10X increase in 'Weird Weather' discussion

Introduction:

In "normal times", the ratio of time people spend talking about 'weird weather', to time spent speaking of a 'farmers market', should remain relatively constant.

In the event that actual "weird" weather was occurring there should be a marked increase in the ratio of 'weird weather' conversation vs. 'farmers market' converstation.

'Farmers market' conversation was chosen at random as a constant; farmer's markets being a fairly universal topic.

Hypothesis:

More people are talking about weird weather these days.

Procedure:

Data was collected through "Google" searches, for example:

'1900 "farmers market"' returns 87900 links
'2000 "weird weather"' returns 25800 links

Yearly data was collected for 1990 through present. Once per decade data was selected 1900 through 1980. 1951,1949,1911, and 1909 data was included to help clarify spikes felt at 1950 and 1910 respectively. A ratio of 'farmers market' to 'weird weather' data was then established; 1990's value set to 1. All data was collected 01/12/2006.

Data

YEAR WW FM Ratio

1900 516 87900 1.29
1909 337 18400 4.04
1910 702 34100 4.54
1911 351 26600 2.91
1920 215 44600 1.06
1930 214 54400 0.87
1940 317 41300 1.69
1949 376 31800 2.61
1950 2420 62300 8.57
1951 448 31400 3.15
1960 559 53900 2.29
1970 662 68400 2.13
1980 739 90700 1.80
1990 612 135000 1.00
1991 540 109000 1.09
1992 621 117000 1.17
1993 517 114000 1.00
1994 602 291000 0.46
1995 992 196000 1.12
1996 9130 257000 7.84
1997 10200 223000 10.09
1998 13700 240000 12.59
1999 14600 329000 9.79
2000 25800 591000 9.63
2001 19700 512000 8.49
2002 22300 559000 8.80
2003 27700 747000 8.18
2004 33400 1440000 5.12
2005 89600 2210000 8.94
2006 60600 1180000 11.33

Results:

In the period of from 1996 to present there is a TENFOLD increase in the co-mentioning of a given date and 'weird weather' as compared 'farmers market'.

There are also lesser spikes in the data at 1950 and 1910.

I can include a logarithmic excel graph that makes all of this more clear if someone could instruct me as to how to upload it. The graph shows a clear change in conversation occuring with regard to the last decade.

Conclusion:

The hypothesis seems to hold water.

Recommendation based upon results:

Head for the hills. Weird weather is here to stay.


-Sri Oracle



[edit on 12-1-2006 by Sri Oracle]

Are you sure it's not just a lack of casual documents cataloged on the internet prior to 1996? I mean, if you're finding links to documents from 1909, chances are they're some kind of official archive and not an indication of what people were actually discussing.

Why not just use the National Weather Service archives if you want to find stats on unusual weather?

Originally posted by mythatsabigprobe
Are you sure it's not just a lack of casual documents cataloged on the internet prior to 1996?

If it was just a lack of documents there would be an equal lack of documents regarding 'farmers market' creating balance.

What is being presented is the RATIO between occurances of 'farmers market' and 'weird weather'. That ratio has shifted over the past decade by a magnitude of TEN.

I mean, if you're finding links to documents from 1909, chances are they're some kind of official archive and not an indication of what people were actually discussing.

Correct, no internet back then. Archive of discussion. Some about apples, some about oranges, some about farmers markets, and some about weird weather.

From 1900 through 1996 the ratio of discussion about farmers markets to discussion of 'weird weather' remained relatively constant. For the past 10 years, there has been ten times as much 'weird weather' discussion, during any given year therein.


Put another way: If discussion was logged on the internet regarding the occurances of any year in the past decade, it is ten times more likely that the discussion mentioned 'weird weather' than if that discussion had referenced any other year in the 20th century.

*I can tell I'm going to loose people here...*

Why not just use the National Weather Service archives if you want to find stats on unusual weather?

An order of magnitude in the frequency of the public's discussion shows growing global intuition of something out of the ordinary occuring. It is another perspective on the situation.

Sri Oracle

[edit on 12-1-2006 by Sri Oracle]

Very interesting theory and experimentation method

One thing you might want to take into consideration though, how often do you hear people talk about farmer's market in any context these days? I personally rarely hear it at all, and never in conversational speech, and it may have nothing to do with weird weather at all. Our society is (and has been) moving away from a more agricultural focus, and most people probably don't have that phrase in their vocabulary anymore.

Like I said, just something to consider. You might want to think about choosing a phrase that would be a bit more common across demographics like age (there's a lot more young people using the internet these days; wierd weather might come out of a 15 year old's keys a couple hundred times before farmer's market) and decades (regardless of any climate changes, the populous isn't as concerned with agriculture as it had been in previous times.)

Just a thought. Again though, kind of a neat idea for the experiment.

Year - Children - Farmers Market
1900 3.38 1.29
1909 7.78 4.04
1910 10.08 4.54
1911 6.77 2.91
1920 2.10 1.06
1930 1.87 0.87
1940 2.35 1.69
1949 3.26 2.61
1950 12.24 8.57
1951 4.37 3.15
1960 2.56 2.29
1970 2.23 2.13
1980 1.88 1.80
1990 1.00 1.00
1991 1.02 1.09
1992 1.08 1.17
1993 0.86 1.00
1994 0.83 0.46
1995 1.03 1.12
1996 8.58 7.84
1997 9.64 10.09
1998 12.56 12.59
1999 11.86 9.79
2000 14.86 9.63
2001 11.28 8.49
2002 12.38 8.80
2003 13.06 8.18
2004 12.72 5.12
2005 16.28 8.94
2006 22.89 11.33


Using "children" instead of "farmers market" as the denominator still shows a 10X spike at 1996-current for "weird weather"

Sri Oracle

Ok...so I did exaggerate a bit... not 10X but 3.89X

Still... much more common to be discussing 'weird weather' with regard to this past decade than with regard to any other.

These are the results when you average the ratio's of "children to WW" with "farmers market to WW". The numbers should be more stable:


1900 2.336767727
1909 5.91237105
1910 7.3093073
1911 4.839451336
1920 1.58371971
1930 1.370262085
1940 2.019378164
1949 2.933673931
1950 10.40239893
1951 3.756377695
1960 2.422165813
1970 2.183219305
1980 1.838014534
1990 1
1991 1.058359754
1992 1.127817198
1993 0.93108872 average stnd dev
1994 0.641989352
1995 1.072374584 2.880986168 2.568879119

2.88 is the average ratio during the 1900=1995 era


1996 8.205895043
1997 9.864849972
1998 12.57429971
1999 10.826613
2000 12.24568421
2001 9.881969694
2002 10.58770531
2003 10.62113791
2004 8.917701582
2005 12.61014717
2006 17.10697955 11.22208938 2.41661942

11.22 is the average ratio 1996-present


11.22/2.88 = 3.895225011 increase in average 1900-1995 compared to 1996-2006