Sunday, February 27, 2011

Gold Stars and Flag Paths: Trust Mechanisms and Social Roles in Online Communities

As a member of various different online communities, I’ve never given much thought to the importance of trust and how it plays out between users. Like most of us I have given feedback on eBay and participating in many other systems of trust building, such as rating fellow users or even just adding them to my list of friends, but this week’s readings made me realize for the first time how important trust really is to the health of any online community. Social roles on the other hands are a concept that seems more obvious at first, bringing to mind both visibly recognized function such as moderators and administrators, as well as loosely defined roles such as newbies, BNFs (big name fans), or lurkers. Yet by the end of this assignment I feel more confident in my understanding of the former, while the somewhat nebulous and unspecific definition of the latter has left me more uncertain about what social roles really are all about, something that was only intensified by my experience in two different online communities.

The communities I chose are Postcrossing, specifically the forum portion of the site, and the Expat Blog. Postcrossing is a community enabling users from all over the world to exchange postcards, and the forum is an extension allowing members to initiate “unofficial” contacts such as private swap or round robins, form communities based on different languages, and more. The Expat Blog is a site offering information, help, and networking to expatriates, people who have left their home country to live in another, and those who are interesting in doing so. I joined both communities, attempted to contribute by leaving comments on the posts of others as well as adding my own content, and tried to observe and compare some of the social roles and trust mechanisms at work there.

"my" two blogs

One of the most noticeable differences between the two communities struck me right after signing up: the level of activity. Despite the wealth of information and features the Expat Blog offered, its forums were almost completely dead, with very little interactivity or user contribution going on. The Postcrossing blog on the other hand seemed a very lively place, featuring many current threads and lots of member activity. With a similar community set-up and comparable audiences (both postcard collecting and expatriate experiences are probably very appealing to a certain group of people while being of minor interest to the general internet population) this start contrast definitely raised some questions, and my next step was to use the concepts of social roles and trust to try and find some answers.

Both sites require a similar sign-up process and offer the usual profile features such as an avatar and a “bio” field to display personal information and interests. Some of these details, along with the number of posts, are visible every time the user participates in the community, and this gives other members a better idea who they are interacting with. Added visual clues such as national flags revealing the member’s country of origin can also lead to a feeling of connection between users, especially in a place like the Expat Blog where the focus is on life in different countries. Their feature of displaying the “path” of their members through a sequence of little flags certainly gave me a sense of interest in and even trust of users whose path mirrored my own. The site offers many such additional features, including the ability to add detailed information and create photo albums for each country in one’s “path”, to have one’s own blog listed, and to be part of a network of expatriates in one’s current location.
Yet despite all these mechanisms to encourage trust and connections, there was very little activity in both the country or city specific forums and the general “Expat CafĂ©” open discussion, and looking at some threads and user profiles it seemed that many members either just lurk or lose interest soon after joining (as hinted at by the very low number of comments of almost all members encountered). The only exception to this rule were official moderators, who acted not only as discussion initiators but at times explicitly rewarded contributions by members by posting “thank you” comments, and a handful of regular users fulfilling the same function of opening discussion and sometimes even keeping it active by contributing their own replies. Still many forums looked pretty sad and dead… like the community for expatriates to Hawaii which was filled with moderator’s encouragements to share and discuss and posts by 1-post-users, some of whom never even bothered to acknowledge the rare reply to their own question. The balance of social roles definitely failed on this site, and mechanisms used to encourage trust between members did little to stimulate participation.

Screenshot of the Hawaii forum... compare the high number of page views to the almost non-existent participation.

The Postcrossing forum presented me with a completely different picture from the very start. Despite offering no special features beyond the regular forum characteristics, active threads and groups are common, and even as a newbie without much social capital I not only felt connected to like-minded people as I had on the Expat blog, but I was able to participate as a part of that group right away. The social roles of the two communities seemed similar, with “official” moderator and administrators sometimes action as discussion starters, and some particularly active regular members doing the same, but instead of just starting random threads much of the activity was organized around certain ongoing threads that received continuing input, such as an “Introduce yourself” thread for new members, and a very current “Question & Answer” thread. Identifying these roles beyond visible labels was made easier by a system of gold stars displayed under the user’s avatar, indicating “member status”. From my observation the most active regular members, who often acted as unofficial “answer people” or “greeting people”, always had a high number of stars. One possible scenario is that by accumulated a large number of comments/posts and therefore stars regular members may eventually “earn” official moderator/administrator status.

example of the Q&A thread... "answer people" and their gold stars

This system of ranking users encourages trust, especially since in this setting threads that help or encourage fellow members seem to be the ones that accumulate a large number of comments, thereby giving users wishing for a special role in the forum a chance to “collect” more stars. Trust becomes especially important in a setting where exchanging personal offline information, in this case mailing addresses, cannot be avoided. This is not quite as risky as Massa’s example, but requires a similar trust statement to make a possibly unsafe practice more secure. (Massa) One mechanism that encourages trust, and makes it possible for forum newcomers such as myself to display social capital, is the practice of many users to link back to their profile on the main Postcrossing site, allowing other members to see the number of cards they have sent and received through the “official” system. The larger the amount of postcards, the more likely it will be for community members to trust a person new to their group (the two are not directly connected, have separate sign-up processes and profiles, and only a fraction of the main site users seem to join the forum).
I encountered this trust mechanism myself when, as a complete newbie with 1 post (in the introductory thread) to my name, I started receiving responses to a trading offer within minutes of my post, and 24 hours later have a private message inbox filled with 30+ replies from users who willingly share their personal information with someone who, apart from a link to their main site profile, is a complete unknown quantity to them. Another measure of trust connected to this is the practice of posting public feedback for trades in the original threads (“got the cards, thanks” etc.), despite all initial interactions, including the exchange of addresses, taking place through private messages (knows as u2u in the forum).

My PM inbox 24 hours after posting a card trade offer as a total newbie.

In general I had the impression that the Postcrossing forum made the best use of the pretty basic tools available to create a balance between very active social roles on one side and the more reactive majority of the member-base on the other, and to encourage trust through simple mechanisms such as providing private messaging and the gold star system with users creating their own additional tools such as linking back to the main site. The Expat Blog on the other hand could not achieve this balance despite its many features and active moderators, maybe because creators or administrators put too much emphasis on those features rather than making sure enough members would actually have a stake in the site and start using them. It seems to me that there was little to encourage newcomers to stay part of the community after their initial information need was fulfilled (or not). Both sites are examples of “bridging social capital”, with users from different backgrounds come together to exchange resources of information (Williams), yet providing clear yet simply trust mechanisms and visible indicators of social roles that allowed users to influence their own status made the Postcrossing forum the more active and successful community.

The idea behind the Expat blog is a good one, and could be an example of Ellison et al.’s bridging social capital as a force that offers new opportunities to members of the community, such as networking and gaining knowledge about a country from someone who has made the experience of moving and adjusting to life there. But unlike Facebook this site doesn’t make use of the immense potential and proved too few opportunities for members to not only initially connect but maintain that connection. Of course unlike the original Facebook there is no location-based link between users from all over the world, but the fact that I did feel somewhat connected to other German expatriates in the US shows that country of origin could be used to create stronger ties. One of the goals would definitely be higher user participation in the forums, which could then lead to members identifying with the site and connecting to each other, building trust and really forming a community. But it doesn’t seem that a simple tool such as gold stars would work in this case, and having “official” moderators fulfill the social role of discussion starter doesn’t work if there are no user who are contributing… this is a good example of Gleave’s role ecology, which in this case is out of balance and therefore does not work.

comparing visible user information:
the Expat blog vs. the Postcrossing forum... the latter gives far more clues about possible status/social role and could inspire trust (you can also see the main profile link I mentioned before)

So a suggestion for the Expat Blog might be to introduce some kind of ranking system of users who contributed, maybe linking it to country or expatriation “path” since those seem to be strong links between users, and focus less on the personal connection and networking features that already exist but are rarely used. The Postcrossing forum, with a strong system of trust mechanisms already in place, may benefit from attempts at deepening connections between users and expand into bonding social capital such as social support. (Williams) I have already seen some examples of this in threads such as “Random Acts of Smileness” which includes sending post cards to people, including non-members, who are ill or in other difficult situations, as well as surprise or birthday cards, in general giving support without expecting anything in return. Adding features such as birthday reminders, as suggested by Ellison et al., or maybe virtual gifts that users can display on their profile or with their avatar (something some Livejournal users are even willing to pay for), could encourage such bonds.

Finally some ideas for my project. My main interest is in how personality (such as introverted/extraverted characteristics) or even gender affect user behavior or choice of communities to participate in, but I haven’t been able to come up with a specific question yet. I personally really enjoy exploring and observing new online communities, so to be able to make that part of the project would be nice, but I would also like to explore more readings on various viewpoints and research that has been done in this area. I’d be grateful for any suggestions to help me to narrow my focus or any general comment/ideas. Thanks!

Sunday, February 13, 2011

Motivations for Participating in Online Communities

As someone who has only ever considered a user’s perspective of online communities, it was really interesting to see a different point of view, and get some insights into why networks, forums, and other online groups functioned the way they do. This week’s readings offered many such insights, including an approach from a business perspective, something I never thought to connect to an activity that seemed so purely personal to me. While the role of business in running a community and making profit from it by selling advertising space or from users directly (such as offering paid accounts with special features), the idea of companies creating and managing communities for their own purposes still seems foreign to me, so it would be really interesting to see anyone who has experience with that sort of thing share some more information about it.

Another aspect of the readings I found curious is the way they all seemed to agree on the types of motivations at work in keeping online communities active, even though thee authors had different mind-sets and used different approaches to reach their conclusions. Ridings and Gefen examined the different reasons for joining a group, the desire for information exchange and friendship being most important, with social support and recreation less significant, and came up with three main motivation factors: being part of a group, reaching goals, and being rewarded. They introduce the concept of lurkers being part of a community as well, despite their lack of interaction, and the idea that while many users enter communities with the goal of gaining knowledge, in line with the original design of the internet as an information exchange, connecting to others with the purpose of gaining social support or finding friendship have become major aspects of online interaction as well.

Ling et al. examine the role of motivation in a more complex way, questioning if social science theories predicting user behavior can be useful to designers in order to create communities that remain active and encourage participation. They introduce the concept of social loafing, something that can be commonly seen in groups when the majority of members remain passive or contribute less than they would as individuals, and the collective effort model, which explains factors that can reduce social loafing, such as user’s belief in the importance, uniqueness and recognition of their contribution, as well as their affinity to the group. One significant conclusion of their research is the finding that motivation needs to happen internally, with extrinsic efforts such as the promise of benefits to users rarely generating lasting participation.
Tedjamulia et al. analyze motivation from a business perspective, in order for community “managers” to create a productive group environment by providing feedback and rewards and prevent loss of funding through problems such as disorganization and unstable membership. They describe several levels of participation from lurker to veteran, suggesting that a balance between them is necessary for a thriving community. In order to increase motivation and sidestep knowledge-sharing dilemma, the authors suggest personal and environmental factors such as high self efficacy, trust in the community, interesting content, and visibility of contribution, as well as goals and rewards, showing some overlap with ideas put forward by Ridings and Ling.

Schrock addresses motivation from a personality perspective, suggesting different user patterns between introverts and extroverts, with the former using text-based communities to overcome offline difficulties of expressing themselves, while the latter prefer more visual resources that offer more disclosure, such as youtube. He also examines gender differences of online community use and motivations, indication that males use such sites to expand their social networks while females tend to maintain close ties to already existing connections.
Java et all concentrate on the new concept on microblogging, which fulfills a need for ever fast modes of communication and higher number of updates. The authors propose a high reciprocity between users, with previously existing social networks making up a majority of the active user base, and point out different user roles depending on this reciprocity, with members acting as information source or seeker. They also show that switching between roles and directing various types of posts at different audiences with the clear division between friends and communities, or categorizing/filtering tools that exists in other online communities such as Facebook or Livejournal, can make interaction confusing.

Having observed many of the motivations and behaviors described by these authors, I find myself agreeing with much of their eventual results, if not all of their original proposals. For example, Ling at al.’s observation of group goals being more effective motivators of contribution that individual goals reminded me of several instances where Livejournal communities I joined initiated fundraising events, for example for victims of the Haiti earthquake, and were able to collect thousands of dollars as a group, while only a few of those who contributed would probably have donated purely on their own incentive (this also confirms the value of a clearly stated goal in motivation, as the this effort also included a set amount of donations to be reached, which probably encouraged members to continually work toward that number).

While personal experiences such as this generally confirmed the concepts and conclusions addressed in the readings, I do disagree with some of the basic approaches and specific premises used. In general, it seems to me that all this research is based on perspectives and interests other than that of the users themselves, who after all represent an essential part of the equation. Managers and designers of communities can only be efficient if user’s needs are taken into consideration, and sometimes the best way to find out what motivates contribution is simply to listen to what members have to say and observe what they do. Social theories and business practices are certainly concepts to take into consideration, but if they clash with the actual needs of a certain member base, sticking to them as the only tools to run a community can do more harm than good. I have experience this numerous times on Livejournal, where the most active member groups (which are usually also the ones who invest the most money in the site) are very vocal about their ideas and motivations yet are repeatedly ignored in the implementation of new features that go against community mentality. This not only leads to an unhappy community and in many instances the need for designers to adjust features in the face of protests, but also to a loss of revenue due to users retracting their financial support of the site.
Specifically, I also found myself to question assumptions such as Schrock’s concerning female users being less tech-savvy and therefore in more danger of falling prey to online predators or similar pitfalls of community interaction. Conclusions such as this make me doubt the usefulness of surveys that identify women and girls as less knowledgeable of technology, and make me wonder if the problem doesn’t lie in a different approach to answering questions regarding such topics, or cultural norms that lead to them assessing their own skills at a lower level then they actually might be at. To point out the high number of female users in online communities and yet continue to rate their skills of using online technologies as less than males should really lead us to question where this discrepancy really comes from rather than what it may lead to.

For the second part of the assignment, I observed the forum of F1 Fanatic, a blog that is usually my news source for anything Formula 1 related (Formula 1 or F1 is an international car racing world championship). While I visit the blog regularly, I’ve never paid attention to the forum until now, so searching posts and examining participation and contributions in this environment that was connected to something familiar but yet completely new was really interesting.

Modes of Participation
I realized quickly that the basic setup of the forum was very simple. This could be part of the close connection to the blog itself, as several aspects were not directly part of the forum itself, but linked through the interactivity of the blog. I included those modes in the list because they’re still important to the participation on the forum. Forum users are able to:

- post content
- comment on posts
- tag posts by subject
- track favorite posts (this is not visibly to other users in the individual posts, only on the user’s profile)
- post guest entries on the main blog
- comment in polls on the main blog

How is participation encouraged?
- users’ contributions are identified
(example: they can create accounts, upload a user icon, profiles show posting activity, date of joining & favorites)

- performance rewards/social recognition
(example: blog owner picks “comment of the day” displayed on main blog)

- forum is organized for easy viewing and use
(example: search function available, users can create rss feed of favorite threads)

- members offer easy ways to become part of the group and contribute
(example: many open-ended questions/discussion, “introduce yourself” post)

- members post interactive content
(example: meme, trivia, “what’s your favorite…” posts)

What content gets the most responses?
These results really surprised me, but the reason for the pattern may be that most news items, and even specific drivers or races are usually discussed in comments to main blog posts by the owner of the site. Numbers would also differ not quite so much if I had taken into account the amount of time for which posts had been up in the forum, but I could find no easy way to include both types of posts, ones that had been up for months but were still active, and those that were more numerous but very short-lived. It was obvious to me though that even if I would divide the high comment numbers by the amount of time the post had been up, they would still outnumber the type of post that stops receiving comments after a day or two. In general, recreational topics had a much higher rate of response and stayed current even after months of the original post dates, while posts relating to news, particular drivers, technology, or other discussions only gathered responses in a short time-frame.

1. trivia,meme, games
# of posts:13
# of comments: 5872

2. requests to share experiences, favorites etc.
# of posts:10
# of comments: 1074

3. news discussions
# of posts:13
# of comments: 185

4.general discussions of individual drivers, races, etc.
# of posts:5
# of comments: 90

5. F1 game discussions
# of posts:6
# of comments: 46

The results were really surprising to me, since I had expected much more technical and general discussion, the content I was used to from reading comments on the main blog. But this definitely makes sense, as followers of the blog obviously use the forum to connect to others and even form friendships rather than repeating the more information-heavy exchange from the blog. This conforms with Riding’s and Gefen’s ideas of communities staring out as places of information exchange, but often leading to further connections and friendship. It also made me remember how, when the forum was unavailable for a few months due to technical site issues, users would continually comment on the main blog with questions about the forum’s status. So for many followers, being able to have those personal connections in addition to the information exchange on the blog itself had already become an essential part of their experience of this particular site.

It also shows that microblogging may constitute a form of online communities somewhat different from traditional forums or journal sites, as the type of connections formed is the direct opposite of Java et al.’s observations of pre-existing social networks with high reciprocity. Here the mode of connection is completely based on a shared interest, with users having no previous connections, and while personal information not directly related to the shared interest (in this case F1) is exchanged and some friendships develop, much of the interaction is more recreational and less personal. This also constitutes an online environment very different from what I personally participate in, where shared interests play a role but social support and the simple daily life interactions make up a far more essential part of the community experience. Which is probably one of the reasons why I have never used this forum, despite being a daily visitor and occasional commenter on the main blog, and why, despite joining many similar communities throughout my years online, I’ve never become an active contributor in any of them.