“Rise Above the Crowd”: A Quasi-Experiment in Journalistic Event Coverage Using Mobile Phones and Billboards

Media companies have embraced the idea of citizen journalism for decades. Little research, however, systematically explores how people use mobile smartphones to engage in acts of reporting.¹ Bridging theory between social and media ritual, this project explores how a crowd used mobile smartphones to report on an organized live event where content submissions were immediately shared on billboard-size screens. A group of interdisciplinary faculty and students built a content management platform that captured texts, tweets, and photos from a crowd of over 30,000 people at a university campus event in May 2011. This paper summarizes the design of the system, as well as the patterns of content contributions, and situates the results in conversation with research on citizen journalism and mobile technology. Results suggest a platform like the one piloted here can prolong engagement at community events. Furthermore, results describe media “super users” whose exceptional engagement could be harnessed for further mobilization or even monetization.

Literature Review

Overview of Citizen Journalism

The idea that average citizens can and should produce journalistic content has roots in the public or civic journalism movement that began in the late 1980s/early 1990s.² Originally a grassroots movement seeking to challenge the transparency and control of so-called “big media,” media companies gradually came to embrace citizen journalism. After tepid first steps, like forming citizen editorial boards, companies began to encourage citizens to post original reportage on their own news sites. Ultimately, however, rather than being hailed as an advancement in transparency, critics cynically interpreted the corporate media’s embrace of citizen journalism as a cost-cutting measure, with the free content of citizen journalists replacing that of professional career journalists.³

Studies comparing the content of citizen-generated news to that of traditional news sites detail the pros and cons of relying on citizen journalism for news. Citizen-generated content generally puts greater emphasis on local news.⁴ Citizen news sites are also most apt to hyperlink to other local sites.⁵ Hyperlinking and the visibility of author profiles contribute to the perceived credibility of citizen-generated content.⁶ Perhaps most promising for advocates of citizen journalism, Kaufhold et al. found evidence that consumers of citizen journalism exhibited higher levels of both online and offline political participation than users of traditional media.⁷ Despite these positives, Lacy et al. concluded that “citizen journalism Web sites (news and blog sites) are generally not acceptable substitutes for daily newspaper Web sites.”⁸ They argue that citizen journalism is less timely and less interactive than online news sites run by daily newspapers.⁹ Accordingly, they suggest citizen websites and blogs are better suited as “complements to daily newspapers.”¹⁰

Like the media industry, academic research on journalism has a hard time keeping up with advancements in consumer technology. It is increasingly clear that tablets and mobile smartphones will have some lasting impact. According to research from the Pew Internet and American Life project, 56% of American adults had smartphones in early 2013, up 46% from 2012. Smartphone adoption grew in every demographic group in the U.S., but the largest yearly increase was in 35–44 year olds. In this group, 69% had smartphones, a 15% increase since March 2012,¹¹ the same month in which the legacy media company Gannett announced it was purchasing one thousand iPhone for reporters.¹²

In the relatively nascent field of research on mobile phones, scholars pay particular attention to how mobile technology redefines time, space, and power, with implications in a wide range of fields, among them medicine, politics, science, and journalism.¹³ Media analyst Henry Jenkins has argued that we have entered an era of “convergence culture,” wherein people expect “media convergence, participatory culture, and collective intelligence.” Together, he argues, these expectations are generating new patterns of consumption and interaction.¹⁴ In contrast to concerns that their usage facilitates social isolation, research suggests mobile phones can be used to coordinate and support social actions.¹⁵

Mobile Media and Group Engagement

Ling argues that everyday mobile communication facilitates “social cohesion” because it capitalizes on ritual interpersonal interactions.¹⁶ The same interpersonal communication behaviors that transpire face-to-face, such as humor, emotional support, and romantic cues, take place via mobile media reinforcing existing social ties. While most mobile interpersonal interactions are seemingly mundane, people can and have accessed their networks to organize protests.¹⁷ Ling argues that mobile media actually enable a different, less formal, less “stiff” traditional interpersonal interactions that may be more amenable to sharing diverse ideas and opinions.

Ling distinguishes his use of ritual from “media ritual.” He intends it to refer to an “interpersonal, point-to-point form of interaction, whereas the term ‘media ritual’ refers to a broadcast ‘one-to-many’ form of interaction.”¹⁸ Indeed, most research on media ritual has focused on television, a medium less inclined to accept audience feedback. Most studies of media rituals take a critical perspective and concern big media events, like high-profile funerals or political events.¹⁹ The significance of these events is read in the context of re-inscribing national identity, a big contrast from the small group emphasis in social ritual.

Less studied is coverage of small local events, which are a staple, or a ritual in the routine sense, of local news. Local event coverage is very formulaic, with obligatory paragraphs about who was there, what they said, and how many people came. Nevertheless, these stories are important snapshots of the interests of the local community. Often they support local or civic causes, like the local chapter of the United Way or other charities. Local event coverage has less conflict than big media events defined by Dayan and Katz and Carey.²⁰ These stories also tend to be episodic in nature, without much follow up coverage or exploration of crowd engagement or impact. Given the promise of mobile media to enhance social cohesion and Jenkins’s assertion that people are coming to expect a convergent culture,²¹ we wondered if we could define a new model of local event coverage that invites engagement outside the physical event. A model of event coverage that invites continued civic engagement potentially invigorates political participation, since civic participation is often its gateway.²²

Augmenting Citizen Journalism

Lastly, in recognition of the fact that, while mobile media use is on the rise, not every person has a smartphone, we considered ways to ensure that our model would be more than just a side conversation between mobile users. We found promise in research on how location-based annotation can play a role in engagement, community and local news.²³ The term Augmented Reality (AR) has been applied to situations where digital overlays are superimposed on the physical environment. These overlays may take many forms, from the simplest digital information displays to the rich visual information presented by AR systems.

Augmented live-event coverage is a novel area that has received very little attention, particularly when the goal is to enhance engagement of citizen journalists. In the present study we opted to augment the mobile experience by presenting citizen-generated content on large digital billboard screens spatially distributed throughout the event. The screens encouraged participants to raise their eyes from the small screens, fostering physical interactions with people standing nearby. This way even non-smartphone users could feel part of the experience.

Research Questions

We posed five exploratory research questions to help us assess the level of engagement in our model of event coverage. First, we wanted to know who used the system and who submitted content (RQ1), as well as what types of content were submitted (RQ2). Given that the system had multiple modes of engagement built in, we also wanted to know how crowd-goers used the many facets of the system (RQ3).

Furthermore, because we were interested in sustaining engagement over time, we asked, what was the time pattern for content contributions and voting? (RQ4). Similarly, we wanted to know where people voted on content—at the event or off-site? (RQ5)

Finally, although we built our own content management system, drawing on the work of Ling²⁴ we rationalized that our model would be more successful if we encouraged participants to access their established social networks. Thus, our final research question: how did social media facilitate content-sharing (RQ6)?

System Design

In order to track and share mobile content submissions during an event, we designed, built and marketed a new content management system, “Rise Above the Crowd” (“Rise”). We launched the system at a large campus event in May 2011. Students, staff, and faculty from across our university contributed to the design and development of the system. The content management system catalogued event-related texts, tweets, blogs, and photographs, which were submitted by any user who completed the online registration. Event-goers could text information to a widely publicized phone number, e-mail in text or photographs to a designated e-mail address, and/or upload photos and text via the Rise website. We also captured all tweets with the hashtag of our university’s name.

The core software development team consisted of ten representatives (four faculty and six students) from Journalism, Design, Entrepreneurship, Interactive Games and Media, and Print Media. In addition to the core development team, more than one hundred students were active contributors through their coursework on campus.

The test event, an annual festival that is free and open to the general public, attracted over thirty thousand people. It showcased demonstrations and exhibits by students and faculty at our university. Upon entry to the festival, event-goers were given mini “press passes” including instructions for how to contribute new content to or to rate existing content on Rise. Furthermore, students wearing Rise t-shirts walked around campus talking to event-goers and modeling how to make submissions. Event-goers were able to submit content via their own smartphone or on laptop and desktop computers across campus. At four designated locations around campus, individuals could learn more about the system or use tablets or other devices to upload content. Additionally, student volunteers met incoming visitors at two key entry points to help them understand and get started with Rise.

The content uploaded by event-goers was visible on big screens strategically placed in heavily trafficked areas across campus, as well as on mobile interfaces (Figure 2) and desktop computers. The big screen interface, pictured below in Figure 1, was displayed on six 52” televisions and six trailer-mounted Barco B-10 outdoor display screens (141.1” x 105.8”). The two columns to the right of the interface displayed the “Most Popular” and “Most Recent” photos, dynamically scrolling through the top twenty photos in each category. The large image to the left provided an enhanced view for one of the most popular photos, including any information entered as a caption with the photo. Across the bottom of the display, a ticker tape of recent text messages and tweets was continually updated.

Figure 1 - Big screen interface

Figure 2 - Smartphone interface - main (left) and viewing screens (right)

Event-goers (whether registered or not) could vote for their favorite photos and text-based news stories. To support voting, every content item was assigned a unique number to include in text messaging. This number, along with a username and vote total, were displayed under each content item on all interfaces.

Prizes and incentives played an important role in the engagement of all stakeholders in the Rise experiment. Registration on the day of the event was rewarded with a T-shirt and/or buttons and stickers given to encourage participation and spread awareness of the project. Prizes, including Amazon.com gift cards, digital cameras, e-book readers, and a tablet computer, were awarded both for participation and skill. Random drawings rewarded registered users at periodic intervals. Skill-based prizes were awarded for best photo and story based on crowd votes and judges’ determination, with separate awards given for journalists and photojournalists who contributed.

Method

To understand the interaction between users and the Rise system, many varied sources of data were brought together. Data were extracted from the database generated by the system using custom queries for specific research questions. Counts and descriptive statistics were calculated for contributions using Excel and SPSS. Cross-tabulation allowed for simple comparisons across groups. Voting information was analyzed similarly, as was additional data from text messaging logs maintained by the SMS gateway (Twilio). Finally, a post-event convenience survey was e-mailed to people who left their contact information in exchange for an event poster by the organizer of the campus event.²⁵ The post-event survey (N=465) asked questions about the use of technology on campus during the festival, which offered additional insights into the overall impressions of users about Rise.

Results

RQ1: Who used the system and who submitted content?

On the day of the event, more than 615 users completed the registration process to join Rise. More than one thousand unique visitors accessed the website to explore and interact with the media created during the event. More than 150 registered users submitted content into the system, accounting for 29% of the total content items submitted. The remaining 71% of the content items were submitted without using a registered account directly. These remaining contributions, which numbered almost nine hundred, included tweets by those who did not connect their Rise and Twitter accounts and text messages from those who did not verify their cell phone number with Rise.

Because we did not collect personal information as part of the registration process, we do not have much demographic data. However, results from the post-event survey (N=465) conducted by the organizers of the campus event suggest the system penetrated equally into all demographic categories (students, alumni, community members), as well as people with and without smartphones. Almost 39% of event-goers reported using their cell phones to connect to the Internet during the event. Of those 39%, 12.3% reported using Rise. However, 30% of all event-goers (including those with and without cell phones) reported being engaged with Rise in some way, such as looking at the big screen displays or text-voting for content. This percentage of use was the same across the student and community demographic categories.

RQ2: What types of content were submitted?

Table 1 lists the number of user contributions by type during the experiment. Users submitted a wide variety of contributions, including photos, tweets, text messages, and longer form written stories. The most frequently submitted types of content were tweets, stories and text, and photos.

Table 1

Analyzing the content submitted just by registered users, we found they contributed 122 texts and/or stories, 108 photos, and 247 votes. Further analysis of who submitted what showed a wide degree of variability among users representing a significant portion of the overall activity. Figure 3, for example, shows the number of photo submissions contributed by each user. The top contributor submitted twenty-five unique photos to be shared with the crowd. Twelve users contributed 30% of all photos, while the average user contributed 4.6 items. The number of photo submissions drops off gradually after these top contributors.

Figure 3 - Photo submission count by user

Similarly, Figure 4 illustrates the number of votes contributed by each user. Twelve users contributed 30% of all votes, while the average user voted 4.03 times. One super voter cast votes for 103 different content items. Three users contributed significantly more than the rest of the crowd. Interestingly, only two of the top twelve photo contributors were also among the top voters.

Figure 4 - Vote count by user

RQ3: How did event-goers use the system?

Aware of the multiple ways people at the event could use the system, we wanted to know how people used the system beyond posting content. The post-event survey had one question on this topic: “What was the primary way you used Rise Above the Crowd?” Table 2 describes the answers of the 30% who reported using Rise. People reported interacting with the large screens more than any other mode.

Table 2

Although voting may not have been the primary way event-goers interacted with the system, we analyzed the votes to see what types of content people enjoyed the most. As demonstrated in Table 3, users overwhelmingly chose to vote for photos. This is not surprising since the digital displays emphasized the photos and photo competition elements in Rise.

Table 3

RQ4: What is the time pattern of contributions and voting?

The time pattern of user activity provides important information about the user experiences and their interactions with the Rise system. The system started operation on May 7 at 10:00 a.m. However, a technical outage prevented correct functioning during the first two hours of the event. Appropriately tagged Twitter messages were recovered from the period before the formal system start. Figure 5 shows the timing of content contributions. The vast majority of activity took place during the event, in the afternoon between twelve noon and five o’clock.

Figure 5 - Contribution count by time

Users’ voting activity similarly peaked during the afternoon of the event. However, significant additional voting did occur after the event was over. This pattern is consistent with users returning to the website in order to view all of the media submissions and users encouraging voting for their own photos in order to win prizes.

Figure 6 - Voting count by time

RQ5: Where did people vote on content?: Campus versus off-campus activity

The Internet Protocol (IP) address serves as a proxy to understanding the users’ location as they used Rise. Users’ contributions did not include IP addresses, but IP addresses were captured for the votes being cast. Unsurprisingly, the voting activity on the day of the event originated primarily (68%) on campus. The day after the event, the majority of the voting (72%) shifted off-campus, as users continued to view and support the content that they enjoyed. On the second day following the event, the majority of the activity (74%) returned to campus.

Table 4

RQ6: How did social media facilitate content sharing?

Social media was essential to the Rise experience. Because our registration system was not ready until the day of the event, social media played a large role in building buzz, alerting people when registration was ready, and facilitating the sharing of content submissions and gathering votes. Table 5 below offers measures of social media engagement. The Google Group and Facebook pages each attracted more than 300 fans suggesting the usefulness of social media to support an event. Notably, however, the event ultimately attracted even more users than were in the social media groups alone. Twitter also proved useful for the event, with people tweeting specifically about the project in over 500 tweets.

Table 5

Further probing of the project’s Facebook page revealed it attracted more than 300 fans before the event began. The demographic information about this group is displayed in figure 7. The majority of fans were in the city where the event was held and were between the ages of 18–24. This data, combined with the results from the post-event survey presented above, demonstrate that while Rise was used by people across different demographic categories, connecting with and about the project over social media was more prominent in the younger, college-aged demographic.

Figure 7 - Facebook fan demographics

Discussion

The Rise experiment provides intriguing information about the behavior of connected individuals during a large public event. It suggests that user communications, cultivated within a common event-specific platform can create a rich, dynamic digital conversation. Engagement represents the ideal for those running public events and was one of the project requirements for Rise. Whereas capturing photos at events and sharing them on social media has become very common, the results of this experiment demonstrate that publicly sharing the experiences of event-goers has the potential to yield significant engagement and resultant social cohesion.

First, one specific metric of the deeply engaging nature of this digital conversation is the fact that nearly 15% of the captured Twitter messages referencing our university also mentioned Rise. In other words, despite being surrounded by over one hundred exhibits, the Twitter conversation frequently mentioned the Rise project, rather than any particular event or display. Second, another indicator of social cohesion we speculated about was activity on the Rise website after the event. The temporal pattern of use makes clear that the engagement with the system (particularly voting) continued for several days after the event. Analysis of IP addresses further shows that this voting activity was largely off-campus during the day immediately following the event. Activity returned to campus on the second day after the event. It is clear that those that enjoyed this experience wanted to stay in touch with the media and the people who were similarly involved. Perhaps even people who did not attend the event visited the Rise website.

In addition to the post-event traffic and the official survey results, the engagement of Rise was evident in other ways: social media conversations about the experience continued for several weeks; multiple videos and visualizations recaptured the experience of students and attendees; one student graduation speaker reflected on the Rise experience as emblematic of the university as a whole; family members contacted us to try to find copies of pictures in the system. Anecdotally, we heard that the moment a contributor’s first content submission appeared on a large screen was particularly powerful. In response to an open-ended question in the post-event survey, one person wrote about Rise: “It added a whole new way to enjoy my visit by turning me into a documentary photographer with a purpose.”

User Behavior and Convergence Culture

The Rise experiment also provided new information about the complex and multi-faceted behavior of individuals in this digital conversation. The demographic group of 18–24 year-olds is already accustomed to capturing and sharing media across platforms. The design of the Rise experiment relied on these expectations for, and facility with, convergence culture. Yet in the end we found evidence that an equal number of students, alumni, and community visitors used the Rise system, even though the younger demographic overwhelmingly dominated the project’s social media groups.

The prizes and incentives that were part of the experiment clearly motivated some users to contribute, vote, and solicit other voters in support of their self-interest. The voting that continued after the event demonstrates this behavior, with more votes being cast for content that was in the running for winning a prize than for those that had too few votes to be competitive. However, other users simply used the system for the experience of sharing and consuming in a novel way. This can be seen in the minimal overlap between the top contributors and the top voters. Such a distinction in use could indicate at least two basic modes of participation: those who chose to contribute content versus those interested only in voting. Alternatively, this trend could simply be a reflection that the voting system (one vote per person per item) did not provide a means to support directly one’s own contributions in pursuit of prizes. Either way, the active use of the system outside of the simple rewards may relate to the intrinsic rewards of participatory culture.

Forecasting Payment Models

Media companies today are experimenting with different payment models in response to diminished demand for traditional print newspapers. Local events have traditionally promoted a number of successful revenue sources, such as concessions, sponsorship, advertising or branded goods. As mobile connectedness advances, the engagement and associated social cohesion provided through digital experiences may become fundamental drivers of these event revenue streams. As such, deploying a system like Rise may become part of the upfront infrastructure costs provided by the venue.

Although the engagement of participants is a clear outcome of this experiment, it is also clear that the value of the digital experiences in Rise derives from the users. Analysis of user behavior provides intriguing new means for identifying payment models to support directly the value of users’ contributions. Consistent with contemporary social media systems, there were a small number of individuals who accounted for a disproportionally high use of the system in both contributions and voting. These users ultimately accounted for almost thirty percent of the contribution and voting activity overall. Like casinos, heavy users (“whales”) of online games and other social media are now being wooed because of their key role in revenue.²⁶

The Rise system captured two useful measures about the consumption of specific media items: (1) the number of unique requests to view a particular item and (2) the number of votes received by each item. These measures could easily be combined to allocate some incoming revenue (e.g. from sponsorship or event sales) to users based on the popularity of specific items. The details of a payment model are beyond the scope of the present work. However, data captured during the Rise experiment helps to identify key users who drive the overall engagement. Future research devoted to identifying their practices and targeting products and services to support their key activities may point to more engagement and more successful revenue sources for an event-media system.

Limitations and Future Research

This research presented many technical and practical challenges. Our system development time was limited to three months due to initial availability of funds, as well as a firm event date. On the day of the event, during the transition from development to production servers, the system experienced two problems. First, an outage occurred, and the system was down at the beginning of the event. Potentially valuable data from early in the event was lost. Second, the texting facility was broken, allowing only a small number of texts to be captured. This inhibited both the contribution of text content as well as the text message voting. The details of these errors were studied and solved during the period immediately following the field test. A subsequent test at a new event three months later demonstrated that these features have been corrected.

A second limitation is the difficulty in making scientific comparisons given the current quasi-experimental methodology. In making design decisions, the priority was for engagement rather than for conducting a pure and unbiased experiment. For example, user interactions with the text-based media items and the photographs were strikingly different in terms of the number of contributions and number of votes. As seen in figure 1, the screen space dedicated to photos is significantly greater than that for text-based items on the large displays. Thus users’ preference for photos in terms of contribution and voting numbers may be explained in part by the attention accorded to each in the interface.

The voting and incentive system may also be construed as a limitation in that it creates motivations to impact the natural preferences of users. There is some evidence, for example, that some users sought to recruit votes using Twitter messages with the system’s hash tag.

In terms of data collection, we would like more demographic and user assessment data regarding the usability of the system and reflecting on the overall experience. The limited set of responses indicates it might be fruitful to allow both audio and video submissions. Furthermore, we received feedback that the registration process was cumbersome. This is another technical feature we would like to revisit.

Another important future opportunity is to gather data to understand the consumption of professionally created media in the context of a large quantity of amateur-created media. Whether consumers appreciate and value the effort and training of professional journalists is an essential question in a diverse media marketplace. Similarly, more direct examination of the impact of incentives on individual behavior offers a potentially useful future research stream. Later studies in this area could help support the monetization of media systems in the future. For this experiment, the balance of the reward systems (number, value, and criteria for prizes) was constructed to maximize user engagement. The basic architecture would support alternative reward systems with different objectives, for instance to distribute income related to an event with media creators.

Finally, Rise was exploratory in nature, establishing a set of baseline outcomes for how individuals use media features. Our context of study was a non-controversial community event. One important area for the future is to examine how such a system would work at different events—particularly issue-related community, governmental, or political events—to learn if the media sharing/consumption behavior is similar in those conditions and, if not, what the boundaries of those behaviors are.

Bibliography

Arminen, Ilkka. “New Reasons for Mobile Communication: Intensification of Time-Space Geography in the Mobile Era.” In The Reconstruction of Space and Times: Mobile Communication Practices, edited by Richard Ling and Scott W. Campbell, 89–107. New Brunswick, NJ: Transaction Publishers, 2010.

Carey, James. “Political Ritual on Television: Episodes in the History of Shame, Degradation and Excommunication.” In Media, Ritual and Identity, edited by Tamar Liebes and James Curran, 42–70. London, England: Routledge, 1998.

Dayan, Daniel, and Elihu Katz. Media Events: The Live Broadcasting of History. Cambridge, MA: Harvard University Press, 1992.

Farman, Jason. Mobile Interface Theory. New York: Routledge, 2012.

Galston, William A. “Political Knowledge, Political Engagement, and Civic Education.” Annual Review of Political Science 4 (June 2001): 217–234.

Gillmor, Dan. We the Media: Grassroots Journalism by the People, for the People. Sebastopol, CA: O’Reilly Media, 2006.

Gordon, Eric, and Adriana De Souza. Net Locality: Why Location Matters in a Networked World. Hoboken, NJ: Blackwell Publishing Ltd., 2011.

Jenkins, Henry. Convergence Culture, Where Old and New Media Collide. New York: New York University Press, 2006.

Johnson, Kirsten, and Susan Wiedenbeck, ”Enhancing Perceived Credibility of Citizen Journalism Web Sites.” Journalism & Mass Communication Quarterly 86, no. 2 (June 2009): 332–348.

Kaufhold, Kelly, Sebastian Valenzuela, and Homero Gil de Zúñiga. “Citizen Journalism and Democracy: How User-Generated News Use Relates to Political Knowledge and Participation.” Journalism and Mass Communication Quarterly 87, no. 3 (September 2010): 515–529.

Katz, James Everett, ed. Mobile Communication: Dimensions of Social Policy. New Brunswick, NJ: Transaction Publishers, 2011.

Lacy, Stephen, Margaret Duffy, Daniel Riffe, Ester Thorson, and Ken Fleming, “Citizen Journalism Web Sites Complement Newspapers.” Newspaper Research Journal 31, no. 2 (Spring 2010): 34–46.

Lacy, Stephen, Daniel Riffe, Ester Thorson, and Margaret Duffy. “Examining the Features, Policies, and Resources of Citizen Journalism: Citizen News Sites and Blogs.” Web Journal of Mass Communication Research 15 (June 2009). http://wjmcr.org/.

Ling, Richard. New Tech, New Ties: How Mobile Communication is Shaping Social Cohesion. Cambridge, MA: The Massachusetts Institute of Technology Press, 2008.

Ling, Richard, and Scott W. Campbell. “Introduction.” In The Reconstruction of Space and Times: Mobile Communication Practices, edited by Richard Ling and Scott W. Campbell, 1–15. New Brunswick: Transaction Publishers, 2010.

Nichols, Sandra L., Lewis A. Friedland, Hernando Rojas, Jaeho Cho, and Dhavan Shah, “Examining the Effects of Public Journalism on Civil Society from 1994 to 2002: Organizational Factors, Project Features, Story Frames, and Citizen Engagement.” Journalism & Mass Communication Quarterly 83, no. 1 (2006): 77–100.

Putnam, Robert D. Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster, 2001.

Smith, Aaron. “Smartphone Ownership 2013.” Pew Internet and American Life Project (March 2012). http://pewinternet.org/Reports/2012/Smartphone-Update-2012/Findings.aspx

Stverak, Jason. “The Pros and Cons of ‘Citizen Journalism.” OJR: The Online Journalism Review, March 12, 2010. http://www.ojr.org/the-pros-and-pros-of-citizen-journalism/.

Tyrsina, Radu. “Gannett Buys 1,000 iPhones for Journalists.” ITProPortal, March 27, 2012. http://www.itproportal.com/2012/03/27/gannett-buys-1000-iphones-for-journalists/.

Weise, Karen. “Facebook’s Payment Whales.” Bloomberg Businessweek, March 8, 2012. http://www.businessweek.com/articles/2012-03-08/facebooks-payment-whales.