In the first four parts of this series it was established that no statistically significant climate change–neither global warming nor cooling–occurred between 1995 and 2009 in the United States, any region of Africa, Australia, or Asia . The following represents the fifth part of this continuing series-an analysis of Europe.
As established in the correlation study below, no statistically significant climate change occurred in Europe between 1995 and 2009. Average daily temperatures in Copenhagen, London, Madrid, Moscow, Munich, Paris, and Rome serve as supporting data for this study.
Null Hypothesis
During the period 1995 to 2009, there was no statistically sigificant correlation between time and average daily temperatures in any of Europe’s major cities.
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 a decrease 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 cities of Copenhagen, London, Madrid, Moscow, Munich, Paris, and Rome.
A link to all of the data used herein is provided in this article’s sources. Negative 99 represents no data available for a given day and requires either omission under the 5% rule or imputation. Neither method makes a material difference to the r statistic. Imputation was used herein.
Critical Value Range
(.15) to .15* Interpretation–in order to reject the null hypothesis above, the r statistic for Copenhagen, London, Madrid, Moscow, Munich, Paris, or Rome must be .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
Copenhagen, Denmark: r = .11
London, England: r = .05
Madrid, Spain: r = (.04)
Moscow, Russia: r = .06
Munich, Germany: r = .08
Paris, France: r = .02
Rome, Italy: r = .10
Conclusion
The null hypothesis cannot be rejected because no r statistic falls outside of the critical value range. No statistically significant climate change occurred in Europe 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 European cities analyzed herein.
Source(s):
“Average Daily Temperature Archive,” University of Dayton
