-
sures, emerged,
citizen, voice,
eect
Complaints on South
Korean government
for being irresponsive
and unaccountable
regarding air pollution
issues, how it has been
reluctant in safeguard-
ing citizen health
‘Moon Jae-in is a disaster he can’t say anything
to the Chinese communist party while people
feel threatened because of the cancer-inducing
ne dust why can’t he demand apology from
China like he did to Japan why can’t he demand
compensation’
‘I started o with talking about why can’t Moon
say anything to China about ne dust problem
is it because of the rice price spikes and coals in
North Korea wonder what the hell is in his head
is it only North Korea and commies who the
hell voted this guy I said this and that guy from
Soon-cheon said all the rest are red necks who
should I vote for so I lost my words’
1 3

E. E. Song
people’s perceptions of the country. Social media data have limits in terms of infer-
ence: although they oer advantages in exposing hate speech, they lack represen-
tativeness. Specically, Twitter users are younger on average, which suggests that
the trend observed above could be limited to a sample of youth citizens. In addition,
selection bias might exist—those expressing negative feelings toward China online
might already have strong anti-Chinese tendencies. To disentangle these problems of
endogeneity and representativeness, I conducted an online survey experiment to test
the causal eect of China-blaming articles on anti-Chinese sentiment. In addition,
I tested whether such articles have impacted foreign policy attitudes by fomenting
resentment here.
The online survey experiment was conducted with 500 South Korean adults aged
20–70 years old and was carried out in March 2022 by Hankook Research, a public
opinion research company in South Korea. The sample was randomly drawn from
the company’s survey panel. The experiment was conducted by briey presenting
respondents with a news article, which was randomly shown to the participant. This
comprised either: (1) a news item attributing blame to China and including the phrase
“air pollutants from China” in the headline or (2) a news item that remained neu-
tral on the issue. Also, following the nding that the majority of the South Korean
population consumes news via online aggregator sites, only a short length of time
(of minimum 10 s) was required for the respondents to be exposed to the treatment
or control before they moved on to subsequent questions. Those presented after the
treatment or control were oered randomly to minimize satiscing. The treatment (1)
and control (2) details are reported in Table 4. Summary statistics are presented in the
Appendix (Table A1). The total number of respondents was 504, and the size of the
treatment (N = 253) and control (N = 251) groups was almost even. The distribution
of covariates such as age, education, income, and gender were similar between the
two groups.
The sampled respondents were randomly split; the treatment group was primed
with a news item that blamed China for air pollution, and the control group was
primed with a news item that was neutral in terms of reporting the causes hereof.
The survey included questions that captured individual covariates such as age, gen-
der, and education level. Questions were included on one’s feelings regarding the
Chinese government and the Chinese people respectively: (1) How favorably do
you feel toward China? (2) How favorably do you feel toward the Chinese people?
The response was aggregated from a 5-point Likert scale to a 3-point Likert scale
(3=“Unfavorable,” 2=“Neutral,” 1=“Favorable”) due to the low response counts
mapping to “Very Favorable” and “Favorable” on the 5-point Likert scale.4 These
two separate dimensions were considered due to the nding in the literature that anti-
Chinese sentiment consists of negative attitudes toward the country’s both govern-
ment and people [4].
To capture the dependent variable, questions on foreign policy attitudes and per-
ceptions of China were added to the survey. More specically, these asked respon-
dents about their view on the rise of China and whether South Korea should strengthen
4 “Very favorable” (2) and “Favorable” (23) accounted for 5% of the total responses. To construct the
3-point Likert scale, the responses “Very unfavorable” and “unfavorable” were also merged.
1 3

Air Pollution Coverage, Anti-Chinese Sentiment, and Attitudes Towards…
relations with that country amid heightened US-China hegemonic competition. These
two topics were included due to recent survey results showing that the top two issues
South Koreans perceive as threatening national interests in the future are (1) the rise
of China and (2) greater US-China hegemonic competition [33].5 The question relat-
ing to the rst dimension asked whether respondents agreed with the following state-
ments: (1) the rise of China will become a serious military threat to South Korea, (2)
the rise of China will become a serious economic threat to South Korea. Responses
were captured on a 5-point Likert scale and then aggregated to a 3-point Likert scale.
The question concerning the second dimension, meanwhile, was constructed as: (3)
What do you think South Korea should do regarding increasing US-China rivalry in
Northeast Asia? Three response options were given: (a) “Strengthen relations with
5 According to the fourth round of a survey on Korean identity in May 2020, carried out by the East Asia
Institute, the top issue that South Koreans perceived as threatening to the national interest was the rise of
China as a great power (41.7%), followed by US-China hegemonic competition (34.9%) [33].
Treatment Control
Title Fine dust levels “bad” all
day, due to smog from China
remaining inland
Fine dust levels 2–3 times
higher than usual – “bad”
until tomorrow morning
Text Fine dust pollutants from
China covers the whole
country, and the ne dust
rating is “bad”. According
to the Korean Environment
Corporation, the average ne
dust level in Seoul at 7 am
this morning is 101μg/m3,
which is more than twice
of annual average ne dust
level (50μg/m3). This level is
equivalent to air quality level
within car tunnels. Aside
from Seoul, ne dust levels
in other areas are: Choong-
book, Jeonbook, Daejeon,
Daegu, Gwangju, Busan,
which are all high levels. A
sta at the National Institute
of Environmental Research
said that “ne dust levels will
be low during the afternoon
but will increase sometime
later in the afternoon”. The
reason why ne dust level is
“bad” these days is because
the smog from China remains
within the South Korean
inlands.
Air pollutants and fog
have been creating smog
for four days. The ne
dust levels have increased
2–3 times than usual, and
the smog will continue
until tomorrow morning.
The fog decreased during
the day time, but the pol-
lutants remained creating
hazy sky view. Fine dust
levels have increased 2–3
times than usual, and the
levels are at “bad” level.
PM2.5 particles, which are
smaller than 2.5 microm-
eters consist of 70–80% of
the entire ne dust mass.
Since these small par-
ticles can enter lungs, the
elderly and patients with
respiratory disease should
refrain from being outside.
National Institute of
Environmental Research
analyzed the cause that
pollutants created within
the country and those that
came from China were
mixed, and since the air
circulation lessened, the
pollutants have accumu-
lated in the air.
Table 4 Treatment and Control
for the Survey Experiment
1 3

E. E. Song
the United States,” (b) “Maintain balance,” or, (c) “Strengthen relations with China.”
In the subsequent analysis, I created a dichotomous variable of 1=“Strengthen rela-
tions with China” and 0=“Other” (“Maintain balance” or “Strengthen relations with
the United States”) to estimate the causal eect of intensied anti-Chinese sentiment
on the South Korean populace’s foreign policy attitudes.
One source of bias potentially arising during the analysis was that respondents
may have been strongly interested in the issue of air pollution and thus might have
sought more information on the phenomenon. This would have potentially exposed
them to a greater number of articles blaming China for the latter. This increased
exposure could have led to selection bias, where respondents were more likely to
harbor negative feelings toward China. A question on one’s interest in air pollution
was included to control for this: “How interested are you in air pollution?” Responses
were 5=“Very interested,” 4=“Interested,” 3=“Neutral,” 2=“Disinterested,” and
1=“Very disinterested.” This variable was also later aggregated to a 3-point Likert
scale prior to the analysis.
As this online survey experiment was devised to test whether exposure to news
articles that attribute blame to China for air pollution in South Korea has had an
actual eect on inciting anti-Chinese sentiment among the latter’s populace, the fol-
lowing hypothesis was formulated:
H1: Exposure to news blaming China for domestic air pollution has increased
anti-Chinese sentiment.
To test whether exposure to China-blaming articles has led to negative foreign
policy attitudes via an increase in anti-Chinese sentiment, a second hypothesis was
then formulated:
H2: Exposure to news blaming China for domestic air pollution has had negative
eects on foreign policy attitudes via anti-Chinese sentiment.
Analysis of H1: Exposure to Articles Blaming China for Air Pollution has
Increased Anti-Chinese Sentiment, Particularly toward the Chinese People, and the
Eect Is Moderated by Age Group.
Because the dependent variable of sentiment (toward the Chinese government
and Chinese people respectively) was categorical (1=“Favorable,” 2=“Neutral,”
3=“Unfavorable”) and ordinal, I used ordinal logistic regression to test H1. The
independent variable is a dummy one indicating treatment and control (1 = Treat-
ment, 0 = Control). I also included the interaction term between age and treatment to
account for recent ndings that anti-Chinese sentiment varies across age groups [58].
Table 5; Fig. 2 above report the results of the log-odds from the ordinal logistic
regression. They indicate some support for H1—although the treatment itself did
not have a statistically signicant eect on the outcome variables, the interaction
term between treatment and age, especially for the 40–49 age group, was statistically
signicant at the 10% condence level. The estimated log-odds suggest that online
survey respondents were 3.8 times more likely to dislike the Chinese people when
they were exposed to a China-blaming article and when they were in their 40s com-
pared to those in the baseline group—those who were treated and were in their 20s.
In addition, the eect of the treatment on the respective age groups was stronger for
sentiment toward the Chinese people than for that toward the Chinese government.
Figure 2 shows that the exponentiated coecient estimates of the interaction term are
1 3

Air Pollution Coverage, Anti-Chinese Sentiment, and Attitudes Towards…
larger than 1 for all those between 30 and 69 years old when the dependent variable
is sentiment toward the Chinese people rather than toward the Chinese government.
In short, those falling within this age range were more likely to dislike the Chinese
people when they were treated with an article blaming that country for South Korea’s
air pollution compared to those aged 20–29, whereas there was no such eect when
the dependent variable was instead sentiment toward the Chinese government.
Analysis for H2: China-Blaming Articles Decreases Support for Strengthening
Relations with China via Increasing Anti-Chinese Sentiment.
Several steps were taken before testing H2. First, although the question is a binary
classication one, the data were heavily imbalanced, where those in favor of strength-
ening relations with China accounted for only 1.5% of the entire pool of online survey
Table 5 Ordinal Logistic Regression Results, Log-Odds (Hypothesis 1)
Dependent Variable Sentiment towards Chinese govern-
ment (3 = Unfavourable 2 = Neutral
1 = Favourable)
Sentiment towards Chi-
nese people (3 = Unfa-
vourable 2 = Neutral
1 = Favourable)
Model I Model II
Treatment 1.010 0.475
(0.02) (-1.14)
Gender 0.887 0.863
(-0.58) (-0.71)
Age = 30s (Baseline = 20s) 0.356* 0.292**
(-1.80) (-2.00)
Age = 40s 0.407 0.175***
(-1.53) (-2.91)
Age = 50s 0.182*** 0.212**
(-3.06) (-2.55)
Age = 60s 0.147*** 0.150***
(-3.48) (-3.18)
Education 0.921 0.904
(-0.77) (-0.95)
Income 1.065 1.045
(1.43) (0.99)
Interest in Pollution (1 = Not Inter-
ested; 2 = Neutral; 3 = Interested)
0.671 0.933
(-1.40) (-0.26)
Treatment#30s 1.345 2.426
(0.36) (1.11)
Treatment#40s 0.624 3.820*
(-0.58) (1.70)
Treatment#50s 0.967 1.564
(-0.04) (0.58)
Treatment#60s 1.098 2.140
(0.12) (0.99)
N 504 504
Note. Exponentiated coecients; t statistics in parentheses
*p < 0 .10, **p < 0.05, ***p < 0.01
1 3

E. E. Song
respondents. In a classication question, this could result in lack of statistical power
when applying the model. To correct this imbalance, I used SMOTENC (synthetic
minority oversampling technique for nominal and continuous [11]), which allows
for correcting imbalances in both numerical and nonnumerical data. Compared to
the classic oversampling technique, which duplicates data points from the minority
class, SMOTENC uses a k-nearest-neighbor algorithm to create synthetic data. First,
random data from the minority class are chosen, then k-nearest neighbors from the
datasets selected. Synthetic data that lie between the random data and the randomly
selected k-nearest neighbor are now created. In this study, the corrected imbalance
ratio was 9%, and the total number of data points equaled 550 (Appendix, Table A2).
I used ordinal logistic regression to estimate the eects of anti-Chinese sentiment
on three dependent variables regarding foreign policy attitudes: (1) the perception of
China as a military threat; (2) the perception of China as an economic threat; and,
(3) attitudes toward strengthening relations with China. For the third, I used logistic
regression because the variable was binary (1=“Strengthen relations with China,”
0=“Other”). Results for Models III and VI in Table 6 below show that there is a
strong negative correlation between the treatment and the alliance choice. The results
Fig. 2 Ordinal logistic regres-
sion results (Hypothesis 1, odds
ratio)
1 3

Air Pollution Coverage, Anti-Chinese Sentiment, and Attitudes Towards…
in Model II show that when a person was treated with an article blaming China for
South Korea’s air pollution, the odds of that respondent being in favor of strengthen-
ing relations with China decreased by 95%, and this eect was statistically signicant
at the 1% condence level.
Although there was a high correlation between the treatment and the alliance
choice, the chosen model specications did not test whether the treatment aected
the alliance choice via anti-Chinese sentiment. The logistic regression specication
tested correlation rather than causation in this regard. To further examine this link, I
used the causal mediation algorithm developed by Tingley et al. [65]. The estimation
process consists of two steps: First, tting the model that predicts the mediating vari-
ables, which here are: (a) sentiment toward the Chinese government and (b) sentiment
toward the Chinese people. Second, estimating a model that predicts the dependent
variable—support for foreign policy that strengthens relations with China—using
treatment variables, the mediating variables, and all other covariates included in the
rst step. Then, the average direct eect (ADE) and the average causal mediation
eect (ACME) are estimated. ADE captures the eect of the treatment variable on
the outcome one less the eect via the mediating variable, while ACME captures the
eect of the treatment on the outcome via the mediating variable [26]. The analysis
was conducted using mediation package in R [65]. Because the treatment was only
statistically signicant when the dependent variable was strengthening relations with
Table 6 Ordinal Logistic Regression & Logistic Regression Results, Log-Odds (Hypothesis 2)
IV: Sentiment towards Chinese
government
IV: Sentiment towards Chinese
people
DV (I) China
will be a
serious
military
threat
(II)
China
will be a
serious
economic
threat
(III)
Strength-
en rela-
tions with
China
(IV) China
will be a
serious
military
threat
(V)
China
will be a
serious
economic
threat
(VI)
Strength-
en
relations
with
China
Sentiment towards Chinese
government
1.795** 1.605** 0.548**
(2.34) (2.50) (-2.38)
Sentiment towards Chinese
people
2.001*** 1.433*0.124***
(2.81) (1.83) (-5.94)
Treatment 0.914 0.966 0.045*** 0.949 0.983 0.035***
(-0.29) (-0.15) (-4.19) (-0.17) (-0.08) (-4.22)
Gender 0.705 0.758 0.0193*** 0.708 0.758 0.0140***
(-1.11) (-1.24) (-3.86) (-1.09) (-1.25) (-3.99)
Age 1.250*1.202** 1.382** 1.242*1.173** 1.238
(1.85) (2.22) (2.24) (1.82) (1.97) (1.24)
Income 1.147*1.016 0.923 1.144*1.018 0.979
(1.92) (0.35) (-0.96) (1.87) (0.38) (-0.22)
Education 0.669** 0.858 1.446** 0.684** 0.860 1.348
(-2.46) (-1.36) (1.97) (-2.32) (-1.34) (1.48)
N504 504 550 504 504 550
Note. Exponentiated coecients; t statistics in parentheses
*p < 0 .10, **p < 0.05, ***p < 0.01
1 3

E. E. Song
China (Models III and VI), and not the perception of China as a military and eco-
nomic threat, I examined whether anti-Chinese sentiment mediated the causal eect
of treatment on alliance choice.
Table 7 shows that anti-Chinese sentiment—particularly hostility toward the Chi-
nese people—had a causal eect on attitudes toward foreign policy. ACME revealed
that the mediating variable—anti-Chinese sentiment—impacted the outcome vari-
able of support for strengthening relations with China. The second column in Table 7
shows that this eect was statistically signicant at the 10% level. However, the
eect was not statistically signicant in the case of negative sentiment toward the
Chinese government. For both mediating variables, under ADE the direct eect of
the treatment on the outcome aside from the mediating variable was statistically sig-
nicant, which implies that there were other unobservable factors aecting alliance
choice via the treatment variable besides anti-Chinese sentiment.
Discussion
The controversy that has arisen over the source of South Korea’s air pollution has
become one of the strongest correlates to anti-Chinese sentiment in the country in
recent years. How has South Korea’s media reported on air pollution, and how has
this contributed to increased anti-Chinese sentiment there? How has this coverage
aected the domestic population’s foreign policy attitudes? By examining how the
South Korean media has reported on air pollution, this study showed that blame for
it was increasingly attributed to China during the 2015–2018 period. It was revealed
how this had the negative causal eect of creating unfavorable sentiment toward
China—particularly toward the Chinese people. Moreover, the article found that
this heightened resentment has had a detrimental impact on South Koreans’ foreign
policy outlooks—especially regarding strengthening China-South Korea relations.
The ndings of the paper both demand discussion and leave room for further
research. First, although the correlation between media bias against China and anti-
Chinese sentiment are understandable as a high degree of association was found
between these aspects both in Twitter data (2015 and 2018) and in the online survey
experiment, it is not clear why the eect of that media bias was stronger among
respondents in the 40–49 age group. One possible explanation for this is the heuris-
tics generated by the China-blaming articles—lack of information on government
responsibility and the overt framing of where the air pollution comes from through
the phrase “ne dust pollutants from China” could have primed individuals to attri-
Mediator: sentiment
towards the Chinese
government
Mediator: sen-
timent towards
the Chinese
people
ACME (average) 0.78 0.076 *
ADE (average) < 2e-16 *** < 2e-16 ***
Total Eects < 2e-16 *** < 2e-16 ***
Table 7 Causal mediation
analysis results (p-values)
Note. Sample Size Used: 550.
Simulations: 500
*p < 0 .10, **p < 0.05, ***p < 0.01
1 3

Air Pollution Coverage, Anti-Chinese Sentiment, and Attitudes Towards…
bute blame to the Chinese people for emissions as opposed to the Chinese govern-
ment, which has a responsibility for managing pollution.
Previous research has shown that media heuristics matter in terms of who people
attribute blame to for public policy failures, and that they tend to take onboard the
apportioning of responsibility laid out in the media [23, 66]. In addition, it has been
found that specifying public ocials’ culpability here by revealing their titles aects
the ways in which the public blames these individuals for said policy failures [44].
In the context of South Korea, partisanship could also be in play considering that
people in their 40s are the strongest supporters of the Democratic Party [69] and that
the Moon administration’s foreign policy was largely pro-China. Thus, the media
heuristics could have pushed respondents in their 40s to attribute blame more to the
Chinese people than to the Chinese government. Future work should further empiri-
cally examine whether variation in blame attribution vis-à-vis the Chinese people
correlates with exposure to information pertaining to the Chinese government’s role
in regulating pollution levels, and how this is mediated by partisanship.
Regarding the link between air pollution, anti-Chinese sentiment, and foreign
policy attitudes, one possible explanation for the nature of it could come from
image theory. The latter stipulates that countries can be considered as being held to
an image, such as an “enemy,” “ally,” or “imperialist” [24]. The increased hostility
toward China in relation to the air pollution issue may reect how the South Korean
public sees it as an enemy, an actor that could potentially pose a threat, reducing their
will to cooperate (i.e. to strengthen alliances/relations). Future research should thus
examine whether this negative causal eect on foreign policy preferences is moder-
ated by particular sets of images and threat perceptions. Lastly, this work has limited
external validity; despite the low salience of the air pollution issue in the presidential
elections that took place in March 2022, it was still covered in preceding debates.
This could have served to heighten the public’s overall awareness of the issue. As the
online survey took place after these presidential elections, that increased familiarity
may have contributed to the results obtained—albeit with less impact on the results’
internal validity.
Supplementary Information The online version contains supplementary material available at https://doi.
org/10.1007/s11366-023-09849-z.
Funding Open Access funding enabled and organized by Projekt DEAL.
Declarations
Conflict of Interest There is no potential conict of interest to report regarding this research.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,
which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long
as you give appropriate credit to the original author(s) and the source, provide a link to the Creative
Commons licence, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line
to the material. If material is not included in the article’s Creative Commons licence and your intended use
is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission
directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/
licenses/by/4.0/.
1 3

E. E. Song
References
1. Air Pollutant Informant. https://www.facebook.com/dusttenki/. Accessed 1 Dec 2021.
2. All News. 2021. S. Koreans with Unfavorable Views of China Rise Sharply This Year: Survey. In
Yeonhap News, 28 September. https://en.yna.co.kr/view/AEN20210928006600325. Accessed 12
Dec 2021.
3. Anon. 2018. Protest against the danger of ne dust and air pollution”, 24 March. https://www1.presi-
dent.go.kr/petitions/174292?page=1. Accessed 12 Dec 2021.
4. Armony, A. C., and N. Velásquez. 2015. Anti-chinese sentiment in Latin America: an analysis of
online discourse. Journal of Chinese Political Science 20 (3): 319–346.
5. Baek, S.-G. 2018. “Protest against China regarding ne dust air pollutants”: 200K people so far
participate in the blue house online petition. In MTN News, 29 March. https://news.mtn.co.kr/news-
detail/2018032916135865232. Accessed 12 Dec 2021.
6. Baum, M. A., and P. B. K. Potter. 2019. Media, public opinion, and foreign policy in the age of social
media. The Journal of Politics 81 (2): 747–756.
7. Bennett, W. L., R. G. Lawrence, and S. Livingston. 2006. None dare call it torture: indexing and the
limits of press independence in the Abu Ghraib scandal. Journal of Communication 56 (3): 467–485.
8. Brooke, A., and M. Anderson. Social Media Use in 2021. In Pew Research, 7 April. https://www.
pewresearch.org/internet/2021/04/07/social-media-use-in-2021/. Accessed 4 April 2022.
9. Brown, A. 2018. What is so special about online (as compared to oine) hate speech? Ethnicities
18(3): 297–326.
10. Cha, J.-M. 2017. The change of korean perceptions toward China since 1992: normalization of the
diplomatic relations. The Journal of Asiatic Studies 60 (2): 10–45. [In Korean].
11. Chawla, N. V., K. W. Bowyer, L. O. Hall, and W. P. Kegelmeyer. 2002. SMOTE: synthetic minority
over-sampling technique. Journal of Articial Intelligence Research 16: 321–357.
12. Chen, Y., H. Zhang, R. Liu, Z. Ye, and J. Lin. 2019. Experimental explorations on short text topic
mining between LDA and NMF based schemes. Knowledge-Based Systems 163: 1–13.
13. Cheong, S.-H. 1991. The politics of anti-japanese sentiment in Korea. 1945–1952. Praeger. Japanese-
South Korean relations under American occupation.
14. China-Korea Environmental Cooperation Center, https://www.chinakoreaecc.org.cn/html/About.
Accessed 12 Dec 2021.
15. Cho, I.-H., and S.-H. Park. 2001. Anti-chinese and anti-japanese sentiments in East Asia: the politics
of opinion, distrust, and prejudice. The Chinese Journal of International Politics 4 (3): 265–290.
16. Cohen-Almagor, R. 2011. Fighting hate and bigotry on the internet. Policy & Internet 3 (3): 1–26.
17. Entman, R. M. 2003. Cascading activation: contesting the White House’s frame after 9/11. Political
Communication 20 (4): 415–432.
18. GlobeScan. BBC World Service Poll. www.globescan.com.
19. Government Information Agency. 1996. Research on Koreans’ perception and value.
20. Gries, P., and Y. Masui. 2022. How history wars shape foreign policy: an ancient kingdom and the
future of China–South Korea relations. Journal of East Asian Studies 22 (1): 1–21.
21. Gu, Y., X. Zhang, and Z. Lin. 2022. “Factors aecting attitudes toward Migrants: an International
Comparative Study”. Chinese Political Science Review 7 (2): 234–258.
22. Gülseven, E. 2021. “Identity, nationalism and the response of Turkey to COVID-19 pandemic”. Chi-
nese Political Science Review 6 (1): 40–62.
23. Hameleers, M., L. Bos, and C. H. de Vreese. 2017. “They did it”: the eects of emotionalized blame
attribution in populist communication. Communication Research 44 (6): 870–900.
24. Herrmann, R. K. 2013. Perceptions and image theory in international relations. In Oxford handbook
of political psychology, eds. L. Huddy, D. O. Sears, and J. S. Levy, 2nd ed., 334–363. Oxford Univer-
sity Press.
25. Hwang, Y. 2011. The eect of Twitter use on civic and political participation. Korean Journal of
Journalism & Communication Studies 55 (6): 56–81. [In Korean].
26. Imai, K., L. Keele, and D. Tingley. 2010. A general approach to causal mediation analysis. Psycho-
logical Methods 15 (4): 309–334.
27. Ismail, M., A. Shah, and M. Idrees. 2022. “The Dynamics of Anti-Americanism among Educated
Youth in Pakistan.” Chinese Political Science Review.
1 3

Air Pollution Coverage, Anti-Chinese Sentiment, and Attitudes Towards…
28. Jang, Y. 2021. Ideological discrepancy in media coverage on ROK-China controversy: comparative
analysis on China’s Northeast Project and the THAAD deployment. Journal of Research and Meth-
odology 6 (3): 77–106. [In Korean].
29. Jee, M.-S. 2001. Lessons from the China–South Korea garlic conict. LG economic Periodical 5
(23): 18–21. [In Korean].
30. Jeong, D.-S. 2017. Presidential candidates promise to “talk” to China, but they need to pre-
pare what they will talk about rst. In Edaily, 1 May. https://www.edaily.co.kr/news/
read?newsId=01292326615924656& mediaCodeNo= 257. Accessed 13 Dec 2021.
31. Kim, H.-W. 2019. China says Korean media exaggerated the role of China contributing to air pol-
lutants in Korea, yet Minster Cho did not refute the claim. In Chosun Ilbo, 9 March. https://www.
chosun.com/site/data/ html_ dir/2019/03/09/2019030900178.html. Accessed 13 Dec 2021.
32. Kim, S.-O. 2018. President Moon tells Yang Jiechi that China is partly responsible for the air pollu-
tion issue. In Yeonhap News, 30 March. https://www.yna.co.kr/view/AKR20180330178000001?inp
ut=1195m. Accessed 13 Dec 2021.
33. Kim, S.-Y. and S-J. Lee. 2020. South Korean perception of the United States and China: United
States, a more favorable partner than China. In EAI Issue Brieng. https://www.eai.or.kr/new/en/
pub/view.asp? intSeq=19850&board=eng_issuebriefing&keyword_option=&keyword=&more=.
Accessed 13 Dec 2021.
34. Kim, Y., H. Lee, Y. Jang, and H. Lee. 2015. How does media construct particulate matter risks?: a
news frame and source analysis on particulate matter risks. Korean Journal of Journalism & Com-
munication Studies 59 (2): 121–154. [In Korean].
35. Ko, S.-M. 2019. Conservative groups wage protests against ne dust pollution in front of the Chinese
embassy – We do it because the government cannot. In Chosun Ilbo, 7 March. https://www.chosun.
com/site/data/ html_ dir/ 2019/03/07/2019030702406.html. Accessed 13 Dec 2021.
36. Korea Customs Service. https://unipass.customs.go.kr/ets/. Accessed 2 Dec 2022.
37. Lai, B. H., and N. F. Martini. 2021. “Nationalism and its Eects on Attitudes about Trade, Coopera-
tion, and Immigration”. Fudan Journal of the Humanities and Social Sciences 14 (1): 1–22.
38. Lee, I-H. 2020. 2.52 million foreigners residing in South Korea … South Korea is now a multicul-
tural society. In Hankyung, 13 April. https://sgsg.hankyung.com/article/2020041010691. Accessed
12 Dec 2021.
39. Lee, O.-S. 2021. Who are the people that dislike everything about China? In SisaIn, 17 June. https://
www.sisain.co.kr/news/articleView.html?idxno=4482. Accessed 13 Dec 2021.
40. Lee, S. 2019. Opposing leftist heaven, pro-North Korean regime Liberty Korea Party wages protest
in Gwanghwamun – approximately 20,000 persons participated. In Yeonhap News, 20 April. https://
www.yna.co.kr/view/AKR20190420032851001?input=1195m. Accessed 2 Dec 2022.
41. Lin, H.-Y. 2021. “COVID-19 and american attitudes toward U.S.-China Disputes”. Journal of Chi-
nese Political Science 26 (1): 139–168.
42. Liu, T., T. Guan, and R. Yuan. 2022. “Can Debunked Conspiracy Theories Change radicalized views?
Evidence from racial prejudice and anti-china sentiment amid the COVID-19 pandemic.” Journal of
Chinese Political Science.
43. Ma, B., Q. Tan, and P. Du. 2022. “Public Opinion and Social Justice in China”. Journal of Chinese
Political Science 27 (4): 619–636.
44. Malhotra, N., and A. G. Kuo. 2008. Attributing blame: the public’s response to Hurricane Katrina.
The Journal of Politics 70 (1): 120–135.
45. Maoz, Z., and B. Russett. 1993. Normative and structural causes of democratic peace, 1946–1986.
American Political Science Review 87 (3): 624–638.
46. Mellon, J. and C. Prosser. 2017. Twitter and Facebook are not representative of the general popula-
tion: political attitudes and demographics of british Social Media users. Research & Politics 4 (3):
1–9.
47. Ministry of Foreign Aairs. 1992. Information on the establishment of diplomatic relations with
China.
48. Moon, H. 2015. Fine dust pollutants at “serious” levels in China – government posts red alert. In
Newsway, 19 December. http://m.newsway.co.kr/news/view?ud=2015121916041480094. Accessed
2 Dec 2022.
49. Moon, K. H. S. 2012. Protesting America: democracy and the U.S.–Korea alliance. University of
California Press.
50. New York Times reports that China has won the war against ne dust air pollutants. 2018. In YTN,
14 March. https://ytn.co.kr/_ln/0104_201803141700071055. Accessed Dec 2022.
1 3

E. E. Song
51. Newman, N., R. Fletcher, A. Schulz, S. Andı, C. T. Robertson, and R. K. Nielsen. 2021. Reuters
Institute digital news report. Reuters Institute. https://reutersinstitute.politics.ox.ac.uk/sites/default/
les/2021-06/Digital_News_Report_2021_FINAL.pdf. Accessed 4 April 2022.
52. Papageorgiou, M., and D. S. N. Melo. 2022. “China as a responsible power amid the COVID-19
Crisis: perceptions of Partners and Adversaries on Twitter”. Fudan Journal of the Humanities and
Social Sciences 15 (2): 159–188.
53. Park, C.-H. 2013. National identities and South Korea–Japan relations. In National identities and
bilateral relations: widening gaps in East Asia and Chinese demonization of the United States, ed. G.
Rozman Woodrow Wilson Center Press.
54. Poland, B. 2016. Haters: harassment, abuse, and violence online. University of Nebraska Press.
55. Rhee, J., and S.-H. Kim. 2018. News frames in the coverage of ne-dust disaster – application of
structural topic modeling. Korean Journal of Journalism & Communication Studies 62 (4): 125–158.
[In Korean].
56. Röder, M., A. Both, and A. Hinneburg. 2015. Exploring the space of topic coherence measures. In
Proceedings of the Eighth ACM International Conference on Web Search and Data Mining, 399–408.
57. Schultz, K. A. 1999. Do democratic institutions constrain or inform? Contrasting two institutional
perspectives on democracy and war. International Organization 53 (2): 233–266.
58. Shin, G.-W. 2021. The rise of anti-Chinese sentiments in South Korea: Political and security implica-
tions. In Live Webcast at Korea Society Forum 2021, https://fsi.stanford.edu/news/ rise-anti-chinese-
sentiments-south-korea-political-and-security-implications. Accessed 13 Dec 2021.
59. Shin, G.-W., H. Gordon, and H. J. Kim. 2022. South Koreans are rethinking what China means to
their nation. In Commentary for Walter H. Shorenstein Asia-Pacic Research Center at Stanford
University. https://aparc.fsi.stanford.edu/news/south-koreans-are-rethinking-what-china-means-
their-nation. Accessed 13 Dec 2021.
60. Silver, L., and C. Huang, and Laura Clancy. 2022. Negative Views of China Tied to Critical
Views of Its Policies on Human Rights. In Pew Research, 29 June. https://www.pewresearch.org/
global/2022/06/29/negative-views-of-china-tied-to-critical-views-of-its-policies-on-human-rights/.
Accessed 2 Dec 2022.
61. Suler, J. 2004. The online disinhibition eect. Cyberpsychology & Behavior 7 (3): 321–326.
62. Tang, Y. and I. Côté. 2021. “How Large-Scale Land Protests Succeed in China: the role of domestic
media”. Journal of Chinese Political Science 26 (2): 333–352.
63. The Institute for Far Eastern Studies. 1993. Korean’s Perception of China.
64. Thomas, C., and S.-H. Choi 2021 Anti-China sentiment and South Korea’s presidential race. In The
Diplomat, September, https://thediplomat.com/2021/09/anti-china-sentiment-and-south-koreas-
presidential-race/. Accessed 13 Dec 2021.
65. Tingley, D., T. Yamamoto, K. Hirose, L. Keele, and K. Imai. 2014. Mediation: R Package for causal
mediation analysis, MIT, http://hdl.handle.net/1721.1/91154. Accessed 4 April 2022.
66. Van Dalen, A., H. Svensson, A. Kalogeropoulos, and E. Albaek, and C.H. de Vreese. 2018. Economic
news. Informing the inattentive audience. New York, NY: Routledge Research in Journalism.
67. Wang, Z., and Y. Tao. 2021. “Many Nationalisms, one disaster: categories, attitudes and evolution of
chinese nationalism on Social Media during the COVID – 19 pandemic”. Journal of Chinese Politi-
cal Science 26 (3): 525–548.
68. Yoo, S.-C. 2019. President Moon asks China to cooperate regarding ne dust air pollutants from
China, China replies “There is not sucient evidence.” In The JoongAng, March, https://www.joon-
gang.co.kr/ article/ 23403331 #home. Accessed 13 Dec 2021.
69. Yoon, H.-W. 2020. The 40s who were the ‘X-Generation’ are loyal supporters of the Democratic
Party of Korea. In Kyunghyang Shinmun, 24 October. https://m.khan.co.kr/politics/politics-general/
article/202010241128011#c2b. Accessed 2 Dec 2022.
70. Zaller, J., and D. Chiu. 1996. Government’s little helper: US press coverage of foreign policy crises,
1945–1991. Political Communication 13 (4): 385–405.
Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in published maps
and institutional aliations.
Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under
a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted
manuscript version of this article is solely governed by the terms of such publishing agreement and appli-
cable law.
1 3