Tinnitus is sometimes caused by depression. Depression can be seasonal, and there's another vaguely cyclical graph that follows the seasons of the Northern Hemisphere. Spiking in the months after Christmas (bills) and another big one in late spring / early summer (student exams?) and then the onset of winter. There's a big reduction over the summer, and interestingly, a negative spike right at Christmas. Lately, there might be some confounding searches for economic depression -- at least that's how I interpret the latter half of 2008.
Stress and depression are also strongly correlated with religion, but less so with God or prayer. There's a drop in the searches for religion around christmas...
There are different patterns in all three regions, but the correlation holds.
I took shoulder as a neutral term, to be sure the variations were not related to some global trend in search volume.
-
Surpisingly, though, the Romney speech spike is bigger in worldwide queries than in US specific ones.
At the world scale, the spike is also very correlated with feet, flower and shoulder...
It's not the case in the US... And the Romney spike is much smaller when compared to the average frequency of each term.
Ok, double weirdness. Earlier I was seeing spikes in late 2007 for almost any query.
At first I thought it was confirming my theory because that's when the housing bubble was really popping. But it was also there for control words, including "the". So I reported a bug in their forum.
But now that's all gone. Either that was speedy corrective action or a transient problem.
I see a general trend on all queries peaking around March and late November. This matches what I see on retail websites I have access to the backend of.
Searches like money, sex and music are quite flat but appear to have a low profile match to this general trend.
"sun cream" and "snowman" are clear counter examples.
Looking at the detailed trafic, year per year, I can't see any such tendency. Some years, there are some bumps for sex around November or March, but never both, and their amplitude isn't bigger than the seemingly random fluctuations.
It's gone here too (I'm from Belgium). Edit: it's back again, but not if you look at the year 2007 specifically.
The regional differences in seasonal patterns are interresting, I think.
They all exihibit a major drop in the summer months. On top of that there's a 3 month periodicity in Australia and a 1.5 month periodicity in the UK. In the US, there's a drop mid november each year.
Are there English or Australian citizens around to explain what happens? I assume it's related to national holidays, but it would be nice to have a confirmation.
This is a US specific plot, but the world wide one is similar. There are plausible explainations for a correlation between the type of breakfast and the school cycle.
The spike at the end of 2007 appears for any queries when you look at the world-wide data.
Cold and flu season: head congestion, medications which may affect the ears.
To test weather-based theories: compare northern hemisphere and southern hemisphere search trends. (Unfortunately, there's not enough search traffic on [ringing ears] in southern-hemisphere english-speaking regions to be sure... though the thin Australia data is vaguely consistent with the idea their highest traffic is mid-year.)
Whenever I get bad sinus/ear infections during winter my ears ring.The ears/face are very closely related and a simple cold/flu can spark an infection.
http://en.wikipedia.org/wiki/Tinnitus#Causes_of_subjective_t...
"Ear infection" has a similar pattern. Although there I'm baffled by the huge spike exactly at Christmas followed by nothing in January.
http://www.google.com/trends?q=ear+infection&ctab=0&...
Tinnitus is sometimes caused by depression. Depression can be seasonal, and there's another vaguely cyclical graph that follows the seasons of the Northern Hemisphere. Spiking in the months after Christmas (bills) and another big one in late spring / early summer (student exams?) and then the onset of winter. There's a big reduction over the summer, and interestingly, a negative spike right at Christmas. Lately, there might be some confounding searches for economic depression -- at least that's how I interpret the latter half of 2008.
http://www.google.com/trends?q=depression&ctab=0&geo...