DATA ANALYSIS

The data collected for this study was strictly de­scriptive in nature. For this foundational study, the individual cases were categorically sorted within and analyzed for patterns within the groupings of the keywords used for the search. Four distinctive groupings were analyzed in this study: groups associated with disabilities or having a disabil­ity, race/ethnicity, gender, and sexuality. Table 1 shows the search terms used for each of these group designations.

Basic average and summation tests were con­ducted to analyze for patterns that would signify that deeper analysis is warranted. Simple com­parison tests were also conducted on membership statistics within these groups (both the groups that were returned during the search and the groups that actually met the criteria).

RESULTS

Comparison of Groups

The total number of groups that were returned within the search varied greatly. The numbers ranged from the Race/Ethnicity grouping return­ing 101 keyword matches to 1283 matches based

Table 1. Key search terms by group Gender Race/Ethnicity Disability Aging Sexuality Female African American Disability Retired Homosexual Woman Hispanic Disabled Aging Heterosexual Male Latino Handicap Geriatric Bisexual Masculine Islamic Impaired Geezer LGBT Transgender Muslim Muscular Dystrophy Old Gay Aspergers Lesbian Blind Learning disorder Aging Wheelchair Queer

Figure 1. Comparison of search returns vs. actual number of groups that met the search criteria

Figure 3 depicts the comparison of the average size of the groups within the Second Life archi­tecture, taking into account that groups with only one member were not counted in this analysis.

on the Gender keywords. The percentage of these groups that actually passed the search criteria discussed above also varied greatly. The Sexuality grouping had the highest passage rate with 87.8% of groups originally found to represent the criteria specified in the questions above. TheAging group had the lowest passage rate of 18.9%. Figure 1 depicts this variance.

Figure 2 highlights that the Gender and Sexu­ality groupings have a greater representation within the Second Life architecture. The Dis­ability groups represented only 5% of the total number of groups that passed the search criteria filter, while the Aging group represented an even smaller amount, 2% of the total number of groups returned in the search.

Disability Grouping

The disability-identified groups had a great deal of in-group variance. Compared to the whole group, the disability group was significantly less represented as compared to the gender and sexual orientation groups. The tables highlight these trends. Table 2 highlights the specifics of the disability associated groups analyzed for this chapter. Table 3 gives a more nuanced description of the specific groups that represent the disability

grouping: a total of 10 keywords were used for the data collection. Of the 65 groups that met the search criteria, the average group size was 30.75 members. There was a difference of 53.3 members between the lowest and highest average number of group members.

For context, an estimated 19.4% of non-insti — tutionalized civilians in the US (approx. 48.9 million people) have a disability, with about half of these individuals being considered to have a severe disability (Kraus, 1996). According to Mobility International and the United States Ac­cess Board, there are at least 53 formal disability groups (real life groups) operating on a national scale in the United States. When comparing esti­mated populations to numbers of estimated dis­ability groups, the numbers are striking. There is one national “real life” group for every 5,000,000

Table 2. Disability group statistics

U. S. citizens, while there is one group for every 215,000 Second Life citizens, who may reflect the comparative ease of identification, and forma­tion of communication based groups in a virtual setting.

For comparison, the same analysis for disabil­ity groups was applied to one of the comparator groups. In the physical world, we find that in the U. S. there are estimated to be around 135 groups representing the Lesbian, Gay, Bisexual, and Transgender (LGBT) individuals (rough estimate

of formal groups by the US Small Business As­sociation and the National Gay and Lesbian Task Force). When comparing estimated populations to the number of estimated LGBT groups, the numbers are even more polarizing than the dis­ability numbers. There is one group for every 2,000,000 U. S. citizens while there is one group for every 29,000 Second Life citizens. One pos­sible explanation for this phenomenon is that in Second Life groups, group names, and group keywords are selected solely by users (Diehl & Prins, 2008). Noting that these group names may reflect “real life” references, it is important to acknowledge that these labels are chosen, not applied. While this pattern seems enlightening, further data collection and analysis needs to be obtained, specifically in relation to groups sizes of the real-world groups, so that populations can be normalized in order for a formal comparison to be made.

Given that this study is exploratory, future research might probe this condition further. The use of a multimodal system of in-world surveys, and interviews, in addition to potential social and cultural modeling and potential software analysis tools appears to be an appropriate next step. The use of purely descriptive data and statistics, while highlighting patterns, does not give us enough to
claim there is a statistically significant correlation that exists.

CONCLUSION

Not surprisingly, if virtual places are any reflec­tion of underlying social construction the number of individuals gendered or otherwise who choose to represent themselves as having a disability is relatively small. The meaning of this observation is somewhat complex to interpret as it is uncertain as to 1) whether this suggests that people with disabilities choose to alter their identity based on characteristics, or abilities (or lack thereof), or al­ternatively, 2) whether people with disabilities are underrepresented in the virtual/gaming population, or both. Statistics on income and information and communication technologies (ICT) use, suggest that people with disabilities have a much lower rate of Internet use, and it would be a valuable exercise to conduct exploratory research to ex­amine the relationship between income and other accessibility issues and the lower Internet usage rates among people with disabilities (Dobransky & Hargittai, 2006).

Of particular interest, this work developed a foundation for future disability research by

extending schema theory to include the issue of disability identity. This new extension, disability schema theory, has both online as well as “real life” application. Application of disability schema theory to interpretation of Second Life research seems to suggest similar perceptional impacts that also occur in the physical world. In the “real world”, the visual cues that activate schemas can serve as an explanation for the stigmas and ensuing isolation often felt by people with visibly apparent disabilities.

In Second Life the visual cues are removed un­less a conscious effort is made to create an avatar with characteristics that mirror real life disability. In the case of an avatar created without disability, study results suggest that Second Life users with disabilities are still associating with others who identify as having disabilities. This finding helps us to probe re search que stion number three: Based on search term identifiers, do people appear to associate with others who identify with specific characteristics (i. e. disabilities) or genders when the cues that reinforce schemas are not apparent or observable? This is an area that is ripe for future research. In world interviews might probe users (avatars) regarding their actual disability status/ identity in order to answer questions regarding the choice to either identify or not identify as a person with a disability. Additionally, in world interviews might seek to answer questions re­garding decisions to associate with others who identify with similar disability status, especially in an environment where visual cues are removed. Because of the exploratory nature of this study, this was in fact a limitation of the current study.

In response to research question number two: Given the lack of visual cues, are there groups cre­ated in Second Life that are specific to people with disabilities? Are any of these groups gendered? Gender appears to play a role in many groups (i. e. “communities”) found in Second Life. This is evident in the number of groups that were iden­tified using the keyword ‘gender’. Interestingly, groups such as “Gimp Girl” are specific to people who identify as both female and disabled. In fact, avatar wheelchairs are available for purchase at the Gimp Girl site and a number of avatars ex­ist that are depicted as wheelchair users. Future research might delve further into refining both gender and disability schema theory in an effort to determine their ability to explain the choices an individual makes when deciding whether or not to identify(online) as a person with a disability as well as within the traditional binary gender paradigm. It is especially interesting to note that groups exist as a “safe space” for those who identify within the criteria of the group under review. Groups such as Gimp Girl exist not only in Second Life, but also have a presence on the Web as well as on the social networking site, Facebook. Another area for future research would be a replication of this study within social networking sites such as Facebook and MySpace.

As noted above, research suggests some 20% of gamers have some degree of disability (Faylor, 2008; Ingham, 2008) — in contrast the results of our empirical examination ofmembership of disability related groups shows that < 1% of total groups in-world self-describe or affiliate with disability. This is an interesting answer to research question number one: Do people with disabilities identify as disabled in Second Life? A reasonable response to these relatively stark observations might be to conduct additional in-world survey and interview research, in Second Life as well as other virtual platforms, with both avatars that self-identify as disabled, as well as with individuals who choose not to so identify but who may in fact be disabled. Another interesting question for future research is ‘Why do some people with disabilities choose to gather within gender specific Second Life sites?’ Additionally, it may be of interest to open a query regarding the stigma of gender in virtual environ­ments. Future research might seek to discover why individuals choose a more fluid representation of gender, or no representation of gender at all.

We believe that both within and outside of the binary gender framework, the differences between the two disability identified groups, those externally classified as having some degree of dis­ability, and those who choose to self identify, or more accurately affiliate “visibly” with disability related groups, have rich import for the sociology of online communities as well as for the design and characteristics of games. More practically, the applied value of this inquiry is in the poten­tial for the development of new employment and community engagement applications as well as providing suggestions about new approaches to developing virtual and online environments to facilitate enhanced participation in society. Ad­ditionally, any research that serves to lessen the stigma felt by those who identify as disabled and those with visible disabilities will help to “level the playing field” in the real world, similar to that which exists in virtual environments such as Second Life.