Patterns of online participation in climate change discussion in Russia
The field of global warming online discussion. Climate change attitudes in Russia. Turning personal experience into political attitudes: the effect of local weather on americans’ perceptions about global warming. Public understanding of climate change.
| Рубрика | Экология и охрана природы |
| Вид | курсовая работа |
| Язык | английский |
| Дата добавления | 24.08.2020 |
| Размер файла | 3,1 M |
Отправить свою хорошую работу в базу знаний просто. Используйте форму, расположенную ниже
Студенты, аспиранты, молодые ученые, использующие базу знаний в своей учебе и работе, будут вам очень благодарны.
Размещено на http://www.allbest.ru/
FEDERAL STATE AUTONOMOUS EDUCATIONAL INSTITUTION FOR HIGHER PROFESSIONAL EDUCATION NATIONAL RESEARCH UNIVERSITY HIGHER SCHOOL OF ECONOMICS
BACHELOR'S PROJECT
PATTERNS OF ONLINE PARTICIPATION IN CLIMATE CHANGE DISCUSSION IN RUSSIA
Field of study: 39.03.01 Sociology
Degree programme: Sociology and Social Informatics
St. Petersburg School of Social Sciences and Area Studies
Polina Semenovna Eilon
Saint Petersburg 2020
Chapter 1. Introduction
Climate Change awareness is rising worldwide and becoming of an interest not only to stakeholders like scientists and politicians but to general public as well. The discussion on the topic has taken place on panels, political agendas, newspapers, television and the Internet, becoming a ground for a debate for users of the Net (Schдfer, 2012).
This discussion around Climate Change and Global Warming in particular has drawn scholars attention due to the controversial attitude towards the issue that occurred around 10 years ago, when climate change was a ground for doubt and misperceptions (Dunlap, 2013). Back then, when the debate was held between those who were sure climate change was happening and those who believed in scientific fabrication and overstatement, Jacques et al. (2008) searched for the reasons of “environmental skepticism”. Conservative elites were proven to be behind the spread of false “scientific” information that resulted in right-wing affiliated public perceiving Global Warming as fraud. Many Western research since then elaborated the connection between political affiliation and the general public attitude towards Climate, concluding that it plays a significant role in public understanding of the issue, where left-wing ideology resulted in higher acceptance and caution rate towards climate change (McCright & Dunlap, 2011).
However, the denial in Western countries (Western Europe, United States, Canada, Australia, etc.) has shown a decline trend since 2010 (Carter & Clements, 2015) and seems like global Warming acceptance has turned into a trending attitude since then. In 2019 “Climate Emergency” was declared as the “Word of the year” by Oxford Dictionary (2019). The scientific and political field seems to reach consensus on the fact that global warming is happening, and more than 20 countries have already announced “climate emergency” in their jurisdiction.
Nevertheless, public concern on global warming is still based on many factors of personal experience (Egan & Mullin, 2012) and political affiliation (Carter & Clements, 2015).
The present research is the first step in types classification of Climate attitudes in Russia. The objective of this study is to classify the existing attitudes in users' comments in news broadcasters' communities in VK social network and compare attitudes found in Russia with those in Western research.
The to show how online discourse around Climate Change and Global Warming is being constructed in comments of Internet News Broadcasters in Russia several questions have to be answered:
What topics are being discussed by users?
What attitudes towards Global Warming exist in online communication?
How these attitudes differ from the ones found in research in other countries?
In the following work I will:
Explore the topics that exist within the corpora of users' comments on Climate Change news with the help of LDA topic modeling
Classify users' attitudes based on their expressed position using framing analysis
Compare users' attitudes and topics within the comment's corpora
Compare results with the results of studies on western data
Chapter 2. Theoretical Background
The field of Global Warming online discussion
Climate change communication is a relatively new research field. However, the interest in public knowledge and attitudes is quickly increasing (Wibeck, 2014). The field of climate communication studies has become a salient issue in many countries. It is a diverse topic that engages and merges with science, politics, economy and culture becoming a complex communicative field (Carvalho, 2010).
Over the years the discussion has spread beyond the borders of political and scientific agenda and involved general public into the discourse (Schдfer, 2012). The research has covered numerous countries, languages and scenes.
One of the most notable fields for communicative studies of science and climate related issues is the Internet and in particular social media scene (Anderson, 2017) which includes media broadcasters (Hmielowski et al., 2014), blogosphere (Elgesem et al., 2015) and social networks (Bloomfield & Tillery, 2019). The field of social media stands for all user - generated social content, which includes social networks interaction, news commenting, blog posts, reviews writing and etc.
In current research, the focus will be directed to one unit of social media - news commenting. Via interactions between users, sharing news, facts, content and information seeking this field encourages participation in the discourse, providing greater knowledge of climate change, place for climate action and debate (Anderson, 2017).
Many studies in the field of Climate change communication online have been done so far. For example, Elgesem et al. (2015) conducted a research in English-language blogosphere. Based on the texts of 1,3 million blogs the authors explored the hyperlink network structure of blogs and topics within the corpora of communities to find out groups of sceptics and those who have different types of acceptive positions. The results of their study showed that in English - speaking blogging environment only one big skeptical community is present, but there is a number of acceptive communities of different levels and types.
Bowen (2008) studied discussion of global warming in online forums and evaluated 8 strategies of climate change denial. Bowen concluded, that denial mostly occurred due to the lack of knowledge or misperception and false understanding. The author also concluded that Internet has a strong influence on the discussion as Internet blogs frequently become a source for scientific information, which sometimes appears to be false and leads to misperceptions.
Comparison of online and offline communication
It is also important to clarify why in particular online discussion and social media is taken into account in present research. (Bowen, 2008) points out, that social media reduce the uncertainty, raises knowledge of users about climate change and (as it will be explained in more detail in the section about boosters of participation) encourages participation in the discussion. This point is crucial for the research on climate attitudes, as more users, that do not hold pro-active positions are involved into the communication. They react to content, share opposite opinions or participate in the discussion in order to provide knowledge.
Anderson (2017) states that there are indicators found in several research that suggest positive relation in social media use and participation in the discussion. Some actors that are not vocal in offline communication may become more salient. Online environment shows a positive influence on communication engagement as it allows some groups to gain power and become visible (Carvalho, 2010).
Online environment was also proven to be a stage that raises environmental knowledge and provides place for environmental dialogue (Rokka & Moisander, 2009) which encourages higher awareness of climate related issues.
Boosters of participation
Personal experience. Weather events
Findings in previous works suggest relation between climate risk perception, knowledge and action with direct personal experience (Whitmarsh, 2008).
Personal experience as a factor of higher climate debate engagement was shown by Egan and Mullin (2012). They studied how personal experience could be transformed into political attitudes. To do this, the authors examined the dependencies of local weather and the awareness of the respondents. Authors state that personal experience is a more significant factor for engagement than factual information. Davis & Fisk (2014) also proved the idea that people show higher levels of engagement when personal risk involved.
Direct experience can make distant and abstract topic of climate change clearer and more understandable and thus, weather experience can lead to higher acceptance that climate is changing. Carmichael and Brulle (2017) while looking for factors that affect levels of anxiety towards climate issue, concluded that extreme weather events is related to public concern.
(Social) media use.
Anderson (2017) studied the connection of social media usage and climate change communication. The research indicates that social media encourages participation in climate change discourse. Social Media becomes one of the most crucial triggers for communication due to its' ability to provide greater knowledge of the issue, filed for activists' action and space for discussion with others. Author also states that media makes the topic of Climate Change more understandable to public with the help of several mechanisms, such as personalization (close-to personal experience news, visualization, gamification) which removes distance between readers and Global Warming.
Demographics. warming climate weather
Some demographical characteristics were also proven to affect the engagement in the climate communication. Honeybun-Arnolda and Obermeister (2019) suggested that young people and millennials are more likely to participate in climate change communication.
McCright et al. (2016) also states that demographic characteristics and the effects of age, education and sex are found to be related to climate change discussion participation.
Public Understanding of Climate Change
Scientific field seems to reach consensus that Global Warming is happening and that anthropogenic causes are significant (Lee et al., 2015), but in general public discourse some are still unsure whether climate is changing and maintain the denial position (Bloomfield & Tillery, 2019). The research in the field has explored different views and positions towards the issue and elaborated different types of climate behavior, attitudes and stances.
Country - specific classification
Based on the controversial background of global warming discussion formation and factors that underlie climate attitudes, many researchers have turned their attention to types of climate change attitudes. This was extremely popular, after the publication of global Warming's Six Americas (Roser-Renouf et al., 2014) that classified American public into 6 groups according to their climate attitudes and perceptions, which included alarmed, concerned, cautious, disengaged, doubtful and dismissive.
Later, this study's methodology was applied to understand Global Warming attitudes in Germany (Metag et al., 2017) resulting in 5 different attitude groups, India (Leiserowitz et al.,2013) showing absolutely different types of climate positions and Australia (Morrison et al., 2013) where segmentation into attitudes differed in comparison to American study. Lee et al. (2015) argues that different predictors are relevant for the climate understanding and the level of awareness in different countries. For example, the temperature change was a predictor for African and Asian countries, when in United States the attitudes towards the government influence global warming risk understanding (here the political ideology is also an important factor - Democrats are more likely to have higher awareness). USA also showed dependency from education, communication access and public engagement (the higher the engagement the higher the awareness of global warming). In China the concern about climate change was affected by the type of the area respondent lived in (rural or urban), education levels and income.
Attitude related predictors
Some of the boosters mentioned in the previous section do not just encourage participation, but are also proven to be related to specific types of taken climate position (denial, skepticism, acceptance).
The research conducted by Carmichael and Brulle (2017) is focused on factors that can affect public attitude towards global warming. Authors built a model where political discourse, extreme weather events, availability of scientific information and economic factors affected media coverage which then shaped public concern about the global warming. The authors are conducting a research to understand which of these factors impact the understanding and found out that the economic factors (rising GDP) has a positive influence on the rise of concern. A model has shown that elite (or political discourse) shows the connection with public concern that happens according the next schema: when congress increases active discussion of climate change issues, media increase the coverage of this political discourse and this leads to increase in public concern.
Russian Case
Countries of Eastern Europe that had been under the Communist Regime, Global Warming discourse are rarely analyzed. Scholars seem not to pay much attention to the discussion in the region and there are several reasons for that. McCright et al. (2016) showed that, due to specific historical formation, different from Western Countries, Global Warming is not an issue of major concern to the general public. It is hard to explain attitudes towards the issue in Eastern Europe from political affiliation point of view as it is mostly done in Western Countries. The distinction into left and right ideologies, where left is mostly associated with freedom and equality whereas right is associated with order is less visible in the countries of former Soviet Union and the common relation between left wing and higher involvement and acceptance of Climate change is weaker. McCright et al. (2016) point out that global warming positions in Eastern Europe can't be explained from left-right division, the authors do not yet suggest what factors may be related to the attitude's formation and what types of climate positions do exist in the former communist countries. Russia is not taken into account in the research, probably due to the language limitations and survey data used by authors. However, Russia is an example of former Soviet Union country and these inferences can perhaps be extended to it (Markowitz, 2000). Hence, this idea needs further investigation.
Based on the literature and specific characteristics of Russia, the hypotheses for the current study are as follows:
1. Personal experience (especially weather) will be used by users as supportive arguments for taken position
2. Stance taking in Russia will not be connected to political affiliation similar to McCright et al. (2016) findings.
3. Topics discussed by users in comments will be related to media coverage in the selected news sources.
4. Attitudes found in the discussion will vary significantly from those found in Western countries.
5. Accepters will support their arguments with scientific information.
Chapter 3. Climate change attitudes in Russia
Data
The VK social network platform, largest social network in Russian Language, was chosen for the following research and searched for the communities of News Broadcasters.
On the first step of selection around 30 communities, marked as Internet Media that specializes in News Broadcasting and followed by more than 450 000 users were selected. Each of the chosen communities was searched for news mentioning either Global Warming or Climate Change. This step was devoted to specifying communities that provided commenting opportunities for users and made at least 1 substantive (not as part of news roundup) post about Climate Change or Global Warming. When the exploratory analysis of the selected media was conducted, only 14 communities were left. A great number of the selected on the first step communities were excluded as they did not publish any news on Global Warming issue or presented posts were not commented by users.
With the help of VKR package Code for data collection can be found in Appendix for R all the posts about climate issue were collected from these 14 communities (using filter that searched for the use of collocations “global warming” or “climate change” and all their forms in Russian language "(глобальн[а-я]+ потеплен[а-я]|изменен[а-я]+ климат[а-я]+)"). After cleaning the amount of collected posts equaled 461 with some communities making only 2 and some publishing almost 50 posts. More precise characteristics of collected media types, amount of posts and comments can be seen in Table 1.
Table 1. Communities of news broadcasters with number of climate change related posts and total amount of comments.
|
Media Name |
Number of Followers |
Total number of Comments |
Total amount of Climate Posts |
|
|
NR NR - Community of Hip-Hop and Rap news that was highly engaged in publishing news of general interest has switched its' focus to music separating general news to an independent community on the 3rd of May 2020 while the research was conducted. However, it was decided to keep collected data. |
5247991 |
4259 |
23 |
|
|
RIA News |
2482211 |
3067 |
42 |
|
|
Lentach |
2273770 |
4461 |
15 |
|
|
Life.ru |
2143825 |
1211 |
48 |
|
|
Post News |
2116105 |
1831 |
37 |
|
|
Mash |
1471857 |
579 |
2 |
|
|
Bad News |
1211145 |
5775 |
10 |
|
|
Kp.ru |
826070 |
931 |
35 |
|
|
RBK |
765404 |
2909 |
33 |
|
|
TJ |
702978 |
842 |
35 |
|
|
«Meduza» |
702115 |
4261 |
48 |
|
|
Vesti |
778230 |
1172 |
44 |
|
|
Lenta.ru |
576075 |
1530 |
49 |
|
|
National Geographic |
1458445 |
1335 |
64 |
|
|
34172 |
483 |
The data is representative as some of the selected communities vary in focus, affiliation and age groups of followers. For example, “RBK” News Broadcaster specializes in Economic and Politics or “RIA” and “Lenta.ru” are federal News Broadcasters, when “Meduza” and “Lentach” represent more liberal opposition. Communities like “NR” and “Mash” mostly attract teenagers and millennials, and “Vedomosti” or “RBK” mostly focus on older users. However, the sample is limited with users of Internet and more precisely those who read and comment news in News Broadcasters in VK social network.
For the selected communities the comments for posts about Climate Change and Global Warming were automatically extracted resulting in 25 780 25 780 out of original 34 172 were left due to the removal of comments with no text, that contained pictures, videos or other attachments; or were deleted in the community comments corpus (356 438 words after removing of stopwords, numbers and punctuation) over 2 years period of time (2018 - 2020). It should be noted that all the available data was collected. In selected communities the topic of Global Warming appears in 2017 / 2018, but due to platform limitations comments for posts published earlier than 2018 were not available for extraction.
For the qualitative part of analysis 20 posts containing climate change and global warming and all the comment beneath were downloaded in PDF files to be uploaded to ATLAS.ti. Posts were chosen randomly for communities that had more than 100 comments for climate change posts and most commented posts were selected for communities that had less then 100 comments. For each community 1 - 2 posts were selected resulting in 2760 units of comments for qualitative analysis.
Table 2. Number of comments selected for framing analysis with commented posts' date, source and of users' reactions.
|
Media Name |
Post date |
Likes |
Reposts |
Comments |
|
|
KP.RU |
12.01.2020 |
112 |
14 |
43 |
|
|
KP.RU |
10.06.2019 |
100 |
9 |
41 |
|
|
LENTA.RU |
22.12.2019 |
118 |
2 |
65 |
|
|
LENTA.RU |
08.09.2019 |
39 |
1 |
58 |
|
|
LENTACH |
30.09.2019 |
2437 |
38 |
382 |
|
|
LIFE.RU |
04.04.2020 |
168 |
13 |
56 |
|
|
LIFE.RU |
13.07.2019 |
53 |
5 |
85 |
|
|
MASH |
09.01.2019 |
56 |
166 |
478 |
|
|
MEDUZA |
18.05.2020 |
210 |
21 |
95 |
|
|
NATGEO |
09.10.2019 |
154 |
7 |
53 |
|
|
NATGEO |
21.11.2019 |
84 |
6 |
83 |
|
|
NR |
12.01.2020 |
4285 |
62 |
355 |
|
|
POSTNEWS |
10.01.2020 |
575 |
25 |
118 |
|
|
RBK |
26.01.2020 |
76 |
12 |
155 |
|
|
RBK |
10.01.2020 |
128 |
7 |
207 |
|
|
RIA |
02.10.2019 |
262 |
50 |
116 |
|
|
TJ |
21.09.2019 |
121 |
4 |
50 |
|
|
VESTI |
06.10.2019 |
166 |
17 |
58 |
|
|
VESTI |
22.03.2020 |
343 |
51 |
62 |
|
|
BAD NEWS |
24.09.2019 |
403 |
14 |
200 |
|
|
TOTAL |
2760 |
Collected data has some important advantages as well as some limitations. The main gain of the selected data is it being nonreactive, due to the fact that users write comments not knowing these comments are collected for the research, and as a result not being affected with the knowledge of being observed (Salganik, 2018). However, this is only true for scientific observation, where users are unaware of being studied. Shmueli (2016) claims that some users modify their behavior in order to avoid legal punishment or embarrassment. Law regulation for Internet media is an important characteristic of Russian reality. For the specific research this aspect of users' environment that is being reconstructed through Internet communication is assertive.
2 main limitations of the data collected quantitively are dirtiness, as spam comments and advertisements are being collected as well, and lost information like pictures, videos and emojis that could be of particular use for understanding the discourse. It is also incomplete (Salganik, 2018), it can't explain meanings and senses are hardly to be found.
It is also possible that discussion is mainly attended by users that have high rate of interest which means that neutral position towards the problem may not be seen at all - this idea has to be precisely careful with when analysis is conducted and biased inferences not to be done.
The present research is a mixed method study. This approach is aiming at reducing limitations of qualitative analysis, like bias and subjectivity.
Methods
The main goal of current research is to present the structure of Global Warming online discussion among general public in Russia and compare it to Western data, received in the field. In particular, the following steps will be devoted to content analysis and framing of the discourse in online communities of News Broadcasters. The results will be compared to those received in other countries and arguments that may influence the structure of discourse will be suggested.
In current study mixed method approach of content analysis is used to make wider inferences about general population. Content analysis is one of the most prominent methods in Climate Change communication research (Metag, 2016).
LDA topic modeling
To explore topics in the corpus of users' comments LDA topic modeling was used. LDA (Latent Dirichlet Allocation) is an unsupervised method for topics detection within documents. It is a hierarchical Bayesian model main assumption of which is that documents are represented as random collections of latent topics that can be explained by words distribution (Blei, 2003). Application of the method provides ability to explore the structure of topics implied in corpus.
However, the method has some limitations. LDA does not suggest the number of topics to be used over the corpora, so the optimal has to be found while applying different numbers in model.
The method has already been used in some studies on Global Warming discussion. Elgesem et al. (2015) used this approach in the research of Global Warming discussion structure and content in blogosphere. The authors apply topic modeling to characterize the discourse differences in topic distribution over communities of skeptics and accepters.
In current research LDA is used for 2 purposes. First, to have a look at the data and predict frames for qualitative analysis in the second part. Due to the fact, that first round of framing is conducted inductively, the possible frames should be explored first. However, the model is also important to expand qualitative inferences to the larger population, when framing is finished. Basit (2003) claims that using quantitative analysis in qualitative research may add important insights to the analysis which may not be found otherwise.
Comments framing analysis
To classify types of attitudes towards global warming in Russian online communities manual inductive framing analysis is used.
This method requires the development of a codebook that is assigned to the units of analysis by hand. Each comment is marked with its' relevance to specific frame. For code-book development several rounds of content and framing analysis are applied.
Framing analysis is widely used in different fields of climate change communication studies (Schдfer & O'Neill, 2017). It has been applied while analyzing media coverage, patterns of communication between stakeholders and public understanding of the issue.
There is one important characteristic of manual framing present. Only comments that contribute to specified frame are coded. This process is explained by Entman (1993) as “selection” and “salience”, which means that some units (even if they contribute to the discussion) can become muted if they are not relevant for sense making.
Manual coding has some limitations, so for framing only 20 posts with comments threads were selected, dated 2019 - 2020.
In current research generic frames Entman et al. (2009) will be used, similar to (Semetko & Valkenburg, 2000) study, where media news were framed and 5 categories were developed, including such frames as “consequences”, “human interest” and “responsibility”.
Approach used in this study was inspired by the idea of attitudes classification used in several works with different methodology. For example, popular “Global Warming Six Americas” study (Roser-Renouf et al., 2014), that spread to take place in different countries including Germany (Metag et al., 2017) and Australia (Morrison et al., 2013) and examined types of attitudes in social media towards global warming in Western countries and compared findings with each other. The research of Elgesem et al. (2015) that used LDA topic modeling to explore the topics in communities of different climate positions, like skeptics and acceptors. And Nisbet (2009) work that framed comments for climate communication research.
The combination of these approaches resulted in a mixed method study for classification of expressed attitudes in online communities of news broadcasters.
Analysis
LDA
After applying different number of topics, optimal of 25 was chosen, where the output was meaningful for the analysis. The model was run multiple times with different number of topics to find the most interpretable one. Irrelevant stopwords were removed after each round of model launch. The best model selected for the analysis is presented below Original model on Russian language without translation and graphs with closer look on some topics can be found in Appendix. .
Figure 3. LDA model for 25 topics with labels for 25 780 comments of climate change news.
|
1 |
russia russian putin great thief army host liberal our everywhere |
Russian politics |
|
|
2 |
fade bee zoo bird titmouse church comment call page sin |
Birds and bees extinction |
|
|
3 |
permafrost perpetual protocol turn out melting learn Kyoto history labor perennial |
Permafrost |
|
|
4 |
news very serious move shit comment good bad bot fake |
Fake |
|
|
5 |
gas emission atmosphere carbon dioxide greenhouse few planet emit oxygen |
Carbon emissions |
|
|
6 |
our let planet time mess mother serious person turn disentangle |
Damage to the planet |
|
|
7 |
putin think power speak election watch vote president your people |
President election in Russia |
|
|
8 |
water ice glacier level ocean sea melt melt arctic leave |
Glacier melting |
|
|
9 |
article scientist write read source data scientific link information fact |
Science. Scientific justification |
|
|
10 |
garbage factory new city water live machine let everyone buy |
Factories and cars pollution |
|
|
11 |
year winter summer snow day warming warmth russia back weather |
Weather experience |
|
|
12 |
eat meat drink disease treat coffee food weight sick fat |
Consumption |
|
|
13 |
money tax pay work russia country want to receive people give |
Taxes and economic situation in Russia |
|
|
14 |
forest fire burn disappear siberia stew die plant bee seriously |
Forest fires |
|
|
15 |
girl greta go *** *** *** *** go *** well done |
Greta Thunberg. Aggressive Original language used can be found in Appendix |
|
|
16 |
man bear animal dying save friend white poor being sorry |
Save white bears |
|
|
17 |
power law usa ukraine russia russian people war day regime |
Conflict with Ukraine |
|
|
18 |
year woman court also center refugee bank letter be darkness |
Incoherent |
|
|
19 |
warming climate global year change earth temperature occur period vary |
Climate change periods |
|
|
20 |
know your write your write can be nonsense understand the word speak |
Understanding climate change |
|
|
21 |
man planet earth live nature our soon humanity god end |
Awarness. Consequenses for humanity |
|
|
22 |
give school strike schoolboy sit out leave russia immediately go wait |
Skipping school for strikes |
|
|
23 |
man can problem talk the most understand thing do nothing think |
Lack of action |
|
|
24 |
child give birth want woman life live right to love every family |
Woman right for childbirth |
|
|
25 |
russia country usa china europe population america air chinese trump |
Comparison of Russia and other countries |
Each topic is labeled with a thematic code. To label the topics top documents for each were explored and if the content still remained unclear the topic was marked as “incoherent”. In the model only one topic is marked with this label.
This model gives interpretable result, which helps to build initial frames for the qualitative analysis.
Several topics related to Russian politics on different levels can be seen in the model. This result was predictable as previous work emphasize that climate change discourse is a politicized field. Here topic #13 stands for political - economic situation in Russia and citizens' income and topic #7 relate to the president election.
Several topics also represent consequences of climate change for nature like forest fires (topic #14), glacier (topic #8) and permafrost (topic #3) melting, danger for white bears (topic #16) and extinction of birds and bees (topic #2). Consequences for people are presented by topic #21 which stands for catastrophic effects of climate change for the humanity.
There is also a topic for personal experience with weather (#11). However, some information is lost here, due to quantitative analysis limitations. Users write about temperature numbers and when data was cleaned all the numbers were deleted, which led to information loss in this topic.
On this step, attitudes, like “concerned about the consequences” can be seen. However, to build initial frames I need a closer look on my data. I will come back to LDA results after framing.
Framing.
1st round. Pilot inductive framing.
To expand results, received in LDA, the pilot manual coding was held. Over the first round, comments were marked according to the topics they contained. This round was devoted to specification what topics, found in LDA were related to climate change or global warming and what attitudes they relate to. Over this step several frequent thematic nodes for 3 general frames were coded (similar to (Metag et al., 2017) framing method)
4 posts with largest number of comments from different News Broadcaster communities 1 post from each of the following community was selected: Lentach, RIA, RBK, NR. They were chosen as they represent different types of audience, have different focus and showed high engagement into climate change discussion under climate related news posts with all comments beneath were coded. Table 4 represents inductive node tree with number of coded categories. 934 comments were read and framed at this step. Out of this number only around 200 were coded. There are several reasons for that. First, not all the comments under climate change news are topic related, and those comments that are not relevant to research were skipped. Second, the first round was a trial and most of the nodes were found during the coding, so some comments could have been missed in process.
Table 4. Initial frames divided into 3 categories with explanation and number of documents.
|
Who is responsible |
||||
|
Russia |
Lack of action in Russia, ignorance of problems (ex. Forest fires in Siberia) |
15 |
||
|
West |
Blaming West. On the grounds of carbon emissions, ignorance or without explanation. |
13 |
||
|
Politicians |
Blaming politics or politicians in lack of action or otherwise overacting in spite of political agenda |
9 |
||
|
People |
People or humanity is responsible for climate change |
11 |
||
|
Factories // carbon emissions |
Factories do not reduce solution and carbon emissions and hence are responsible for climate change |
5 |
||
|
No one |
Climate change is not anthropogenic, and even if it is happening, the process is natural |
17 |
||
|
What is the problem |
||||
|
Weather change |
Weather experience interpretation - both acceptive and denial |
44 |
||
|
Problem does not exist |
Complete denial. Climate Change does not exist |
9 |
||
|
Doubt if problem exist |
Skepticism if climate is changing |
21 |
||
|
Existance of the humanity |
Catastrophic change, that will lead to people's death |
13 |
||
|
Nature damage |
Nature and animal world damage |
15 |
||
|
It is not a problem in Russia |
Climate change either will not affect Russia or will be beneficial to it (better weather, standards of living) |
15 |
||
|
Life change |
Life will change together with climate (migration, famine, lack of drinkable water etc.) |
8 |
||
|
What is the solution |
||||
|
People awarness |
People can become more responsible and reduce damage |
7 |
||
|
Political action |
Politicians can take action and save the planet |
4 |
||
|
Inavitable |
Either change is natural, and people can't do anything, solution does not exist |
11 |
||
|
Technological proress |
People can invent solution |
2 |
||
|
No need |
Belief that solution is not needed, and problem will solve itself |
3 |
Second step of this round was devoted to closer read of coded comments, results from LDA model and classifications in previous research. This resulted in 5 categories for the second round of framing (Table 5).
Table 5. Pilot coding classification of climate change attitudes with definition and examples 2nd round. Pilot framing.
|
Classification of the attitude |
Definition |
Example |
|
|
Skepticism |
Doubt in climate change existence, uncertainty in information provided, questioning the effects, size and consequences of climate change |
“Глобальное потепление это оч раздутая и преувеличенная тема...”“Я жду, ?огда льды Антар?тиды начнут таять... вот, тогда можно будет реально говорить о глобальном потеплении...” |
|
|
Denial |
Firm position that climate change does not exist |
“Неужели ?то-то деи?ствительно верит в миф о глобальном потеплении”“Глобальное потепление,это миф придуманныи? мировым правительством, Барухами и Ротшильдами.” |
|
|
Catastrophism |
Alarmed position of catastrophic consequences of climate change |
“Сегодня австралия выгорит. Завтра Китаи?. Потом и до нас дои?дет.”“Осталось лет 20,потом большие проблемы,у тех, ?то близо? ? уровню мирового о?еана” |
|
|
Acceptance |
Agreement that climate is changing |
“Глобальное потепление по всему миру и не толь?о в России, в дубае потоп в буэнос аи?рес снег” |
|
|
Beneficial |
Position, that climate change is beneficial for Russia. |
“Очень хорошо, может ?лимат у нас на?онец станет нормальныи?.”“Единственное что сделает с Сибирью потепление это повысит среднегодовую температуру что позволит использовать земли в сельс?ом хозяи?стве” |
Created classification was applied to posts selected for framing analysis. At this step 50% of the posts were framed and new frames were extracted. Acceptance frame appeared to be broader and had to be divided into 2 - anthropogenic acceptance and position, that climate change happens due to natural reasons. This division appeared in comments mostly in form of argument, where users promoting different attitudes quarreled about climate change grounds. One more frame was found - disengaged. This frame includes only comments that expressed position of indifference (For example, “Не доживу раньше в ящи? сыграю...?а? с?азал один французс?ии? ?ороль Людови? "после нас хоть потоп"”).
Two more frames were added - blame and belief. Blaming position includes several categories - blaming politicians (most frequent), blaming factories and carbon emissions and blaming people. Belief, the least frequent attitude stands for the position that problem will be solved or will solve itself.
3rd round. Final framing
Final codes that were used for coding included 9 categories of attitudes that were presented in comments of users that expressed position on climate change issue. Attitudes, their definitions and examples can be seen in Table 6.
Table 6. Classification of users' expressed positions of climate change news in VK communities of news broadcasters with definitions and examples.
|
Classification of the attitude |
Definition |
Example |
|
|
Skepticism |
Doubt in climate change existence, uncertainty in information provided, questioning the effects, size and consequences of climate change |
“Глобальное потепление это оч раздутая и преувеличенная тема...”“Я жду, ?огда льды Антар?тиды начнут таять... вот, тогда можно будет реально говорить о глобальном потеплении...” |
|
|
Denial |
Firm position that climate change does not exist |
“Неужели ?то-то деи?ствительно верит в миф о глобальном потеплении”“Глобальное потепление,это миф придуманныи? мировым правительством, Барухами и Ротшильдами.” |
|
|
Catastrophism |
Alarmed position of catastrophic consequences of climate change |
“Сегодня австралия выгорит. Завтра Китаи?. Потом и до нас дои?дет.”“Осталось лет 20, потом большие проблемы,у тех, ?то близо? ? уровню мирового о?еана” |
|
|
Acceptance (anthropogenic) |
Acceptance of anthropogenic causes on climate change. |
“тюлени не размножаются из за тон?ого льда, поэтому медведям нечего ?ушать и лишь челове? виновен в этом глобальном потеплении”“Плохо ?огда челове? засирает атмосферу и ломает баланс, ?оторыи? был достигнут в природе за миллионы лет.” |
|
|
Beneficial |
Position, that climate change is beneficial for Russia. |
“Очень хорошо, может ?лимат у нас на?онец станет нормальныи?.”“Единственное что сделает с Сибирью потепление это повысит среднегодовую температуру что позволит использовать земли в сельс?ом хозяи?стве” |
|
|
Non - anthropogenic climate change |
Position that climate change is happening, but reasons for this change are not anthropogenic. |
“Изменение ?лимата - это естественныи? процесс ?оторыи? продолжается миллионы лет и регулируется самои? природои?.”“глобальное потепление существует, толь?о челове? ? нему ни?а?ого отношения не имеет. ?лимат всю историю земли менялся ци?лично” |
|
|
Blame |
Any blaming attitude. Blaming politics or politicians for profit from climate change or ignorance, blaming factories and cars for emissions and people for lack of action. |
“власти подделывают отчеты об э?ологичес?ои? ситуации в та?их регионах ?а? челябинс?, они завышают различные нормы для выбросов хими?атов, чтобы заводы могли еще увеличивать объем производства. В общем мерзости вся?ие творят. А мы ?аждыи? раз за них голосуем снова и снова” |
|
|
Belief |
Belief that climate change will either resolve or solution will be found. |
“Да все будет нормально ? этому времени все равно технологии будут намного лучше, чем сеи?час. У нас еще? есть марс))”“Выделят триллионы долларов на борьбу с этим ужасным недугом и все будет снова хорошо” |
|
|
Disengaged |
Expressed attitude of indifference. (potential comments, from users that have seen the post, but did not comment are not included) |
“тебе не до ?лимата, если ты живешь в раш?е”“Проблема реальна толь?о для следующего по?оления. Я представитель нынешного и для меня не существует проблемы.” |
Over the final coding only 699 comments (out of 2760 that were selected initially) were coded as position expressing. Figure 7 shows the percentage distribution of attitudes in coded comments.
Figure 1. Percentage of attitudes in comments that express position towards climate change.
Out of 699 comments coded in frames the most frequent attitude was acceptive. This category included experience-based acceptance (agreement with climate change due to weather experience), knowledge-based acceptance (if the acceptive attitude was supported arguments) and acceptance with no clear grounds (this position is pro-acceptive, but it is hard to say if users' attitude pro-anthropogenic or not).
Blaming category took the second place. It included all types of blaming statements, where 57% of blaming attitude was towards politics and politicians (58 political blaming statements out of 104 blaming comments). This category also included West - blaming attitude (25% of blaming frame) and factories - blaming position (14%).
Catastrophism is also an acceptive position (tough, it is unclear if acceptance is anthropogenic) that is compared to “alarmed” attitude found in research on Western countries (Metag et al., 2017). It is presented in online discussion in news broadcasters in 14% of all coded position expressing comments. This attitude shows the highest concern rate and is presented in form of seeing severe consequences and damage from climate change.
Non - anthropogenic climate change is a stance of natural influence on climate change and cyclic patterns. This attitude is presented in 11% of all coded comments and mostly appears together with anthropogenic acceptance being a ground for the debate of different attitudes. This attitude is knowledge-based in 73% of cases, users rarely share personal experience when expressing this position, on the other hand, users prove the stance explaining periods in planet and climate formation.
Beneficial attitude is the most unexpected, as it wasn't present in western research to any extent. This position is presented in 10% of coded comments and is characterized with seeing profit for Russia in climate change, like better weather, standards of living and new economic advantages, like crops growth in yet frosted areas or new trade routes in yet frosted North Sea.
Skepticism is presented in the same proportion as beneficial attitude. This attitude stands for questioning climate change existence or seeing it as exaggeration. It is closely connected to denial attitude, which represents firm position that climate change does not exist, this position is represented in 9% of position expressing cases. Meanwhile, denial is more often experience based (weather events), while skepticism is knowledge oriented.
One more unpredictable position found in comments is disengaged similar to the one found in American study (Roser-Renouf et al., 2014). This attitude was not expected to be found, as the current research is not interview - based and such position was expected to be muted and just not presented in the discussion. Nevertheless, some communities have ideology (Gershon, 2010) of participation (it could be seen in comments threads where users are familiar with each other) and this ideology encourages even disengaged attitudes to be expressed. This position takes 9% of coded units ahead of denial position (which rejects Williams et al. (2015) idea of tendency to express strong opinion of belief or disbelief in climate change rather than having a neutral position on the fact that climate change is happening).
The smallest attitude found was belief, which stand for the ideas that nature will return the balance itself, or that humanity will find solution. This attitude was not coded into any other frame due to the fact, that it is the most optimistic and differs significantly from any other. It is presented only in 3% of cases (only 20 comments out of 699 coded).
LDA topics related to different types of attitudes.
When framing was done and 9 attitudes towards climate change classified, LDA topic modeling was once again explored, to specify topics related to positions found in framing analysis.
Acceptive position is represented in topics about anthropogenic contribution to climate change.
|
# 5 |
gas emission atmosphere carbon dioxide greenhouse few planet emit oxygen |
Carbon emissions |
Carbon emissions often used as proving argument in acceptive attitude as well as in blaming.
|
#6 |
our let planet time mess mother serious person turn disentangle |
Damage to the planet |
#6 topic also stands for anthropogenic damage to the planet that people are now responsible for.
|
#9 |
article scientist write read source data scientific link information fact |
Science. Scientific justification |
Scientific information topic represents both man-made and non-anthropogenic climate change positions (#9).
|
#19 |
warming climate global year change earth temperature occur period vary |
Climate change periods |
Idea of natural climate change is often supported by cycles of climate and climate periods and eras (topic #19).
|
#21 |
man planet earth live nature our soon humanity god end |
Awarness. Consequenses for humanity |
End of humanity or severe consequences that will affect people's lives, planet, nature etc. characterize catastrophism attitude.
Topics #10, #25 and #23 represent the blaming position, where lack of action, politics, factories and West (and other countries) are blamed
|
#10 |
garbage factory new city water live machine let everyone buy |
Factories and cars polution |
|
|
#23 |
man can problem talk the most understand thing do nothing think |
Lack of action |
|
|
#25 |
russia country usa china europe population america air chinese trump |
Comparison of Russia and other countries |
Some attitudes are not presented in LDA topic, such as disengaged and beneficial. These attitudes, probably, merge with other topics, but methods limitations can't provide a clear answer for that.
However, some topics found in LDA are not connected to any attitude either, which are meaningful for understanding climate change discussion in Russia.
When framing was conducted each of these topics were also coded in separate nodes, to then explore what is their connection to the discussion and expressed attitudes.
#13 topic is of particular interest for the analysis. This node was not connected to any of the attitudes and was a separate attitude itself. Comments that contributed to this topic were aimed at changing discussion focus to more “relevant” issues in Russia. This topic is very frequent (for some documents up to 30% coverage) and probably means, that not expressed disengaged position was related to the fact that financial situation is of grater interest for some users.
