Data Data Data: Interesting FB Findings

We have finally finished the arduous task of coding and putting variables on our data and can now look at the actual trends the data reveals; and let me say there is some pretty cool stuff for December!  The first thing I looked at was a general breakdown of the major categories of posts to see what was most popular:

As you can see, information sharing is the largest segment with 40% of posts.  One can also add the posts categorized as information seeking to this percentage as I found that these posts tended to be answered in the comments section.  Thus, the total for information based posts was 61%.  This indicates that this page was an effective disaster response page as it was most often used to ask for or provide the information.  By comparing those posts categorized under the umbrella of information, a further breakdown of the topic/type of information is possible:

As you can see the most common topics of information that were discussed in posts and comments were aid (25%), followed by institutions (15%), and planning for the future (10%).   When examining these three types of posts (which make up 50% of the information topics) more closely by looking at actual examples, the majority of these posts deal with some sort of recovery effort.  In fact, of all of the information topics only 10% (grievances, memories, Obama, weather, disaster mythology, devastation, and Katrina) do not deal with some sort of attempt for recovery of long beach; this further speaks to the success of the page.

Now I am going to highlight some of anomalies in the data by doing side by side comparisons.  I have chosen to focus on the anomalies to both identify the general trends of the topics in general (where they exist) and point out some areas where these trends do not hold.  I also attempt to explain why, in certain circumstances, these disturbances may occur.

Gender: Throughout the topics females tend to be the primary posters.  In fact, in information sharing, information seeking, and grievances around 2/3 of posters are women.  The only place where there exists an anomaly is in memories, where 2/3 of posters are men.  This anomaly, however, is not significant because only 3 total posts were categorized as memories.

Admin v Other: Overall the majority of posts are not made by the admin.  This margin is great in memory (66.67% other), information seeking(89% other), grievances (83.33% other).  The margin is slim (under 20% gap) in solidarity (56.86% other), and information sharing (52% other).  The only place where this trend does not hold is in contemporary documentation where the admin posted at a rate of 60.00%.  The anomaly, and indeed the close margins can be explained by the fact that the admin tended to re-post certain posts that I believe he/she found to be extremely important.  Contemporary documentation and solidarity seem to be re-posted with the intent of highlighting the strength and recovery in Long Beach.  Information sharing posts were re-posted to provide helpful information to as many people as possible.

Hyperlink: The majority of posts did not have hyperlinks.  The only topic with a close margin (margin gaps under 20%) was solidarity (58.82% no, 41.18% yes).  This can be explained by the amount of posts that had to do with outside sources (YouTube videos, products for sale), though the majority of creations seem to be directly displayed on the page as uploaded photos or written text.  Information sharing had a reversed trend with 60% of posts having hyperlinks.  I believe this can be explained by the fact that much of the information shared led to government or local websites or the Long Reach foundation, as these were the major sources of aid.

For the rest of the post I am going to examine the average likes, shares, and comments on the topics and explain why certain posts seem to have received more attention or certain types of attention than others. Note that in the numbers in the following charts are the averages of posts that had likes comments or shares, therefore those posts that received no attention are not included. This was done because some posts seem to have not received attention simply because of the programming in facebook that causes those posts that do not get immediate attention to be moved further down the page in groups.  Also note that series 1=average likes, series 2=average shares, series 3=average comments

Looking at the first graph, the things that jump out are the high number of average likes for memories and the high number of average shares for information seeking.  The average likes for the memory is likely so high simply because there is a smaller pool of memory posts (3 only) that we are taking our data from.  The high number of shares for information seeking is likely due to the fact that other citizens saw the importance of getting an individuals question answered and thus shared it to reach that end.  This indicates that the page was an effective informative site.

The second graph is a composite where series 1-3 are put together so one can easily see which type of post received the most overall attention.  As presented in the graph that was contemporary documentation.  This would seem to indicate that the page was not as effective in sharing information and actually aiding in recovery.  It is, however, important to take a look at the content of the attention these posts received, by far and large they seem to be focused on a sense of community building and showcasing the beauty of Long Beach despite Sandy.  So although this type of information is not directly aid related, it may be helpful in preserving the strong sense of community and keeping the spirits of Long Beachers up post Sandy, which may be valuable in itself.

I also think it is important to note that grievances received the least amount of attention.  There were 12 grievance based posts and they received the attention of under 10 individuals on average, none of which were shares so they did not spread farther than the Long Beach Hurricane Information page.  In my opinion this shows the positive focus of both the page and the citizens.