Introduction
In times marked by social media activism and movements, Deferred Action for Childhood Arrivals (DACA) took centre stage on social media in general and Twitter in particular (Alam et al., 2020). The DACA hashtag is one of several hashtags on Twitter, a social networking service, that create online forums for discussion concerning the U.S. Immigration Policy, a program that provides protection from deportation to undocumented immigrants who entered the country as children. A majority (523,970) of the nearly 700,000 active DACA recipients residing in the United States are from Mexico (USCIS, 2019), with thousands of others from several other Latin American countries. Even though the DACA policy did not only affect Latin Americans and Latinx families, this became a salient issue in the community, and many took to social media to discuss and protest the uncertain and unstable situation that recipients faced in 2017, when USCIS ceased accepting new applications after Trump announced the program would end (Valverde, 2018). Thousands of U.S. born children have at least one parent who is a DACA recipient and family separation would potentially result from the ending of DACA protections (Svajlenka, 2019).
This research conducts a content analysis of random samples of tweets from the DACA hashtag on March 5, 2018, the day DACA was supposed to end and congress was supposed to come up with a replacement policy (Cruz, 2018), as well as a day before and a day after. While discussion on this divisive issue ensued for the next couple of months up until the end of April, when a federal judge ordered the continuation of DACA (Sacchetti, 2018), the March 5th date was pivotal in organizing supporters and opposers in a social media realm. This study investigates the uses surrounding DACA hashtag, analyzing engagement of stakeholders (individuals, media organizations, private organizations and the government) and those who oppose, support or remain neutral (groups) of the policy. The study also examines activism or call to action by stakeholders and groups. Applying the uses and gratifications theory, this study examines the use of features, such as attaching links, hashtags and visual media, by different stakeholders and groups in the context of the debate.
Background
In 2017, when the Trump administration announced it would rescind DACA, the U.S. had approximately 10.5 million undocumented immigrants, with an estimated 5.6 million undocumented young adults who were brought to the country as minors (Mallet-García and García-Bedolla, 2021). The introduction of the DACA executive order by President Barak Obama, allowed undocumented immigrants brought as minors to apply to defer deportation through a process that protected and gave a path to legal permanence to such individuals (Cruz, 2018). The issue remained divisive and the debate over immigration, with DACA alongside the building of a wall between the U.S. and Mexico, acquired prominence in the 2016 Presidential election (Valdez et al., 2021). And while immigration concerns flow from several different countries spanning many continents, the focus of much of the debate in that election referred to immigration from Latin America and Mexico in particular into the United States.
Coverage of Mexican immigration to the United States has been traditionally marked by negative stereotypes, with immigrants being portrayed as having "backwardness, peasantness, and lack of modern sophistication" (Chavez, 2001: 260). Chavez, Whiteford, and Hoewe (2010) identified salient topics associated with Mexican coverage by mainstream news organizations as crime, drug trafficking and violence. Politicians have equally relied on such stereotypes, and in 2016, images of gangs and criminals surrounded the immigration debate. However, the coverage of Dreamers and DACA recipients has been more positive. DACA has revitalized the debate on whether the United States needs to reform the way visas are allocated, and DACA was intensely discussed during the 2016 presidential campaigns. Politicians and activists took to social media to engage with the quickly shifting decisions on DACA. DACA youth in particular used social media as a form of activism by sharing their stories and mobilizing their communities (Rodriguez et al., 2019).
Theoretical Framework
Uses and gratifications (U&G) of media have been studied since the 1940s (McQuail and Windahl, 1993) to know more about how audiences use media and what motivates them to use said media. In the 1970s and later, the theory was further developed to include other variables that impact its uses such as: social situations, socio economic factors, and personality traits. More recently, researchers have used this theory to examine the uses of digital media. Ruggiero (2000) suggested that researchers should "include concepts such as interactivity, demassification, hypertextuality, asynchroneity, and interpersonal aspects of mediated communication," (Ruggiero, 2000: 29) in order to expand the U&G theory. With the rapid diffusion of social media that are now used by millions of people, the U&G theory is all the more relevant. According to Auxier, and Anderson (2021) 81 percent of American adults use YouTube, 69 percent use Facebook, 40 percent use Instagram and 23 percent use Twitter. However, a larger percentage of adults in the 18-29 age group use these and other social media.
For example, 95 percent use YouTube, 65 percent use Snapchat, 71 percent use Instagram, 42 percent use Twitter and 48 percent use TikTok (Auxier and Anderson, 2021). Sundar and Limperos (2013) use the term "modality" to describe the various ways that content may be presented in different media such as: textual information, images, audio and video formats, with each of these having a different type of appeal to the receiver. The affordance of "agency" allows the users of these social media features to be sources or agents of information and be involved in community-building, while the affordance of navigability enables the users to browse (Sundar and Limperos, 2013). Rathnayake and Winter (2018) suggest that the interactive character of social media, the various features of social media platforms and the control users have over the content need to be considered while measuring U&G.
In our study, we will examine: the use of different features of Twitter such as textual content and visual media (images, graphics and video) on the DACA hashtag; the differences in the use of Twitter between stakeholders (individuals, news/media organizations and the government) who are sources or agents of information; and the use of navigability features such as attaching links and hashtags to tweets. While we know that hashtags surrounding the DACA policy were salient when its viability was being decided, different stakeholders, such as politicians, DACA youth, media and others, made different use of social media to achieve different gratifications. Given the ramifications of the portrayal of Latin American and Mexican immigrants in mainstream media and the intensity of the debate over DACA on social media, this study seeks to further our understanding of uses, engagement and portrayal of this divisive issue by different stakeholders in a time of social media activism.
The current socio and political environment that the U.S. and Mexico experience is profoundly marked by changes in technology. Social media has provided the tools necessary for groups to mobilize and to satisfy sought gratifications, while it has also altered the political environment in fundamental ways. Technopolitics, as presented by Rodotà (2004), focuses on the transformations brought by technology, specifically by communication technologies, which both enable citizen participation while also being a resource for citizen surveillance and manipulation. The connection between Twitter, and its mobilizing potential, with citizen participation and group manipulation is an important area of understanding (Treré and Carretero, 2018).
Bennett and Segerberg (2013) place the focus on digital media as organization agents. The authors claim that the ability to personalize issues and stories to then have those personal narratives broadcasted to others in the networks, multiply the ability to engage in collective action (Bennett and Segerberg, 2013). The personification and individual role in building communities and engaging others for social change is then different from centralized, classic organizational political communication. Our study assesses some differences in engagement with the DACA hashtag by different stakeholders: individuals, news media and private organizations and the government.
Literature Review
Social media have often been discussed in relation to "hashtag activism" campaigns that associate a hashtagged term with a political or social cause (Wonneberger et al., 2021). Researchers note that hashtag activism campaigns are able to produce counter-publics that support public action and build solidarity among participants (Choi and Cho 2017). Further, for organizations associated with social movements, hashtag activism can be a way to influence policy and promote democracy (Xiong et al., 2019). Pavan (2017) has argued that social media also play a role in longer term processes of institutionalization, since actors must switch strategies over time in order to engage both within and outside of institutional arenas.
Despite these reported benefits, scholars argue that hashtag activism must be evaluated in context and without assuming that social effects are inherent in the communication technology (Lindgren, 2013). Challenges to hashtag activism include: the potential for "noise" to dilute or disrupt a communication around a hashtag (Anderson 2013); the presence of trolls and other harmful actors on social media (Phillips and Milner 2018); and the possibility that actions taken by social media users are merely examples of "slacktivism," or token actions such as liking a social media post or changing a profile image that signal support for a cause, while lacking direct political efficacy (Christensen 2011). However, as Kristofferson, White, and Peloza, (2014) argue, acts associated with slacktivism can also help to predict more meaningful future actions.
Social Media Engagement
Twitter is a social media platform characterized by short messages, hashtags and the ability to connect strangers. Messages were initially limited to 140 characters, but by 2017 the limit was increased to 280. Hashtags are essential for Twitter's potential to connect strangers within common interests, topics or threads, building conversations across geographies and spaces. The connections through hashtags and the ability to amplify and diversify recipients of messages make Twitter particularly useful in social movements. Social media in general and Twitter in particular have played a key role in the distribution and organization of people around issues promoting social, cultural and political change (Housley et al., 2018).
Researchers have studied a wide range of communications and interactions within an online community, such as participating via Twitter by information-sharing, attaching links to URLS, attaching images and videos, retweeting (Murthy and Gross, 2017; Pang, and Ng, 2016). Twitter users are more likely to gratify their need to connect with others if they are active participants in an online conversation rather than if they use only a specific function of Twitter such as retweeting (Chen, 2011). A better understanding of Twitter use by organizations has led to strategic communication with stakeholders. For example, Guo and Saxton's (2018) study revealed that nonprofit advocacy organizations strategically attracted the attention of and connected with their target audiences via Twitter by sending direct messages, also known as public reply messages, retweeting, favoriting and using other tools such as hyperlinks, hashtags, and visual content. Park and Kaye (2017) found that online civic participation strongly impacts Twitter opinion leadership. Kim, Wang, and Lee (2016) found that Twitter's interactive tools allow journalists to post information on current issues, connect with the public and be influential, among other things.
Other researchers have studied Twitter use during public debates and found that activists opposing the issue being debated were the most engaged group in terms of the number of messages (Moe, 2012), and that live participation, such as tweeting, during the debates was strongly connected with political engagement (Vaccari, Chadwick, and O'Loughlin, 2015). Hengst (2017) examined the levels of engagement in tweets of those who opposed and those who supported Wendy Davis when she attempted to filibuster Texas Senate Bill 5 in 2013 and found that supporters demonstrated a far higher level of engagement, measured in terms of favorites and retweets, than those who opposed her. The tweet that had the highest number of retweets (12,624) over an eight-day time frame had a link leading to a livestream of the filibuster.
The above discussion leads us to the following hypotheses, examining the level of engagement on the DACA hashtag by different stakeholders:
H1: There will be a difference in the level of engagement between the three groups (those who oppose, those who support and those who are neutral).
H2: There will be a difference in the level of engagement between the different stakeholders (individuals, news/media organizations, government and private organizations).
H3: There will be a difference in the level of engagement between the three days (March 4, March 5 and March 6, 2018).
Definitions: In our study, we defined the level of engagement as the number of hashtags, visual media (including images, graphics, and video) and URLS/links attached to a tweet. A higher number of hashtags, visual media and links attached to an original tweet indicates more action from the user than if they were ONLY to retweet another person's tweet. We categorized user groups based on whether they supported, opposed or were neutral toward DACA. In our research, tweets from supporters included ones that used positive words toward DACA and DACA recipients, defended the dreamers, opposed ending DACA, provided helpful information, offered help to DACA recipients and said that families deserve to stay together. Tweets from those who opposed included ones that were negative toward DACA, supported the ending of DACA, were against illegal immigration, supported deportation, supported building the wall and were against sanctuary cities. Neutral tweets were those that did not clearly take a side either to support or to oppose. Based on past research (see for example, Rao et al., 2018) we categorized stakeholders as individuals, news/media organizations, private organizations and the government. The three days considered in this study were March 5, 2018, the day DACA was supposed to end; March 4, the day before and March 6, the day after that. Activism was measured by coding the tweets as call to action (such as asking people to contact the Congress or to assist those impacted by DACA) or no call to action.
Social Media and Activism
Activism is generally referred to as groups of individuals acting together in order to bring about political, economic, or social change (Cammaerts, 2007). Among online communication spaces, social media sites, such as Twitter and Facebook, have been found to be important places for strengthening social movements and sharing information. Scholars have found positive relationships between social media use and activism (Sandoval-Almazan and Gil-Garcia, 2014). Tang and Lee (2013) found that individuals who were exposed to shared political information, and had connections with public political actors, through Facebook were more likely to have participated in political activities. Lindgren and Lundström (2011) found that sharing information, by including a link to a newspaper or television news items and documentaries, followed by a call for action (such as making donations and taking political action) supported freedom of information. Meyer and Bray (2013) found that social media users believed that online activism was most effective in creating awareness. Participants also said that online discussions could reach large and diverse audiences beyond borders.
Furthermore, scholars found that social media sites offered new ways to engage in activism. For example, Sandoval-Almazan and Gil-Garcia (2014) found that social media tools not only allowed activists to promote social and political activism, but also to change their strategies for political activism and social movements. Zimmerman (2016) found that undocumented youth abandoned the earlier practice of making their claims anonymous and expressed their undocumented legal status through videos and podcasts on social media. Hengst (2017) found that #SitDownWendy, a hashtag that suggests it was being used by those who opposed the filibuster, was "hijacked" by supporters who may have perceived the hashtag as a threat.
On the other hand, research studies have shown that social media sites are not effective for alternative journalism. For example, Poell and Borra (2011) who explored the use of Twitter, YouTube, and Flickr during the G20 protests in Toronto, Canada, found that only 20 percent of the users of the hashtag #g20report posted about 50 percent of the tweets and that most users focused on the spectacle and violence rather than the issue of protest. Harlow and Guo (2014) felt that word of mouth, fliers and radio were impactful while communicating with immigrant communities that lacked access to digital media. They also found that there was a digital divide between the immigrants and activists that might be leaving out many immigrants from the online discussions. The authors further found that different stakeholders such as the general public, activist organizations, official/authorities/power holders and immigrants or refugees used the social media for a wide range of purposes such as: increasing awareness, recruiting, mobilizing, sharing information, influencing legislation and coordinating offline and online action.
Informed by previous literature that states differences exist in the nature of tweets by different stakeholders and groups, particularly during activist movements and public debates, we formed the following hypotheses:
H4: There will be a difference in the type of tweets (support, oppose, neutral) between the different stakeholders (individuals, news/media organizations, government and private organizations).
H5: There will be a difference in the tweets that call to action or do not between the different stakeholders.
H6: There will be a difference in the tweets that call to action or do not between the three groups (those who oppose, those who support and those who are neutral).
Researchers have used both qualitative and quantitative methods to study social media engagement and activism. Some scholars have used the online survey method to study user engagement on Twitter (see for example, Kim et al., 2016; Chen, 2011). Other scholars have analyzed Twitter content including videos and photos attached (see Hengst, 2017; Poell and Bora, 2011). Zimmerman (2016) examined case studies of transmedia testimonios (personal narratives shared via various media platforms such as YouTube, Facebook and Twitter) of undocumented youth who "came out" and stated their illegal status and discussed immigrant detention and deportation issues.
Methodology
Although the DACA issue has been debated for a long time, we chose to examine the tweets on March 5, 2018 as this was the date DACA was set to end. We also wanted to examine the tweets a day before and a day after March 5 to see if there were any differences in the levels of engagement and activism between the different stakeholders and among those who supported DACA, those who opposed it and those that remained neutral.
We identified at least 30 hashtags connected to the issue, including the following hashtags: #DACA; #DACANS; #dacadeal; #SaveDACA; #DefendDACA; #Stand-WithDACA; #Pray4DACA; #EndDACA; #StopDACA; #NoDACA; #DACAprotest; #DACASolidarity; #DACArepeal; #DACAmented. We chose the #DACA hashtag because it was top-ranked for this topic by Twitter.
In order to access tweets between March 4 and March 6, 2018 we used the Twitter Streaming Application Programming Interface (API) to collect tweets containing #DACA. While researchers have pointed to small biases in the Streaming API (see, for example, Morstratter et al., 2014), it is generally thought to produce data that is largely representative and is often considered the optimal method for large-scale, ongoing collection (Rafail, 2018). All data were collected using the Digital Methods Initiative Twitter Capture and Analysis Toolset (DMI-TCAT) software tool (Rieder and Borra, 2014) installed on an Amazon Elastic Computer Cloud (EC2) server. No rate limits or server interruptions were experienced during the collection period.
Between March 4 and 6, we collected 47,859 tweets from 31,884 distinct users. Prior to coding and content analysis, we performed basic descriptive analysis of our data, calculating the most frequent hashtags, visual media and URLS present in tweets, as well as the tweets that received the most retweets during our collection period. At this point, we also removed retweets from our dataset and created random samples of 500 tweets for each of the three days in our collection period.
Two independent coders were trained to code the tweets as supportive, opposed or neutral and also to determine whether there was a call to action or not. Both coders coded 300 tweets (first 100 from each of three samples) which is 20 percent of the total sample of 1,500. Intercoder reliability was calculated by using Holsti's method. Holsti's coefficient was 0.99 for both the type of tweets (support/oppose and neutral) and the tweets that called to action or not. According to Neuendorf (2002) inter-coder agreement of .90 or higher is generally acceptable. The coders independently coded the remaining tweets. The coders both knew Spanish and translated the 63 tweets that were in Spanish to English. In addition, there was one tweet in Arabic and one in German; Google Translate was used to translate these.
Results
A descriptive analysis of the 47,859 tweets generated for the three days indicated that the highest number of tweets were generated on March 6 (21,711), the day after the DACA deadline (See figure 1). This was followed by 21,651 tweets on March 5 and 4497 on March 4. The number of hashtags attached to tweets was also the highest on March 6 (21,703) with just a few less on March 5 (21,643) and a much lower number on March 4 (4495). The number of links attached were the highest on March 5 (5419) followed by March 6 (4200) and March 4 (723).
The number of visual media attached was also the highest on March 5 (1286), followed by March 6 (855) and lowest on March 4 (215). In the total sample of this study consisting of 1,500 tweets (500 for each of the three days), the number of hashtags (1912) and the visual media attached (65) were highest on March 4 and the number of links was the highest on March 6 (454).
Chi Square statistics was applied to test the six hypotheses of the study using the data from the samples of 500 tweets for each of the three days. H1 sought to examine if there were any differences in the level of engagement (number of hashtags, visual media and links attached to tweets) between the three groups (those who support, those who oppose and those who are neutral). Chi Square statistics revealed that there were no significant differences between the groups. H1 was not supported. Although the differences were not significant, Table 1 shows that those that were neutral had the highest number of hashtags (1686), visual media (78) and links (626) attached to their tweets compared with the other two groups. Those that opposed had the lowest number of hashtags, visual media and links attached to their tweets.
Groups | Hashtags Number/Percentage |
Visual Media Number/Percentage |
Links Number/Percentage |
---|---|---|---|
Support | 1058 (30.52%) | 56 (35.90%) | 442 (33.31%) |
Oppose | 723 (20.85%) | 22 (14.10%) | 259 (19.52%) |
Neutral | 1686 (48.63%) | 78 (50%) | 626 (47.17%) |
Total | 3467 | 156 | 1327 |
X2=34.24, p-value=.099265. The result is not significant.
Source: Developed by the authors.
H2 focused on finding out whether there were any differences in the level of engagement between the different stakeholders (individuals, news/media organizations, government and private organizations) irrespective of whether they supported, opposed or were neutral toward the DACA issue. A significant difference was found between the different stakeholder groups. As Table 2 indicates, individuals constituted the biggest group of participants and had a significantly higher level of engagement based on the numbers of hashtags, visual media and links attached to their tweets than the rest of the stakeholders. Governmental organizations constituted the smallest group with the least participation.
Stakeholders | Hashtags Number/Percentage |
Visual Media Number/Percentage |
Links Number/Percentage |
---|---|---|---|
Individual (N=1195) |
2823 (81.42%) | 114 (73.08%) | 998 (75.21%) |
News/Media Organizations (N=132) |
281 (8.1049%) | 26 (16.67%) | 141 (10.63%) |
Government Organizations (N=10) |
20 (.58%) | 1 (0.64%) | 10 (0.75%) |
Private Organizations (N=163) |
343 (9.89%) | 15 (9.62%) | 178 (13.41%) |
N=1500 | 3467 | 156 | 1327 |
X2=34.24, p <0.00001. The result is significant at p < .05.
Source: Developed by the authors.
H3 stated that there would be a difference in the level of engagement between the three days. This hypothesis was supported. Tweets on March 4th had the highest number of hashtags and visual media and the second highest links (see Table 3). The overall average number of hashtags per tweet for the three days combined was 2.3. The tweet with the highest number of hashtags of any tweet was 14 and corresponded to a neutral tweet in the March 6 sample. Most tweets did not have any visual media attached and the average for the three days was 0.10. The highest number of visual media attachments was four for each day. The average number of links or URLS attached for the three days was 0.88.
Days | Hashtags | Visual Media | Links |
---|---|---|---|
MARCH 4 | 1276 | 65 | 443 |
MARCH 5 | 1136 | 50 | 430 |
MARCH 6 | 1055 | 41 | 454 |
TOTAL | 3467 | 156 | 1327 |
X2=10.2739, p <0.036059. The result is significant at p < .05.
Source: Developed by the authors.
H4 sought to find out differences, if any, in the type of tweets (support, oppose, neutral) between the different stakeholders (individuals, news/media organizations, private organizations). Government organizations were not considered as there was at least one count of 0. The differences in the type of tweets were found to be significant, supporting H4 As can be seen in Table 4, the largest percentage of tweets in the three categories (support, oppose, neutral) came from individuals. Examples of tweets supporting DACA include the following:
Stakeholders | Support Number/Percentage |
Oppose Number/Percentage |
Neutral Number/Percentage |
Row Total |
---|---|---|---|---|
Individual |
326 (69.36%) | 282 (89.52%) | 587 (83.26%) | 1195 |
News/Media Organizations |
49 (10.43%) | 9 (2.86%) | 74 (10.50%) | 132 |
Private Organizations |
95 (20.21%) | 24 (7.62%) | 44 (6.24%) | 163 |
Column Total | 470 | 315 | 705 | 1490 |
X2= 81.713. The p < 0.00001. The result is significant at p < .05.
Source: Developed by the authors.
"More than 27,500 NC residents have been granted #DACA status. Many of those choose to seek #highereducation through community colleges. But with a looming deadline, DACA remains in jeopardy for now. #NCed https://t.co/y8VJ0MMwxj."
"800,000 Dreamers. Dream Act expiring. Families deserve to stay together. #DACA #DreamActNow."
Examples of tweets opposing DACA include the following:
"I would go to one of the #SanctuaryCities and say give me what you all are giving #DACA and all the other illegals across the nation. States are protecting illegals but sending our Police into a Mc'D's to shake down a homeless man over a hamburger."
"Save America, End #DACA #TriggerALiberalIn4Words."
There was a significant difference in the tweets that call to action or do not between the different stakeholders. H5 was supported (see Table 5). Overall, only 5.33 percent of the tweets (80 out of 1,500) had a call to action, while the other tweets expressed an opinion or shared information but did not point to a specific action. The largest percentage of tweets that called to action was by individuals, closely followed by private organizations. Most tweets from news/media organizations did not have a call to action.
Call to Action Number/Percentage |
No Call to Action Number/Percentage |
|
---|---|---|
Individual | 39 (48.75%) | 1156 (81.4084%) |
News/Media Organizations | 3 (3.75%) | 129 (9.08%) |
Government Organizations | 1 (1.25%) | 9 (0.63%) |
Private Organizations | 37 (46.25%) | 126 (8.87%) |
Column Total | 80 | 1420 |
X2= 110.3825, p < 0.00001. The result is significant at p < .05.
Source: Developed by the authors.
Examples of tweets calling to action include the following:
"Ending #DACA means that between March 6th & November 6th of 2018, our country will lose 300,000 jobs. Congress must take action & pass legislation to #ProtectDreamers. Call Congress today: https://t.co/S3MbPx4Nxw https://t.co/8AH6guPGti."
"Thanks to @UnitedWeDream's #DACA Renewal Fund, you can directly help undocumented immigrant youth with their fees: https://t.co/ypmPqdTgVL #HereToStay."
H6 that sought to examine if there was a difference in the tweets that call to action between the three groups (those who support, those who oppose and those who are neutral) was supported. As can be seen in Table 6, of the 80 tweets that had a call to action, the largest percentage of tweets corresponded to the support group. The largest percentage of tweets that did not have a call to action corresponded to the neutral group. Except for one tweet, none of the tweets by those who opposed had a call to action.
Groups Number/Percentage |
Call to Action Number/Percentage |
No Action Number/Percentage |
---|---|---|
Support (476) | 72 (90%) | 404 (28.45%) |
Oppose (315) | 1 (1.25%) | 314 (22.1126%) |
Neutral (709) | 7 (8.75%) | 702 (49.4366%) |
Column Total | 80 | 1420 |
X2= 132.6307, p < 0.00001. The result is significant at p < .05.
Source: Developed by the authors.
Discussion and conclusion
This study examined the discussion that took place on the DACA hashtag on the important issue of DACA that impacts not only the almost 700,000 recipients, most of whom are from Latin American countries, but thousands of their children and other members of their families. Ending DACA protections would also have potential economic implications, as recipients are community members who pay $8.8 billion in federal and local taxes, about $614 million in mortgage payments and $2.3 billion in rental payments annually (Svajlenka, 2019).
This study contributed to an increased understanding of microblogging behavior on the day DACA was set to end on March 5th, the day prior and the day after. The results of this study need to be interpreted in the context of similar discussions about DACA taking place through a number of different hashtags. In addition, thousands of tweets are generated using these hashtags and these discussions take place over a period of time.
This empirical research carried out within the framework of uses and gratifications theory, provides an understanding of the level of engagement and activism via Twitter by stakeholders and groups that support, oppose or remain neutral on the crucial days pertaining to ending DACA. The high use of hashtags shows an interest in engaging in the discussion in a public manner, sharing perspectives with groups wider than their personal networks, and pointing to other related information. Our study demonstrates that hashtags were more generally used by individuals (as opposed to government, news/media organizations and private organizations), who made use of Twitter to engage and disseminate experiences and positions about the future of DACA policy. Online activism has great reach and impacts awareness of issues among diverse users (Meyer and Bray, 2013). Social media is also used for purposes such as influencing legislative changes and offline action (Harlow and Guo, 2014).
Finding that only 5.33 percent of the tweets in our samples had a call to action was surprising considering that DACA was set to end on March 5th. This could be due to the Supreme Court ruling at the end of February 2018, just a few days before the March 5th deadline, that ensured that DACA would remain at least until the fall of 2018 (Gomez, 2018). It is also possible that studying a larger sample over a longer period of time would yield different results. Besides, social media is one of the many ways to voice one's opinion. As Harlow and Guo (2014) found, using traditional media channels have been found to be more effective by some members of immigrant organizations. In order to include more voices while analyzing a public debate, an analysis of social media content is recommended, in addition to other forms of gathering information about public opinion.
This study did not find any significant differences in the level of engagement between those who opposed, supported and were neutral toward DACA. Previous research on Twitter engagement during public debates has shown contradictory results, in part due to the differences in how "engagement" was measured. For example, Moe (2012) found that activists who opposed the issue being examined were the most engaged users in terms of the number of messages while Hengst (2017) found that those who supported the issue in a political debate were most actively engaged in favoriting and retweeting. In our study, we measured "engagement" in terms of the number of hashtags, visual media and links attached to a tweet.
The level of engagement found in the total number of tweets generated for the three days (47,859) differed from the level of engagement found in the random samples drawn from these tweets for each day (1,500). For example, overall the highest level of engagement in terms of the number of hashtags attached to tweets was on March 6th but in our samples, it was the lowest on March 6th. In our samples, the level of engagement in terms of the number of hashtags as well as visual media attached to tweets was the highest on March 4th, the day before the March 5th deadline. This appears to be logical. The differences found between the total number of tweets and the samples may be due to the fact that we had removed all the retweets before selecting the samples. According to Chen (2011), the gratifications obtained from connecting with other participants is higher when engaging in conversation with an original tweet than when retweeting.
Of the features of Twitter examined (hashtag, visual media and URL) in this study, hashtag was the most widely used feature, followed by links and visual media. These features provide gratifications such as allowing users to interact and participate in community-building (Sundar and Limperos, 2013). Sundar and Limperos further state that the modality of the content could impact how the content is perceived. From an agentic perspective, digital media allow participants to be senders of information. A feature gives them control over the content they put out and provides them with motivation to build community (Sundar and Limperos). Twitter enables navigability in different ways. One way is to use hashtags to inform others of where to find more information relevant to the conversation and further interact with the users of those hashtags.
The results of this study indicate that establishing hashtags is an important way for stakeholders to promote theirs interests in public forums.
The finding that government organizations formed the smallest group (10) and Individuals formed the largest group of stakeholders (1195) with the highest level of engagement was not unexpected. Although the results of our study may be skewed due to the fact that we analyzed tweets only for three days, low government participation is supported by past research (see for example, Spence et al., 2015; Rao et al., 2018). It is possible that governments use their own Twitter accounts to provide information and interact with the public. It may also be to their benefit to increase participation in other popular online public discussion forums on current issues.
Future studies could carry out a content analysis of social media at different points in time during ongoing discussions to get a deeper understanding of not only what is being discussed, and what aspects of the evolving social media are being used, but also to examine whether there are any changes that occur in these public forums over a longer time frame. Similarly, this study acknowledges the contextual information and intentionality of co-occurring hashtags. Hashtags are often used in conjunction with other hashtags within the same post. Understanding this context, of surrounding meanings and associations when other hashtags are used, is important to understand immigration movements and social media activism, and future studies should contextualize hashtag co-occurrence. With the rapid diffusion of Twitter and other social media, it is crucial for stakeholders to better understand how users are making use of these platforms.