Slept in on Monday, well just a bit, so I missed the first session. Rolled into the second session to hear my friends Jennifer Stromer-Galley and Anna Martinson talk about their research. They are using a topic visualization tool designed by Susan Herring and Andrew Kurtz at IU. Research by friends using friends research, synchronicity is wonderful. Program information says:

Author(s): Jennifer Stromer-Galley, Anna M. Martinson

Title: "Coherent Argument or Fragmented Flaming: Comparing Entertainment and Political Chat Online "

Abstract: One of the pressing questions of the social impact of the Internet focuses on democracy. Citizen deliberation on political and social issues and current events is a key component of a strong democracy (Barber, 1984; Habermas, 1984). What impact, if any, does the existence of the Internet have on democratic processes and structures? This is a big question, and it has been tackled from multiple angles over the years. This study focuses on deliberation in synchronous chat from the perspectives of communication and information science. Our theoretical foundation is in the area of e-democracy. That is, the larger theoretical concern is in determining if and when the Internet is used in ways that benefit democracy in the strong sense (Barber, 1984).

Much of the research of online citizen deliberation has focused on asynchronous, threaded discussion (Davis, 1999; Graham & Witschge, 2003; Hill & Hughes, 1998; Schneider, 1997; Selnow, 1998). These studies of Usenet are important, in part because Usenet has been around longer that the HTML protocol that enabled the World Wide Web, and because Usenet has provided a vibrant space for political conversation by early adopters of the Internet. Now that the Web has experienced a decade of adoption, newer channels for conversation have been developed. Although Internet Relay Chat (IRC) has also been in existence since before the Web, it has been used more for social than political interaction. Even today, there are few places for synchronous political conversation. America Online has rooms devoted to politics, but they are lightly populated. MSN, Microsoft’s Web portal, has restricted access and use of their chat rooms. Yahoo!, however, has by comparison a vibrant set of political chat rooms open 24 hours a day, 7 days a week, 365 days a year.

Studies of political conversation on Usenet may not be generalizable to political chat (i.e., synchronous political conversation). Prior research indicates that the user base for Usenet is older and is more dedicated to the discussion space than users of Yahoo!’s political chat (Stromer-Galley, 2002). Research on political conversation in non-mediated settings suggests that those who talk are likely to be older than those who do not. If Yahoo! has a relatively younger user base, then it is possible that the substance of these discussions is different from those on Usenet.

Further, synchronous interaction is a different experience than asynchronous interaction. Synchronous interaction has minimal delay in sending and receiving messages; as such, in temporal terms, it feels more like an unmediated interaction. Chat messages tend to be shorter than asynchronous Usenet messages. There is a general perception that political chat is fragmented, shallow, incoherent, and filled with ad hominem attack. One study of political chat suggests that the conversations are not filled with flame wars—participants calling each other names rather than engaging in political debate (Hill & Hughes, 1998).

More research is needed in this area, however. One weakness in prior studies is a failure to compare political chat with chat of other topics. Without comparison it is difficult to know whether political chat differs from chat on other topics. If other topics compare similarly to political chat, then we have a basis for surmising that synchronous chat is a poor technology for promoting healthy political deliberation.

The aim of this study is to better understand how focused or fragmented political conversation is, employing Dynamic Topic Analysis (DTA) developed by Herring (2003) in combination with methods adapted from Cappella, Price, and Nir (2002), Graham and Witschge (2003), and Kuhn (1991).

The study tests 6 hypotheses:
- H1: political chat it is less coherent than non-political chat
- H2: political chat has more instances of personal attack than non-political chat
- H3: political chat provides more sourcing of information than non-political chat
- H4: political chat provides has a higher level of reasoned argument than non-political chat
- H5: Political chat will exhibit less agreement than entertainment chat.
- H6: political chat will exhibit more questions and requests for information than entertainment chat.

For this comparative study, 10 hours of Yahoo’s political chat and 10 hours of Yahoo’s Entertainment Lobby chat were archived between January 23 and January 27, straddling the New Hampshire Primary. After an initial observation of the discussion spaces, it was determined that the evening hours were the most active. Recording was slightly staggered so that each night of recording started one hour later than the prior night in order to capture the range of evening conversations. For example, recording on the 23rd began at 6pm and ran until 8pm. Recording on the 24th began at 7pm and ran until 9pm.

A preliminary test of a small sample of political and entertainment chat data suggests that political chat is more coherent, provides more reasoned argument, has more instances of agreement, and more instances of questions and requests for information than entertainment chat.

At lunch I again hung out with the gang from the night before. Laughter abounded and was good.

In the afternoon I went to a session on "Identity Online." Three papers bear some comment. Lewinsky Anet's paper applying; Bechar-Israeli, H. (1995). From to : Nicknames, play, and identity on internet relay chat. Journal of Computer-Mediated Communication, 1. It was a good presentation that makes me think about my own work with adolescent nicknames applying the same article. Program information:

Author(s): Eimi Lev, Lewinsky, Anat

Title: "The Presentation of Self in Online Life: The Importance of Nicknames in Online Environments. "

Abstract: The names people carry have major significance both for their own personal identities and as identifiers in social interactions. Though in most cases one cannot choose his/her own name, the name has an enormous symbolic, psychological and social significance. Lieberson (2000) emphasizes the symbolic quality names have. Names, according to this perspective, can be seen as one of the best examples of the theoretical approach known as Symbolic Interactionism. According to this approach, human communication is based to a large extent on symbolic representations rather than on actual features and characteristics. Hence, the name a person carries represents him/her when interacting with his/her surrounding. Goffman’s Dramaturgical approach (Goffman, 1959) is the one most relevant and helpful in the matter discussed here. Goffman emphasized the fact that people are constantly displaying characters, and expecting other people to take seriously the impression that is fostered before them. People are asked to believe that the character they see actually possesses the attributes he/she appears to possess. The same applies for names as “Identity markers” on the Internet. People do believe that the name (or nick name for that matter), has a meaning, and that many things can be learned from the name on the personality and characteristics of the person. Attributing meaning to the names, enables us, as Goffman claimed, to understand the situation and adjust our behavior accordingly. A common distinction in the academic literature related to names, is between names and nicknames. Nicknames are informal names, that are subjected to more changes than formal names. Nicknames are thought to better reflect the person’s identity than the formal name. Both names and nicknames play an important role in determining how other people would perceive the person carrying that name/nickname (Anderson, 1997). Social interaction on the Internet, especially in interfaces that support transient acquaintances (e.g. chat-rooms), must be affected by nicknames chosen by participants . In these interfaces, the nickname is, many times, the only detail that is available to other people to recognize the person by, and thus the image of a person and other people’s decision as to whether to interact with him/her or not, depends on the nickname he/she carries. On the Internet, choosing a nickname usually involves thinking of two different goals: hiding the “real” identity, and attract attention to the virtual one (Danet, 1996). Empirical research done by Bechar-Israeli (1995), on the distribution of nicknames on the IRC revealed that most nicknames (45%) are referential to the personal identity of the user , though only a small number of users used their real name. The second largest group included nicknames that refer to the medium, and the smallest group included nicknames that related to sex or were otherwise provocative. The focus of the current research was to achieve a better understanding regarding the effect nicknames have on internet mediated interaction, and to provide an updated description of the distribution of nicknames along different interfaces and types. To achieve that purpose, two different studies were conducted. The first investigated the effect of nicknames on the way other people perceive the person, other people’s willingness to interact with the person, and the type and length of interaction developed. Significant findings will indicate the importance of nicknames in online environments, and their effect. Results of this study indicated that nicknames indeed affect other people’s response. Based on this conclusion another study is presented that sought to provide an up-to-date mapping of the popular nicknames in different online interfaces including chat rooms, forums and instant messaging software. This study used as a departure point the categorization of online nicknames created by Bechar-Israeli (1995). Study # 1: This study used a vignette design. It examined the connection between the nickname and participants’ reaction to it in online interaction. Findings indicated that the chosen nickname is a strong factor regarding the reaction of the user. Study #2: A samle of 800 nicknames was collected from 4 different internet tools: forums, chat rooms and popular Israeli portals, Israeli chat instant messaging program- “Odigo” and IRC- Internet relay chat program. This sample of nicknames was divided into groups based on the categorization in Bechar-Israeli (1995) study. The findings of this study indicate that most nicknames (42.8%) were ordinary names from real life. The second largest group (24%) included nicknames refering to the person’s identity. The smallest group (2%) included provocative or sexual nicknames. Given the unique characteristics of the internet in general, and especially when talking about names, the current study sought to examine whether the importance of names is significant in Internet based interactions as well. The results of the current study strongly indicate that this is indeed the case. Study # 1 clearly supports the importance of names online, while study # 2 adds another important stratum, providing an up to date mapping of the names used in different online interfaces. Analyzing the two studies together, gives rise to several insights: As noted, the findings of study # 2 indicate that despite the choice given to people on the internet, with regard to their name, nearly half of them chose to represent themselves with a nickname that is based on a regular name. The second largest group was names that include some kind of reference to the person's identity, and the third largest group included names that are based on either an animal name or an object. The current study clearly indicates that most people do not use the opportunity given to them on the internet, to change the name representing them, and to choose unconventional names. The choice of names is usually conservative in nature, and the percentage of regular names is the highest of all. This interesting finding can be better explained when taking into consideration the results of the first study, where it became apparent that regular names are the names to which the response is the most positive of all. Bibliography: Andersen, C, P.(1997). The name game. Simon & Schuster: New-York. Bechar-Israeli H, (1995). From to : nicknames, play, and identity on internet relay chat. Dept. of communication & journalism. Hebrew University of Jerusalem: jerusalem. Danet, B. (1996).Text as mask: gender and identity on the internet. Paper prepared for the coference on "masquerade and gedered identity". Vebich. Italy, Feb 1996. Goffman, E. (1959). Presentation of self in everyday life . Garden City, NY: Anchor. Lieberson, S. (2000). A Matter of Taste: How Names, Fashions, and Culture Change. By Stanley Lieberson. (New Haven: Yale University Press, pp. 334.

Lori Kendall's paper "The Narrated Self: Self Presentation and Image Management on Live Journal" is a wonderful companion piece to my own work. I will be pulling the paper off the archive as soon as possible to read and cite it. Program information for Lori's presentation is not available online.

Finally Naomi Baron presented work she had discussed when we last talked - in New Orleans at ICA. Her project, titled "You are What They Post: Third-Party Identity Construction on the Internet" is a great look at who we appear to be when our name is inserted in a search engine. I know I have a very amusing set of tracks online some that I placed, often not thinking about the persistence of such tracks, and some that have been posted by others. Program says:

Author(s): Naomi Baron

Title: "You Are What They Post: Third-Party Identity Construction on the Internet "

INTRODUCTION The Internet is recognized as a medium that allows individuals to construct identities of their own choosing. Chat rooms, MUDs and MOOs, and Web pages and Blogs enable biological males to morph into online females or teenagers to appear as suave thirty-somethings. Such repackaging of self is hardly novel. The theme of disguise permeates literature; wigs have long camouflaged baldness; and bending the truth to “make a good impression” crosses time and medium. Erving Goffman’s analysis of how people “present” themselves in everyday life was standard fare among social theorists decades before Internet scholars recognized the applicability of Goffmanian constructs to life online. But if computer mediated communication invites individuals to craft virtual identities, the emergence of powerful Internet search engines has introduced a new phenomenon we might call third-party identity construction. Instead of individuals crafting their own public identities, users of the Internet (“third parties”) form conceptions of others on the basis of information gleaned from Web sites on which those individuals’ names appear. As the number of Web pages continues to soar and as search engines become growingly sophisticated, end-users are becoming increasingly dependent upon the Internet not only for gathering “objective” information but also for shaping their opinions regarding unfamiliar domains. University students commonly google their would-be professors before selecting courses for an upcoming semester. Academic professionals often go online to size up job applicants or unknown correspondents from whom they have received unsolicited email. The vetting procedure once entrusted to Who’s Who or journal indices is moving to Web searches. After perusing the first dozen or so hits produced by a major search engine, Web users begin constructing mental profiles of the individuals referenced in those pages. However, outside of the homepages those individuals may have created (or approved) themselves, the contents of Web sites on which their names appear are beyond the individuals’ control. Like our construction of virtual public selves, third-party identity construction is not unique to the Internet. Radio listeners imagine what favorite announcers actually look like. Public figures are plagued (or blessed) by images projected by the media or word of mouth. Third-party identity construction on the Internet is similar to off-line formulations in a number of ways: End-users need to be prepared to receive different information from different sources (e.g., conservative vs. liberal news media; Google vs. Alta Vista); end-users must decide whether to accept information uncritically or with a grain of salt. However, third-party Internet construction seems unique in that first, the end-user is generally working alone (not part of a mass media audience); second, search engines often produce tens of thousands of hits, only the first few of which most end-users pursue; third, the heuristics determining the order in which hits appear vary across search engines, thereby all but guaranteeing discrepancies in the kinds of information (and hence profiles) yielded in the top entries; and fourth, since yields from even the same search engine change over time, the profile emerging on an individual may shift even though the individual’s accomplishments or activities haven’t altered. Finally, if Internet search engines enable end-users to construct identity profiles of others, the same tools can be used by individuals to discover information about themselves that they either had not previously known or had not been aware was known to others. While the former information may prove useful (comparable to hiring a private investigator to gather personal information), discovery of instances of the latter can lead to concerns that the Internet is becoming a Panopticon, severely compromising privacy. EMPIRICAL STUDY To understand how third-party identity construction on the Internet works in practice, an empirical study will be undertaken analyzing the results of multiple searches on the present author’s own name. Using the search engines Google, Yahoo!, Alta Vista, and Lycos, the first 150 hits (from each search engine) will be collected during the months of February and May, 2004, respectively, yielding eight data sets. After removing duplicate hits and hits referring to people with only the same first name or only the same last name, the entries will be categorized by content type (e.g., references to publications, notices of academic lectures, other people sharing the same first and last names). Taking rank order in the search results into consideration, the content categorizations will be used to construct eight identity profiles. These profiles will then be compared across search engines and across time. The personal use of search engines for information about oneself will be studied using data from two sources: the above corpus plus pilot data collected by the author over the past 12 months. The pilot data suggest the following categories of analysis will prove relevant: (1) Doppelgänger: other individuals with the same first and last names as the author (2) Revelations: personally relevant information about which the author was previously unaware (3) Stalkers: email messages from unknown interlocutors referring to forthcoming activities of the author (presumably accessed through a search engine) (4) Space Junk: long-outdated Web pages referencing the author’s past activities (5) Transience: hits from a search at time T1 that no longer appear at time T2 (6) Lost in Translation: citations on Web pages in other languages for which even rough online translations are unavailable FUTURE RESEARCH This study is the first step in an attempt to understand how information and knowledge are created on the Internet. Simple extensions of the present research include (a) looking at third-party identity construction of individuals in non-academic settings and (b) interviewing end-users regarding their perceptions of how they use search engines to create profiles of others as well their perceptions of Internet constructs of their own identities. A more far-reaching enterprise will explore how Web pages and search engine usage are changing the ways members of the educational enterprise understand the nature of research and of the knowledge we assume derives from it.

Monday night was the conference banquet which I spent with the same core group I had been hanging out with previously. The food was good and the service was outstanding. Of course the real fun started after the banquet when many of us retired to the adjoining bar. Laugher, cheers, and comradery ran as freely as the Guinness. Great conversation with lots of self-deprecation - toward all our chosen careers and research areas - was had by all.

Posted by prolurkr at September 20, 2004 10:03 PM

Trackback Pings

TrackBack URL for this entry:
http://www.professional-lurker.com/cgi-bin/mt-tb.cgi/222