Understanding Discourse on Gender Differentiated Impacts of Climate Change for a Social Change Amit Pariyar, Narayanan Kulathuramaiyer Institute of Social Informatics and Technological Innovations Universiti Malaysia Sarawak, Malaysia
[email protected],
[email protected] Abstract The move towards sustainable development has intensified with growing ramifications of climate change leading to extensive online and offline promotions. Since climate change is a generalised theme under sustainable development there is a possibility that some of its vulnerability contexts such as gender differentiated impacts may go unnoticed in social discourse. The reasons are meagre role of civil society in global environmental governance, excessive media coverage on the observable destructs obscuring gender vulnerabilities and opposition to the contemporary social norms rooted with gender inequalities. The discourse is also hindered due to limited access to scientific and official reports on women vulnerabilities and virtuousness in relation to climate change. The tendency to assert gender vulnerabilities in climate mitigation as a problem of developing countries also understates the importance of gender issues in climate change and a reason for limited global discourse. This research investigates how online communities associate gender and climate change in their discourse. We use keyword based query method to extract Twitter datasets and examine user engagement, demographic, geographic coverage and views expressed. Our finding suggests the need for extensive online awareness campaigns and involvement of higher male participation alongside female and youths. We also suggest aggregated views among organization and media required to sensitize climate change and gender issues. Region specific campaigns to target online user communities from climate vulnerable developing countries is also suggested. This research contributed by uncovering online policies to raise awareness on gender differentiated impacts of climate change as an enabler for social change. Keywords: Climate Change; Gender Differentiated Impact; Public Awareness; Discourse; Social Media. 1. Introduction Society is deemed prosperous when its path to development not only focuses on the present needs of the people but also on challenges that the future generations would need to face. One important social responsibility therefore is to protect the environment so that our children and their grandchildren have the resources to survive and evolve. But the drive for social change especially for environmental protection has not been that profound. Climate change is already preeminent and threatening to be the sixth wave of human extinction. Some of its ramifications are already felt in sustainable developmental areas which includes food shortage, declining biodiversity, scarce water resources and disease outbreaks [1]. In a world economic and social survey 2016, the UN reported that climate change impacts are faced disproportionately and unevenly among different sections of society due to various vulnerability contexts such as socio-economic inequalities, increasing development gaps between urban and rural, inadequate resilience and adaptation mechanisms [2]. Sustainable development that is inclusive of all
social and cultural groups in a society has become ever so important in combating climate change. Gender is one of the many vulnerable components of climate change and a pressing issue for sustainable development arising from the unequal power relation between men and women in the society. According to the Intergovernmental Panel on Climate Change (IPCC) Fifth Assessment report, women compared to men are bearing higher burden of climate change [3] which explains for the importance of mainstreaming gender in climate change policy and the need for extensive discourse on this subject to trigger sustainable social change. Though climate change policy responses in the past were criticized to be overly focused on scientific and economic solutions [4], it was on December 2007 that the human side particularly the gender dimension of climate change gained attention with the establishment of the Global Gender and Climate Alliance (GGCA) in Bali, Indonesia at the UN Climate Change Conference [5]. The gender insights into climate change mitigation and adaptation practices has since then become central which is marked by the Gender Day event [6] at the recently held Conference of the Parties (COP23) in Bonn, Germany to highlight the importance of gender responsive climate policy and actions. In the wake of development agencies and government bodies becoming gender sensitive in achieving sustainable development, it is dubious whether societal perceptions have changed over time to assimilate gender differentiated consequences of climate change in the social discourse. This is because when it comes to global environmental governance, study has stressed meagre role of civil society in comparison to the major players, often the national and international caretakers [7]. This leads to knowledge gap with society divided into ‘wellinformed’ and ‘uninformed’ of the many prevalent issues of climate change including gender. The ignorance on sensitive issues is further reinforced by the excessive coverage of media on the observable destructs caused by climate change, serving society with the surficial view and obscuring the hidden vulnerabilities, for instance gender issues may go unnoticed in the discourse which is a hindrance to inclusive sustainable development. The link between climate change and gender is intricate and difficult to ideate with for the general public because it involves defying the societal and cultural norms imposed on women by the society. Studies that do link climate change and gender [8][9] primarily focus on two aspects: vulnerabilities one which positions women as physically disadvantaged, socially underprivileged and highly vulnerable compared to men when facing the brunt of climate change, and another is virtuousness which unmasks the differential role played by men and women when it comes to environmental protection. The challenge is in confronting the contemporary social norms rooted with gender inequalities and establishing the climate change and gender linkages in a discourse, which is a gradual process. Women’s vulnerability to climatic stresses is mostly explained by their higher mortality rates than men during disasters, which is experienced strongly in societies where women do not enjoy economic and social rights equal to men [10]. Women and children are fourteen times more likely to die than man during disaster [11]. In the aftermath of climatic stress, women are also highly exposed to sexual abuse, trafficking, malnourishment, poor sanitation and hygienic condition, lack of pre and post-partum facilities and so on. The reason being unequal conduct of actions towards women who are treated as secondary when it comes to delivering emergency responses. Other than this, the high dependence of women on natural resources also explains
their vulnerabilities to climate change [12]. Many societies in developing countries still typecast women’s role as safeguard of water, food, fuel for their family. Climate change has reduced their ability to secure these indispensable resources due to declining biodiversity [13]. Women and girls have to walk long distances to collect wood and water which results in less time to fulfil their domestic chores, engage in recreational activity and for their education [14]. This also leads to increased risk of sexual assaults due to exposure to unknown territories and injuries from carrying heavy loads to long distances. On the contrary, the virtuousness side of women to climate change is explained by their potential as an agent of change for climate mitigation and adaptation [15] because of their extensive practical knowledge of the environment and resource conservation. Women are also found to be more prepared than men for behavioural change, are conscious of the environment and more likely to support drastic policies and measures on climate change [16]. The above logical narratives are strongly voiced to support gender sensitive climate change policies. The dilemma is that most of the documented reports on hardships faced by women due to changing climate are made on subjective basis as case stories, and in many cases produced by development agencies, government bodies and scientific communities. There is a possibility that the reach of such mediums to general public may not be impactful as it should be to create awareness about gender issues in climate change. Studies have shown discrepancies in people’s awareness and their behaviour when it comes to taking action, for instance purchasing eco-labelled products [17] despite of government measures in dispersing information about environmental hazards. This also leads to doubt as to whether the society at large is aware of how climate change is becoming a burden to women more than men. Apparently there are also no factual evidences on whether community beyond the domain of climate change are aware of its gender differentiated impacts. The public discourses on this sensitive subject offers an important clue on whether society is forming a gender aware opinion towards climate change. The discourse on gender issues also becomes increasing important because of disintegrated opinions on gender and climate change found to occur beyond territorial boundaries. Studies [18] have shown that women from the countries in the South are projected as vulnerable while women from the countries in the North are shown to possess virtues of environmentalism. It is still debated that developed countries consider gender differentiated impacts of climate change as a problem of developing countries. For instance, the Swedish Bill on climate and energy policy regards gender and social inequalities as an issue of developing countries to aggravate climate change problems. Gender as a global problem facing climate change was only recognized in April 2009 by the House of Representatives in the United States [19]. This hints on the possibility of diverse opinions among global community when contextualizing gender in their climate change discourse. Further in societies where climate change is still believed to be hoax by certain sections, gender differentiated consequences of climate change may be understated in the social discourse. This research investigates how online communities in social media platform associate gender and climate change. The purpose is to understand the awareness of communities in linking climate change and its gender differentiated consequences via their online discourse, and uncover policies for social integration in response to climate change. Over the last decades Twitter has turned out to be the most pervasive platform for communities to openly debate, converse with varied perspectives. The enormous data generated from Twitter has been used
in many studies to understand social dynamics of interaction on various prevalent issues in society such as racial discrimination [20], political injustice and violence [21] [22]. Examining Twitter dataset can provide crucial information on user engagement and attitudes in their online discourse on climate change and gender, to imply their awareness in this sensitive subject. This paper is organized as following. Section 1 establishes the linkage of climate change and its gender differentiated consequences as prevalent in the development, government and scientific studies. The gap in the current studies of social media targeted at gender and climate change is highlighted in Section 2. We present keyword based query method to analyse user engagement, demographics, views expressed and geographic coverage in the topic of interests in Section 3. Finding are summarized in Section 4 along with discussion in Section 5. Limitations are highlighted in Section 6 and a conclusion is made in Section 7. 2. Discourse on Climate Change and Gender The topic ‘climate change’ and ‘gender’ are both full-fledged domain on their own with plentiful literatures, yet there are limited studies on their linkages especially in the literatures of social media discourse. In the climate change domain, the research on global public discourse focuses on polarities in perception [23] [24], spatial-temporal patterns [25], and coverage [26] in response to natural and socio-economic events. The COP events are a common target for studies on social media discourse. The democratic extension of the physical world to the online world is revealed in these studies. For instance, the extent of polarization in online communities emerging from citing the ideological profiles is revealed in a study of several English language blog posts related to Paris meeting [24]. Similar to the physical world, studies have supported peer influences in affecting individual perception, attitudes, behaviours and shaping the public discourse on climate change, for instance likeminded people collectively form an activist or sceptic view [27]. There is also empirical support for the classic two-step model of communication in climate change discourse characterized by high concentration of discussions around a relatively small number of twitter handles of influential personalities, agencies and news media [24] [28]. Studies have also stressed huge differences not only in the volume of coverage but also on the emphasis of various themes around climate change via social media outlets compared to traditional newspaper [26]. Similar trends of research on public discourse in social media are also found in the gender domain. The sociological, policy and computational analyses of gender issues in social media are also driven by similar motivation to investigate demographics, geography of user engagement. The studies topically revolve around online campaigns against gender-based violence (GBV) contrasting and comparing user engagement patterns [29]. There is consistent support for higher female participation in anti-GBV campaigns suggesting better coordination mechanisms for framing the social message [30]. Socio-cultural influence is also prevalent on user engagement with gender issue [31]. There is also evidences of two-step model of communication with emphasis on public figures such as celebrities as central in encouraging people to share their personal stories of gender violence [30]. It can be gathered that there is a growing interest of research community to investigate climate change and gender discretely. Social media studies that do involve gender context in climate change only revolve around gender differences in communication patterns [32], male/female opinions about anthropogenic impacts of climate change [33]. In this light, the study of public perceptions on the gender-differentiated impacts of climate change is at nascent stage. Gender
impacts remains marginalised in studies of online discourses on climate change. This presents an opportunity for our work to bridge this gap between these seemingly unrelated worlds by empirically investigating the public perception on the gender-differentiated impacts of climate change. While past studies have limited gender issues topically to GBV in online discourse, this study goes further by placing gender issues at the centre of the climate change discourse and investigates user engagement, demographics, geographic coverage and views expressed. To examine the public discourses around climate change and its differentiated impacts on gender we focus on primary keywords ‘climate change’, ‘gender’ and ‘violence’ both discretely and jointly, and seek answers to following research questions. RQ1: What is the demographics of user participation? Age, gender distribution. RQ2: How characteristics of user engagement vary with topic of interest? Influential, popular user and hashtags. RQ3: How views expressed by user vary with the topic of interest? RQ4: How geographic coverage of user participation vary with the topic of interest? Close to our work is the study of public engagement, user attitudes on online campaigns specific to climate change [27] and gender-based violence [29] [30]. We differ in the approach (hashtag vs keyword based) and purpose of the research (isolated vs joint investigation).
3. Methodology We use keyword based query method to extract data sets made available to us by the Crimson Hexagon [34] under the data philanthropy initiative for climate action. The platform provides an interface to query large datasets of posts from online social media such as Twitter, Facebook, Google page and Instagram. It also provides functionalities such as social monitor and buzz monitor to mine posts from individual Facebook profiles and twitter handles or across social media accounts. In this research we are interested in mining Twitter datasets to extract posts related to “climate change” and “gender based violence”.
Table 1: List of Discussion Topics and Query
Discussion Topics General 1. Climate Change 2. Gender 3. Violence Specific/Aggregated 4. Climate Change and Gender 5. Gender and Violence 6. Climate Change and Violence 7. Climate Change, Gender and Violence
Keywords Based Query
Total Posts
(climate OR climatechange) (women OR woman OR girl OR gender) (violence OR trafficking)
97460602 4622325160 114584907
(climate OR climatechange) AND (women OR woman OR girl OR gender) (women OR woman OR girl OR gender) AND (violence OR trafficking) (climate OR climatechange)AND (violence OR trafficking) (climate OR climatechange) AND (women OR woman OR girl OR gender) AND (violence OR trafficking)
924972 9719361 149990 11542
3.1 Query Formulation As an initial step we prepare keywords to match our topic of interest and scope of analysis. The main focus is to understand user’s discourse on the topic of climate change and gender both discretely and then jointly. Table 1 lists the discussion topics and the keywords used to query Crimson API for accessing posts related to each topic. The discussion topics include general topics (‘climate change’, ‘gender’, ‘violence’) and specific topics (‘climate change and gender’, ‘climate change and violence’, ‘climate change, gender and violence’). The specific topics are supposed to give detail account of user engagement when associating the topics of interest in their online conversation. For instance, the topic ‘climate change and violence’ is supposed to relate conversation about climate change leading to violence in the form of trafficking. Similarly, the topic ‘climate change, gender and violence’ is supposed to give even more details of climate change and its impact on various forms of gender violence mostly trafficking. We formulate query by using logical connectives ‘AND’, ‘OR’ and keywords for each topic.
3.2 Data Collection By using Crimson API, we extract Twitter posts for a period of five years from August 1, 2011 to August 1, 2017. Table 1 gives the quantity of posts extracted for each discussion topics returned with the query. The general topics have large number of posts compared to the specific topics. The returned posts also form the data sets for each topic which are analysed for user demographics, geographical coverage, user engagement and views expressed.
We also collected Twitter posts from a selective twitter handles between December 19, 2013 to August 1, 2017. First from organizations working in specific areas of interest, @UNFCCC for climate change, @UN_Women for gender, @UN_Nepal for sustainable development areas in Nepal. Second from global broadcast media @BBC, @CNN and national media in Nepal called @thehimalayan. We extracted self-authored posts which refers to posts that were made by selected twitter handles. Table 2 gives the quantity of self-authored posts extracted from the selected twitter handles.
Table 2: Self-Authored Posts of Selected Twitter Handles
Twitter Handles Organization UNFCCC UNWomen UNNepal Media BBC CNN HimalyanTimes
Total SelfAuthored Posts 13999 26938 3295 104968 106124 29282
4. Analysis and Finding The extracted datasets for each discussion topics in Table 1 and 2 are used for the analysis on demographics, user engagement, views expressed and geographic coverage. 4.1 User Demographic To examine the demographics of user engagement such as gender and age, we referred to Crimson API to filter data set into male/ female categories and different age-groups. Fig 1 visualizes the gender distribution of the participants for each discussion topics. The data reveals significant trends in user participation. The female participation is comparatively higher than men (above 50%) for all topics that include gender. In response to RQ1 this suggests that female population is more sensitive about the issues they are facing such as increasing violence, hardship due to climate change, and climate induced violence such as trafficking. The male participation is higher (above 50%) for topics that lack gender sensitivity such as climate change and violence. This suggests that male population is unable to fully comprehend the gender differentiated impacts caused by climate change. As shown in Fig 2 the age distribution of user also shows higher participation (above 50%) of matured age groups above 35 years for each topics except ‘gender’. The high participation (above 60%) of youth age groups below 24 years to discuss topics on gender is due to reference to pornographic materials. The low participation of youth age groups on each topics is a serious concern as they represent the future caretakers of this global crisis of climate change. In response to RQ1, this means that youth participation has to be encouraged and made aware that women are more vulnerable to men due to negative consequences of climate change.
Fig 1: Gender Participation across Discussion topics.
Fig 2: Age Group Participation across Discussion topics. 4.2 User Engagement In Twitter platform user engagement happens in several ways. Tweeting means posting a content by using a twitter handle. Mention ‘@’ means engaging a particular user in conversation by referring to their twitter handles. Reply means posting a response to a content. Retweet ‘RT’ means re-sharing content beyond the content creator’s network of followers. HashTags ‘#’ are used to reference a particular topic or event and enable twitter users to contribute content to that specific topic. We use Crimson API to filter datasets in Table 1 to extract mentions, retweets and hashtags for each discussion topics. 4.2.1 Popular Mention Table 3 lists the top twitter handles mentioned across discussion topics. The data reveals the twitter handle ‘@realDonaldTrump’ is mentioned in six out of seven discussion topics. This shows the popularity of Donald Trump amongst the users to engage in discussion for topics
from climate change, gender and violence. This could mean the possibility for either positive or negative conversation with Donald Trump on the topics of interest. Further text mining is needed to determine the tonality of conversation. Twitter handles ‘@guardian’ ‘@UN’ are also mentioned in majority of discussion topics which shows the increasing role of journalism and international bodies such as UN perceived by users to engage in conversation of global crisis. But their absence in topics associating gender and climate change means lack of interest for user to engage such bodies in discourse about climate change and its impacts on women. The reason could be that the users have little knowledge on the gender aware advocacies made by such entities.
Table 3: User Distribution Spanning Discussion Topics
Popular Mention @realDonaldTrump @guardian @UN @YouTube @POTUS
Out of 7 Influential Topics Author 6 4 4 4 3
Out of 7 Topics
@AP @BarackObama @BBCNews @BBC @billboard
6 4 3 3 3
@ABC
3
Geographic Out of 7 Coverage Topics Australia Canada India Indonesia United Kingdom United States of America Nigeria South Africa France Brazil Kenya
7 7 7 7 7 7 6 5 4 3 3
Further investigation revealed that out of 50 unique popular mentions majority of twitter handles (40%) are that of Individuals. Even among individuals, politicians and celebrity twitter handles are found most preferred to engage in discussion. Twitter handles representing Media and Organization make up 24% and 22% of popular mentions respectively. In response to RQ2 this means that politician and celebrities are perceived as a better ambassador for creating global awareness on crisis of climate change and gender violence, and serve as point of contact for users to share opinions and engage in conversation. Thus it can be implied that political parties and their views on the global issues have a great impact on the user attitude. In current scenario the withdrawal from the Paris Climate Agreement accord by the Trump administration is a reflection of the political stand on issue of climate change and is a serious concern because of its negative repercussions on the online communities who engage ‘@realDonaldTrump’ in their conversation. As shown in Fig 3, it is also revealed that categories of twitter handles mentioned vary significantly for each discussion topics. 70% of twitter handles mentioned in discussing topics on gender are Individual mostly celebrity while the aggregated topic on climate change and gender is discussed by majority of twitter handles (60%) belonging to organization. In contrast Media twitter handles are frequently mentioned (60%) to discuss aggregated topic on climate change and violence. In response to RQ2 this suggests that organization have additional impact
to sensitize climate change and gender issues while media can play a part in publicizing increasing violence such as trafficking from climate change. It is also worth noting that out of 5 twitter handles representing campaigns, all of them are mentioned for gender violence. This suggests that campaigns are mostly preferred to initiate and advocate discussions on topics of gender based violence. For the aggregated topic of climate change, gender and violence there is no definite choice but Media twitter handles are frequently used (50%) to engage in discussion.
Fig 3: Distribution of Twitter Mention across Discussion Topics. 4.2.2 Influential Author We also use Crimson API to extract influential twitter handles from the dataset. By ‘influential’ we mean twitter handles that have the ability to motivate individuals or masses to bring about the change for a common good. For this we use klout score of twitter handles provided by API which is computed by taking into account various factors such as user scalability, network scalability, interaction graph, strength and reach of influence. Table 3 lists the top influential twitter handles across the discussion topics. The data reveals that twitter handle of Associated Press ‘@AP’ appears in six out of seven discussion topics. This illustrates the significance of ‘News’ medium among other mediums for the spread of topics of global concern. Four other news medium ‘@BBCNews’ ‘@BBC’ ‘@billboard’ ‘@ABC’ also make it to the list of influential twitter handles appearing in more than two discussion topics. In response to RQ2 this shows the power of digital news medium to trigger a social change. This is also because news medium is believed to be the authentic source of information and possibly the primary source referred at crucial times. The data reveals twitter handle of Barack Obama ‘@BarackObama’ as influential and appearing in topics of gender, climate change, violence. It is also worth noting that none of news media and handle for Barack Obama appeared as influential for the aggregated topic on gender consequences of climate change. This means much of focus has been on tackling issues of climate change, gender and violence in isolation. A rigorous
advocacy and consolidated view of climate change and its impact is required from the ‘news’ medium and political leaders to create awareness of gender differential impacts. From 47 influential twitter handles it is also found that higher majority (47%) belong to ‘Media’ and 32% belong to ‘Organization’ such as UN, UNWomen, UNESCO, UNICEF, WHO and WorldBank. This signifies the role of international bodies and media in initiating a conversation and reaching to large number of audiences, influence their attitude and behaviors and be an agent of change to fight the global crisis. In this move the role of individual that make up 11% of influential twitter handles cannot be ignored. Most notably political leaders especially ‘@BarackObama’ who is known for advocating on environmental concerns have an inspiring role to play for collective action. As shown in Fig 4 it is also found that categories of influential twitter handles differ significantly across the discussion topics. The majority of twitter handles belonging to Media are found to be influential to sensitize topic of climate change (60%), violence (70%) separately and jointly (60%) as well as gender based violence (80%). While twitter handles for Organization (40%) and Media (30%) are both comparatively influential to spread aggregated topic of climate change, gender and violence. Since influential twitter handles has an important role to play in igniting social reformation, the role of Media and Organization is suggested to be most impactful. The role of individual influential twitter handles (10 to 20%) is uniform across all discussion topics.
Fig 4: Distribution of Influential Twitter Handles across Discussion Topics.
4.2.3 Popular Hashtag We also extracted twitter hashtags from the dataset and categorized them as official and unofficial. Hashtags that are launched by an organization as a campaign or have a twitter handle were considered as official. While hashtags that were used without any specific mission and used randomly to make a point by an individual were considered as unofficial. That hashtags extracted are listed in Table 3. Out of top 46 hashtags extracted, the data reveals that 35 hashtags (76% of hashtags) are unofficial while only 11 (24% of hashtags) are official. In response to RQ2 this raises a concern that we need more official hashtags in order to sensitize the topic of global concerns.
Table 4: Compilation of Popular Hashtags
Official HashTags (11) Unofficial HashTags (35)
#16days #1billionrising #Women4Climate #WomensMarch #YesAllWomen #orangetheworld #ParisAgreement #COP21 #COP22 #ActOnClimate #auspol #ActionsMatter #autism #climate #climatechange #debate endVAW #environment #fashion #Ferguson #Fracking #free #gender #girl #globalwarming #GOPdebate #hot #India #indigenous #KONY2012 #LRA #news #porn #science #sex #sexy #tcot #trafficking #Trump #VAW #VAWA #violence #waleg #woman #women #xxx
Fig 5: Distribution of Popular Hashtags.
4.3 Views Expressed We also performed qualitative content analysis of the posts to identify the views expressed. The views that are consistent with the scientific consensus and/or promoting action to prevent climate change, gender or violence were classified as Activist. The views in opposition to the scientific consensus and/or opposing action to prevent climate change, gender or violence are classified as Sceptic. For view that could not be identified based on information provided we classified as Unknown. To identify views expressed in the posts we first sampled top 10
retweets from each discussion topics. The posts with maximum retweets are certain to reach wider network of followers and draw more attention. The views expressed in such posts give a glimpse of whether supportive attitudes proliferate in online platforms. Fig 6 visualizes the views expressed in discussion topics. In response to RQ3 the data reveals that climate change and gender violence have higher activist views about 80% and 90% respectively. The reason could be plentiful coverage and importance given to climate change and gender violence such that it has raised awareness among the general population. While views over how climate change is resulting in gender differentiated impacts is low (40%) which means more awareness programs are needed to educate the general population. 4.4 Geographic Coverage To understand geographical distribution of users participating in discussion topics we also filtered datasets from the originating region. As shown in Table 4, data reveals that Australia, Canada, India, Indonesia, United Kingdom, United States of America have population who are highly interested in discussing topics of climate change, gender and violence either jointly or separately. Geographically the discussion is concentrated in the industrially developed countries. It is worth noting that Spain has only single entry for having user population who are interested in discussing aggregated topics of climate change and gender violence. In response to RQ4 the data also reveals that countries with serious consequences of climate change on increasing violence mostly faced by women are underrepresented when it comes to user engagement in discussing topic of interest. This suggests that more and more user engagement is necessary for population from counties which is highly vulnerable to climate change.
Fig 6: Distribution of Views Expressed across Discussion Topics.
Table 5: Distribution of Geographic Coverage across Discussion Topics
Discussion Topic Climate Change Gender Violence Climate Change and Gender Gender and Violence Climate Change and Violence Climate Change, Gender and Violence
Top 10 Geographic Region United States of America United Kingdom Canada Australia India Indonesia France Germany South Africa Netherlands United States of America United Kingdom Canada India Indonesia Philippines Nigeria Brazil Australia South Africa United States of America United Kingdom India Canada Australia Nigeria South Africa Indonesia Pakistan Brazil United States of America United Kingdom Canada Australia India France Indonesia Nigeria South Africa Kenya United States of America United Kingdom Canada Australia India South Africa Nigeria Indonesia Kenya Pakistan United States of America United Kingdom Canada Australia Indonesia India Nigeria France Brazil Mexico United States of America Indonesia United Kingdom Canada Australia India France Spain Kenya Nigeria
4.5 Organization vs Media We further examined twitter posts from specific organization dedicated to climate change and gender such as UNFCCC, UNWomen and UNNepal as well as news media outlets such as BBC, CNN and HimalayanTimes. Table 2 gives the quantity of self-authored posts by selected Twitter Handles. As shown in Fig 7 the data reveals that quantity of posts made on aggregated topics are significantly less in comparison to the total posts made. There are null posts for topics on climate change and gender consequences. This is a matter of concern as UN organization are regarded as key in mainstreaming various vulnerability concerns of climate change of which gender is one of the component. It is also noted that UN organizations make posts that are relevant to their theme of action. Of the seven discussion topics UNFCCC has majority of posts (25%) on climate change and UNWomen has majority of posts (49%) on gender. This indicates that isolated views are still prevalent among organization and a need for a consolidated view to educate general public of the gender differentiated impacts of changing climate. The results are even discouraging for media outlets. There are negligible posts made by media outlets such as CNN, BBC for the discussion topics. The reason could be that media outlets as a whole have to cover broad areas and climate change or its consequences is just one part. Since we have already hinted that media outlets are influential to trigger online discussions, it is important to encourage media for intensifying their coverage on climate change and its various vulnerability contexts.
Fig 7: Distribution of Self-Authored Posts by Organization 5. Discussion By examining Twitter datasets on general topic (‘climate change’, ‘gender’, ‘violence’) and specific/aggregated topic (‘climate change and gender’, ‘climate change and violence’, ‘climate change, gender and violence’) we revealed interesting findings on user engagement, demographics, geographical coverage, views expressed, role of media and organization.
Active Participation of Female and Matured age groups: The varied gender participation illustrates that female population is highly supportive of the topics where gender references are integrated such as gender violence, climate change impacts on gender. The results are in congruent with past studies [30][31] which suggested higher female participation in topics that address gender issues. While male population are found interested in discussing topics on general such as climate change, violence and seem to be gender insensitive. Though past studies have stressed tendency for women to be more concerned about environment, we find that female participation in climate change discourse is highly inclined towards discourse over its gender impacts. This suggests for launching gender based awareness campaigns involving male participation alongside female to initiate conversation on topic of interests. It is also found that matured age groups above 35 years are highly participative in discussing topics associating climate change gender and violence, in comparison to youth. This is a concern as youths are going to lead advocacies on various vulnerability issues of climate change. Youth participation has to be encouraged with incentives such as volunteer, workshops, raising prospects for their job in the topic of interest.
Danger of Political Views: We also found that majority of Twitter handles mentioned in the discussion topics are that of Individuals comprising of politician and celebrities. Consistent with past studies [24][28] the two step communication model still holds on in online discourses of climate change and its gender impacts. This implies that political parties and their views on the global issues have a great impact on the user attitude as they serve as point of contact for users to share opinions and engage in conversation. We also note that online communities have a preference to engage with ‘@realDonaldTrump’ for discussing topics whose administration is criticized for their stern America first policy sidelining environmental concerns. This hints on possibility for negative repercussions but detailed study is needed to reveal the tonality of conversation shared with ‘@realDonaldTrump’ to determine how it may impact online communities. We also found that twitter handles of Organization such as UNDP, UN have impactful role in sensitizing climate change and gender issues while Media are significant in publicizing increasing violence such as trafficking from climate change. We also found that campaign based twitter handles are mostly preferred to initiate and advocate discussions on topics of gender based violence.
Significant Role of Media and Organization: We found that Media and Organization UN, UNWomen, UNESCO, UNICEF, WHO play important role to reach masses in spreading awareness and having behavioral influences on masses to address global concerns. Specifically, it is found that Media are most influential to spread topic of climate change, violence separately and jointly as well as gender based violence.
Presence of Unofficial hashtags. From extracted twitter hashtags we found majority are unofficial hashtags without any definite mission. This stresses that plentiful official hashtags to address specific topics of interest is required to be launched by organization.
Lesser Activist views on Aggregated Topic. We noted majority of activist views on discrete topics of climate change and gender, but their aggregated topic over how climate change is resulting in gender differentiated impacts received less activist views. This suggest that more awareness is needed to educate about hardships that climate change has created for women mostly in developing countries.
Negligible User participation from Climate Vulnerable country: The results show that limited countries from Southeast Asia (India and Pakistan) and Africa (Kenya, Nigeria, South Africa) have active user engagement in topics of climate change and its consequences. Most of the discussion are concentrated in developed countries especially from North America and Europe as also revealed in [25]. Several factors such as language spoken, population size, rates of internet penetration and overall twitter popularity can account for this differences. For instance, internet access is still poor in developing countries some even ban Twitter platform such as China. However, this finding is significant since both Africa and Southeast Asia regions have majority of least developed countries vulnerable to climate change consequences. This suggests that region specific online campaigns are needed to attract user engagement from climate vulnerable regions.
Need for Intensifying Aggregated View. Though organization roles are important we revealed that some are focused only on specific theme of action such as UNFCCC advocating information on climate change while UNWomen spreading post on gender. Since climate change has consequence spanning all sections of society including gender differentiated impacts there is a need for organizations to highlight interrelatedness of climate change and various vulnerability context in their online posts. Additionally, twitter handles for popular media outlets such as CNN, BBC even showed significantly less posts related to topic of interest. Organization in Nepal (UNNepal) showed more focus on gender issues and seem to have less coverage on climate change and its consequences. This suggest that organization and media need to intensify their aggregated views on climate change and its consequence on violence faced by women.
6. Limitation The data for this research is extracted by accessing Crimson API with keyword based query prepared to align with our topics of interest and scope of analysis. Thus the datasets created for demographics, geographical coverage, user engagement are derived based on the algorithms implemented by API. There could be possibility for some differences in datasets if another platform is chosen.
The limitation of our work is that the query formulated for each discussion topics are explicit and to the point, for instance (climate OR climatechange) to extract post discussion about climate change, and (climate OR climatechange) AND (women OR woman OR girl OR gender) AND (violence OR trafficking) to extract posts discussion climate change and its gender differentiated impacts. There are rooms for improving the query formulated by adding more logically implicit contexts and increasing the keywords, for instance adding keywords such as global warming, anthropogenic, policy, women trafficking, sexual harassment, and physical violence. Past studies [29] have listed keywords to represent gender and climate change contexts in the content analysis which can be used to formulate detailed query.
Another limitation is that twitter dataset from English language is used for this study which creates a language bias since studies have shown that online user engagement is increasingly taking place in non-English languages. Our work is only representative of English language twitter user communities.
7. Conclusion The topic climate change is generalized theme under sustainable development and there is possibility that some of its vulnerability contexts such as gender differentiated impacts may go unnoticed in the social discourse. Though gender dimensions are now central components in climate change policy and strongly advocated by development agencies and government bodies, it is still dubious whether the general public perception has changed over time on the gender biased nature of climate change. In other words, whether the society is aware of women bearing higher burden than men due to climate change is still questionable. The reasons are (a) meagre role of civil society in global environmental governance (b) excessive coverage of media on the observable destructs obscuring vulnerabilities such as gender (c) challenge in confronting the contemporary social norms rooted with gender inequalities (d) limited access
to documented reports explaining women vulnerabilities and virtuousness in relation to environment (e) disintegrated views between developed and developing countries. There is no evidence on whether community beyond the domain of climate change or layperson are aware of its gender based consequences. Additionally, gender differentiated impacts remains marginalised in past studies of online discourses on climate change. This research examined twitter dataset on topics of climate change, gender and violence jointly and separately and expanded our understanding on user engagement pertaining to influential and popular twitter handles, demographics, geographical coverage and views expressed. The findings suggest the need for extensive gender based awareness campaigns to involve male participation alongside female, and target youth participation. The findings also suggest that twitter handles of organizations to be influential in sensitizing climate change and gender issues while Media to be influential in publicizing increasing violence such as trafficking from climate change. Further within organization it is found that the focus is on specific theme of action, for instance UNFCCC focus on climate change and UNWomen on gender. This calls for organization to have aggregated view to spread information about climate change and its various vulnerability contexts, gender is one of them. We also found presence of unofficial hashtags and suggest organization to launch official hashtags for a wide reach. The findings also stressed region specific campaigns to target online user communities from the least developed countries vulnerable to climate change. This research contributed by giving future directions of online user engagement policies in creating awareness on climate change and its gender differentiated impacts and enable a social change for a sustainable future.
References [1] Wright, H., Huq, S., & Reeves, J. (2015). Impact of climate change on least developed countries: are the SDGs possible. IIED Briefing. [2] United Nations World Economic and Social Survey (2016). Climate Change Resilience: an opportunity for reducing inequalities. https://wess.un.org/wp-content/uploads/2016/06/WESS_2016_Report.pdf [3] Mach, K., & Mastrandrea, M. (2014). Climate change 2014: impacts, adaptation, and vulnerability (Vol. 1). C. B. Field, & V. R. Barros (Eds.). Cambridge and New York: Cambridge University Press. [4] Aboud, G. (2011). Gender and climate change. Institute of development studies. [5] Aguilar, L. (2009). Training manual on gender and climate change. International union for conservation of nature (IUCN): and UNDP. [6] Gender and Women Events at COP 23 / CMP 13 / CMA 1-2 http://unfccc.int/gender_and_climate_change/items/10517.php Accessed on November 20, 2017 [7] Gemmill, B., & Bamidele-Izu, A. (2002). The role of NGOs and civil society in global environmental governance. Global environmental governance: Options and opportunities, 77-100.
[8] Bäthge, S. (2010). Climate change and gender: economic empowerment of women through climate mitigation and adaptation. Eschborn, Germany: GTZ. [9] Rodenberg, B. (2009). Climate change adaptation from a gender perspective. [10] Neumayer, E., & Plümper, T. (2007). The gendered nature of natural disasters: The impact of catastrophic events on the gender gap in life expectancy, 1981–2002. Annals of the Association of American Geographers, 97(3), 551-566. [11] Brody, A., Demetriades, J., Esplen, E., 2008. Gender and Climate Change: Mapping the Linkages. A Scoping Study on Knowledge and Gaps. BRIDGE, Institute of Development Studies (IDS), Brighton. [12] UN factsheet Women, Gender Equality and Climate Change. http://www.un.org/womenwatch/feature/climate_change/downloads/Women_and_Climate_C hange_Factsheet.pdf Accessed on September 01, 2017 [13] UNDP. Sustainable Energy Services: The Gender Dimension. http://www.undp.org/women/mainstream/SustainableEnergyServices.pdf [14] Lambrou, Y et al. (2006) Energy and Gender in Rural Sustainable Development. FAO. Rome ftp://ftp.fao.org/docrep/fao/010/ai021e/ai021e00.pdf [15] UNDP. Gender and Adaptation. Policy Brief (2013) http://www.undp.org/content/dam/undp/library/gender/Gender%20and%20Environment/PB2 -AP-Gender-and-Adaptation.pdf [16] Hemmati, M., Ro¨hr, U., (2007). A Huge Challenge and a Narrow Discourse. Women and Environments. Women and Gender Studies Institute, New College, University of Toronto, Toronto. [17] Mitomo, H., & Otsuka, T. (2012). Rich information on environmental issues and the poor reflections on consumers’ green actions: A behavioral economic approach. Telematics and Informatics, 29(4), 400-408. [18] Arora-Jonsson, S. (2011). Virtue and vulnerability: Discourses on women, gender and climate change. Global Environmental Change, 21(2), 744-751. [19] Lee, B., McCollum, B., Eschoo, A., (2009). Recognizing the Disparate Impact of Climate Change on Women and the Efforts of Women Globally to Address Climate Change. House of Representatives, Resolution 98. [20] De Choudhury, M., Jhaver, S., Sugar, B., & Weber, I. (2016, May). Social Media Participation in an Activist Movement for Racial Equality. In ICWSM (pp. 92-101). [21] Lotan, G., Graeff, E., Ananny, M., Gaffney, D., & Pearce, I. (2011). The Arab Spring| the revolutions were tweeted: Information flows during the 2011 Tunisian and Egyptian revolutions. International journal of communication, 5, 31. [22] Olteanu, A., Weber, I., & Gatica-Perez, D. (2016). Characterizing the demographics behind the# blacklivesmatter movement. OSSM. http://arxiv. org/abs/1512.05671.
[23] Segerberg, A., & Bennett, W. L. (2011). Social media and the organization of collective action: Using Twitter to explore the ecologies of two climate change protests. The Communication Review, 14(3), 197-215. (lets see) [24] Elgesem, D. (2017, September). Polarization in Blogging About the Paris Meeting on Climate Change. In International Conference on Social Informatics (pp. 178-200). Springer, Cham. [25] Kirilenko, A. P., & Stepchenkova, S. O. (2014). Public microblogging on climate change: One year of Twitter worldwide. Global Environmental Change, 26, 171-182. [26] Painter, J., Kristiansen, S., & Schäfer, M. S. (2018). How ‘Digital-born’media cover climate change in comparison to legacy media: A case study of the COP 21 summit in Paris. Global Environmental Change, 48, 1-10. [27] Williams, H. T., McMurray, J. R., Kurz, T., & Lambert, F. H. (2015). Network analysis reveals open forums and echo chambers in social media discussions of climate change. Global Environmental Change, 32, 126-138. [28] Nisbet, M. C., & Kotcher, J. E. (2009). A two-step flow of influence? Opinion-leader campaigns on climate change. Science Communication, 30(3), 328-354. [29] Karuna, P., Purohit, H., Stabile, B., & Hattery, A. (2016). On the Dynamics of Local to Global Campaigns for Curbing Gender-based Violence. arXiv preprint arXiv:1608.01648. [30] ElSherief, M., Belding, E. M., & Nguyen, D. (2017). # NotOkay: Understanding GenderBased Violence in Social Media. In ICWSM (pp. 52-61). [31] Purohit, H., Banerjee, T., Hampton, A., Shalin, V. L., Bhandutia, N., & Sheth, A. P. (2015). Gender-based violence in 140 characters or fewer: A# BigData case study of Twitter. arXiv preprint arXiv:1503.02086. [32] Holmberg, K., & Hellsten, I. (2015). Gender differences in the climate change communication on Twitter. Internet Research, 25(5), 811-828. [33] McCright, A. M. (2010). The effects of gender on climate change knowledge and concern in the American public. Population and Environment, 32(1), 66-87. [34] Crimson Hexagon API for Twitter Datasets. https://www.crimsonhexagon.com/ Accessed on August 01, 2017
