Eye tracking and social media MPC Thesis Miller 2013 Page | 1 Eye tracking and social media MPC Thesis Miller 2013 Page | 2 Acknowledgements I want to thank a number of people who were instrumental in helping me complete this two-year journey of graduate school and completing this thesis. First of all, I am indebted to Professor Sheree Josephson, whose guidance and passion for eye-tracking research elevated my project to a level that I never would have been able to achieve alone. I also want to thank Ron Hendricks for his key contribution to this research in operating the eye-tracking system, along with Professor Shane Farver, whose expertise in social media and journalism was invaluable to me as I completed this project. I want to thank my family and friends for their support, and those who volunteered as participants for my eye-tracking study. I am also indebted to my colleagues and editors at The Salt Lake Tribune, who worked around my school schedule while I somehow maintained a full-time job. Lastly, I want to thank my husband, Kelly Miller, for his love and support, even when I was frazzled and overused the phrase, "I have to work on my thesis." I love you, and am proud that we worked through this adventure of school together. Eye tracking and social media MPC Thesis Miller 2013 Page | 3 Standing out in the social media crowd: Using eye-tracking technology to examine the effectiveness of journalists’ Facebook and Twitter messages Jessica Schreifels Miller Master of Professional Communication Class of 2013 Weber State University Abstract: Studies have shown that many people turn to digital content when seeking news. That digital content often comes in the form of social media. With more than 140 million active Twitter users recorded in 2012 and 901 million monthly active Facebook users recorded in April 2012, social media is teeming with information seekers. For news organizations, it’s become a place to communicate news and market stories to a wide audience. For many news outlets, social media feels like the “Wild, Wild West” of journalism, as no strict guidelines or rules about how to make a successful post exists. Little research has been conducted to distinguish what phrasing techniques are most effective at catching the eyes of readers. The hunches of some journalists are that posts containing a question or quote are more effective than a straight-forward summary headline. To literally see if one headline approach is more eye-catching than another in social media, an eye-tracking study was designed. This exploratory study revealed that most people fixated most often on questions posts on both Twitter and Facebook. The majority of participants most often fixated on a question post, left the area, and then returned – true for both social media sites. Eye tracking and social media MPC Thesis Miller 2013 Page | 4 Social media has become so pervasive, it can no longer be ignored. Twitter reported in March 2012 that it topped 140 million active users, who send 340 million tweets per day (Twitter blog, 2012). In April 2012, Facebook officials wrote in its IPO report that the social media site had grown to 901 million users. The network experiences 3.2 billion comments per day, and the site hosts 125 billion “friendships” (Ebersman, 2012). While the number of people using social media and technology soars, print circulation numbers continue to plummet. According to State of the News Media 2011, a report by the Pew Research Center’s Project for Excellence in Journalism, weekday print circulation for newspapers is estimated to have fallen 4% on weekdays and 1% on Sunday (Pew Research State of the Media, 2011). Researchers say those declines were on par with 2010, but only about half of what they were in 2009. Despite the decline of paper-and-ink circulation, Pew researchers noted that newspapers’ digital audiences are growing. With that digital growth, many journalists have begun using social media to connect their audiences to the news. Stassen (2010) likens social media for print journalists to literally reaching out to your audience. In exploring the role of social media in journalism, she concluded that social media facilitates a type of journalism where the audience is more involved in creating the news: a place where feedback happens in real time and users can interact with one another (Stassen, 2010). Some journalists fully embrace this new way to interact with readers, while others are resistant to adopt the changes. Many journalists have failed to see the utility of Twitter in their work (Lariscy et al., 2009). The researchers reported that while 79% of the 200 business journalists surveyed used online resources for reporting, less than 1% said they used social media networks, message boards, and Twitter as a primary source. A third of those surveyed reported that they spent no time on social media in their daily professional lives. In contrast, a 2011 Pew Research study (Holcolb et al., 2011) concluded that, on their own individual accounts, reporters were not more likely than a news institution account to use Twitter as a reporting tool. According to their findings, only 3% of tweets from the individual journalists’ accounts Eye tracking and social media MPC Thesis Miller 2013 Page | 5 solicited information, which was similar to the reports from accounts maintained under the news organization’s account. Six percent of the individual journalists’ posts were retweets from outside entities, compared to 1% of the news organization’s feeds (Holcolb et al., 2011). In contrast, a 2011 Pew Research study showed that many journalists are using social media in their work. That study, a content analysis of a week’s worth of tweets from national news outlets, (Holcolb et al., 2011), concluded that reporters tweet mostly links to their organization’s site. Pew also reported that many national news outlets rarely used retweets to spread news, with retweets accounting for only 9 percent of the examined tweets. A retweet is commonly defined as when a user re-publishes something that another Twitter user has written. It functions much like “forwarding” an email, but goes to all of the user’s followers. Guidelines While several documents or books exist that dictate how journalists should carry themselves in their profession, a guidebook dedicated solely to helping journalists navigate social media does not exist. In 2010, the Associated Press Stylebook, a writing guide commonly used by journalists, added a social media section to its guidebook; however, the section merely offers definitions for social media terms matched with policies on using social media sites, not recommendations of how to create an effective or eye-catching post. While some anecdotal success cases have been offered, or “experts” have solicited advice, little academic exists that explores what sort of social media messages are most effective for journalists or news industries. Evolution of print journalism This evolution to a new digital world has caused those in the print industry to take a hard look at how they are producing their media. Hansen et al. (2010) warns that this new economic reality will require the newspaper industry to “adapt or die.” In order to fully understand where print journalism and social media have intersected, it’s important to grasp where print journalism came from and where it is going. Eye tracking and social media MPC Thesis Miller 2013 Page | 6 There was a time in print journalism where it seemed as if the sales model was nearly untouchable. Fahri (2008) explains what he calls a “scarcity advantage”: In most cities, there was only one outfit that could profitably collect, print, and distribute the day’s news, and it could raise prices even as it delivered to fewer readers each year. Indeed, monopoly daily newspapers enjoyed enormous profit margins -- sometimes as much as 25 percent or more – until very recently. But the scarcity advantage has faded; the Internet has essentially handed a free printing press and a distribution network to anyone with a computer (pg. 15). Traditionally, news organizations faced fairly low levels of competition. The news business traditionally separated the institutionalized role of journalism from its market value; however, with the shift to digital content, researchers have wondered whether people are willing to pay for the news (Heikkila et al., 2010). Looking at the source of income for print newspapers, the quick answer would be negative. Pew Research’s 2011 State of the News Media report stated that several newspapers – including the New York Times – are attempting a pay-wall system is order to monetize their digital content (Nielsen Online, 2011). But some researchers believe that current consumer habits would not support payment for news product, as consumers have been taught for a decade that unless you are seeking a niche product, content on the Web is generally free (Carter, 2009). Even if print products were able to turn a profit through offering their product in Internet form, Maier (2010) said the change in platform could affect the public’s view of the news media’s authority to report. “Migration of audiences to the Internet poses an economic threat to legacy news media,” the researcher wrote. “But less evident is how the transition affects the public’s perspectives on the news” (pg. 551). Today fewer and fewer of the younger generation read a print product. Jeff Jarvis, director for the Tow-Knight Center for Entrepreneurial Journalism at the City University of New York, was quoted as saying few people age 12 to 25 ever read a newspaper, and wireless technologies for accessing news are becoming cheaper, more accessible, and more convenient (Tucker, 2009). Eye tracking and social media MPC Thesis Miller 2013 Page | 7 In a study about downsizing’s effects on personnel, Reinardy (2010) noted in Pew Research’s State of the Media in 2010 – the latest at the time of Reinardy’s study – it was estimated that more than 13,000 full-time journalism jobs have been lost since 2006. With news of bankruptcies and cut staff, many believe that the newspaper industry is out-of-date and tired. However, Carter (2009) declared in her research: “Newspapers aren’t dying, they are transitioning. But what they are transitioning to is anyone’s guess” (pg. 80). Journalists’ social media use Because of this paradigm shift into a digital world, most newspapers experiencing success are beginning to use multiple platforms to get their news stories to the public. Franklin (2008) wrote: It is undeniably a pivotal moment in the history of newspapers understood literally as ‘news’ printed on ‘paper,’ but also understood as new content on multiple media platforms which is now delivered by the Internet, podcasts and mobile telephony, more often than by newspaper delivery boys and girls (pg. 631). Journalists are dabbling in many kinds of social media, including Twitter, Facebook and blogging. Fangfang and Martin-Kratzer (2010) studied how journalism blogs are used in China. Through the study of 54 “j-blogs” in mainland China, they discovered that nearly 40% of all posts had no reader comments. Almost a third of posts had one or two comments while only about 20% had five or more comments. Though audience participation was low in most of the blog posts, the researchers observed: “A very important finding of this study was that reader comments were positively related with on-site hyperlinks, videos and pictures. The more pictures, on-site hyperlinks and videos the j-blog posts had, the more reader comments appeared in the j-blog posts” (pg. 79). It should be noted in this study, however, that cultural differences may have some effect on these numbers, with government supervision and Internet censorship being prevalent in China. Nielsen Online reported that in 2011, Twitter was the fourth-most-popular social media site behind Facebook, Blogger, and Tumblr (Nielson Online, 2011). According to a 2012 report by Experian Hitwise, Twitter had the second-highest total of visits in March 2012 at 182 million. Facebook reportedly Eye tracking and social media MPC Thesis Miller 2013 Page | 8 had 7 billion visits during that time, and Pinterest had the third highest with 104 million visits (Wasserman, 2012). While Twitter may not be the powerhouse of social networks, it is a powerful tool. In an article printed in a 2011 issue of Economist, Nic Newman, former head of future media at the BBC, said, “The audience isn’t on Twitter, but the news is on Twitter” (pg. 9). McCarthy (2010) also identified Twitter as a major platform contributing to successful social media interaction between the public and the news media. The researcher wrote that it was undeniable that Twitter has emerged as platform where people can not only comment and share news, but they could often function as a source for journalists when they post reports of what is happening around them (McCarthy, 2010). This is one advantage journalists have from using social media: they are able to garner new sources and story tips from citizens tweeting about the world around them. Many citizens act as journalists when events occur, tweeting new information or even sending photographs. Murthy (2011) said, in a day and age of budget cuts and smaller newsrooms, print journalists can use these citizen journalists to their advantage when the citizens can tweet at a moment’s notice from their phones or other mobile devices. Some researchers (Ahmad, 2010) believe that print journalists actually have the upper hand when it comes to media messages displayed on Twitter. Because of the text-based format of Twitter, it offers more to those who specialize in the written word, compared to television and radio journalists who rely on different media to deliver the news. With this increased usage of social media, the journalists’ roles are also shifting. Because so much of news is “broken” through social media, journalists are now not only valued as reporters, but they are also being called upon to interpret of explain events, or to speculate about how those events could change the future (Harrington & McNair, 2012). The first federal proceedings to be live-tweeted by a reporter took place in 2009 (Fahri 2009). The reporter, Ron Sylvester of the Wichita Eagle, used Twitter to report on a trial of six suspected gang Eye tracking and social media MPC Thesis Miller 2013 Page | 9 members. He tweeted frequently, providing an eyewitness account that streamed nearly as the events were unfolding in the courtroom. While many reporters have begun to tweet events live from mobile phones, Fahri (2008) warns that many reporters are also using the platform in ways that may not be as useful. The researcher said that not all tweets are equally useful, using a recent political conversation to illustrate that many journalists fill their social media newsfeed with plenty of snark or trivia but offer little in the way of important news (Fahri, 2008). No hard and fast rules for what is appropriate for a journalist to post on social media exist, and some liken social media to the “Wild, Wild West” of journalism. While some adhere to rules of posting only story links or information about a story, other reporters freely give their opinions or become somewhat of a personality, posting personal messages about haircuts or where they went to lunch. How social media messages affect Web traffic While Pew researchers reported that in 2012 news sites only received 9% of their traffic from social media sites, that number is up 57% compared to the last two years (Olmstead et al., 2012). Olmstead et al. also noted that the percentage of website hits generated from search engines is declining. While it represents 21% of Web traffic, this is down 9% since 2009. The researchers concluded that social media now bring in almost half of the traffic to news sites as search engines. This shift to a social media audience has caused American magazine “The Atlantic” to focus less on search engine optimization and more on social media interaction (Indvik 2012). Scott Havens, senior vice president of finance and digital operations at The Atlantic Media Company, was quoted in Indvik’s Mashable article saying that 40% of its traffic comes from social media as of May 2012. Effective phrasing techniques According to research by Facebook employees (Lavrusik & Cameron, 2011), feedback on journalists’ Facebook pages came most often when the journalist posed a question to followers. The researchers wrote that while only 10% of journalist pages studied posed questions on their Facebook pages, posts with a question received twice as many comments and 64% more feedback overall than the Eye tracking and social media MPC Thesis Miller 2013 Page | 10 average post. According to the researchers, this question phrasing technique solicited the most interaction, compared to asking the reader to take a closer look, behind-the-scene posts or posts with catchy or clever language or tone. Twitter researchers have also provided a best-practice guide for journalists. However, they did not offer any advice to journalists on what sort of phrasing techniques were effective; rather, they advised reporters to tweet their beat or coverage area, use hashtags in their posts and to share what they are reading with their followers (Luckie, 2012). In a column published in the Columbia Journalism Review (Friedman, 2012), Ann Friedman offers readers tips on how to get more Twitter followers. Among those tips, she advises journalists to quote the most interesting fact from an article, rather than post a headline. She reasons that while headlines “can” work on Twitter, it’s most often a shocking fact or quote that will prompt people to click on the link. The Poynter Institute, a non-profit school for journalism in Florida, recommends that instead of tweeting headlines, reporters should try starting a conversation about their story by posing a question, sharing a detail about the reporting process or talking about the reporter’s favorite part of the story (Tenore, 2011). Steve Buttry, digital transformation editor for Digital First Media, advised in his blog that sharing merely headlines on Twitter should be avoided. He advised journalists to create a tweet that is more conversational and will intrigue people to click on the link (Buttry, 2012). Eye tracking While the journalistic experts can make recommendations based on first-hand experience or study the effectiveness of techniques by noting click-through rates, eye tracking allows research to observe participant’s eye movements and gain a better understanding of exactly where people are looking on their social media pages. The method is useful for providing insight into what people pay attention to, as well as how they scan and search for information online. Eye tracking and social media MPC Thesis Miller 2013 Page | 11 Eye-movement technology uses a video camera paired with an infrared light that shines into the participant’s eyes. The camera records exactly where the participant is looking at, and can precisely measure fixations, or the brief periods of time when a participant’s eyes are fixed on a point (Josephson, 2008). Levy-Schoen (1983) explained the importance and relevance of eye-tracking research in a way that is still pertinent today: “Since oculomotor activity (eye movement) is an overt behavior that is accessible to recording and measurement, it opens a door to the scientist interested in the organization of perceptual and cognitive processing. To the extent that eye movements are reliable correlates of the sequential centering of attention, we can observe and analyze them in order to understand how thinking goes on” (p. 66). The eye-tracking method is beneficial because it allows researchers to measure where a participant’s point of interest is. Even when participants can verbalize what they are observing with a “think-aloud protocol,” eye-tracking data will provide more accurate data than verbalization can (Cooke, 2005). This ability to track a participant’s natural behavior and observe which elements capture his or her attention is beneficial. While self-reporting methods can provide a deeper thought process, eye-tracking data gives researchers a direct insight to how the participant is interacting with the media they are observing (Bucher & Schumacher, 2006). Furthermore, this method allows researchers to observe intentional movements, as well as non-intentional movements that the participant might not verbalize. Participants only acquire information from what they read during a fixation, which is defined as an average of about a quarter of a second, or 250 milliseconds (Rayner, 1993). The study of observing eye movement is of interest because researchers can use the information to describe the characteristics of eye movements, as well as use the data make inferences regarding a participant’s perceptual and cognitive processes while they are observing a medium (Rayner, 1993). Only a small visual area can be observed during a fixation. Levy-Schoen (1983) points out that this may protect the central processors in the brain from overload. These visual fixations are physiological indicators of attention that can be linked to cognitive processing (Just & Carpenter, 1980; Rayner, 1978). Eye tracking and social media MPC Thesis Miller 2013 Page | 12 Eye-tracking hardware can either be mounted on the head or mounted remotely. These sorts of systems measure the orientation of the eye relative to a fixed unit (Cooke, 2005), and in some systems require a participant to remain as motionless as possible, since the system cannot measure pupil glint from multiple angles. Surveys or questionnaires, wherein the participant gives further data about their opinions or view behaviors, often follow eye-tracking exercises. Research studies on eye tracking Though academic eye-tracking studies detailing people’s observations of journalists’ use of social media is scarce, a number of studies have been published relating to how participants view information. Chu et al. (2009) used a Tobii 1750 eye tracker to observe 96 participants viewing a Minneapolis Star Tribune website. The researchers were interested in how different approaches to designate updated or breaking news aided or impeded the likelihood that users will find, read or remember news items. They found that 89% of participants fixated on a box marked with the words “Breaking News” on the newspaper’s website, 48% fixated on a timestamp, and 49% fixated on a headline that was prefaced with “New!” In the same study, Chu et al. (2009) concluded personal interest in a story significantly affects how much time people spend viewing an element on the page. They concluded that the best design may not impact the reader’s attention if the observer is not interested in the topic. Another study utilized eye-tracking research to observe on-screen legibility of sans serif and serif typefaces (Josephson, 2008). Six participants read four Associated Press wire stories on a computer screen, each with a different typeface. Though the results were not statistically significant, Josephson discovered that participants were able to read the story displayed in Verdana typography more quickly and with fewer regressions. A regression is defined as when a participant’s eyes move back in the text to look at material that he or she has already read (Rayner, 1993). Bojiko (2006) used eye tracking to compare Web page designs for the American Society of Clinical Oncology, a website intended to give general information about the organization and cancer care. A Tobii 1750 system was used on 12 participants. Participants were asked to perform tasks on both the Eye tracking and social media MPC Thesis Miller 2013 Page | 13 existing Web page and a new proposed design. Through observing the participant’s eye movements, researchers were able to conclude that the new proposed design better met user needs and expectations on multiple fronts, including a less cluttered and distracting design and link placements. Another communication study used eye tracking to assess participants’ concentration of attention during Internet video advertisements. Using a Tobii 1750 eye tracker, they collected data from 58 participants (Teixeira et al., 2012). They observed: Avoidance of television advertising has become a major problem for the advertising industry. This is one reason advertisers are migrating to the Internet. However, viewers exhibit avoidance behaviors online as well, either through lack of attention concentration or simply by various forms of zapping, clicking or scrolling the advertisements (p. 156). Hervet et al. (2011) observed whether Web viewers were actually “blind” to banners, in other words, whether users could completely divert their attention to areas where a banner ad typically resides on a Web page. Forty-eight participants observed eight websites that described how to choose a digital camera. All of the websites viewed contained banner ads that contained only text. Eye-tracking information was gathered using a Tobii 1750 eye tracker. The researchers concluded that while a majority of participants fixated on a banner ad at least once during their website visit, the ads were not fixated upon as long as the editorial content regarding digital cameras. Following the eye-tracking test, participants were asked what they remembered about the ads. Through this survey, researchers concluded that while the participants may have fixated on the ads, it did not guarantee that ad content will be remembered. Though Leckner (2012) did not conduct her own research, she examined eye-tracking research relating to people’s reading behaviors of printed and online media. Her discussion focused largely on how presentation factors on both print newspapers and websites affect people’s reading behavior. One of her main conclusions was that, while photos are main entry points in printed newspapers, they have generally ambiguous results as entry points online and do not elicit major visual attention online (Leckner, 2012). Eye tracking and social media MPC Thesis Miller 2013 Page | 14 Theoretical relation Uses and gratification theory can be applied to understand a users’ relationship to social media sites and what drives them to participate in such sites. The theory, which explains that people use media purposefully to meet their desires and needs to achieve gratification, emphasizes that people are intentional and selective in their media choices (Katz et al., 1974; Raacke & Bonds-Raacke, 2008; Althaus & Tewksbury, 2000). For social media users, their desire and reasons for participating in the medium are plentiful. Popular uses and gratifications would be to keep in touch with old friends, to keep in touch with current friends, to feel connected, to learn about news or events, or to post about social functions (Raacke & Bonds-Raacke, 2008). Quan-Hasse & Young (2010) applied the uses and gratification theory to social media by probing into what gratification users received from using Facebook. The researchers found that most participants studied were motivated to join Facebook because a friend suggested it. Top gratifying reasons for using the social media included that Facebook was a pastime or a time-waster, in order to express friendship to others, because it is trendy, to share their problems and to observe social information. The uses and gratification theory applied to social media usage directly relates to this study as the researcher explores how people see and interact with social media in their everyday lives. A neurological approach to perception paired with visual communication theory also provides insights into how people interpret visual media, such as television or a computer screen (Barry, 2002). According to Barry (2002), when people watch television or a film, the unconscious part of the brain sees what is happening as reality and continues to learn from and to respond to what it sees. In other words, while people stare at a piece of glass on the television, their brain tells them the images they are seeing are real people and places, not an array of glowing dots (Pinker, 1997). This theory can be extended to the reaches of social media as well. While people see a computer screen full of megapixels, their brain interprets the social media messages as meaningful and real events. Eye tracking and social media MPC Thesis Miller 2013 Page | 15 Research questions While researchers have probed whether journalists use social media in their daily habits, little academic research has been introduced that helps journalists understand how to make their social media interactions successful. There are a myriad of ways to present a link to a story or article on social media, but is there a phrasing technique that makes the message more eye-catching? The purpose of this study is to explore two phrasing techniques and determine if one is more eye-catching than the other, and whether what is eye-catching on Twitter is the same on Facebook. The techniques examined were a “statement headline,” such as “Blizzard hits NYC,” and a question format, such as “Were you prepared for the blizzard that hit NYC?” The research questions for this exploratory study are as follows: Q1. What headline phrasing technique (statement or question) do participants fixate on most? Q2. What headline phrasing technique (statement or question) do participants look at the longest? Q3. What headline phrasing technique do participants visit most? (In other words, do participants revisit the headline after looking away, and if so, how many times?) These three questions will be answered for each social media platform individually, ultimately leading to the next research question: Q4. Are there any differences (fixation number, fixation duration, fixation returns) between the two social media platforms? In other words, what is the most effective headline phrasing technique for Facebook? What is the most effective headline phrasing technique for Twitter? Methodology This study was designed to investigate eye-movement behavior in relation to viewing social media messages on a computer screen. Two headline phrasing techniques— statement and question—were selected for observation. The two techniques were selected because of their common usage on social media networks and also because they are commonly discussed by experts providing advice to journalists using social media. Eye tracking and social media MPC Thesis Miller 2013 Page | 16 Participants Thirty-seven participants were recruited for the study, which was conducted on the campus of Weber State University, a mid-sized university of more than 26,000 students, located in Ogden, Utah. The participants included 21 women and 16 men. Their ages ranged from 19 years old to 56 years old, with the average age being 28. All participants were regular users of the Internet who had both a Twitter and Facebook account. Apparatus The eye-movement data were collected using an ISCAN RK 426PC Pupil/Corneal Reflection Tracking System. The same system was used by Josephson (2008), who described the system as one that uses a “corneal reflection system to measure the precise location of a person’s eye fixations when looking at a visual display” (pg. 75). In this case, that visual display involved two social media newsfeeds. The system is a PC card-sized, real time digital image processor that automatically tracks the center of the pupil and reflection from the corneal surface. The eye is illuminated with a low-level infrared source to allow the system to track exactly where an individual is looking at (ISCAN, INC. Operating Manual, 1998). The processor operates at a sample rate of 60 Hz and the subject’s eye position may be determined with an accuracy typically better than 0.3 degrees over a +/-20 degree horizontal and vertical range using the pupil/corneal reflection difference. Stimulus materials Facebook and Twitter newsfeeds were created using Photoshop. The eye-tracking system allowed users to observe messages on the screen, but it did not allow for users to click on links or scroll (See Appendix 1). The newsfeed used in the study contained four stories from a fictional news outlet, “The Associated Times.” A fictional news outlet was used to eliminate the risk that participants may have seen an authentic news outlet’s logo or social media handle. Eye tracking and social media MPC Thesis Miller 2013 Page | 17 The Associated Times posts were placed in between other posts by made-up “friends.” This mimicked the social media experience of following a news outlet but also seeing messages from friends or other people of interest on one’s newsfeeds. In an effort to not specifically draw attention to the fabricated friend’s posts, stock photos of three men and three women were used as profile pictures, and their identities were chosen from common names in the United States. In order to create the content for the “friend’s” posts, the researcher gathered generic posts from the most-followed accounts on Twitter and Facebook, which as of January 2013, were celebrities Justin Bieber, Katy Perry, Taylor Swift, and Rihanna. These accounts were used because no research regarding common content in social media posts was found, and the most-followed accounts could be deemed “successful” at using Twitter and Facebook. The story links posted from The Associated Times used in this study were gathered from the Associated Press newswire in January and February 2013. The headlines were re-phrased as questions and also shortened for use in this study. The social media posts were between six and nine words long and contained between 29-34 characters. The posts were purposely short so eight posts from “friends” and the Associated Times could be included on the page. Participants were asked to look at both a Twitter newsfeed and a Facebook newsfeed, each containing two statement headlines and two questions interspersed with social content. If a participant saw a statement headline listed first on a Facebook newsfeed, for example, they saw a question as the first entry on the Twitter newsfeed. In an effort to prevent ordering bias, the four posts by the Associated Times were rotated through both social media newsfeeds so that each headline appeared in each of the four headline positions. To accomplish this, eight versions of Facebook and eight versions of Twitter were created in Photoshop. Eye tracking and social media MPC Thesis Miller 2013 Page | 18 Procedure: Though 37 participants intended to complete the study, because of issues with the eye calibration, data could only be recorded for 33 participants. To keep the participant groups even, data from 32 participants were examined for the study. Participants were randomly assigned into two groups, with one group observing a Twitter newsfeed first, and another viewing Facebook first. As detailed above, the four posts by the Associated Times were rotated throughout both social media newsfeeds in an effort to prevent ordering bias. When participants arrived in the area where the study was being held, they were asked to read and sign an informed consent form. Then, they were individually taken into a small room where they were asked to sit in front of a computer so they could comfortably read the screen. Recording Eye Movements Before participants could view the two newsfeeds, the eye-movement apparatus was calibrated to each participant’s eyes using a procedure that requires the system operator to instruct the participant to fixate on one of five numbered dots displayed on their computer screen. The dots are positioned in the center, upper left, upper right, lower left, and lower right sections of the participant’s computer screen. The system calculates horizontal and vertical correlating factors for each of the four quadrants of the participant’s field of view. After correlation, these correlating factors are used to compute the point of regard for subsequent eye movements. Data collection Eye movements were recorded while participants viewed two social media newsfeeds. They were allowed to see each newsfeed for 15 seconds. Data was recorded by the computer program, as well as was captured on a video file. A survey was distributed to participants after they observed the two social media pages, which probed questions regarding the participants’ media habits and their ideas of what made a social media post successful. Fixation Criterion Eye tracking and social media MPC Thesis Miller 2013 Page | 19 A criterion for a fixation was selected before eye-movement data was analyzed. This study used the same definition of a fixation as Josephson and Holmes (2002): a pause of at least 100 milliseconds where an area of 10 pixels by six pixels was processed. This fixation criteria is widely used in eye tracking research, including researchers Fischer et al. (1989) and Baron (1980), who defined 100 milliseconds as the minimum amount of time necessary for a pause to be considered a fixation. Descriptive Statistics In order to answer the research questions, the number of total fixations a participant had on a post, how long they fixated on a post, and how many different visits they made to each the post. For the initial statistics, fixation data for each participant was totaled for both summary headlines and question headlines to determine the total number of fixations, the total duration of fixation, and the total number of visits for each headline phrasing technique for Facebook and then for Twitter. Once those numbers were totaled, the researcher compared the two numbers to determine which headline phrasing technique received the most attention from each individual. From there, the researcher could conclude which phrasing technique was observed most often for each of the three measurements. Descriptive statistics (percentages) were also computed to show how many participants fixated at least once on a post, as well as to analyze what effect ordering had on the results. “Visual areas” were defined in order to establish what areas were defined as an individual post. Each of the boxed areas below represented one “visual area” where fixations were recorded and grouped together. Eye tracking and social media MPC Thesis Miller 2013 Page | 20 Results Number of fixations: Facebook The total number of fixations for the two statement headlines were added together for each participant, as well as the total number of fixations for the two questions that each participant observed on Facebook. Once those numbers were totaled, the researcher compared the two numbers to determine which kind of headline each participant paid the most attention to. Among the 32 participants with useable data, 16 participants (50%) fixated more times on questions. Ten (31%) had the highest number of fixations for headlines, while six participants (18 percent) fixated equally on statement headlines and questions. Eye tracking and social media MPC Thesis Miller 2013 Page | 21 ParticipantTotal headline fixationsTotal question fixationsViewed most132Headline244Equal314Question412Question523Question676Headline723Question871Headline955Equal1123Question1303Question1443Hedline1545Question1631Headline1723Question1822Equal1911Equal2012Question2142Headline2222Equal2303Question2442Headline2544Equal262Question2743Headline2843Headline2934Question3023Question3123Question3234Question3311Equal3442Headline3643HeadlineAverage2.822.5Number of fixations of FacebookEye tracking and social media MPC Thesis Miller 2013 Page | 22 Length of time: Facebook A similar method was used to observe which phrasing technique participants spent the longest amount of time looking at or the total fixation duration. Participants were split evenly: 16 participants spent the longest amount of time looking at questions, while 16 participants looked longest at statement headlines. Participant Total time on headlines Total time on questions Most time viewed 1 2416 599 Headline 2 2666 4566 Question 3 1766 5033 Question 4 1366 2850 Question 5 1416 2416 Question 6 4116 2032 Headline 7 2683 1200 Headline 8 4816 1450 Headline 9 2066 2416 Question 11 2449 3866 Question 13 0 2132 Question 14 5532 2666 Headline 15 333 5166 Question 16 3400 3133 Headline 17 5099 3499 Headline 18 3499 2899 Headline 20 4650 3033 Headline 21 2483 3583 Question 22 3483 1200 Headline 23 0 4600 Question 24 4700 3600 Headline 25 3400 3666 Question 26 1200 3550 Question 27 1749 3266 Question 28 4433 2882 Headline 29 3282 2866 Headline 30 1216 2583 Question 31 4416 3850 Headline 32 2483 2333 Headline 33 300 1266 Question 34 3166 4483 Question 36 3933 2199 Headline Average 2766.16 2965.09 Time spent viewing on Facebook (in milliseconds) Eye tracking and social media MPC Thesis Miller 2013 Page | 23 Times participants returned to a post: Facebook Again, the same method was used to observe which headline phrasing technique participants most often returned to after looking away. Thirteen participants (40%) most often returned to questions, nine (28%) came back to statement headlines, and five (15%) returned to both phrasing techniques equally. Six participants (18%) never returned to a headline after looking away. ParticipantTotal time on questionsTotal time on headlinesMost time viewed102Headline320Question521Question7None914Headline111Question1310Question1533Equal17None2103Headline23None2522Equal2712Healine2921Question3110Question3300None211Equal410Question645Headline805Headline1411Equal1602Headline1800None2010Question2210Queston2413Headline2610Question2812Headline3020Question3221Question3402Headline361HeadlineAverage1.141.48Number of times participants returned (Facebook)Eye tracking and social media MPC Thesis Miller 2013 Page | 24 Number of fixations: Twitter The same method was used to calculate number of fixations on Twitter. Fifteen participants (46%) fixated more on questions on Twitter, while 12 participants (37%) had the largest number of fixations on statement headlines. Five participants (15%) had the same number of fixations for statement headlines as questions. ParticipantTotal headline fixationsTotal question fixationsViewed most132Headline242Headline321Headline466Equal532Headline623Question731Headline814Question956Question1133Equal1322Equal1414Queston1524Question1614Question1724Question1825Question2011Equal2124Question2210Headline2311Equal2402Question2532Headline2647Questio2725Question2810Headline2934Question3043Headline3125Question3226Question3310Headline3424Headline3653HeadlineAverage2.383.13Numbr of fixations of TwitterEye tracking and social media MPC Thesis Miller 2013 Page | 25 Length of time: Twitter A similar method was used to observe which phrasing technique participants spent the longest amount of time looking at. Participants were split equally for Twitter as well as Facebook. Sixteen participants spent the longest amount of time looking at questions, while 16 participants looked at statement headlines the longest overall. ParticipantTotal time on headlinesTotal time on questionsMost time viewed142822783Headline237331233Headline336001833Headline421832250Question523491666Headline612332316Question727501400Headline8161500Question930661999Headline1134491266Headline1312492400Question1421834399Question1535663916Question169833416Question173049376Question1816333083Question2022831300Headline2147832183Headline2214660Headline2334002033Headline2402966Question2534662416Headlie2617165200Question2736162132Headline281660Headline2934501850Headline3033822900Headline3113163633Question3217664166Question331330Headline3418162299Question3534661483Question3633833566QuestionAverage2431.272342.52Time spent viewing on Twitter (in milliseconds)Eye tracking and social media MPC Thesis Miller 2013 Page | 26 Times participants returned to a post: Twitter Again, the same method was used to observe which phrasing technique participants most often returned to after looking away. Seventeen participants (53%) most often returned to questions, five (15%) came back to statement headlines, and five (15%) returned to both phrasing techniques equally. Five participants (15%) looked at the content in only one visit. Order and position The researcher also examined whether position of the headline on the page had an effect on the three variables of interest. On Twitter, 28 of the participants (87%) fixated on the first Associated Times post at least once. Four participants did not have a single fixation on the first post. ParticipantTotal times returned to headlinesTotal times returned to questionsReturned to most often111Equal211Equal310Headline435Question511Equal611Equal710Headline803Question924Question1111Equal1301Question1401Question153Question1602Question1701Question1803Question2000Neither2102Question2200Neither2300Neither2401Question2510Headline2636Question203Question2800Neither2912Question3021Headline3103Question3214Question3300Neither3402Question3631HeadlineAverage0.721.66Number of times participants returned (Twitter)Eye tracking and social media MPC Thesis Miller 2013 Page | 27 In the second Associated Times post listed on the page, 29 participants (90%) fixated at least once. Three participants did not have a single fixation. In the third Associated Times post on the Twitter page, 23 participants (71%) fixated on the post at least once, while 9 never fixated on the post. In the last Associated Times post, only 14 participants had at least one fixation on the post. Eighteen did not have a single fixation on this post. Results were similar on Facebook. Twenty-nine participants (90%) fixated on the first Associated Times post at least once. Three participants did not have a single fixation on the first post. In the second Eye tracking and social media MPC Thesis Miller 2013 Page | 28 Associated Times post on the Facebook page, 30 participants (93%) fixated at least once, while only two participants did not have a single fixation. Twenty-seven participants (84%) fixated at least once on the third Associated Times post on the Facebook page, while five never fixated on that post. In the last Associated Times post, participants were split evenly: 16 had at least one fixation on the post, and 16 never fixated on the post. Looking at the four Associated Times posts on both Facebook and Twitter, the post in the top position on the page had the largest number of fixations. On Twitter, the participants fixated on the first Associated Times post 70 times. In the second Associated Times post, 65 fixations were recorded from all participants. For the third Associated Times post, 36 fixations were recorded, while 25 fixations were recorded for the fourth and final Associated Times post. On Facebook, the participants fixated on the first Associated Times post 68 times. In the second Associated Times post, 66 fixations were recorded from all participants. For the third Associated Times post, 46 fixations were recorded, while 23 fixations were recorded for the fourth Associated Times post. It should be noted that in both Twitter and Facebook, the first “friend” post had the highest number of fixations on both social media pages. On Twitter, 116 fixations were recorded for the first “friend” post, compared to 70 fixations on the first Associated Times post. On Facebook, 77 fixations were record for the first “friend” post, compared to 68 fixations on the first Associated Times post. On both Facebook and Twitter, participants looked at the first Associated Times post for the longest amount of time. On Twitter, the participants looked at the first Associated Times post for 61,708 milliseconds. In the second Associated Times post, the participants looked at the post for 47,340 milliseconds, compared to 37,275 milliseconds on the third Associated Times post. On the fourth and final Associated Times post, the participants spent a collective 19,293 milliseconds looking at the post. On Facebook, the participants collectively looked at the first Associated Times post for 66,057 milliseconds. In the second Associated Times post, the participants looked at the post for 62,973 Eye tracking and social media MPC Thesis Miller 2013 Page | 29 milliseconds. On the third Associated Times post, participants collectively looked at the post for 45,507 milliseconds, compared to 25,158 milliseconds on the fourth and final post. Next, the researcher analyzed how many times the participants looked at a post, looked away and looked back. For Twitter, participants returned to the first Associated Times post 36 times, 28 times on the second Associated Times post, and 10 times for both the third and fourth Associated Times post. On Facebook, participants returned to the first Associated Times post 28 times, 35 times on the second Associated Times post, 15 times for the third post and five times for the fourth and last Associated Times post. Survey results Once participants completed the eye-tracking portion of the study, they were asked to fill out a short questionnaire on a different computer. The survey was compiled using a Google form. (See Appendix 4). Seventy-three percent of participants reported that they used social media daily as a tool to get information about news. Forty-three percent of participants reported that they follow over 20 media outlets on various social media accounts. After participants were asked demographic questions, including their social media habits, they were asked an open-ended question: “What phrasing techniques would catch your attention on social media?” Twenty-eight respondents included “a headline” in their response, while 14 said they thought quotes would catch their attention on social media. Seven respondents indicated that a question could catch their attention, while five said a photo could prove eye-catching to them as well. After they answered the open-ended question, they were asked four Likert-scale questions asking how effective a question or a headline would be in Facebook and in Twitter. Twenty-two respondents (59%) said a Facebook post phrased as a question would be “somewhat effective,” with only four rating a question as “very effective.” Six responded with “somewhat Eye tracking and social media MPC Thesis Miller 2013 Page | 30 ineffective,” while five said a question phrasing in Facebook would be “neither effective nor ineffective.” No participants rated a question phrased in Facebook as “very ineffective.” If a Facebook post was phrased as question, how effective would that be at catching your attention? Very effective Somewhat effective Neither effective nor ineffective Somewhat ineffective Very ineffective When they were asked how effective a statement headline would be in Facebook, 18 participants (48%) said it would be “somewhat effective,” while 13 participants (35%) rated it as “very effective.” No participants rated a post phrased as a headline as “very ineffective” or “somewhat ineffective,” though six said they would find the phrasing technique neither effective nor ineffective. Eye tracking and social media MPC Thesis Miller 2013 Page | 31 If a Facebook post was phrased as a statement headline, how effective would that be at catching your attention? Very effective Somewhat effective Neither effective nor ineffective Somewhat effective Very ineffective When asked how effective a Twitter post phrased as a question would be, 19 participants (51%) said it would be “somewhat effective,” with only one participant responding that it would be “very effective.” Seven said they would find it “neither effective nor ineffective,” while another seven rated a question phrasing question as “somewhat ineffective.” Three said they would find it “very ineffective.” Eye tracking and social media MPC Thesis Miller 2013 Page | 32 If a Twitter post was phrased as a question, how effective would that be at catching your attention? Very effective Somewhat effective Neither effective nor ineffective Somewhat ineffective Very ineffective Eighteen respondents (48%) said they thought a Twitter post phrased as a statement headline would be “somewhat effective,” while 16 participants (43%) rated a statement headline being “very effective” on Twitter. Three said they would find it neither effective nor ineffective, and no participants rated a statement headline as “very ineffective” or “somewhat ineffective” on Twitter. Eye tracking and social media MPC Thesis Miller 2013 Page | 33 If a Twitter post was phrased as a statement headline, how effective would that be at catching your attention? Very effective Somewhat effective Neither effective nor ineffective Somewhat ineffective Very ineffective Discussion The first research question attempts to determine what phrasing technique had the highest number of fixations for each social medium. For both Twitter and Facebook, the majority of participants had the highest number of fixations on posts phrased as questions, with 50% for Facebook and 46% for Twitter. The second research question addresses which phrasing technique participants looked at longest. For both Twitter and Facebook, the number of participants who looked longest at questions was equal to those who focused longest on statement headlines. The third research question asks what phrasing technique participants looked at on multiple occasions. According to the data, the majority of participants most often returned to a question post after looking away. With these three questions answered, it allows for the researcher to conclude that posts phrased as questions generate more attention on both Facebook and Twitter than those phrased as statement headlines, with the exception of the amount of time spent looking at the post. This conclusion indicates Eye tracking and social media MPC Thesis Miller 2013 Page | 34 that the most effective phrasing technique is the same for both Twitter and Facebook, which answers the fourth research question. Interestingly, most of the participants indicated in their survey that they felt statement headlines were more effective in getting their attention than questions were. Though they responded that they preferred statement headlines, the eye-tracking research showed that they may be more drawn to posts phrased as questions than they are aware of. Their preference may have more to do with not what they feel is more eye-catching, but what phrasing techniques are most pleasing to them. For instance, one participant wrote that while writing a post in social media with the caps lock on would be eye-catching to him or her, that person finds that method annoying. The same could possibly be true for why participants did not feel questions were effective in getting their attention. It appears that ordering may have significance in regards to how long a person looks at a post and how many times they fixate on the post. The posts in the first two positions of either Facebook or Twitter are looked at longest and most often. Each participant looked at each social media page for 15 seconds. That amount of time may not have given the participants enough time to view the bottom of the page, as evidenced by the fact that only half of participants fixated on the last Associated Times post on the bottom of the page. However, it should be noted that slightly more participants fixated on the fourth Associated Times post on Facebook, perhaps indicating that people read through Facebook newsfeeds more quickly than Twitter. This finding regarding ordering is meaningful to note, though it may not assist social media users seeking to create better or more eye-catching content. Though a user can control what they post on other’s newsfeeds, they cannot control where that post appears on anyone’s social media feed. Limitations and Future Research One of the biggest limitations of this study was the ability to analyze only two variables. Another limitation of the study was that it is impossible to truly re-create someone’s experience on social media in a controlled environment. Though the simulated newsfeeds looked the same as the authentic social media pages, it is safe to assume that participants may read their “friends” posts more closely than they did Eye tracking and social media MPC Thesis Miller 2013 Page | 35 during this research, simply because they have a personal attachment to that friend. The same can be said about a news outlet: if that news outlet is their personal favorite, they may read posts more thoroughly because of their attachment or favor towards that outlet. Another limitation of the study was that participants could not click on any of the links. This was a limitation because it did not mirror an authentic social media experience. Similarly, the participants were not able to scroll down the page, as they would on their own social media pages. Researchers with more advanced eye-tracking equipment may want to research what effect scrolling and the ability to click on links has on how people fixate on posts on Twitter and Facebook. In future research, it would be beneficial to run a similar study for other phrasing techniques, such as quotes or photos, which are suggested by journalism “experts” and bloggers as also being effective to catch a social media user’s attention. With the trends of social media always changing, it would also be beneficial for a similar study to be conducted on a different social media – such as Google+ or whatever the next trend may be. Conclusion Though this research shows neither phrasing a social media post as a question or a statement headline affects how long a person looks at the post, journalists and those working in news media can conclude through this research that phrasing social media posts as questions may help to garner more attention to their story links than a statement headline would. This revelation supports anecdotal theories offered by journalists and experts – such as Friedman (2012), Tenore (2011), and Buttry (2012) – who advise using questions or other phrasing techniques over straight forward statement headlines. It would be intriguing to see if their guidelines in using other phrasing techniques, such as quotations or personal reporter anecdotes, would also prove to be more effective in garnering attention to posts. Of course, just because a person may observe or fixate on a post does not necessarily mean they will click on the story and read about it on a news outlet’s Web page. As Chu et al. (2009) wrote in their eye-tracking findings, design (or in this research, phrasing techniques) may have no impact if the user has Eye tracking and social media MPC Thesis Miller 2013 Page | 36 no interest in that topic. However, with an overwhelming amount of information being directed at social media users in today’s technology-enriched climate, it is beneficial to take a thoughtful approach to social media in an attempt to make posts stand out among the crowd. If journalists can produce posts that are more eye-catching, it can help them to draw more eyes to their publications. As mentioned previously, 9% of Web traffic is coming from social media sites in 2012, according to Poynter (2012). While this number pales in comparison to other sources, such as search engines, it still up 57% compared to the last two years. While the trend of sharing links and stories through social media is still in its infancy, it will continue to gain momentum and become more popular as social media becomes more and more integrated into peoples’ lives. Through effective phrasing techniques, journalists can help draw more eyes to their publication’s website, a small number, perhaps, but the effort is worthwhile when one considers the masses of people shifting to the Internet and social media and away from traditional and legacy media outlets. Though the social media audience is by no means the majority of Web traffic to news websites, tailoring a message to maximize effectiveness can still help to boost the number of readers on the Web. Eye tracking and social media MPC Thesis Miller 2013 Page | 37 References Ahmad, A. (2010). Is Twitter a useful tool for journalists? Journal of Media Practice, 11(2), 145-155. doi:10.1386/jmpr.11.2.145_1. Althaus, S. L., Tewksbury, D. (2000). Patterns of Internet and Traditional News Media Use in a Networked Community', Political Communication, 17: 1, 21-45 Ball-Rokeach, S.J. (1985). The origins of individual media-system dependency: A sociological framework. Communication Research, 12(4), 485-510. Barry, A. (2002). Perception and Visual Communication Theory. Journal Of Visual Literacy, 22(1), 91-106. Baron, L. (1980). Interaction between television and child-related characteristics as demonstrated by eye movement research. Education, Communication and Technology: A Journal of Theory, Research and Development. 28(4), 267-281. Bucher, H., & Schumacher, P. (2006). The relevance of attention for selecting news content: An eye-tracking study on attention patterns in the reception of print and online media. Communications: The European Journal Of Communication Research, 31(3), 347-368. doi:10.1515/COMMUN.2006.022. Buttry, S. (2012). Advice and examples on how and what journalists should tweet. [Blog Post.] Retrived from http://stevebuttry.wordpress.com/2012/08/06/advice-and-examples-on-how-and-what-journalists-should-tweet/ Bojiko, A. (2006). Using eye tracking to compare web page designs: A case study. Journal of Usability Studies, 3(1), 112-120. Carter, A. Y. (2009). Global Perspective: Seeking a News Business Model for Newspapers. Journal Of New Communications Research, 4(2), 80-92. Chu, S., Paul, N., & Ruel, L. (2009). Using eye tracking technology to examine the effectiveness of design elements on news websites. Information Design Journal (IDJ), 17(1), 31-43. doi:10.1075/idj.17.1.04chu Conati, C., Merten, C., Amershi, S., & Muldner, K. (2007). Using eye-tracking data for high-level user modeling adaptive interfaces. Association for the Advancement of Artificial Intelligence. Conference Proceedings, 1-4. Cooke, L. (2005). Eye Tracking: How It Works and How It Relates to Usability. Technical Communication, 52(4), 456-463. Ebersman, David. (2012). Facebook Inc. Amendment No. 4 to Form S-1 United State Security and Exchange Commission. http://www.sec.gov/Archives/edgar/data/1326801/000119312512175673/d287954ds1a.htm Eye tracking and social media MPC Thesis Miller 2013 Page | 38 Emmett, A. (2008). Networking News. American Journalism Review, 30(6), 40-43. Retrieved from EBSCOhost. Farhi, P. (2009). The Twitter Explosion. American Journalism Review, 31(3), 26-31. Retrieved from EBSCOhost. Farhi, P. (2008). Don't Blame the Journalism. American Journalism Review, 30(5), 14-15. Retrieved from EBSCOhost. Fangfang, G., & Martin-Kratzer, R. (2011). New scheme of communication: an exploratory study of interactivity and multimedia use in Chinese j-blogs and the implications. Asian Journal Of Communication, 21(1), 69-83. doi:10.1080/01292986.2010.524232 Fisher, P. M., Richards, J. W., Berman E. F., & Krugman, D. M. (1989). Recall and eye tracking study of adolescents viewing tobacco ads. Journal of the American Medical Association. 261, 90-94. Franklin B. (2008). The Future of Newspapers. Journalism Studies [serial online]. October 2008;9(5):630-641. Available from: Academic Search Premier, Ipswich, MA. Accessed October 10, 2011. Friedman, Ann. (2012). The tweet life: How journalist can reap the benefits of Twitter without falling into its traps. Columbia Journalism Review. Retrived from http://www.cjr.org/realtalk/the_tweet_life.php?page=all. Hansen, K. A., Paul, N., DeFoster, R., & Moore, J. E. (2011). Newspaper Training Program Shows Gains in Social Media. Newspaper Research Journal, 32(3), 40-51. Harrington, S., & McNair, B. (2012). The ‘New’ News. Media International Australia (8/1/07-Current), (144), 49-51. Heikkilä, H., Kunelius, R., & Ahva, L. (2010). From Credibility to Relevance. Journalism Practice, 4(3), 274-284. doi:10.1080/17512781003640547 Hervet, G., Guérard, K., Tremblay, S., & Chtourou, M. (2011). Is banner blindness genuine? Eye tracking internet text advertising. Applied Cognitive Psychology, 25(5), 708-716. doi:10.1002/acp.1742 Holcomb, J., Gross, K., Mitchell, A. (2011). How Main Stream Media Outlets Use Twitter. Pew Research Study. Available from: www.journalism.org. Accessed on December 4, 2011. Josephson, S. (2008). Keeping Your Readers' Eyes on the Screen: An Eye-Tracking Study Comparing Sans Serif and Serif Typefaces. Visual Communication Quarterly, 15(1/2), 67-79. doi:10.1080/15551390801914595 Josephson, S.; Holmes, M. (2002). Visual attention to repeated Internet images: Testing scanpath theory on the World Wide Web, In Eye Tracking Research & Applications (ETRA) Symposium Just, M. A. & Carpenter, P. A. (1980). `A theory of reading: From eye fixations to comprehension', Psychological Review, 87(4), 329-354. Eye tracking and social media MPC Thesis Miller 2013 Page | 39 Katz, E., Blumler, J. G., & Gurevitch, M. (1974). Utilization of mass communication by the individual. In J. G. Blumler & E. Katz (Eds.), The uses of mass communications: Current perspectives on gratifications research. 19-32. Lariscy, R., Avery, E., Sweetser, K. D., & Howes, P. (2009). An examination of the role of online social media in journalists’ source mix. Public Relations Review, 35(3), 314-316. doi:10.1016/j.pubrev.2009.05.008. Lavrusik, V., Cameron, B. (2011). Study: How People are Engaging Journalists on Facebook & Best Practices. [Blog post.] Retrieved from http://www.facebook.com/notes/facebook-journalists/study-how-people-are-engaging-journalists-on-facebook-best-practices/24577514876784. Leckner, S. (2012). Presentation factors affecting reading behaviour in readers of newspaper media: an eye-tracking perspective. Visual Communication, 11(2), 163-184. doi:10.1177/1470357211434029. Luckie, M. (2012). Best Practices for Journalists. [Blog post]. Retrieved from http://blog.twitter.com/2012/09/best-practices-for-journalists.html. Maier, S. (2010). All the News Fit to Post? Comparing News Content on the Web to Newspapers, Television and Radio.Journalism & Mass Communication Quarterly, 87(3/4), 548-562. McCarthy, Anna. (2010). From TV to Twitter: How Ambient News Became Ambient Journalism. M/C Journal, 13(2), 7. Murthy, D. (2011). Twitter: Microphone for the masses? Media, Culture & Society, 33(5), 779-789. doi:10.1177/0163443711404744. O’Regan, J.K., & Levy-Schoen, A. (1983). Integrating visual information from successive fixations: Does trans-saccadic fusion exist? Vision Research, 23(8), 765-768. State of the Media: The Social Media Report. (2011), Nielsen Online. Accessed December 4, 2011. Pew Research. (2012). The State of the News Media 2012: An Annual Report on American Journalism. Retrieved from http://stateofthemedia.org/ Pinker, S. (1995). How the mind works. New York: W.W. Norton & Co. Posnansky, C.J. & Rayner, K. (1978). Visual vs. phonemic contributions to the importance of the initial letter in word identification. Bulletin of the Psychometric Society, 11, 188-190. Quan-Haase A., Young A. Uses and Gratifications of Social Media: A Comparison of Facebook and Instance Messaging. Bulletin of Science, Techonology & Society [serial online]/. October 1, 2010;30(5):350-361. Available from: ERIC, Ipswich, MA. Accessed April 11, 2013. Eye tracking and social media MPC Thesis Miller 2013 Page | 40 Raacke, J., & Bond-Raacke, J. (2008). MySpace and Facebook: Applying the Uses and Gratifications Theory to Exploring Friend-Networking Sites. Cyberpsychology & Behavior, 11(2), 169-174. Doi:10.1089/cpb.2007.0056 Rayner, K. (1993). Eye movements in reading: Recent developments. Current Directions in Psychological Science, 2, 81-85 Reinardy, S. (2010). Downsizing Effects on Personnel: The Case of Layoff Survivors in U.S. Newspapers. Journal Of Media Business Studies, 7(4), 1-19. Stassen, W. (2010). Your news in 140 characters: exploring the role of social media in journalism. Global Media Journal: African Edition, 4(1), 1-16. Squiers, L. B., Holden, D. J., Dolina, S. E., Kim, A. E., Bann, C. M., & Renaud, J. M. (2011). The Public's Response to the U.S. Preventive Services Task Force's 2009 Recommendations on Mammography Screening. American Journal of Preventive Medicine, 40(5), 497-504. doi:10.1016/j.amepre.2010.12.027: The people formerly known as the audience. (2011). Economist, 399(8741), 9-12. Teixeira, T., Wedel, M., & Pieters, R. (2012). Emotion-Induced Engagement in Internet Video Advertisements. Journal Of Marketing Research (JMR), 49(2), 144-159. doi:10.1509/jmr.10.0207 Tenore, M. (2011). 10 ways journalists can use Twitter before, during and after reporting a story. Poynter. Retrived from http://www.poynter.org/how-tos/digital-strategies/146345/10-ways-journalists-can-use-twitter-before-during-and-after-reporting-a-story. Tucker, P. (2009). Newspapers Face the Final Edition. Futurist, 43(5), 8-9. Retrieved from EBSCOhost. Twitter blog. (2012). Twitter Turns Six. [Blog Post]. Retrieved from http://blog.twitter.com/2012/03/twitter-turns-six.html Wasserman, T. (2012). Pinterest is Now the No. 3 Social Network in the U.S. [Blogpost.] Retrieved from http://mashable.com/2012/04/06/pinterest-number-3-social-network/ Eye tracking and social media MPC Thesis Miller 2013 Page | 41 Appendix 1: Mock newsfeeds created for the study Displayed for participants 1, 9, 17, 25 Eye tracking and social media MPC Thesis Miller 2013 Page | 42 Eye tracking and social media MPC Thesis Miller 2013 Page | 43 Eye tracking and social media MPC Thesis Miller 2013 Page | 44 Displayed for participants 2, 10, 18, 26, 35, 36: Eye tracking and social media MPC Thesis Miller 2013 Page | 45 Eye tracking and social media MPC Thesis Miller 2013 Page | 46 Displayed for participants 3, 11, 19, 27, 34: Eye tracking and social media MPC Thesis Miller 2013 Page | 47 Eye tracking and social media MPC Thesis Miller 2013 Page | 48 Displayed for participants 4, 12, 20, 28, 33: Eye tracking and social media MPC Thesis Miller 2013 Page | 49 Eye tracking and social media MPC Thesis Miller 2013 Page | 50 Displayed for participants 5, 13, 21, 29: Eye tracking and social media MPC Thesis Miller 2013 Page | 51 Eye tracking and social media MPC Thesis Miller 2013 Page | 52 Displayed for participants 6, 14, 22, 30: Eye tracking and social media MPC Thesis Miller 2013 Page | 53 Eye tracking and social media MPC Thesis Miller 2013 Page | 54 Displayed for participants 7, 15, 23, 31: Eye tracking and social media MPC Thesis Miller 2013 Page | 55 Eye tracking and social media MPC Thesis Miller 2013 Page | 56 Displayed for participants 8, 16, 24, 32 Eye tracking and social media MPC Thesis Miller 2013 Page | 57 Eye tracking and social media MPC Thesis Miller 2013 Page | 58 Appendix 2: Informed Consent INFORMED CONSENT FORM The purpose of this study is to look at how people view social media messages from a news outlet. The research is being conducted by Master’s of Professional Communication student Jessica Miller and Professor Sheree Josephson of Weber State University. RISKS: There are no risks in this study. PROCEDURES: The simulated Facebook and Twitter feeds you will observe were created specifically for this study. Most of the data will be collected using a standard eye-tracking apparatus. The apparatus will record the general area of the screen that the eye is focused on. DURATION: The eye-tracking portion of this experiment will take no longer than 30 minutes in one session to complete. The questionnaire will take about 10 minutes more. CONFIDENTIALITY: Your name is not important to us, and will not be recorded on any of the forms that are used to collect the data. You will be given an identification number that is unique to this study so that your responses can be recorded consistently for all measuring instruments. But this number will not be used to identify you in any way. Your eye-scan patterns while you are viewing the social media newsfeeds will be recorded, but these recordings will not be of your person (your face or any other identifying features). WITHDRAWAL: Participation in this study is voluntary and you have the right to withdraw from participation at any time without penalty. You may also refuse to answer any questions that you find invasive or objectionable. CONCERNS: If you have any concerns or questions about your participation in this study, please contact Professor Sheree Josephson at 801-626-6164, the Office of Sponsored Projects at 801-626-7537, or the Institutional Review Board chair, Dr. Teresa Kay, at Weber State University. CONSENT: I have read this form. I understand the material presented in this document and agree to participate in the study. Name: _________________________________________________________ Date: ________________ Eye tracking and social media MPC Thesis Miller 2013 Page | 59 Appendix 3: Verbal instructions For this study, we will show you a Facebook newsfeed and a Twitter newsfeed. Look at the pages as naturally as possible, as if you were looking at your own social media newsfeeds. You will not be quizzed on any material, so you don’t need to study the content. Look at it as if you were browsing your own newsfeeds on a normal day. You will not be able to scroll down the page or click on any of the links. Do you have any questions before we begin? Eye tracking and social media MPC Thesis Miller 2013 Page | 60 Appendix 4: Survey Gender: * Age: * How often do you use Facebook? How often do you use Twitter? How many news outlets or employees of news outlets do you follow on social media? How often do you use social media to inform yourself about news? What phrasing techniques would catch your attention on social media?(example: a headline, question, quote, etc.) If a Twitter post was phrased as a question, how effective would that be at catching your attention?Example: Do you think the blizzard will hit hard in NYC?  Very effective  Somewhat effective  Neither effective nor ineffective  Somewhat ineffective  Very ineffective If a Twitter post was phrased as a statement headline, how effective would that be at catching your attention?Example: Blizzard hits NYC  Very effective  Somewhat effective  Neither effective nor ineffective  Somewhat ineffective  Very ineffective Eye tracking and social media MPC Thesis Miller 2013 Page | 61 If a Facebook post was phrased as a statement headline, how effective would that be at catching your attention?Example: Blizzard hits NYC  Very effective  Somewhat effective  Neither effective nor ineffective  Somewhat effective  Very ineffective If a Facebook post was phrased as question, how effective would that be at catching your attention?Example: Do you think the blizzard will hit hard in NYC?  Very effective  Somewhat effective  Neither effective nor ineffective  Somewhat ineffective  Very ineffective Eye tracking and social media MPC Thesis Miller 2013 Page | 62 Appendix 5: Survey Responses Gender:Age:How often do you use Facebook?How often do you use Twitter?How many news outlets or employees of news outlets do you follow on social media?How often do you use social media to inform yourself about news?What phrasing techniques would catch your attention on social media?Male33DailyMultiple times a day10Multiple times a dayHeadlines, names of prominent individuals (sports stars of teams I like), quotes from specific individuals Female27hourly (multiple times per day)multiple times per day40+multiple times per dayI pay most attention to headlines that regard issues I am passionate about. Science, Social Issues, Class warfare. headlines that leave me guessing catch my interestmale31alotalotfiveoftenGlitter, sparkles, fireworks, or creative usageof recognized logos, similar to google.Female43Daily2 times a week20dailyBolded headings, urgent, exclusive, breaking news anything like thatFemale50dailyweekly1weeklyGood news! Guess what happened? Or an inspirational quoteMale22Once daily3 times dailyThreeThree times weekly.Indicative statements, keywords relating to interests, words relating to importance or significance.Male28DailyHourly15-OctHourlyBreaking news, headline, unusual quote, a caption informing me of an image or video (ex. PIC)Female21multiple times a dayat least every other day10 or moreweeklya photo or a catchy headline that makes me curiousMale46Dailymonthly2 or 3rarelyHeadlines more than anything but only if it comes from a source that I trust.Eye tracking and social media MPC Thesis Miller 2013 Page | 63 Gender:Age:How often do you use Facebook?How often do you use Twitter?How many news outlets or employees of news outlets do you follow on social media?How often do you use social media to inform yourself about news?What phrasing techniques would catch your attention on social media?Male35DailyDaily12DailyLogos, headlines, pictures.f55everydaynot that often2somewhat, just at that moment like storm warningsheadlinequotesMale26Every dayEvery dayOver 50Every dayHeadlines most often. All CAPS does, too, although annoying.Female19Not oftenVery ferquentAbout 5Very oftenSomething that I can relate to.female21once a dayfew times a weekthree newspapersmost of the timeI usually look at the headlines to see what the story is about, so they catch my attention. Female21At least once an hourAt least once an hourSo many I can't even countEvery dayA headline, question, or quote could all catch my attention. It depends on the type of social media. For example, questions catch my attention better on Facebook because I can see what others wrote in response and I'm more likely to respond and click through to the link. Male27Few days a weekLike 40,000 times a day10ishEvery dayDefinitely headlines. I absolutely hate quotes and questions.. headlines always get me to click.Female27DailyWeekly20Dailyquotes, headlines, Female52Every dayOnce in awhile3Multiple times a dayPhotograph, headlinemale22DailyRarely, once a month4EverydayHeadlines, Question, Opinion, "Hahaha", Female21Hourlymultiple times a day10hourlyHeadlines, pictures, articles, questionsEye tracking and social media MPC Thesis Miller 2013 Page | 64 If a Twitter post was phrased as a question, how effective would that be at catching your attention? Very effective Somewhat effective Neither effective nor ineffective Somewhat ineffective Very ineffective If a Twitter post was phrased as a statement headline, how effective would that be at catching your attention? Very effective Somewhat effective Neither effective nor ineffective Somewhat ineffective Very ineffective Eye tracking and social media MPC Thesis Miller 2013 Page | 65 If a Facebook post was phrased as question, how effective would that be at catching your attention? Very effective Somewhat effective Neither effective nor ineffective Somewhat ineffective Very ineffective If a Facebook post was phrased as a statement headline, how effective would that be at catching your attention? Very effective Somewhat effective Neither effective nor ineffective Somewhat effective Very ineffective Eye tracking and social media MPC Thesis Miller 2013 Page | 66 Appendix 6: Charted results Participant Headline 1 Question 1 Headline 2 Question 2 Total time on headlines Total time on questions Most time viewed 1 1,416 2,783 2,866 0 4,282 2,783 Headline 3 3,600 1,833 0 0 3,600 1,833 Headline 5 1,166 0 1,183 1,666 2,349 1,666 Headline 7 2,150 1,400 600 0 2,750 1,400 Headline 9 1,950 1,383 1,116 616 3,066 1,999 Headline 11 1,333 800 2,116 466 3,449 1,266 Headline 13 1,133 2,400 116 0 1,249 2,400 Question 15 3,450 3,916 116 0 3,566 3,916 Question 17 1,616 1,350 1,433 2,366 3,049 3,716 Question 21 933 700 3,850 1,483 4,783 2,183 Headline 23 3,400 2,033 0 0 3,400 2,033 Headline 25 1,850 1,200 1,616 1,216 3,466 2,416 Headline 27 1,183 266 2,433 1,866 3,616 2,132 Headline 29 1,950 950 1,500 900 3,450 1,850 Headline 31 166 1,050 1,150 2,583 1,316 3,633 Question 33 0 0 1,333 0 1,333 0 Headline Average 1,712 1,385 1,358 774 3,070 2,159 Time spent viewing on Twitter (in milliseconds): Group 1 Participant Question 1 Headline 1 Question 2 Headline 2 Total time on headlines Total time on questions Most time viewed 2 1,233 2,783 0 950 3,733 1,233 Headline 4 2,250 1,933 0 250 2,183 2,250 Question 6 1,783 1,233 533 0 1,233 2,316 Question 8 1,500 116 0 0 116 1,500 Question 14 2,333 2,183 2,066 0 2,183 4,399 Question 16 2,850 983 566 0 983 3,416 Question 18 1,600 1,633 1,483 0 1,633 3,083 Question 20 0 2,283 1,300 0 2,283 1,300 Headline 22 0 1,466 0 0 1,466 0 Headline 24 2,966 0 0 0 0 2,966 Question 26 5,200 1,716 0 0 1,716 5,200 Question 28 0 166 0 0 166 0 Headline 30 1,350 966 1,550 2,416 3,382 2,900 Headline 32 3,066 1,766 1,100 0 1,766 4,166 Question 34 2,083 750 216 1,066 1,816 2,299 Question 36 2,466 2,333 1,100 1,050 3,383 3,566 Question Average 1,918 1,394 620 358 1,753 2,537 Time spent viewing on Twitter (in milliseconds): Group 2 Participant Headline 1 Question 1 Headline 2 Question 2 Total times returned to headlines Total times returned to questions Returned to most often 1 1 1 1 1 Equal 3 1 1 0 Headline 5 1 1 1 1 Equal 7 1 1 0 Headline 9 1 4 1 2 4 Question 11 1 1 1 1 Equal 13 1 0 1 Question 15 3 0 3 Question 17 1 0 1 Question 21 1 1 0 2 Question 23 0 0 Neither 25 1 1 0 Headline 27 1 2 0 3 Question 29 1 1 1 1 2 Question 31 1 2 0 3 Question 33 0 0 Neither Average 1 1.555555556 1 1.333333333 0.5625 1.375 Number of times participants returned (Twitter): Group 1 Eye tracking and social media MPC Thesis Miller 2013 Page | 67 Participant Question 1 Headline 1 Question 2 Headline 2 Total times returned to headlines Total times returned to questions Returned to most often 2 1 1 1 1 Equal 4 5 3 3 5 Question 6 1 1 1 1 Equal 8 3 0 3 Question 14 1 0 1 Question 16 2 0 2 Question 18 1 2 0 3 Question 20 0 0 Neither 22 0 0 Neither 24 1 0 1 Question 26 6 3 3 6 Question 28 0 0 Neither 30 1 1 1 2 1 Headline 32 4 1 1 4 Question 34 2 0 2 Question 36 1 2 1 3 1 Headline Average 2.6 1.714285714 1.25 1 0.875 1.9375 Number of times participants returned (Twitter): Group 2 Participant Headline 1 Question 1 Headline 2 Question 2 Total headline fixations Total question fixations Viewed most 1 2 2 1 0 3 2 Headline 3 2 1 0 0 2 1 Headline 5 2 0 1 2 3 2 Headline 7 2 1 1 0 3 1 Headline 9 3 5 2 1 5 6 Question 11 1 2 2 1 3 3 Equal 13 1 2 1 0 2 2 Equal 15 1 4 1 0 2 4 Question 17 1 2 1 2 2 4 Question 21 1 2 1 2 2 4 Question 23 1 1 0 0 1 1 Equal 25 2 1 1 1 3 2 Headline 27 1 2 1 3 2 5 Question 29 2 3 1 1 3 4 Question 31 1 2 1 3 2 5 Question 33 0 0 1 0 1 0 Headline Average 1.4375 1.875 1 1 2.4375 2.875 Number of fixations of Twitter: Group 1 Participant Question 1 Headline 1 Question 2 Headline 2 Total headline fixations Total question fixations Viewed most 2 2 3 0 1 4 2 Headline 4 6 5 0 1 6 6 Equal 6 1 2 2 0 2 3 Question 8 4 1 0 0 1 4 Question 14 2 1 2 0 1 4 Question 16 3 1 1 0 1 4 Question 18 2 2 3 0 2 5 Question 20 0 1 1 0 1 1 Equal 22 0 1 0 0 1 0 Headline 24 2 0 0 0 0 2 Question 26 7 4 0 0 4 7 Question 28 0 1 0 0 1 0 Headline 30 1 2 2 2 4 3 Headline 32 5 2 1 0 2 6 Question 34 3 1 1 1 2 4 Headline 36 2 3 1 2 5 3 Headline Average 2.50 1.88 0.88 0.44 2.31 3.38 Number of fixations of Twitter: Group 2 Eye tracking and social media MPC Thesis Miller 2013 Page | 68 Participant Question 1 Headline 1 Question 2 Headline 2 Total time on questions Total time on headlines Most time viewed 1 2 0 2 Headline 3 2 2 0 Question 5 2 1 2 1 Question 7 None 9 2 1 2 1 4 Headline 11 1 1 Question 13 1 1 0 Question 15 3 3 3 3 Equal 17 None 21 3 0 3 Headline 23 None 25 1 1 1 1 2 2 Equal 27 1 2 1 2 Headline 29 2 1 2 1 Question 31 1 1 0 Question 33 0 0 None Average 1.50 2.00 1.25 1.33 1.21 1.38 Number of times participants returned (Facebook): Group 1 Participant Headline 1 Question 1 Headline 2 Question 2 Total time on questions Total time on headlines Most time viewed 2 1 1 1 1 Equal 4 1 1 0 Question 6 1 4 4 4 5 Headline 8 2 3 0 5 Headline 14 1 1 1 1 Equal 16 2 0 2 Headline 18 0 0 None 20 1 1 0 Question 22 1 1 0 Question 24 3 1 1 3 Headline 26 1 1 0 Question 28 2 1 1 2 Headline 30 2 2 0 Question 32 2 1 2 1 Question 34 2 0 2 Headline 36 2 1 1 Headline Average 1.75 1.45 2.50 1.00 1.06 1.47 Number of times participants returned (Facebook): Group 2 Participant Question 1 Headline 1 Question 2 Headline 2 Total time on headlines Total time on questions Most time viewed 1 216 2,416 383 0 2,416 599 Headline 3 5,033 1,766 0 0 1,766 5,033 Question 5 2,416 1,416 0 0 1,416 2,416 Question 7 1,200 1,350 0 1,333 2,683 1,200 Headline 9 1,466 1,133 950 933 2,066 2,416 Question 11 2,766 883 1,100 1,566 2,449 3,866 Question 13 1,916 0 216 0 0 2,132 Question 15 2,700 333 2,466 0 333 5,166 Question 17 1,866 2,416 1,633 2,683 5,099 3,499 Headline 21 950 2,483 2,633 0 2,483 3,583 Question 23 3,900 0 700 0 0 4,600 Question 25 2,083 1,650 1,583 1,750 3,400 3,666 Question 27 3,150 1,216 116 533 1,749 3,266 Question 29 1,083 1,266 1,783 2,016 3,282 2,866 Headline 31 1,500 2,400 2,350 2,016 4,416 3,850 Headline 33 0 300 1,266 0 300 1,266 Question Average 1,980 1,326 1,115 755 2,081 3,095 Time spent viewing on Facebook (in milliseconds): Group 1 Eye tracking and social media MPC Thesis Miller 2013 Page | 69 Participant Headline 1 Question 1 Headline 2 Question 2 Total time on headlines Total time on questions Most time viewed 2 2,350 3,250 316 1,316 2,666 4,566 Question 4 1,366 2,850 0 0 1,366 2,850 Question 6 1,150 1,816 2,966 216 4,116 2,032 Headline 8 1,766 1,450 3,050 0 4,816 1,450 Headline 14 4,166 1,200 1,366 1,466 5,532 2,666 Headline 16 3,400 3,133 0 0 3,400 3,133 Headline 18 1,833 1,816 1,666 1,083 3,499 2,899 Headline 20 0 3,033 4,650 0 4,650 3,033 Headline 22 1,783 1,200 1,700 0 3,483 1,200 Headline 24 4,700 3,600 0 0 4,700 3,600 Headline 26 0 3,550 1,200 0 1,200 3,550 Question 28 2,583 1,166 1,850 1,716 4,433 2,882 Headline 30 116 2,583 1,100 0 1,216 2,583 Question 32 150 1,200 2,333 1,133 2,483 2,333 Headline 34 1,133 2,683 2,033 1,800 3,166 4,483 Question 36 2,850 1,333 1,083 866 3,933 2,199 Headline Average 1,766 2,302 1,615 582 3,382 2,884 Time spent viewing on Facebook (in milliseconds): Group 2 Participant Headline 1 Question 1 Headline 2 Question 2 Total headline fixations Total question fixations Viewed most 2 3 3 1 1 4 4 Equal 4 1 2 0 0 1 2 Question 6 2 5 5 1 7 6 Headline 8 3 1 4 0 7 1 Headline 14 3 2 1 1 4 3 Headline 16 3 1 0 0 3 1 Headline 18 1 1 1 1 2 2 Equal 20 0 2 1 0 1 2 Question 22 1 2 1 0 2 2 Equal 24 4 2 0 0 4 2 Headline 26 0 2 1 0 1 2 Question 28 3 1 1 2 4 3 Headline 30 1 3 1 0 2 3 Question 32 1 3 2 1 3 4 Question 34 1 1 3 1 4 2 Headline 36 3 2 1 1 4 3 Headline Average 1.80 2.07 1.47 0.53 3.27 2.60 Number of fixations of Facebook: Group 1 Participant Question 1 Headline 1 Question 2 Headline 2 Total headline fixations Total question fixations Viewed most 1 1 3 1 0 3 2 Headline 3 4 1 0 0 1 4 Question 5 3 2 0 0 2 3 Question 7 1 1 2 1 2 3 Question 9 3 2 2 3 5 5 Equal 11 2 1 1 1 2 3 Question 13 2 0 1 0 0 3 Question 15 4 4 1 0 4 5 Question 17 2 1 1 1 2 3 Question 21 1 4 1 0 4 2 Headline 23 2 0 1 0 0 3 Question 25 2 2 2 2 4 4 Equal 27 2 3 1 1 4 3 Headline 29 1 1 3 2 3 4 Question 31 1 1 2 1 2 3 Question 33 0 1 1 0 1 1 Equal Average 1.89 1.61 1.17 0.67 2.28 3.06 Number of fixations of Facebook: Group 2