- 23/07/2021
- Category: Digital Intelligence Lab EN
Author: Safira Tafani Cholisi
Introduction
Discourse surrounding one of the most prominent e-commerce platforms in Indonesia, Shopee, emerged and spread rapidly on Twitter since the evening of April 9, 2021. It began when user @CibiWiranegara posted a screenshot of their conversation with a courier of Shopee Express, the platform’s exclusive delivery service.
Image 1. Tweet from user @CibiWiranegara
regarding the alleged strike by Shopee Express couriers.
The conversation revealed an alleged strike held by Shopee Express couriers in operation warehouses due to a sudden reduced wage policy, causing delays on order deliveries to customers. On April 11, Twitter user @arifnovianto_id followed up on the issue and confirmed that the strikes had taken place for 5 consecutive days. According to them, Shopee Express couriers are protesting the significant reduction in wages which are deemed to be inappropriate with the amount of work that has to be done by the couriers (Damanik, 2021). The hashtag #ShopeeTindasKurir was used by Indonesian netizens to raise concerns on Shopee’s exploitative practices and quickly became a trending hashtag on Indonesian Twitter.
Responding to the issue, Shopee management denied that strikes took place in their warehouses. While they acknowledged delay on deliveries, it was argued that this was caused by an overload of orders from their 4.4 promotion event. Executive Director of Shopee Indonesia, Handhika Jahja, explained that Shopee has always followed the existing wage scheme regulations in every region. As an illustration, Jahja mentioned that the average wage for Jakarta-based couriers by other logistics ranges between IDR 1,700 to IDR 2,000 for each delivered package. Shopee itself claims to pay IDR 2,213 per package for couriers in Jakarta (Astutik, 2021). Additionally, Jahja asserts that Shopee ensures a mutually beneficial partnership with couriers and is always willing to receive suggestions in order to guarantee partner’s welfare. The e-commerce platform provides insurance for their Shopee Express courier drivers and allows flexibility on operation working hours.
Nevertheless, the commotion failed to die down despite the first hand clarification from Shopee’s top management figure. Several journalism outlets as well as a group of researchers from Universitas Gadjah Mada released investigative reports and articles on the exploitation of driver and courier partners under the gig economy system. The Finery Report (2021) interviewed ex-Shopee Express courier drivers and revealed that drivers are required to cover operational and collateral fees on their own, including transport fuel, driver uniform, and parking costs. The partnership scheme which supposedly puts the company and drivers on the same level for cooperation is argued to be disadvantageous for drivers. Corporations are able to escape from their responsibilities to ensure the welfare of their drivers as they are not tied in an employer-employee relationship, but rather a partner-partner one (Novianto, Wulansari & Hernawan, 2021). As critical responses kept pouring in despite Shopee’s clarifications, the issue has continued to become a popular topic of discussion for Indonesian netizens.
This article aims to investigate how the information on #ShopeeTindasKurir spread across different groups and communities of Indonesian netizens and became a popular topic on Twitter by identifying the role of community bridgers using Social Network Analysis (SNA). The research will first collect data and subsequently analyze the results using two SNA measurements called modularity and betweenness centrality in identifying communities and the bridging actors between them, including the social identity and role of the actors. Analyzing how a particular issue or information spread through social media can further give insight on understanding patterns of communication and changes in the digital society.
Research Design
This section explains the implementation of the research by elaborating on data collection and visualization methods.
The data is primarily taken from Twitter search results based on a built query list. Gephi is used to calculate and visualize results of Social Network Analysis (SNA) measures in this research. Two measures are investigated and visualized: modularity and betweenness centrality.
Modularity is a measure of SNA that helps identify community structures. Based on the strength and density of connections between nodes, nodes with strong connections with one another are grouped together as a module and those with weak connections to each other are distanced away and grouped into separate modules (Ji et al., 2015). Through identifying the varying modules in a particular network, in this case #ShopeeTindasKurir issue on Twitter, we can see the extent to which the issue is being discussed among various communities.
Betweenness centrality is a measure of SNA that identifies important actors who act as a bridge or bottleneck between communities. This measure is particularly significant in investigating how information can be passed around between separate social groups and communities which otherwise would not interact with each other (Tsvetovat & Kouznetsov, 2011). Following the identification of various modules in the discourse of #ShopeeTindasKurir, betweenness centrality can point out which accounts play an important role in spreading the issue to different groups in Twitter and subsequently expanding the reach of the issue itself.
The following steps are performed in data collection and visualization:
- Build Twitter query by identifying related keywords and set a time frame. The final list of the query keywords is as follows:
#ShopeeTindasKurir OR #TegaSamaKurir OR (Shopee Express AND Kurir) OR (Target AND Kurir) OR (Mogok AND Kurir) OR (Tarif AND Kurir) OR (Oren AND Kurir)
The time frame of the query is 9 April 2021 to 16 April 2021 or a week since the first tweet of the issue was posted.
- Scrape search results of the built query using Twint (Twitter scraping tool written in Python). The types of interactions included in these results are both direct mentions and engagements (reply to, user to mentioned user, and quote retweet; likes and retweets excluded). A total of 4,718 tweets were obtained from scraping the built query and set time frame.
- Convert list of scraped Tweets into separate nodes and edges CSV files for Gephi use.
- Import nodes and edges CSV files and run statistics calculation on Gephi (all measurements under network overview, node overview, and edge overview).
- Results are visualized based on the results of betweenness centrality analysis and modularity.
A total of 8 clusters are identified based on the minimum accumulative modularity class of 50%. 50% is set as a baseline as it represents the largest and most dense clusters where interactions related to the built query takes place. After mapping out the clusters, users with high betweenness centrality measurement from each cluster are identified and displayed.
Results and Discussion
Graph 1. Visualization of Gephi statistics calculation results.
While @lalaaphonee is shown in the graph, the user is not discussed in this section
because the account has been suspended as of the time this research is conducted.
As seen from the graph above, @ShopeeID is both a member of the largest module and also the user with the highest betweenness centrality measurement. Through investigation of the obtained tweets from scraping, the majority of @ShopeeID’s interactions are classified into direct mentions instead of engagements. This means that @ShopeeID’s high betweenness centrality measure is caused by the fact that while discussing the issue of #ShopeeTindasKurir, other users mention @ShopeeID in their tweets. It is also important to note that @ShopeeID is the account of the e-commerce platform in question, which explains the large amount of interactions by other users with this account, particularly on the topic of #ShopeeTindasKurir. In this case, @ShopeeID acts as the main subject of the issue.
While both are members of the same module colour with @ShopeeID, @taroshotaro00 and @bekasi_duren are notably much further apart from the centre of @ShopeeID’s community. This can be explained by investigating the tweets connecting @taroshotaro00 and @bekasi_duren.
Image 2. Tweets from user @bekasi_duren related to the built query.
Tweets from @bekasi_duren mainly consist of replies to accounts under @ShopeeID’s giveaway which was held on 11 April 2021, two days after #ShopeeTindasKurir issue emerged, and repetitively contained the hashtag #ShopeeTindasKurir. It also has connections with @lalaaphonee and @arifnovianto_id from different module colours. These interactions are of a similar pattern, where @bekasi_duren reply their tweets with #ShopeeTindasKurir hashtag. The role of this user as a bridge of communities is rather difficult to assess aside from its consistent usage of the #ShopeeTindasKurir hashtag and replies to other displayed accounts.
Image 3. Tweets from user @taroshotaro00 related to the built query.
Meanwhile, @taroshotaro00 has a more interesting position as the user is connected to user @txtdarionlshop and @detikcom. @taroshotaro00 notably contributes actively to the discourse of Shopee’s courier issues. Other than promoting the hashtag #ShopeeTindakKurir, the user also discussed with other users regarding their experiences with Shopee’s service and compared it to other e-commerce platforms. Most of the connections between @taroshotaro00, @txtdarionlshop, and @detikcom are from direct replies by @taroshotaro00 or indirect mentions through the discussions the user had with other users. This user plays a role in connecting communities by actively participating and replying in discussions started by a range of different users.
An interesting pattern of modules is seen from @AREAJULID, @txtdarionlshop, and @tubirfess. All three users belong to separate modules from each other and from @ShopeeID. They’re also notably far apart from other existing modules in the network. This begs the investigation towards their role as bridges between the communities they are a member of in expanding the discourse of #ShopeeTindasKurir. All three users (@AREAJULID, @txtdarionlshop, and @tubirfess) are called ‘menfess’ accounts, an abbreviation for ‘mention confess’. These accounts are run anonymously and utilize a bot to post anonymous tweets sent by other Twitter users through direct message. The purpose of such accounts is for Twitter users with similar interests to post a topic of discussion, ask questions, and share information without having to reveal their identity. Menfess accounts have particularly become popular among Indonesian Twitter users.
Image 4. Tweet from user @AREAJULID related to the built query.
Image 5. Tweet from user @txtdarionlshop related to the built query.
Image 6. Meme attached in the tweet from user @txtdarionlshop.
Image 7. Tweet from user @tubirfess related to the built query.
As tweets from these accounts reflect, they are mainly messages or satirical jokes calling out Shopee’s mistreatment of their couriers and misdirected strategies, such as holding giveaways and inviting Korean Pop stars as guests. These accounts also have a large number of followers who are specifically interested in the menfess anonymous messaging methods and in interacting with the tweets. Thus, it is expected that many users are drawn to interact with the tweets and jokes, especially when it comes to emerging major issues. It is also interesting to observe that some of the most replied to tweets of these accounts related to the issue do not directly mention the related keywords. Instead, the replies by the followers contain those related keywords. In this sense, menfess accounts play a role in connecting communities to learn about #ShopeeTindasKurir through its intended creation and usage as a public space that connects people through discussions stimulated by the anonymous messages.
Another module that can be identified from the graph is the green module which consists of user @kurawa and @detikcom as the members with highest betweenness centrality measurement. The connection between the two is shown in the following tweet.
Image 8. Tweet from user @kurawa related to the built query.
Image 9. Tweet from user @detikcom related to the built query.
As seen, @kurawa quote-retweeted @detikcom’s tweet of an article regarding #ShopeeTindasKurir. @detikcom is a news site that released one of the earliest articles on the issue, which may explain the large amount of interactions with the tweet. @kurawa also has a large number of followers and often tweets about social issues, thus interactions from the user’s followers are expected.
@innokribow is another user which seems to be connected to different displayed users and modules. The user is connected to @kurawa, @txtdarionlshop, and @lalaaphonee, all of which are members of different modules. However, the user’s tweets only consist of replies to other users’ tweets containing an article link discussing #ShopeeTindasKurir. Although @innokribow seems to be well-connected to a lot of other users from other modules, the interactions between these users are not significant enough to classify the user as an important community bridge as there is no feedback from the other users.
Image 10. Tweet from user @innokribow related to the built query.
Lastly, @arifnovianto_id is another user with notable betweenness centrality measurement. The user’s tweets contain essential information regarding the strike held by Shopee courier drivers, the reasons and demands for the strikes, and the general conditions faced by the drivers. The user’s main thread on the issue also elaborates on the existing laws with regards to e-commerce partnership schemes and the possible exploitation faced by Shopee courier drivers.
Image 11. Tweet from user @arifnovianto_id related to the built query.
@arifnovianto_id’s position as a researcher in a university research institution as indicated in the user’s bio explains their knowledge and expertise on the issue. The user’s thread also has 20,000 retweets, 4,700 likes, 1,949 quote retweets, and 749 replies as of 4 July 2021, showing the extent to which other users have engaged with the tweets. The user’s possible role in spreading the information across communities lies in the extensive first-hand and primary source of information displayed by the user’s tweets through their position as a member of an epistemic community.
Conclusion
Based on the analysis results, it can be seen how various actors play different roles as a bottleneck or bridge between communities and contribute to spreading information about #ShopeeTindasKurir issue on Twitter. Academics, anonymous messaging accounts, social media public figures with a large mass of followers, active regular users and news sites can be classified as the significant actors. Although #ShopeeTindasKurir first became highlighted due to a tweet from a regular user (@CibiWiranegara), various users with different identities and roles joined the discussion on the topic.
As the directed party, it is expected that @ShopeeID receives a high intensity of interactions. Academics (@arifnovianto_id) and news sites (@detikcom) become important actors who report and share extended information on the issue. Social media public figures (@kurawa) can also give more exposure to the issue through their interactions even without providing additional information. Regular users (@taroshotaro00) also engage with the issue by direct discussion and interaction with other regular users. The growing prevalence of menfess accounts as represented by @txtdarionlshop, @tubirfess, and @AREAJULID provides a new method of sharing information and community building among users with common interests. Meanwhile, other regular users (@innokribow and @bekasi_durent) also contribute by sharing news or hashtags related to the issue, although the extent of their significance is difficult to determine. These varying roles can reach different audiences with a range of interests and thus expose and raise the awareness of a large number of people to the issue of #ShopeeTindasKurir.
This research can be further expanded by doing a qualitative study analysis of the content of the tweets by the key community bridge users as well as the interactions from their replies and followers. Such research will be able to inform us the ways in which individuals respond and engage with the issue. The results can be used as a starting point to map out how social identities may affect an individual’s response to a particular social issue, in this case #ShopeeTindasKurir.
Author: Safira Tafani Cholisi
Editor and Reviewer: Josia Paska Darmawan
Data Collecting: Nadia Elaesiana
REFERENCES
Astutik, Y. (2021). Ternyata Upah Kurir Shopee Express Rp 2.213 per Paket, Wajar?. CNBC Indonesia. Retrieved 12 July 2021, from https://www.cnbcindonesia.com/news/20210412183211-4-237257/ternyata-upah-kurir-shopee-express-rp-2213-per-paket-wajar.
Damanik, M. (2021). Kurir Shopee Express Mogok, Warganet Soroti Tagar ShopeeTindasKurir. IDN Times. Retrieved 12 July 2021, from https://www.idntimes.com/news/indonesia/margith-juita-damanik/kurir-shopee-express-mogok-warganet-soroti-tagar-shopeetindaskurir/5.
Ji, X., Machiraju, R., Ritter, A., & Yen, P. (2015). AMIA Annual Symposium Proceedings (pp. 1927–1936). American Medical Informatics Association. Retrieved 12 July 2021, from https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4765615/.
Novianto, A., Wulansari, A., & Hermawan, A. (2021). Riset: empat alasan kemitraan Gojek, Grab, hingga Maxim merugikan para Ojol. The Conversation. Retrieved 12 July 2021, from https://theconversation.com/riset-empat-alasan-kemitraan-gojek-grab-hingga-maxim-merugikan-para-ojol-159832.
The Finery Report. (2021). Pahlawan tanpa tanda jasa Industri 4.0: Kurir dan pengemudi. The Finery Report. Retrieved 12 July 2021, from https://www.thefineryreport.com/articles/2021/4/28/pahlawan-tanpa-tanda-jasa-industri-40-kurir-dan-pengemudi?utm_campaign=later-linkinbio-thefineryreport&utm_content=later-17000171&utm_medium=social&utm_source=instagram.
Tsvetovat, M., & Kouznetsov, A. (2011). Social Network Analysis for Startups. O’Reilly Media.
