![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png){kind=small}
Hottest Summer in more than ten years
Aug. 6th, 2008 05:00 pmI caught the visual edge of one weather news show that compared the number of 100+ days we have had this year vs previous years. This year looked bad, but certainly not the worst.
What bothered me about their report was that they were comparing total number of 100+ days in an entire summer to this year... And this summer is far from over. They should be comparing the same set of time, year to year. I cannot be the only math-inclined person bothered by their analysis...
So I dug out temperature highs and lows going back to January 1997 -- as far back as I could get from my source. My source into the present only goes up to July 16. So I took a subset containing 1997 to 2008, January 1 to July 16 of each year. I have no idea why the news show didn't do this, since they clearly have this same information.
Most years have three or less. 1998 had 15, which was the largest in this dataset until... This year! We are at a statistically unusual 29 days at 100+. Wildly unusual.
For people who like percentages, this means that most years have fewer than 1% of the days up to July 16 as being over 100F. 1998 was unusual because it had 3.8%. 2008 is 7.3%. Statistically, anything over 3.3% is considered "unusually high" (reference: UCL of p control chart).
I hate summer. At least with cold weather I can wear more clothes.
What bothered me about their report was that they were comparing total number of 100+ days in an entire summer to this year... And this summer is far from over. They should be comparing the same set of time, year to year. I cannot be the only math-inclined person bothered by their analysis...
So I dug out temperature highs and lows going back to January 1997 -- as far back as I could get from my source. My source into the present only goes up to July 16. So I took a subset containing 1997 to 2008, January 1 to July 16 of each year. I have no idea why the news show didn't do this, since they clearly have this same information.
Most years have three or less. 1998 had 15, which was the largest in this dataset until... This year! We are at a statistically unusual 29 days at 100+. Wildly unusual.
For people who like percentages, this means that most years have fewer than 1% of the days up to July 16 as being over 100F. 1998 was unusual because it had 3.8%. 2008 is 7.3%. Statistically, anything over 3.3% is considered "unusually high" (reference: UCL of p control chart).
I hate summer. At least with cold weather I can wear more clothes.
