Assignment 8: Coding Social Support

Group Members:
Brianne Gilbert (blue blog)
Jennifer Niesluchowski (red blog)

Rebecca Wilson-Flewelling (brown blog)


We decided to test Braithwaite et al’s theory regarding social support by selecting a site that was dedicated to people who have “experienced the loss of a loved one to cancer”. See group thread at http://members.boardhost.com/grief/ - specific messages from “Just really missing my friend this week”, “Mom, I miss you. Need you a lot right now”, and “What a week: a wake, a funeral, physio post surgery, an MRI, Herceptin”. There were many posts by an individuals who have lost a loved one to cancer, their friends supporting and/or praying for them. There were also people who seemed to frequent the site and comment on specific posts to offer their support/ understanding of that particular person’s situation (assuming they too have lost a loved one to cancer). Lastly, there were people who are currently diagnosed with cancer who have also added their grief to this site as well, in search for some support.

We based our analyses on Braithwaite et al’s coding scheme is based on five categories: Information, Tangible assistance, Esteem support, Network support, and Emotional support. Our results are as follows:

Our results vary greatly (aside from tangible assistance) from Braithwaite et al’s conclusion where the categories read as: 0.313, 0.027, 0.186, 0.071, and 0.040 respectively. The majority of our posts were unquestionably recognized for their emotional and network support. Many people would show their emotional support by offering their prayers, showing sympathy, empathy, understanding, even affection with posts sealed with “hugs and kisses” as a “signature line” or within the message itself. Network support was showed by people (friends of the posters) giving the poster a sense of their presence, and messages ensuring that they are easily assessable in these times, and, of course, companionship was shown as well. Our lowest frequency and percentage of messages came from Tangible assistance (similar to Braithwaite et al’s conclusion, but different amounts). There was no concrete evidence regarding any direct tasks to help anyone; however, there were some instances where people would express their willingness to help the person with their grief and letting them know they are there for them.

Overall, we found that we had a 40.83% inter-rater reliability amongst each other based on these messages which showed that there is low reliability involved. Braithwaite et al had declared that any percentage greater than 70% would prove a vast amount of reliability and, like in our case for instance, anything less that 70% would prove that there is a lot of room for discrepancy.