I am restarting my science blog. I hope to write one entry a week on a current topic in science. References will be given. At this point I am keeping comments on these to friends only
( Read more... )
google flu chartseraphimsigristNovember 18 2009, 20:03:14 UTC
a question would be to what extent the presence of swine flu in addition to seasonal flu has increased the number of hits aritificially? my guess would be that any hits from your own research would be additional to your pattern of previous years. so precisely this year in particular I suspec their chart. also note how it jacked up in september when they went to a new basis somehow and yet how little (to my observation) that correlates to what I see looking around which would be that we still are early in the flu season without any noticeable spike. I am thinking of how certain computer football rankings would skew crazily at the beginning of the season and gradually level off to what human observors would estimate on team strength etc
Re: google flu charttempus_aeternaNovember 18 2009, 20:34:17 UTC
The huge spike that you see, is as you point out, most likely corresponding to the swine flu and not the normal seasonal flu. Swine flu is just another strain of flu and in most places tests are not done to see if patients have the swine flu or the seasonal flu and so both the google program and the CDC data incorporate both the swine flu and the seasonal flu data together. The google monitoring system does appear to match the CDC data with the swine flu included. However separating the seasonal flu from the swine flu is problematic. There does appear to be some issues with the CDC and google data matching early in the swine flu spike, probably because of the "scare" factor which led a lot of people to search online without having any relation to the actual occurrences of the flu. This swine flu will definitely test the system and hopefully they will do some testing on the data to see how the system responds to the "scare" and then the real change in seasonal pattern from the swine flu.
yes the scare and curiosity factor account for hits I think. but as to accounting by adding the two together my sort of sense of it is that there is no special amount of flu going around this year in total even, I can be wrong...my contact is limited etc
The rates of seasonal flu probably won't change that much, but swine flu is going around and greatly increasing the flu number especially because swine flu hits people who normally don't get flu (people in their 20s and 30s) as well as children. I had it this year and I've never had the flu before. It's just bad in a lot of places. Not as bad as the initial predictions or the scare, but a lot of people who don't get flu normally are getting it now and then will be susceptible as usual to the seasonal flu later in the season.
It's really interesting how many fields are finding google searches useful! Linguistics has been treating the internet as a massive corpus for a while now, using searches to determine how often certain phrases are used.
Comments 5
presence of swine flu in addition to seasonal
flu has increased the number of hits aritificially?
my guess would be that any hits from your own research
would be additional to your pattern of previous years.
so precisely this year in particular I suspec their chart.
also note how it jacked up in september when they went
to a new basis somehow and yet how little (to my observation)
that correlates to what I see looking around which would be
that we still are early in the flu season without any noticeable
spike.
I am thinking of how certain computer football rankings would
skew crazily at the beginning of the season and gradually
level off to what human observors would estimate on team
strength etc
Reply
Reply
account for hits I think. but as to
accounting by adding the two together
my sort of sense of it is that there is
no special amount of flu going around this
year in total even, I can be wrong...my
contact is limited etc
Reply
Reply
Reply
Leave a comment