scavenger hunt that helps build professional networks. This web and text messaging game encourages academic conference attendees to create an online ...
2011 IEEE International Conference on Privacy, Security, Risk, and Trust, and IEEE International Conference on Social Computing
Snag’em: Creating Community Connections Through Games Evie Powell, Felesia Stukes, Tiffany Barnes, Heather Richter Lipford University of North Carolina at Charlotte College of Computing and Informatics Charlotte, NC, USA {Empowell, F.Stukes, Tiffany.Barnes, Heather.Lipford }@uncc.edu cards that met certain criteria. The game moderator recorded game interactions in a local database at the game registration site, and posted game statistics on a whiteboard every hour. The game encouraged many new conference goers to interact with each other to hone their networking abilities. Unfortunately, this system was low on technology, with players waiting in line to get their points manually recorded.
Abstract— It is difficult for new community members to make the connections that would be most beneficial to them - connections with seasoned members of the community, or members of other groups. To address this problem, we have created Snag’em, a web-based social networking game that helps people create, monitor, and strengthen connections with one another. In contrast to existing social games, Snag’em facilitates offline interaction to create an online social network. This paper discusses the design of Snag’em, the use of game mechanics to engage academic community members in social interaction, and the results of two preliminary studies along with the design implications of each.
Conference attendees need pervasive and customizable ways to capture networking data, to maximize the networking and community effects over a short time. The 2009 Grace Hopper Celebration of Women in Computing conference used an RFID system and website called CONNECT to track contacts for conference attendees. Attendees first register on a web site and indicate their networking goals for the conference, and then when they meet conference-goers who meet their goals, both people scan their RFID badges with a scanner. Later users can download virtual business cards but no context other than the time of “connecting” is saved.
Keywords-component; Community; Games; Social Networking; Design; Pervasive
1
INTRODUCTION
Snag’em is an online social networking game we have built to promote interactions in academic communities [9] since conferences do not always create an equally beneficial community for all members. Certain individual factors such as level of extroversion, position within the community, and even native language can cause social inequality, influencing the growth of the community [4]. Although several computing applications have been built to address this problem [4][5][6][10], little work has been done to systematically evaluate these techniques. Introducing oneself to a stranger as part of a game, rather than for personal or selfish reasons, can make it easier to do things outside of the norm for the sake of game play [7]. This change in perception can increase confidence and make it easier to approach others.
Joseph McCarthy’s “Ticket2Talk” (T2T) application was designed to encourage users to initiate conversations based on similar networking interests in a relaxed setting. T2T uses tags as in Snag’em, where users tag themselves with personal interests, and are later encouraged to interact with other nearby players with similar interests. T2Ts displays, situated in a food or beverage areas, show relevant data about other conference goers located within the area [6][8], helping people leverage time they already spend in these locations to meet others. 3
Social networking sites have become very popular and games have been linked to considerable increases in social network traffic, giving users a reason to return to the network site [7]. However, we know of no games that are the primary mechanism for growing a social network. Snag’em is a human scavenger hunt that helps build professional networks. This web and text messaging game encourages academic conference attendees to create an online profile and forge connections with other attendees. Snag’em keeps track of game connections, or “snags,” made within the community over the duration of the conference. We believe that iteratively changing Snag’em’s game mechanics based on player interactions make Snag’em as a flexible, adaptable way to promote desirable communities. 2
In academia, sense of community has a strong positive correlation with retention [16]. Research indicates that students who do not feel as if they are part of a larger academic community are less likely to participate in extracurricular activities and organizations. This leads to lower retention rates, especially amongst minority students who suffer without a strong student support group [16]. A feeling of community can be nurtured with small group activities that augment the individual's role within a setting and helps students to foster connections [17].
RELATED WORK
Game Lab’s “Destroy all Developers” is a business card trading game created for the 2007 Game Developers Conference, where players earned points by finding business This work was partially supported by NSF grant CNS-0851745, REU Site: Socially Relevant Computing Research - Visualization, Virtual Environments, Gaming, and Networking, CNS-0739216 BPC-AE: The STARS Alliance: A Southeastern Partnership for Broadening Participation in Computing (20082011), IIS-0757521, Game2Learn: Creative Computing Education, and the Computing Research Association Distributed and Collaborative Research Experiences for Undergraduates
978-0-7695-4578-3/11 $26.00 © 2011 IEEE DOI
SNAG’EM: CREATING A SOCIAL NETWORK
Snag’em was initially created to provide networking opportunities for college students who attend the annual STARS conference; an NSF funded computing diversity and retention initiative. While the conference effectively promoted a sense of community [1], attendees needed a way to make stronger, longer lasting connections with other STARS members.
591
Snag’em was designed as a pervasive game to encourage valuable professional networking and promote sense of community. Snag’em’s pervasive features are designed to help players translate their in-game networks directly into realworld peer groups. Snag’em was designed to work on any web browser, including those on most smart phones. The game is also playable via SMS or text messaging.
considered a “healthy” network with breadth (meeting a lot of people), depth (getting to know some well), and mutuality (people snagging one another). Instead, avid players snagged as many people as possible, making it unlikely that they had meaningful conversations or made lasting connections. If a player interacts only once with another player, asking for example “Do you like video games?” and collects a Snag ID, neither player is likely to remember the connection or contact one another again. To prevent these shallow requests, we added connectivity bonuses to encourage players to snag one another more than once and introduce your friends to one another.
3.1
Snag’em Overall Game Play Players start the game by registering on the Snag’em game website (www.snagemgame.com) and receiving a four-digit Snag identification number (Snag ID). The player must provide the system with an initial list of “tags,” interests and affiliations they are comfortable sharing. These tags form the basis of the user networking mechanism. After adding tags, the player receives “missions” to complete. A snag is accomplished by texting or entering the found player’s Snag ID into the game.
4
RESEARCH METHODS
Snag’em was deployed in two academic communities, the STARS Alliance, and the UNC Charlotte College of Computing and Informatics (CCI). The STARS Alliance is a consortium of 32 colleges and universities, who engage computing faculty, undergraduates, and graduate students in outreach, research, and service activities to broaden participation in computing [1]. The community is composed of computing students, faculty and some individuals from industry and the community. Over half of the members are women, and about half are African American. We deployed our first study of Snag’em at the 2009 STARS Celebration members’ conference, where there were 168 students, 53 faculty and 31 partners (e.g. non-academic). We pre-registered attendees, but players had to tag themselves online or at the recruitment table to begin receiving missions.
The Snag’em site includes a network visualization page where players can view their custom-created avatars and how their snags connect them to other players (as shown in Figure 1). This allows both players and moderators to dynamically access and analyze player networks over time. Players can use this visualization to discover where their network is weakest so that they can strengthen the connections between the members of their network. Moderators can use the visualization to identify when new missions are needed to promote community growth.
Our second study of Snag’em was in March 2010 at UNC Charlotte’s College of Computing and Informatics (CCI) to promote networking among its 1200 students. We posted paper flyers and digital ads on plasma displays on all 4 floors of the CCI building. Every few days (3-5 days) the game developers set up a table at the building entrance to recruit new players, explain how to play, and help with initial missions. During game play, the Snag’em database stores information that helps game moderators understand how the player community interacts. This information includes login times, snag times, and tag update times. We also tracked recruitment times (when a user recruits another user), player computer platform data, and when users moved to a new level in the game. We collected statistics, such as the number of people that registered, how often people sign in, and how many times an average person snags or is snagged.
Figure 1. Players have a network page that visualizes their current network.
3.2
Evolution of Snag’em Game Mechanics We tested an initial Snag’em prototype at the 3-day 2009 STARS Conference, where 28% of attendees participated (80 out of 280), with half successfully scoring points in the game [9]. Although our intent was to promote social networking, some played Snag’em in unfavorable ways. For example, some players engaged in death match/free-for-all competition, hid nametags, and begged for Snag ID numbers. We believe these were due to the zero-sum nature of the initial prototype, where each player is “every man for himself”. After this prototype test, we added several features to improve the game. New “cupcake” points make snags mutually beneficial, where both the snagger and snaggee get points. We also added tag trees and “create-your-own” tags so that there would be too many for players to poll other players about all possible tags. At STARS 2009, Snag’em networking graph visualizations showed that the existing design did not promote what might be
Figure 2: Activity in 6 venues, # attendees below each column.
592
5
outside the classroom or student organizations. In this much larger context (1200 people), it was also much more difficult to find other players, since we did not provide badges or buttons to mark players. Our registration was also a very long process, leaving people with no time until their next class.
RESULTS
Snag’em was played regularly at the 2009 and 2010 STARS Celebrations, but very little in CCI. We first describe STARS results, and contrast them with those from CCI. 5.1
Snag’em Gameplay at STARS Conference Venues After the 2009 pilot, Snag’em was redesigned for STARS Celebration 2010, to include new recruitment and connectivity bonuses. Compared to STARS 2009, slightly less played the game at 66 of a potential 252 players (26%). Players made more snags (about 4.91 snag interactions per player, compared to less than 2 in 2009) and 100% of players scored points in the game. There were 55 total of 66 players who made at least one snag compared to only 40 of 80 in 2009. The other 11 players got points for being snagged. The top three players made 57, 33, and 32 snags respectively. Every player on the top-15 leader board completed five or more missions. The top nine players completed 10 or more missions. These results were encouraging, in the context of low accessibility at the conference: internet access was intermittent at best, and cell phone signals were low in the hotel.
In April 2010, we implemented Snag’em at CHI 2010 but did not conduct a formal study. CHI results reflected those in CCI, with only about 30 players of more than 2000 attendees. Our lower participation at large venues suggests that in order for people to adopt this new way of making friends and connections, it needs to be sufficiently visible to the potential players, and they need to quickly see a benefit to playing. One potential benefit of professional networking is to meet leaders and mentors, but at STARS 2010 (where we have more data on attendees) less than 1% of faculty (e.g. leaders) registered for Snag’em. This suggests that future designs should incorporate ways to entice potential leaders and mentors to play. At STARS 2010 Snag’em engaged 66 players who made about 5 snags each, but many attendees played passively or did not play at all. The post survey, with 12 players and 30 nonplayers gave us insight about attendee motivation. Non-players would have been more willing to participate in a social networking activity with no game elements (with mean 3.57 for non-players vs. 2.64 players on a 7 point scale gwith 1=No and 7=Yes). Non-players, like players, least liked: low internet connectivity, web design, and lack of friends playing the game. Non-players were more interested in mobile play, which was severely limited because of low cell phone coverage in the conference venue.
5.2
Summary of Results within the CCI Community Ninety-one of the 1200 potential students in CCI registered to play Snag’em, with 20 women and 71 men registered. Nine players (5 women, 4 men) made 5 or more snags during the 4week game. All 9 of the active players were game moderators or development members. These Snag’em administrators were responsible for a large portion of game interactions, and almost all gameplay occurred at our registration table. We hypothesized that game moderators could start the game off, and others would adopt and spread the game, but players did not seem to think about Snag’em outside of our advertised area.
7
Though several aspects of Snag’em worked well, in Table 1 we offer several design guidelines for new or improved game mechanics for creating social connections in academic communities. For example, at the STARS 2010 conference, many players from the previous year were excited to play, but were discouraged because wireless internet access was costly, and the conference hotel had bad cell phone reception indoors. This may seem obvious, but this detracted greatly from the game play, since players could only make snags outside the conference venue.
5.3 Snag’em at other venues After each Snag’em implementation, participants asked us to provide Snag’em for other events to help break the ice. We implemented Snag’em at the following venues, ordered from largest to smallest: CHI 2010 (CHI), CCI, STARS Celebration 2009 (ST2009) and 2010 (ST2010). Spelman’s Geek Week 2009 (SP2009), and UNC Charlotte’s summer REU 2011 (REU). Figure 2 shows the percentage of potential players, versus people registered to play, and people who completed snags in the game. These data show that the smaller the venue, the more successful Snag’em has been. 6
LESSONS LEARNED
The absence of leaders and potential mentors, or “elite” members, from Snag’em was one the biggest shortcomings of our system. Though the elite community members have not really played in the past, we found that in post-surveys, the elite were in some cases, more willing to provide us with survey feedback than the active players. This may have been because of their better understanding of why a sense of community is important. We are considering ways to better engage the elite, recognizing higher demands for their time and attention. We’ve since implemented “insta-snags,” a utility snag system that allows players to insert others into their network without completing a mission. Though the exchange does not count for points, it does keep record of people you’ve met at the conference and allows the player to use our system as more of a networking tool. Strong professional networks that make a player more likely to advance within the community cannot be realized unless those players interact with leaders.
DISCUSSION
Snag’em has evolved considerably over the course of these studies and other implementations of the game. At STARS 2009, a few harmful players made it less attractive to play, but participation was still at 28%, and people asked to use Snag’em at other venues. For example, in October 2009 we implemented Snag’em for Spelman’s Geek Week where 10 out of 80 attendees actively played during this week-long festival. For our CCI study in March 2010, we implemented new features, such as cupcake points for being snagged, and bonuses for recruitment and connectivity, but engaged less than 10% of students. At a conference, where everyone is expected to network, Snag’em is a natural fit. However, in a college setting, students don’t actively pursue meeting other students
593
Lastly, location-based missions would make our system invaluable for conference organizers as they would not only be able to create and monitor networking topics but also be able to track what sessions/events people attend and track and change conference traffic in real time. We have created systems for location based systems via QR codes and have started development on GPS mission. Table 1: Lessons learned from Snag’em Recommendation
Snag’em implementation or future plans
Make game visible & personally useful
• Signs and Snag’em kiosks keep it visible.
addition of mutually beneficial snags, interest missions, and recruitment bonuses. The effectiveness of these additions is due to not only the removal of zero-sum competitive play, but also due to the increased viral nature of the game, and the positive correlation between player percentage in a community and game activity. However, there are specific game mechanics that are needed to be sure to engage members of mixed social status in game play, and to leverage the strong connections among community members once they become engaged.
ACKNOWLEDGEMENTS
• We plan to mark players with badges / ribbons. • We added insta-snags, allowing people to collect virtual business cards while playing.
This work was partially supported by National Science Foundation grants 0739216, 1042468, 0757521, and 0851745.
REFERENCES
• We give small point boosts for being snagged and repeat snags to promote mutuality.
[1]
• We added interest missions (missions based on tags) to increase the number of meaningful interactions in the game.
[2]
Ensure wireless & phone connectivity
• Dedicated Snag’em computer encourage play in short breaks.
[3]
Make network connectedness transparent
• Snag’em shows how networking bonuses affect calculated score in score breakdown.
[4]
• Moderators use the network page visualization to quickly show where other network connections are possible.
[5]
• Snag’em and other pervasive games can easily slip from our attention. Snagging special locations can tie the game back to reality and would help track conference traffic in real time.
[6]
• QR code (camera-phones), GPS (smartphones), and ID code (SMS) missions are planned.
[8]
• New mechanics are needed to encourage
[9]
• Insta-snags allow serious participants to use the
[10]
Design and moderate to prevent undesirable play
Leverage local features and special events
Provide motivation for leader/mentor participation
kiosks
leader/mentor participation.
[7]
system as a business card exchange.
8
[11]
CONCLUSION AND FUTURE WORK
“After playing… Snag'em … I have become even more outgoing and more comfortable with approaching random strangers/ potential friends. I have grown leaps and bounds in my networks … [but] I had to go all the way to Tallahassee to meet people who already live close by.” -Student Member STARS 2009
[12]
Snag'em represents a new type of pervasive game built specifically to facilitate both personal and organizational-level goals for social and professional networking. There is more to learn about this type of game, but we hope that the two pilot studies we present here provide insight on how to design a similar system and evaluate its effectiveness. So far, our studies reveal that Snag’em gameplay can effectively facilitate social networking interactions for some community members. In order to ensure success of Snag’em across venues and audiences, however, we must look for new ways to make the system useful for all members of academic communities. Our most effective additions have been the
594
[13] [14]
[15] [16] [17]
Dahlberg, T., T. Barnes, A. Rorrer. “The STARS Leadership Model for Broadening Participation in Computing,” Proc. IEEE Frontiers in Education Conference, Milwaulkee, WI, Oct 10-13, 2007. Ducheneaut, N., Yee, N., Nickell, E., and Moore, R. J. 2006. "Alone together?": exploring the social dynamics of massively multiplayer online games. CHI '06. ACM, New York, NY, 407-416. J. Westheimer and J. Kahne. Building school communities: An experience-based model. Phi Delta Kappan, 75:324-328, 1993. J. McCarthy, D. McDonald, S. Soroczak, D. Nguyen, and M. Rashid. Augmenting the social space of an academic conference. Computer Supported Cooperative Work, 2004. D. Cox, V. Kindratenko, and D. Pointer. Intellibadge: Towards providing location-aware value-added services at academic conferences. In Ubiquitous Computing, pages 264-280, 2003. J. McCarthy, T. Costa, and E. Lionsgosari. Three steps toward ubiquitous peripheral displays. Ubiquitous Computing, p 332-345, 2001. Pew Internet and American Life Project. http://www.pewinternet.org/ Reports/2010/Social-Media-and-Young-Adults.aspx McCarthy, J. F., Nguyen, D. H., Rashid, A. M., and Soroczak, S. 2002. Proactive displays & the experience UbiComp project. SIGGROUP Bull. 23, 3 (Dec. 2002), 38-41. Powell, E. Finkelstein, S., Hicks, A., Phifer, T. et al. SNAG: Social Networking Games to Facilitate Interaction. CHI 2010, Apr 10-15, 2010. N. Villar, A. Schmidt, G. Korteum, and H.-W. Gellersen. Interacting with proactive community displays. Computers Graphics Magazine, 27:849-857, 2003. Nazir, A., Raza, S., and Chuah, C. 2008. Unveiling facebook: a measurement study of social network based applications. In Proc. 8th ACM SIGCOMM Conference on Internet MeasurementIMC '08. ACM, New York, NY, 43-56. http://doi.acm.org/10.1145/1452520.1452527 Nonnecke, B. and Preece, J. 1999. Shedding light on Lurkers in Online Communities. Ethnographic Studies in Real and Virtual Environments: Inhabited Information Spaces and Connected Communities. 24-26 January, Edinburgh. Ed. K. Buckner. 123-128. Ostrom, E. 1990. Governing the Commons:The Evolution of Institutions for Collective Action. New York” Cambridge University Press. Scholte, J. "Beyond the buzzword: towards a critical theory of globalization", in Globalization: Theory and Practice. pp 43 – 57. Pinter, London: Eds E Kofman, G Youngs, 1996. J. Stenros, J. Paavilainen, F. Mayra. The many faces of sociability and social play in games. MindTrek 2009. ACM, Finland. pp 82–89. V. Tinto. Taking retention seriously: Rethinking the first year of college. NACADA journal, 19:5-9, 1999. White, S. and Smyth, P. 2003. Algorithms for estimating relative importance in networks. In Proceedings of the Ninth ACM KDD SIGKDD international Conference on Knowledge Discovery and Data Mining (Washington, D.C., August 24 - 27, 2003.). KDD '03. ACM, New York, NY. pp, 266-275. http://doi.acm.org/10.1145/956750.95
