Making Friends with Strangers!!! Online Social Support (8)

Alison Wollenberg and I (Talia Wissner-Levy) met up together to decode social support messages on Google groups. We found that many of the threads in the support groups had spam e-mails or news articles were simply copied from Reuters and pasted in the thread. Because of difficulty finding threads that had in-depth analysis and many responses to a user's query, and the abundance of spam, we analyzed support messages from different social groups and threads.

We categorized the messages according to 5 categories: informational support (which includes giving advice, referral to expertise, situation appraisal, and teaching), tangible support (any physical, clear benefit which performs a direct task), esteem support (elevates the sense of self with compliments, validation and relief from blame), network support (which provides access through members to professional services, helps with presence and companionship of other members), and emotional support (which includes the use of relationships, affection, confidentiality, understanding, empathy and any use of emotions). In addition to these characteristics, we also looked at the frequency of humor (which we defined as: 'anything intended to amuse'). These different types of messages are not mutually exclusive, therefore many different categories can be found in one message.

Our results found that we had a significant rate of inter-rater reliability (75.8%). This means that there was a significant amount of agreement between the two of us on the coding of the types of support behavior. Similar to the study by Braithwaite et al. (1999), we found that emotional and information support were two of the most common types of support. Tangible and network support were the two types of support that were least frequent in social support messages. Unlike the study by Braithwaite where emotional support was the most frequent behavior (40% of total messages) and information support came a close second (31.3% of total messages), we found that emotional support was not as frequent as information support. In our study, information support was by far the most frequent support behavior (80% chance to be found within a given message) while emotional support was a more distant second (60%). In both studies, esteem support was the intermediary social behavior. Network support was the least frequent method of support in our study (10% chance to be found withing a given message), while tangible assistance followed with 25%. In contrast, tangible assistance was found to be the least frequent support behavior (2.7% of total), with network support second to last (7.1% of total).

A possible reason as to why we obtained different results than Braithwaite may be because some support groups we looked at contained advice and information on drugs used for medical purposes. Anti-depressant and anti-anxiety medication support groups contained more information that used more objective advice of dosage, side effects, and referral to proper psychiatrists who could prescribe a different medication. For example, Anne is giving advice to Tony who is having adverse side effects to Paxil, to switch to Celexa "With Celexa I haven't had to increase (20 mg) at all. So it's probably worth trying a switch if your pdoc thinks the med is the problem."

We viewed quite a few tangible assistance behaviors under one thread discussing a widow's loneliness whereby some users referred the 49 year old widow "ensoul" to specific church groups, giving the website URL, and other specific information to come into contact with the Unitarian Universalist Church. This may have affected our data in emphasizing tangible assistance since we used a number of messages from this thread in our analysis.

Walther and Boyd (2003) have four factors into why people use online social support, and this is manifest in some of the messages we analyzed. Social support online is used due to the social distance, in which users appreciate the distance that the Internet allows as well as the wider range of expertise. For instance, in one post we noted a man named Tony ask for advice on continuing his current dosage of Zoloft prescription he finds it no longer effective. Clearly, he trusts the expertise of the wide variety of Internet group users enough to ask them for advice about such important matters as his medication. Another factor of why online social support is attractive to users is due to anonymity. Many users may feel that their problems are too embarrassing, or their advice is not valid to ask or answer to those close to them in face to face situations. In a post, Emily acknowledges that her fear of clowns "may be weird", but she urgently requests the help of others. Interactive management also allows a user to discontinue the online relationship and phrase their problem as they would like it to be evaluated. Interactive control is also found in a post we analyzed that looked asked for a strange sort of advice: "tell me your life story that is hopefully more pathetic than mine thus making me feel better" says user Omgyjya, asking the posters to exercise interactive control over the type of comfort messages they provide him/her. The last factor is 24/7 access to computers, making a user feel comfortable in reaching out for help any time of the day or week, so that synchronous or time related problems ina sking for advice are not an issue.

Last but not least, humor played a role in our study as it did in Braithwaite's. Some humorous messages included advice to a 27 year old who felt sexually inexperienced to get "a lot of imaginery practice" as well as to a widow to take advantages of toast in not having to cook for two.

Links
OCD
27 Year Old Virgin
Fear of Clowns
Sex Addiction
Zoloft
Depressed 21 year old
How to know you're depressed
Paxil mania
Lonely widow
Weight Loss