I analyzed two types of data. I archived data from the facebook site and I surveyed some facebook participants. This was appropriate to my question because I wanted to see people’s behaviors online versus their thoughts and feelings about what they were doing.
Retrieving Archival Data:
Before I even started my archiving, I formed some operational definitions of key terms I would be analyzing. A facebook user “liking” a status update is defined by an overt pressing of the “like” button beneath each status update which allows others to see your response to the update. These “likings” are showed underneath each status update on my newsfeed. My operational definition of “attended to” was if a status update was commented on or “liked”. It is important to note that I could only record physical proof of a status being attended to. It is highly possible and probably that status updates were read and then not commented or “liked”.
First I set out to retrieve archival data. I needed to look at people’s status updates in order to make claims about them. Most importantly, I had to look at if people had commented or “liked” each status update. To do this, I went to facebook.com and signed on to my own account at 3:50pm on Tuesday, January 26th. I then went to “status updates” and copied and pasted the status updates that were recorded on my newsfeed within a two hour time span. I included all of the comments and “likings” in my archive. I then proceeded to code how many status updates were included in the two hours, which of those were commented on and how many comments were made. Additionally I included in the data how many times each comment was “liked” by someone. I then analyzed the average comments and “likings” of each comment and also calculated the percent of status updates “attended to”.
I repeated this process two more times (Sunday, January 31st at 11am and Friday, February 5th at 6pm). I wanted to get a variety of archival times so that day of the week or time of day wouldn’t be a confounding variable in my data. I postulated that perhaps people would attend to status updates more or less at certain times.
I only included the status updates on my newsfeed. This is problematic because the only status updates seen on my newsfeed are those of my friends. This biases my sample considerably. It is possible that my friends attend to status updates more or less than the average person. Further research on the subject would require a randomized sampling of facebook profiles.
Retrieving Survey Data:
On February 15th I created a survey on surveymonkey.com. The actual survey can be accessed via link on the “Multi-media, Links, Relevant information ect.” Page. The survey questions and possible answers were as follows:
1. How often do you guess that you update your status?
· More than once a day
· About once a day
· About once a week
· Rarely or never
2. How often does someone comment or "like" your status?
· Every single time you post
· Almost every time
· Occasionally
· Rarely or never
3. What do you typically put as your status update. check all that apply.
· Quotes
· Jokes
· What you’ve done lately
· What you are going to do later
· Random thoughts
· Personal opinions
· Other (Please specify) ___________________________
4. What kind of status updates do you like to read?
· Quotes
· Jokes
· What you’ve done lately
· What you are going to do later
· Random thoughts
· Personal opinions
· Other (Please specify) ___________________________
5. How often do you comment on other people's status updates?
· Every time I read one
· Often
· Sometimes
· Never
· I don’t read people’s status updates
After making the survey I attached the link to my status update at 2:09pm on February 15th. I made my status read:
Retrieving Archival Data:
Before I even started my archiving, I formed some operational definitions of key terms I would be analyzing. A facebook user “liking” a status update is defined by an overt pressing of the “like” button beneath each status update which allows others to see your response to the update. These “likings” are showed underneath each status update on my newsfeed. My operational definition of “attended to” was if a status update was commented on or “liked”. It is important to note that I could only record physical proof of a status being attended to. It is highly possible and probably that status updates were read and then not commented or “liked”.
First I set out to retrieve archival data. I needed to look at people’s status updates in order to make claims about them. Most importantly, I had to look at if people had commented or “liked” each status update. To do this, I went to facebook.com and signed on to my own account at 3:50pm on Tuesday, January 26th. I then went to “status updates” and copied and pasted the status updates that were recorded on my newsfeed within a two hour time span. I included all of the comments and “likings” in my archive. I then proceeded to code how many status updates were included in the two hours, which of those were commented on and how many comments were made. Additionally I included in the data how many times each comment was “liked” by someone. I then analyzed the average comments and “likings” of each comment and also calculated the percent of status updates “attended to”.
I repeated this process two more times (Sunday, January 31st at 11am and Friday, February 5th at 6pm). I wanted to get a variety of archival times so that day of the week or time of day wouldn’t be a confounding variable in my data. I postulated that perhaps people would attend to status updates more or less at certain times.
I only included the status updates on my newsfeed. This is problematic because the only status updates seen on my newsfeed are those of my friends. This biases my sample considerably. It is possible that my friends attend to status updates more or less than the average person. Further research on the subject would require a randomized sampling of facebook profiles.
Retrieving Survey Data:
On February 15th I created a survey on surveymonkey.com. The actual survey can be accessed via link on the “Multi-media, Links, Relevant information ect.” Page. The survey questions and possible answers were as follows:
1. How often do you guess that you update your status?
· More than once a day
· About once a day
· About once a week
· Rarely or never
2. How often does someone comment or "like" your status?
· Every single time you post
· Almost every time
· Occasionally
· Rarely or never
3. What do you typically put as your status update. check all that apply.
· Quotes
· Jokes
· What you’ve done lately
· What you are going to do later
· Random thoughts
· Personal opinions
· Other (Please specify) ___________________________
4. What kind of status updates do you like to read?
· Quotes
· Jokes
· What you’ve done lately
· What you are going to do later
· Random thoughts
· Personal opinions
· Other (Please specify) ___________________________
5. How often do you comment on other people's status updates?
· Every time I read one
· Often
· Sometimes
· Never
· I don’t read people’s status updates
After making the survey I attached the link to my status update at 2:09pm on February 15th. I made my status read:
Then I reposted my status at 6:46pm that same night making my status read: 
I then waited 24 hours to close the survey. 29 people completed my survey and 6 people commented on my status claiming they had participated in the survey.As, taking the survey posted on a link of my status was used as an additional means of assessing attention given to status updates I did not ask or seek out participants in the survey. The participants chose themselves based on who had 1. read my status update and 2. wanted to participate and 3. actually did so. This obviously biases the sample of responses because as these people defined themselves as people who attend to status updates simply by taking the survey posted on mine.
My only concerns in terms of ethics was keeping the archived status updates anonymous. I did so by deleting the names of the people whom I archived. I also kept my data completely clear of all personal information including what was actually said in the status updates. I obtained no informed consent to archive the status updates as it would have been difficult to contact each person after the fact. In addition I am not friends with all of the people who commented or “liked” my friend’s status updates so I would have been unable to obtain consent. Because of this, I had to limit my data collection and analysis to examining the attention given to each update, not what was said in each update. Further research could include a component on what is being said in each update and how people are commenting, but this will require consent.
My survey was completely anonymous and I don’t have any information about who answered what or who even participated. Interestingly, 6 participants offered comments on the link giving an unsolicited response claiming they had taken the survey.
