Session 4: Role/Capitol/Trust

Social Capitol
The main concept for the readings this week was social capitol. Ellison et al (2007) defines social capitol as "the sum of the resources, actual or virtual, that accrue to an individual or a group by virtue of possessing a durable network of more or less institutionalized relationships of mutual acquaintance and recognition" (3). The range of connections that can be made through online formats are frequently more beneficial to the user that many offline contacts. There are two types of social capitol, including bridging, "weak ties" that expand the breadth and worldview of the user, or bonding, "strong tied" connections that support the user emotionally (Williams, 4). The bridging effect increases in online environments primarily because of the low entry and exit boundaries. Users make connections with people they wouldn't otherwise in real life (Williams, 17). By expanding the connections made through online networks, the user has a wider scope of access to diverse information and opportunities.

Interesting to note, most users join social networks to maintain offline relationships, despite the benefits to networking with weak ties (Ellison, 12). Users also attempt to solidify offline relationships that would otherwise be temporal and superficial, which links them to further opportunities later on (Ellison, 14). It can also be a means for building social capitol with out the cost of interaction that takes place in offline formats. Social networks provide a security measure for those less likely to make connections or exert themselves into a crowd. (Does this also promote isolationism to an extent? Does it affect the abilities of young people to effectively interact and adapt to offline social situations?)

Williams (2006) states that by using social capitol, a user gets more of it (2). However, I suggest that the development of social capitol would not grow or succeed without social roles and trust.

Social Roles
Gleave et al. (2009) describes social roles as "recognized, accepted, and used to accomplish pragmatic interaction goals in a community" which "operate both sources of constraint and resources for action" (1). Learning the rules of interaction are developed primarily around the contributions of members. For instance,"social roles are often inherently defined in relational terms; a role only exists in relation to others who are likewise enacting social roles" (Gleave et al, 9). This leads me to question whether we create roles in order to facilitate control of our surroundings. Do roles help us assimilate to a new online culture?

Although Gleave et al. (2009) used the social structure of Usenet to describe different roles of interaction, it seems appropriate for several social networks. Usenet roles are broken into three separate characteristics: Answer People, Discussion Person, and Discussion Catalyst (Gleave et al, 6). Answer people respond to multiple questions, providing most of the content generated, while a Discussion Person contributes high content, but it is more focused on keeping a string of conversation going with multiple people. A Discussion Catalyst makes interesting and pointed comments that facilitate long threads of conversation. These three roles make up the environment to several question/answer formats of social networks.

It can be inferred that social roles can play into the development of trust as well. For instance, an expert is far more likely to be trusted than a laymen. Eryilmaz et al (2009) describe this element as the "authority" of a user, which stems from the desire for the most accurate and well informed responses.

Trust Mechanisms
A key component of developing trust within a social network is the presence of "trust statements," the declaration of one user's opinion concerning the worth and interest in another (Paolo, 53). There are different contexts to trust, which should be discovered and noted in the initial stages of interaction within a social network. For instance, some networks build trust through reliable product reviews, ability to be a good friend, or even interesting posts (Paolo, 55).

Timeliness can also be a factor in determining trust. If a social network is especially dependent on current proceedings, like an emergency or news environment, the user is more likely to trust (and search for) new information. Erylizmaz et al (2009) states "users trusted a piece of information more when it received more updates since there would be corroborative evidence" (8). Timeliness contributes to minimizing the risk of internalizing false information for the user.

It should be noted that the quality of a post or response can also be determined by the "number of edits (rigor) and the total number of unique editors (diversity)" (Erylizmaz et al, 4). Although this is primarily the way Wikipedia develops quality articles, I suggest that this is relevant for other social networks. Questions or post that get more attention are usually more trusted. For instance, in Goodreads.com, reviews of books that users uprate are sifted to the top of the page. This information becomes more and more trusted as it gains more attention.

Examples of Trust Mechanisms
In order to practically apply the concepts mentioned above, I joined a couple social networks that used different modes of trust. I should first mention that this proved to be somewhat difficult because most of the social networks I joined ended up having several of the same trust mechanisms. I will conclude this section with a specific example of site that has different means to develop trust, but that I did not join because of the cost of interaction.

Experience Project (http://www.experienceproject.com/index.php)
This social network is based on users sharing and connecting through experiences. The primary means of interaction is through joining groups that have similar interests, for instance "I love to cook" or "I can't believe I'm a grandmother," and then writing stories (like mini-blogs) about your experience.

Trust is developed through a couple different mechanisms. First, users interact based on their own preferences and opinions. They have no reason to interact with other users that have interests outside their own. This builds trust primarily because of the centralized themes of interaction. Users trust themselves the most, so why not trust others that have the same ideals (Allen et al, 1)? As a support to this concept, users can comment on whether the stories made them feel happy, sad, surprised, etc. to further develop groups with the same interests (see Figure).

Secondly, users are given point scales to both disseminate to peers as well as to gain social capitol. Experience Project is unique because it makes use of a "gift shop", where the user can gain points by interacting on the site and then use the point to buy gifts for other users. This is an interesting motivation for interaction. Gifts are also a status symbol, so the more gifts you have on your profile, the more likely you are to have built trust with other users. Roles are also built through interaction (much like Answerbag) because the EP team gives you trophies for each level of interaction you complete.

One interesting way of building trust is actually anonymity. As mentioned before, users interact based on experience. There is a section in Experience Project that allows you to make "confessions," which can be left completely anonymous. You are encouraged to let all your secrets out. People can leave comments and support. If you appreciate their comments, you can make an additional contact and add them to your "circle" of friends.

I'd also like to note one additional form of building trust. I submitted a story to the group "I want to lose weight" and titled it "Dark Sexy Jeans." This is actually a blog I wrote a couple years back, and it just describes my motivation for exercising. When I posted it, it came up with a "18+" adult rating (see Figure)! I was shocked. There is absolutely no adult content in the blog--but it has "sexy" in the title. This actually helps to build trust in the site because it filters content for those who do not want to be inundated with questionable content.


It would be helpful if users could see who uprated the posts, and who considers the post "happy" or "sad." This could develop friendships based on similar interests. To be honest, the use of trophies and gifting is not as well developed as it could be. For instance, giving gifts to others is not rewarded. If the user could gain status or additional access to some aspect of the site, it would motivate higher contributions and collaboration (Allen et al, 3). The incentive for use is not as concentrated as with Answerbag or other social networks. EP almost has too much going on at once, which lowers trust because the user cannot grasp full control of all of the elements of the network.

Goodreads (http://www.goodreads.com/)
Goodreads is a social network that is based on comparing, reviewing, and sharing books. Users can form book clubs and groups based on favorite readings as well as rate and review favorite books. The focus is on finding reading materials that may be of interest.

Trust is built on simple principles. While friends can be made based on reviewing profiles and finding similar tastes in books, it is not the primary focus. Users write reviews, make recommendations, and select ratings from 1 to 5. Other users can "make informed opinions about the product in order to decide about [reading] it or not" (Paolo, 60). This is an essential function of a social network based on opinions. User achieve trust based on giving points to reviews. Reviews that are appreciated by many users move to the top of the page. Users can either uprate or make comments (see Figure).

All contributions are tracked and displayed on the profile. The more contributions you make, the more trust is built. For instance, Graham is now considered the #24 best reviewer of books within Goodreads. This status builds trust because other users voted him into his position (see Figure). On a brief side note, trust is also built by adding friends as comparing books in their library. Other users can build trust based on friends of friends (Paolo, 61).


This trust mechanism would be especially helpful if users could follow a person's reviews. The only option for subscribing to a user's interaction is by adding them as a friend. I attempted to add a few people as friends, but they have to approve your friendship (much like Facebook), and if they don't have any connection to you, they can easily refuse. To find them, you have to keep track by searching for their profile. In other words, there is no easy way to subscribe superficially, which inhibits the spreading of trust.

Other Trust Mechanisms
Just really brief (because I know I'm running long), I wanted to point out another social network with a specific trust mechanisms. Meezoog.com is a dating/relationship social network that is exclusive in its membership. The only way to join is to be invited (much like Okurt), which builds trust because everyone you meet has some degree of connection. Meezoog actually has a social proximity meeter to indicate how closely connected you are (see Figure). It is based on the principle that you will not have to meet strangers and you are likely to have something in common. The trust statement is based on personal recommendation. Has anyone joined this site? Can you comment it's success?

Final Thoughts

I've had some trouble deciding on a final topic. However, because I work in Special Education, I've been considering researching the effect of social networking on families with disabilities.
The question came out of our Session 2 readings concerning motivation for interaction and social companionship.

Question: What are the possibilities for using social networks to enhance emotional support and access to resources for those who have disabilities as well as their families? Do social networks provide the emotional support for special needs that otherwise would not be provided offline?

Methods: Researching scholarly articles, such as "Families with disabled children: Stress and social networks in three samples", as well as other resources. Also, I will collect evidence from interaction based social networks geared toward special needs communities. I will likely start researching with a special needs network call "One Place" (http://www.oneplaceforspecialneeds.com/). Interviews are also a possibility.

Session 3: AnswerBag and Motivation for Online Communities

Key Findings from the Readings
Before joining Answerbag.com, I had to form a strategy for interaction. The readings gave a strong basis for the motivations for users to get involved in online communities, which developed ideas for enhancing the experience with Answerbag (AB). I took away a couple of main points that were repeatedly stressed in the articles:

• Ridings and Gefen emphasized that the motivation for joining online communities was (1) to find information and (2) to find social support and relationships.
  
• Users participate because of common interests, personal feelings, and daily experience (Java et al, 6).
  

Along those lines, experience can be enhanced with the knowledge of computer systems, which Schrock describes as computer self-efficacy (6), as well as the ability to self-disclose. Extroverts are more likely to participate because of the shift in social networks from text based to media rich formats. Because the trend for social networks now depends on exposure and visability, introverts are not as comfortable with participation (Schrock, 7). Schrock emphasizes that the colaboration between media makes the social networking system more popular because users can get a range of information in one format.

Although the Java et al. article was primarily about Twitter and other microblogging systems, it seems that AnswerBag also functions on some of the same principles. Microblogging has a few dominate reasons for participating: status updates, conversations, sharing information, and reporting news. Users typically either predominatly contribute information or take away information, and sometimes the user functions as both depending on the community (Java et al, 6).

Strategy for Interaction
Based on the findings above, I decided to try a couple of different techniques. I developed a set of questions that appealed to the different purposes for use. For example:

1. What's the draw to American Idol? (Cross-media information and participation)
2. What are some creative ways to get out of debt? (Information sharing)
3. How do you avoid gossip at work? (Experience sharing)

I also created a few questions based from lurking the community. I noticed that a large portion of the questions are "philosophical" or commentaries on social life. A link to all questions can be found on my profile here: http://www.answerbag.com/profile/static/qa/questions/864789

In an attempt to generate a unique question that would also garner interaction, I decided to survey a controversial issue. This would inspire debate and allow users to draw from their unique experiences, which was researched as one of the major draws to join online communities. My attempts at asking light-hearted questions (How do I get a bird out of my house? What makes a good office plant? How do you avoid gossip at work?) did not get frutiful contributions. I noticed that the more serious questions (Particularly, Do you avoid special needs children/adults? Why?) attracted more attention, possibly because people could draw on self-disclosure and gain user support for their experiences. Hence the decision for my final question: Would you report a rape if the rapist is someone you know?

Results and Thoughts about Interaction
For the purpose of this post, I'm going to use the above question as the best example for interaction. This post gained 24 answer posts and 46 rating points. The question can be viewed here: http://www.answerbag.com/q_view/1274660

Perhaps part of the reason AB is so successful is the fact that users are required to post unique or original questions. This likely motivates users to contribute primarily because they are required to produce new content. Ling et al stated:

"If [users] believe that their contributions are redundant with those that others in the group can provide, then there is little reason to contribute, because their contributions have little likelihood of influencing the group" (4).

I also noticed that questions and answers received ratings from annonymous users, which suggests that their sentiments were already expressed. This also appeals to the introvert as well as the extrovert, because the interaction is largely text based, but also leaves room for vreative profiles and visability. The posts are also a form of expressing personality, gaining popularity, and even broadcasting experiences (like with a microblog) in the form of a question or comment.

Along those lines, the ratings system was fairly motivating. My whole goal was to reach the 40 points count for the question, but I got absorbed in the question during the process. While the conversation and comments were interesting and kept me involved, the ratings gave me reason to continue posting questions. Tedjamulia et al suggests that "non-monetary rewards like social recognition can be extremely powerful incentives" (8), which includes compliments and recognition. The points system on AB is an efficient way of letting users know they are appreciated. Users are more likely to continue contributing and develop new content if they feel accepted and praised for their efforts to support the community information pool.

As a counter point to this system, do the points seem more controlling than informative (Tedjamulia et al, 6)? The contributions seem superficial when it's obvious that the user is participating for points. In light of the trend/competition to get as many friends as possible on Facebook, and sometimes even pay strangers to be-friend you, doesn't the points system seem to run along the same lines? Especially with the recent acknowledgment of a user reaching level 100, was this user motivated intrinsically or extrinsically? (Some of the questions on this user's profile seemed cookie-cutter to generate points...)

The use of participation levels, with the graphic ribbons, are helpful, since most individuals work well and participate more regularly when given a goal to reach (Ling et al, 19). AB also offers small rewards, like additional rating points to give out or level badges, but these aren't as motivating as trying to get more points on a question.

Ridings and Gefen stated that "virtual communities must have compelling conent, and that they might fail if they do not have good standards for this content" (4). However, I suggest that the content of these online communities don't necessarily need to be compelling, it just needs to be interaction. For instance, I found that the majority of answers and comments to questions on AB were attempts to be humorous, crude, or flippant. Questions were often answered in jest, no matter how serious or trivial the question may be. One answer to my most popular question ended with the user apologizing for the attempt at humor (see Figure 1). If people are just looking for entertainment, socializing, and friendship, the content is not as important as the interaction that takes place. Another development out of this question was the bond of emotional support between users. Because this questions was controversial and troublesome in nature, I noticed that users were more apt to share personal stories and offer condolences. This could stem from the idea that users contribute when they feel they have something unique to offer (Ling et al, 4)--personal experience tops the original factor. For instance, one user admitted to not reporting her rapist, although she thought she would. (I've worked with rape victims before and also shared that I've had second thoughts about reporting. This comment cited sympathy from the answers and comments that followed from other users. Because this was a social experiment, I decided not to reveal that I was not the victim, but allow the conversation to flow as if I were.) Ridings and Gefen found in their research that this is a typical experience in online communities. Social support is a primary influence for getting involved, with some users joining to "support others going through a rough time" and "to let others know that I have gone through it too" (12). My experience confirmed this research (see Figure 2).It was helpful to view the online community from the eyes of a opportunist, with the suggestions from Ridings and Geffen. It is beneficial to the design commuinty interaction after the idea that people are not only looking for information, but also for interaction (17). The content will support itself as long as the community can interact around it. AB does a good job of supporting this principle. Ridings and Geffen suggest that the virtual community have advanced search capabilities, links to non-member-generated material, user profiles, the ability to search profiles for activity, and use of experts in a particular field (17)--all of which AB facilitates.

It was also telling that the questions with the most interaction was also one that produced conversations from people who have had personal experience with the topic. The postings from women who have experienced rape, or others who know victims of rape, confirms the findings in the readings that people draw together because of common interests and the need for support. The questions that asked for personal experience or opinion were overwhelmingly more popular. It seemed to be understood that if you ask the question, than you've experienced the situation yourself. Users reacted with natural support for the issue raised. It immeadiatly formed a small community within the boundaries of the question. Did anyone else notice this?

I also noticed that if I commented on the answers of other users, they were more likely to continue the thread of conversation in other areas of the post. The more comments I made, the more people were drawn to the question and motivated to comment. Visibility is based on frequent participation. I should say that I did end up answering my own question (despite Gazan's warning not to "game the system"), because I felt that the comment would enhance the fruitful contributions. I found that people quickly uprated my answer and the conversation entered a deeper level of communication. Once I started getting to the heart of the issue, I found that other people were more comfortable with sharing their experiences (it built trust).

Overall, the experience was positive. I found myself motivated to keep asking questions just to see if I'd get an answer. The most difficult part was trying to find a question that was not already asked with the added pressure of gaining high ratings. My focus was more on asking a great question than making a great comment. My next attempts with interaction will be to see what kind of comments can gain high ratings.