8: Saving Marriages, Online

Group: Steven Matthews and Ben Finkle (of the yellow blog)

Source Thread:  
http://groups.google.com/group/alt.support.marriage/browse_thread/thread/625e9a6c38b75565/0b20232db627c227?q=troubled+marriage&lnk=nl&


Our Data:

%inter-rater reliability	0.9
	frequency	% of msgs	Braithewaite
Information	11	55	31.3
Tangible Assistance	0	0	2.7
Esteem Suppor	4	20	18.6
Network Support	6	30	7.1
Emotional Support	8	40	40
Humor	2	10

We chose to look at a marriage-support thread.  We did this because we assumed we would have the most luck with a support thread (naturally), and marriage was one of the active topics on google (says something about marriage).  The specific thread we chose was titled "Troubled Marriage" where people can post specific difficulties their marriages are having, and get support and advice from the community.

Braithwaite created a method for analyzing group messages. In his methods there are several characteristics that supportive messages can have. These are:
Information
These include advice, referrals, teachings, and, well, information
Intangible Assistance
These are offerings of actual assistance, ex: “I will come over and help you do _____”
Esteem Support
These comments work to support the author’s individual self-esteem. Ex: compliments
Network Support
These kinds of messages tie in others with similar circumstances.
Emotional Support
This one is pretty self-explanatory. Offering support through affection, understanding, and motivation.

We agreed on the vast majority of our coding, and when we talked it over, it was usually one of us had erred, and read over something too quickly, or missed a more subtle meaning in the text.  We didn't dispute over any of our codings.   

Our data matches the findings of Braithewaite pretty well.  Our only statistically significant deviation fell in the line of network support.  We found a lot more network support than Braithewaite.  We attributed this to the fact that many people are married, and its easy for people to relate their own experience, and the experience of their spouses to other peoples problems.  We also found more informational messages than Braithewaite and this is due to a similar reason.  On this thread, people were specifically asking for advice, not just support.  This lent itself to information based responses.  

Our emotional support was right on target with Braithewaite, and we tied emotional support into the SIDE model studied in class.  A lot of the contributors to the thread had a lot in common whether it was positive relationships, negative relationships, or a career as a custodian (for what ever reason there were a lot on this thread.) This certainly created in groups and out groups, but there remained a high degree of group salience within each faction.  This lead contributers to be even more empathetic of their group, and even more prone to flaming those who disagreed.  

Overall, we found there to be some contributers who were very sympathetic and supportive, and some contributors who were very callous and antagonizing.  There was very little or no middle ground, and this is reminiscent of the cycle of over-attribution -> reconfirmation behavior.  

One interesting thing of note is the fact that we chose to also look at humor. While Braithwaite doesn't look at humor, it is interesting to look at as humor is an important part of human interaction. In our thread we found that about 10% of the messages included humor in them. At first glance, this is surprising considering that the posts were responding to a call for help about a struggling marriage. After reevaluating first impressions, perhaps humor isn't that surprising after all. With such an intense topic, perhaps some thread members felt it necessary to lighten the mood a bit before contributing advice. With regards to theories, we weren't quite sure what to attribute this to, but perhaps humor emerges from difficult situations, especially in interactive group settings.