The Facebook Project

Background And Description

Arguably one of the two most influential SNS websites on the internet, Facebook.com is a comprehensive and encompassing clustering of networks based on universities and colleges, high schools, work places, and geographic areas.  These membership networks are independent of one another but based on the same interface and systems of interaction. Intersections and overlaps between each network are possible but they often have crucial and intentional barriers to access in between.  Started originally in February of 2004, Facebook hit its first tipping point in the late summer of that year with the introduction of groups and public posting ‘walls.’  A second surge in growth resulted from Facebook’s introduction to the global public – the site went from consistently hovering around 14 million unique visitors per month to over 26 million (Lipsman 2007c).  In the span of a little over 3 years - from 2005 to 2008—the user count has grown over 10 times in size.¹  As of June 2007 collectively Facebook claims over 67 million members (users who have returned to the site in the past 30 days) and remains one of the fastest growing websites on the internet (Wakabayashi 2007, Lipsman 2007b, Abram 2007, Facebook Statistics 2008).  Sources vary, but membership saturation ranges between an average of 85% and 95% (Golder et al. 2006, Arrington 2005, Ellison et al. 2005, Jones and Soltren 2005, Facebook Statistics 2008); the last network-based count for the UIllinois Network placed a 92% membership rate among the undergraduate population.²  Responses from the survey featured in this paper estimate something closer to a 97% coverage though this number may be inflated due to the possibility that Facebook users may be more likely to respond to email surveys as they are quite possibly more intensive users of the internet in general.  Regardless, the sheer number of users and level of penetration makes the site a pretty big deal.

Facebook ranks as one of the most visited websites on the internet, with sources claiming as high as the 3rd most visited based on page views, and they now account for at least 1% of all time spent on the internet (Freiert 2007b, Abrams 2007, Alexa.com 2008).  Among college students the website is an even more common stop than Google and outpaces MySpace by a significant margin (Anderson Analytics 2007).  More than 60% of members log in daily and many sign on multiple times a day while the average visitor spends over three hours of time on the site each month (Holahan, Hof, and Ante 2007, Arrington 2005).  The most common activities (based on time spent) overall are in descending order: browsing profiles, interacting with applications, browsing pictures, joining or visiting groups, searching for members and groups, and joining and browsing networks (Freiert 2007a). In 2007 most users were between the ages of 12 and 24, however nearly an equal number amass in the age demographic of 35 and up (Lipsman 2007a, 2007b).  The website in its entirety boasts more than 16 million page views and well beyond 600 million searches per month (Lipsman 2007a, Abram 2007).  The most recent count of average daily visitors is nearly 15 million, with the vast majority (85%) connecting from the US and Europe (Lipsman 2007b).  UIllinois is by comparison to other networks is relatively large, ranking in at nearly 60,000 profiles³.  Facebook is the most viewed website by both females (69%) and males (56%) ages 17-25 in the United States, even surpassing MySpace.com (eMarketer Survey 2007).  Facebook has essentially hit full saturation amongst most colleges and commands a high usage rate in many western nations.  It continues to expand internationally and diversify its audience by adding more and more country-based networks.

Facebook has a definitively different feel than most websites, even ones that might be considered similar like dating websites or professional job seeking networks.  The entire system is organized around exploring and engaging other participants.  The idea is to help you access and display as much information about yourself and others as you want and at the same time pursue connections between the heterogeneous mess of people, components, and ideas.  Facebook is a source of information, place of communication and a multifaceted arena of performance.  It is fundamentally a socio-technical mesh; a blending between human actors, echoes of abstract individual personalities and social perspectives, and code-powered, semi-automated visual interfaces.  Access is mediated by both cell phones and computers of all types.

Upon logging on visitors are greeted with the impression of activity by looking at the newsfeed, their latest application updates and the published shifting statuses of their friends.  On some level it’s almost comparable to the buzz of a city (Stutzman 2007b) or the front page of a newspaper. One can see some of what’s going on amongst their friends on the site and do things like track specific recent changes in their friend’s profiles or upload new media to share with their classmates.  A student might chase down classmates or find that person they ran into on the quad earlier, or seek the social hub of a campus group they’ve just joined. Often all it takes is an impartial set of information – a first name and a major, a year and a club membership, an email address or AIM handle – to find specific people in the system.  In most cases this sharing of media, identities and knowledge is desirable.  Students can keep in touch with family members and distant friends, see pictures of someone they wish to take out on a date, or download the latest song their buddy many miles away composed in his bedroom.  The process is much easier than it would be to normally accomplish such tasks without the help of Facebook largely because just about every venture in social exchange is a function available through the system.  Facebook is notably conducive to one-sided activity and browsing, or ‘stalking’ as most users refer to it.⁴  Surely most of the aforementioned tasks have run rampant throughout programs and websites for years prior, but with such high logon rates, deep integration, and the ability for anyone to conduct them unbeknownst to others as well as in an overt fashion, one can safely say the intensity has changed.

Over the summer of 2007 Facebook spilled into the global scene, and expanded its user base to include many types of people beyond students (Lispman 2007c).  Within the US various adult populations began to employ the network for post-college social grooming, such as searching out old classmates and as a sort of dating service, and the Web 2.0 and business community has begun to adopt it as a new job search social networking tool⁵.  After all, employers usually check up on potential employees online, why not overtly search on Facebook too?  And the group that’s probably roused the most ruckus is the substantial number of older adults, such as parents and administrators, concerned with watching youthful users.

Studies have begun to surface showing just how important Facebook can be in the production of social capital.  For instance, Ellison, Steinfield, and Lampe (2006) have explored the influence of the SNS in terms of both bridging and bonding capital.⁶  Results from a multiple regression analysis of a survey disseminated to the MSU undergraduate population indicate that Facebook has a significant impact on students’ ability to maintain bridging social capital at college.  General internet use, as compared, did not make a significant difference in determining social capital.  The social capital scores of students who reported low satisfaction with MSU life and low self-esteem were most positively impacted by intensity  of Facebook use.  Interestingly enough white students were more likely to benefit in this way than non-white students, which when held in consideration with Mayer and Puller’s (2007) finding that social networks did not show a great deal of connections between racial groups, could suggest a new disparity for digital divide research.  Eszter Hargittai (2007) echoed this possibility in her work surveying student perceptions of SNS in finding that certain racial populations preferred certain networks more than others.  Ultimately having more friends who use Facebook, using Facebook to connect with offline contacts, and using Facebook for fun accurately predicted rates and trends of bridging social capital, but not bonding social capital (Ellison et al. 2006).  In short, students who use Facebook actively may have an advantage in regards to social capital, or more properly framed, the very few not on Facebook will be at a relative disadvantage.

As the paper notes in its general discussion, the relationship between Facebook and social capital does not determine causation – students bring with them a plethora of connections and resources to the SNS environment and consequently take away new ones.  The point is that the two worlds are interconnected and coproducing of one another – invariably linked and dependent reproducing both weak ties, potential and realized, as well as strengthening social bonds.

Most publications, be they news media or academic articles, fail to accurately capture the essence of these exchanges, nor do they typically bring attention to the positive uses of SNS, just the nightly news doesn’t talk about all of the good things happening on the same night of a crime.  Yes, as we so often read in the newspaper or in privacy evaluation reports there are mishaps and negative interactions that occur as a result of Facebook.  Future qualitative studies will hopefully verify⁷ that much of the reason students aren’t afraid of the possible consequences is because they are outliers and students feel they have a good measure of control over the system and their presented identity.  What’s more forthcoming in their usage and interests is Facebook’s enhancement of strong and weak ties and virtually limitless set of opportunities for digital social engagement and entertainment.
Over time Facebook has experienced a complicated evolution of privacy controls and options outside of the initial separation of networks and original educational email id requirement⁸.  Participants were initially only able to adjust what types of people (such as friends or faculty) could see their profile but these options were later expanded to include direct control over what areas were released to others, even down to a specific individual basis.  The introduction of the ‘Newsfeed’ wrought a near-catastrophic response when user habits were published overtly for the first time to the general populace (boyd 2006).  Users could logon and see exactly who their friend broke up with the night before and the pictures another friend put up five minutes ago.  The community rebelled against this change levying all sorts of complaints and threats – many of them even disabled their accounts in response.  After a bit of a “calm down” (almost condescending) response by Mark Zuckerberg (2006), the solution came in the addition of more privacy options.  Users were given the ability to adjust who saw information about them on the newsfeed and of what type. They could customize their newsfeed to tell them about the Facebook activities they cared about most about the people they cared about most. Despite Facebook’s recent focus on addressing privacy concerns and fears in regards to the non-adult user population (Kelly 2007), which was likely in response to numerous complaints accrued over the last year and even a subpoena for information from the State Office of the Attorney in New York (Giannone et. al 2007), a new controversy exploded on the scene right around the time of this paper’s writing. 

Beacon,⁹ a name that will forever live in infamy in the minds of Facebook privacy advocates, was a service meant to be invisible to general users that would keep participants informed of their friend’s browsing activities on the general internet (but only for sites that were participating).  The main issue with Beacon’s deployment was that it was default opt-in, and allowed no ability for users to opt-out globally.  This effectively meant marketers, friends, stalkers, and really anyone in between could very plainly see a given user’s activities on websites such as Amazon.com.  After a slow start MoveOn.org and numerous efforts by groups around the web including bloggers and petitions managed to capture Facebook’s attention and provoke an official apology from Mark Zuckerberg (2007).  The difference this time, however, is that the effort wasn’t just on account of Facebook users – but to a larger media and web community collective (Stutzman 2007a).  As Facebook has grown up into a large company they have become increasingly accountable for their actions.

I’d like, for a moment, to take a step out of the third person into the first and reconcile my perspective in regards to Facebook research.  The purpose of my research is a logical result of a conglomeration of agendas and interests.  My role as a natural participant in the system nearly from its beginning bestows me with the benefits typical of insider ethnography.  I’ve been immersed in both face to face undergraduate life at UIUC as well as the Facebook side of it.  Without even thinking about it I wield an array of understandings of indigenous meanings and can aptly enact as an ethnographic researcher in an online world.  As a social science researcher I have a definitive and distinct interest in authentically describing and understanding the social systems of Facebook, including their benefits and drawbacks, egalitarian aspects and disparities, and other extensions of the face to face world. With my research I hope to inform a sometimes undereducated and misled populace and ultimately aid the use of Facebook for positive social change.
Though this paper is not about my unending ethnographic analysis of Facebook, I employ a great deal of knowledge informed by it in a general sense.  Acknowledging bias is the hallmark of the modern day ethnographer, and to be sure I’ve got plenty of strong opinions.  I myself am a very connected individual, subscribing to Malcom Gladwell’s idea of connectors and social epidemics.¹⁰  I’m an assertive user of Facebook and do my best to familiarize myself with all of its nuances and intricacies, dealing with everything from interface to user perceptions to types and methods of use.  As a result of my personality and undergrad years at UIUC I boast over 820¹¹ UIUC network friends at the time of this writing and can tell you where I know every single one of them from and the context of our meeting or relationship.  I check the site almost as much as I do my email - which is kept up in Outlook 2007 on one of my displays 24/7 – and that effectively means I’m visiting at least a dozen of times a day.  I act as an administrator for multiple groups corresponding to student organizations, message my students about assignments, post videos about causes to fight cancer, drop my sister funny wall posts, scope out potential love interests, and have even been feeding this stupid Fluff Friend application lately.  Suffice to say Facebook is an important facet of my life. 

I know many readers are squirming with a foul impression of me—the behaviors I listed aren’t supposed to be those of an upstanding studious graduate student who spends his Friday nights reading tomes of Karl Marx and Barry Wellman¹².  How could a PhD student in one of the highest-ranked Library and Information Science schools in the US be playing around on Facebook?  And that’s’ just it – at age 23 I can effectively bridge the world of formal academic research and cultivation of professionalism with the social transformation and identity formation catharsis most undergraduates fall through during college.  I’m in touch and I’m motivated.

So where does this leave us in terms of my bias?  First and foremost, I have an obvious focus on Facebook and not other systems of SNS.  I’ve found more than enough to study on Facebook alone and comparatives between the different evolving systems of SNS are something I hope to delve into later in my studies.  My world of research is also limited to just the scope of UIUC.  I haven’t traveled the country talking to individuals from all over like danah boyd, nor do I have massive amounts of data on all of Facebook’s multiple networks like the HP Information Dynamics Lab did in 2005.  Consequently, my examples only reflect Facebook and student life at UIUC.

From a sociological perspective I’m about as advantaged as they come.  I’m a white, heterosexual, able-bodied male from an upper-middle class background.  I’m an extraordinarily assertive extrovert and aspire to embody the antithesis of apathy.  I have few, if any, irrational fears and generally have little to hide from the world about myself or my daily activities.  I hail from a westernized, feminist American viewpoint and prefer a multidisciplinary (or sometimes anti-disciplinary) approach to social science.  My resultant perspective on feminism, gender inequality, and the importance of digital discourse is at least somewhat influenced by this bias. I don’t feel, however, that this invalidates or compromises my lived experience as a researcher or taints the questions of sociological analysis I ask.  If anything it provides a great background to contrast them with. 
At various times I will slip in and out of the first and third person in this work.  I do this on account of my training in traditional writing style, where first-person was outright disallowed.  Most of the section on theory is thus written in the third person, whereas my ethnographic analysis and observation will be relayed more often in the first person vernacular.  With this out of the way, I’d like to move on to introduce my site of research.

The study featured in this paper looks into the happenings within a specific zone on Facebook, known as a group.  Groups are essentially web pages that can be created by any user and are used for a variety of purposes.  Some groups might be silly congregations, such as those who really liked the videogame Oregon Trail,¹³ others might be jokes, like “Eating Babies for Fun and Profit,”¹⁴ and others might correspond to offline organizations, like the College Republicans.¹⁵  Groups can be bound to a specific network on Facebook (schools, geographic locations, workplaces) or span all of them as a global entity.  Generally the interface encourages them to be searchable and categorized and group creators are able to invite people to become members.  When someone becomes a member they are listed as so in the group members section (with a name and picture) and an entry for the group is added to the group memberships section of their profile.

People use and participate in groups in a variety of ways.  Some members are passive, and might just read posts, whereas others may never visit the group again (after joining) and instead consider it to be a quick identifier to convey information about themselves to others viewing their profile (think Goffman (Goffman, 1959) impression management—groups are almost equipment or props in the image of a profile).  A few members are very active and post media in the form of pictures, links, and videos.  Many members post shout-out statements on the wall (a publicly and easily viewable comment zone) and others discuss topics in the forum area.  Activity really varies by group and while the vast majority have few members and only a little activity¹⁶ the lively ones can still be host to a great deal of performance of identity and roles.  The average group size on Facebook is yet undetermined, though groups with over a million members are not uncommon, such as fans for Stephen Colbert,¹⁷ and those numbering in the hundred of thousands seem moderately common. 

Facebook group web pages feature the same sort of interconnectivity that is found everywhere else on Facebook.  Many components are linked and menus are limited to clean easy to read typefaces and separators.  They have a main central column and a side column.  Upon visiting a group an observer will quickly notice its title at the top of the central column followed by the main information below, including name, type (such as common interest, used for categorical searches), description and other contact information.  Below this resides recent news in short text format, then photos, videos, and posted items all with potential thumbnail previews, a compressed view of the discussion board with a preview of three topics and post data and then the wall, a sort of simple guest-book like form that users can fill out to leave their remarks publicly on the page.  Sitting neatly between the wall and discussion board section is a member listing area, with 6 linked thumbnail previews of random¹⁸ members and the total members listing, which is linked to a search return for all members in the group.  Each wall post contains the poster’s name, time and date information, response options, and linked thumbnail picture preview, giving a robust impression as users glance about the page.  Really just about everything is linked and tied to the face to face world with pictures.  The right column has perhaps the most noticeable element of the group profile, its picture, which is generally pretty limited in size.  Right beneath this are navigation and action options, such as the ability to view the discussion board, join the group, or if you are an administrator recruit or manage members and edit the group.  Officers are listed below this, with linked names and subtitles pertaining to their position in the group.  Related groups are found beneath this, with a link listing and category subtitle.  Finally at the bottom comes the official group-type information and administrator(s).  The layout is both organized and friendly, and adheres to sound principles of graphic design, information retrieval and display, and human-computer interface (HCI).  The group architecture is both dedicated to linking people together, but preserves the normal separation of profiles and privacy seen on the rest of Facebook.

So I’m not exactly sure when the idea for this project came up, but it had been on my mind for some time prior to the actual ethnography.  I first noticed the research site when surfing about the profile belonging to a person I had just added as a friend on Facebook.  At the time I had this person to be someone fairly progressive-minded… until I ventured down to take a gander at their group memberships.  She had a rather strikingly strong one listed, “There are just some things guys should do for girls. Period.”  I thought to myself that perhaps she was just a little bit misguided, being a freshman and all, and that perhaps the group was satirical in nature.  I clicked on it to investigate and was rushed with the image of countless almost 10-commandments type rules distinguishing how men should act in regards to women.  I shook my head in dismay, thinking that it’s strange how such sexist people could still exist among youth in this day in age.  Then, as I routinely scrolled down and scanned the group I happened upon a number that made my jaw drop.  One hundred and fifty thousand some members.  Not a few hundred, like I was expecting, not even a hefty tens of thousands like many of the big groups on Facebook boast.  No, this concentrated little digital atrocity played home to the membership of one hundred-fifty thousand.  My shock was punctuated by initial anger and then mixed in with bewilderment—how could this many people possibly believe in something like this?  I may be a feminist, but I thought this sort of thing was by and large on the way out…

After exploring some more I discovered that the sibling group, “There are just some things girls should do for guys. Period.” was created in response and had only a mere thirty thousand some participants.¹⁹  Immediately I could see that there was a disproportionate set of believers in this sort of sexist extravaganza.  I tossed them my sister’s way, who also had a similar reaction and we jointly decided that it would be the sort of thing worth a real set of critical eyes… should we ever find the time to do it.

Enter this semester, whereupon when offered an opportunity to pry into a new world with ethnography this group came to mind in a powerful way.  I signed on, joined the group, deleted the trail from my newsfeed (in a bout of impression management – we wouldn’t want my friends thinking I was suddenly gone crazy), and put my sociology hat on.  The description of the site reveals a number of interesting aspects of mediation.

The most immediately striking point of the page for me is the Facebook group name: There Are Some Things Guys Should Always Do For Girls. Period.  The title is in bold at the top of the page.  The word ‘period’ falls on a completely separate line.  Aligned on the right side is the category for the group – it is global and now boasts over 204,000 members, 177 of them being new.  It’s a rather forward and forceful name –it’s clear that the group creator (and presumably group members) believes strongly in the statement. Separating ‘period’ likely denotes importance.  The fact that this is the first item my attention goes to indicates my level of immersion as a user of the site – the interface and navigation elements have become almost transparent to me. 
The page, like other groups on Facebook, has a sort of boxed layout, with areas of interest separated and organized.  The top banner is dark blue with the ‘Facebook’ logo anchored at the bottom on the left, an area where is actually clips behind the main page frame. The position gives the appearance of layers.

Links having to do with interactions that traverse the network are found up on top.  These include profile, edit (for your profile), friends, networks, and inbox.  The Facebook logo is linked and brings you to your home page.  The upper right includes links to home, account, privacy, and one to log out.  The Friends link leads to a drop down containing the options: status updates, online now, recently updated, recently added, all friends, invite friends, and find friends.  The networks link has a drop down leading to UIllinois (the UIUC Facebook network), browse all networks, and join a network.  Finally, the inbox drop down includes the options: message inbox, sent messages, notifications, updates, and compose message.
A random ad is positioned below the left-side navigation.  Since the ads change every time you visit the page it didn’t seem useful to describe it in detail.  The secondary links on the left are set to a light grey background.  There is a search box with a drop down arrow next to it leading to more search options: basic search, find classmates, find coworkers, profile search, and browse.  Next to the search box is the ‘go button’ which looks like a little magnifying glass, set to a light blue background.

Every group, event or profile has the capability of having a picture.  This group has a picture of a couple, presumably (what looks to be) a male and female, sitting in the snow.  They are positioned in the lower right of the picture frame.  The male squats forward facing her as she sits back looking at him.  The male is dressed in a dark blue coat and has long black hair.  Zooming in reveals he is probably White or Asian, and looks to be pretty young.  The female is in a turquoise jacket, with blue pants (or perhaps boots) with a white border around the hood that is likely soft fur-like material.  She seems to have short hair and could easily be White or Asian, or perhaps Latina, it’s quite hard to tell from the small picture.  The background of the picture involves a long stretch of field covered in snow with a wall traveling into the background on the right side.  Snowflakes are falling in the picture.  A line of trees decorates the top of the wall.  The male appears o be holding the female’s hands in her lap.

Beneath the picture are some more navigation options.  These change depending upon your group membership. Since I have joined the group as an ethnographer my options are as follows: view discussion board, invite people to join, and leave group.  There is also an option button beneath these called ‘Share’ which allows you to post a preview of the group to your profile or send in a message to a friend or friends.
Below the group operations navigation are officers, which in this case includes two links to profiles whose names are not to be revealed here.  The group type is below this, which states “This is an open group. Anyone can join and invite others to join.”  Administrators are listed beneath this, which are the same as the officers in this case.

There’s also a footer section.  The footer is at the very bottom of the web page and is a sort of norm among websites.  This page has a single link, ‘report group’ that is part of the self-policing setup of Facebook.  Users can click on it to message tech support with issues.  Beneath this are some site navigation items: the Facebook copyright stamp, and links for advertisers, businesses, developers, about facebook, terms, privacy, and help.
The massive main frame extends deeply into the page.  It includes an information section, which lists group info including: name (stated before) and the group type, which in this case is ‘common interest – friends,’ which is also linked to other groups of this same category.  Underneath this is the description.  Posted items follows this up, which at the time of observation was empty.

Then there is the discussion board.  It displays the 3 most recent topics out of several thousand (3386 total at the time of the observation) discussion topics.  Additional information is given below each topic header, indicating the number and age of responses.  Members can click on the ‘start new topic’ link if they so choose.  The Members section comes next indicating the group membership total.  6 random thumbnail pictures of member profiles are shown beneath, with a link option to see all.

Finally there is the wall section, which lists the 10 most recent posts out of a total of 25,175 wall posts (at the time of observation).  Visitors are able to choose between writing something on the wall or seeing all of the posts.  The most recent post at the time was written while I was doing my observation.

Of greatest importance to this analysis is the ‘Description’ section of the page.  It contains a list of 40 items in sequence numbered from 0 to 39, each stating rules or suggestions that guys should abide by when dealing with girls.  The full list is available in the appendix.

The analysis of both this list and the values embedded in the interface will follow in the interpretation and analysis section.

---