Furthermore, no statistically significant climate change occurred in the cities of Anchorage, Nashville, Miami nor Salt Lake City, all of which were included in the study to add further regional diversity.
Null Hypothesis
During the period 1995 to 2009, there was no statistically signficiant correlation between time and changes in average daily temperatures in any of the ten most populated cities in the United States.
Statistical Correlation Measure
Pearson product-moment correlation coefficient (denoted by r) measures the linear correlation between two variables X and Y, giving a value between -1 and +1. The closer the r statistic is to +1, the greater the positive correlation between two variables. The closer the r statistic is to -1, the greater the negative correlation between two variables.
For example, in the instant study, an r statistic of +1 would indicate a perfect correlation between time and increases in average daily temperatures. Similarly, an r statistic of -1 would indicate an exact correlation between time and decreases in average daily temperatures. An r statistic of 0 indicates no statistical correlation.
Data
X is 1 to 169 ascending representing the first day of each month from January 1, 1995 to January 1, 2009. Y is the average daily temperature on the first day of said months in the ten most populated U.S. cities: (1) New York City; (2) Los Angeles; (3) Chicago; (4) Houston; (5) Phoenix; (6) Philadelphia; (7) San Antonio; (8) Dallas; (9) San Diego; and (10) San Jose (San Francisco Bay Area).
As mentioned above, data and analysis has also been added for the cities of Anchorage, Nashville, Miami and Salt Lake City.
A link to all of the data used herein is provided in this article's sources. A value of -99 indicates that no data exists for that particular day. Data for those days can either be omitted under the 5% rule or included by imputation. Imputation is used herein.
Critical Value Range
(.15) to .15* Interpretation--in order to reject the null hypothesis above, the r statistic for any U.S. city must either be < (.15) or > .15. Any r statistic between (.15) and .15 indicates no statistically significant correlation between X and Y.
*(167 degrees of freedom at an alpha of .05).
Findings
New York City: r = .11
Los Angeles: r = (.02)
Chicago: r = .05
Houston: r = .07
Phoenix: r = .06
Philadelphia: r = .09
San Antonio: r = .09
Dallas: r = .09
San Diego: r = .01
San Jose (San Francisco Bay Area): r = (.11)
Anchorage: r = (.01)
Nashville: r = .06
Greater Miami Area: r = .05
Salt Lake City: r = .03
Conclusion
The null hypothesis cannot be rejected because no r statistic falls outside of the critical value range. No statistically signficant climate change occurred in the United States during the period 1995 to 2009; that is, there was no statistically significant correlation between time and changes in average daily temperatures in any of the 14 major U.S. cities analyzed in this study.
Source(s):
"Average Daily Temperature Archive," University of Dayton
Published by J.C. Grant
A writer interested in education, finance, health, history, law, music, polemics, politics, satire, sports, statistics, travel, and trivia. View profile
- Depression Support Groups in Salt Lake City, Utah Your guide to depression support groups in the Salt Lake City, Utah area and beyond.
-
Best Kept Secret of Salt Lake City: It's Not Just for Mormons Anymore
Living in Salt Lake City for two years gave me a whole new perspective about Utah, Mormons, and fun.
-
The Wells, Nevada Earthquake Shook My Salt Lake City Apartment
I was sitting in Salt Lake City at my computer writing an article for AC when my chair, my computer and the upstairs apartment floor started shaking.
- Best Irish Pubs and Irish Restaurants in Salt Lake City Utah Check out St. Patrick's Day in Salt Lake City with green beer specials, live bagpipe music performances, best places to eat Corned Beef & Cabbage or Sheppard's Pie, and where to take the little leprechauns.
-
Best Sports Bars to Watch the NBA Playoffs in Salt Lake City, Utah
Lumpys and Port O'Call are among our favorite Salt Lake City sports bars that we will visit to eat, drink, and enjoy the NBA Playoff games.
- Climate Change and Its Effects on the Poor and Disadvantaged Groups
- Correlation, Causation, Relation
- Statistics: Hypothesis Testing and Correlation
- Correlation vs. Causation: Differences Between Likely and Direct Causes
- Job Relocation Costs: Moving from New York City to Los Angeles
- Unions in the United States: An Overview
- A Few Hours in Salt Lake City - Perusing Sam Weller's Bookstore
|
|
24 Comments
Post a CommentExactly, stas.
Increasingly the scientific evidence of the 21st century is that the whole CO2 GHG, sky-is-falling idea, is disproved.
That is why Climategate happened. They would not have had to resort to all the nefarious lying, cheating, stealing, data destruction, character assassination, and suppression of the evidence, if the scientific results were truly bad, as they advertise.
It was because it increasingly appeared as the results of 21st century measurements, unavailble heretofor, that there was no Warming problem, that they had to do so.
Ask yourself, Why else LIE??.
Thomas Ross: "Enough said" is right. Stick to what you know, which as far as I can tell, is very little.
I didn't suggest I was right. I just don't believe you are right either. I suggested you put your numbers to the test and get published.
Enough said.
Thomas Ross: Your post is absurd. Your "average" temperatures yield an r value of .24 at 11 degrees of freedom. The Critical Value at an alpha of .05 is .55. It's not even close to being a statistically significant correlation. Parenthetically, Twain was speaking for the uneducated; count yourself among them.
I just did a quick use to the data to look at the yearly average for LA for the same time period:
1995 59
1996 63.29863388
1997 64.30082192
1998 61.79307479
1999 61.94862637
2000 62.39808743
2001 61.11178082
2002 61.30415512
2003 62.58539945
2004 62.64617486
2005 62.34657534
2006 63.02082192
2007 62.04340659
2008 62.97945205
Looks like an upward trend to me. But that proves little. As Mark Twain is attributed with saying there are 'lies, damn lies, and statistics'. The point being you can prove just about anything you want with the right twist of statistical analysis.
Good science requires more than just a look at some data. It requires a thorough, peer-reviewed, analysis of lots of data.
I suggest that if you wish to partake in the science of climate change, get published in a real journal, not just self-published on the web. On the web, anyone can say anything, that makes it opinion, not fact.
Thanks for the post, Stan B.
"My guess is that" - yes, that's right mj, we need to re-jigger the entire world energy system based on your "guess." Why do we insist on meeting scientific studies with opinion and conjecture and assume the latter stands on the same footing as the former - as if your guess invalidates the entire thrust of this study.....
GENERAL POST: Notice that not one person has come forward with a contrary study in response to either part 1 or part 2 despite being provided with links to the data used. The funniest theoretical criticism I've seen by one left-wing grad school kid is that you should detrend seasonal TEMPERATURE data before conducting a linear correlation study--postively hilarious. To belabor the obvious, if you detrend the data the linear correlation will be even smaller.
And now Al Gore is up for an Academy Award. I knew all along it's balderdash.