March 29, 2013 at 11:29Internet Explorer vs. The number of homicides in the USA
- Transfer
The thesis that “ correlation is not equal to a causal relationship ” was again illustrated in late January by a wonderful graph comparing the number of homicides in the USA and the share of Internet Explorer among browsers that users prefer.
According to the author of the chart: “Of course, I know that correlation does not mean a causal relationship, but it still seems to me that this is definitely something ...”
In fact, of course, we have before us an example of a funny (and provocative) correlation. His amusement is how cleverly the correlation between two completely unrelated meanings was noticed; the trouble is that he is trying to compare percentages and absolute values.
But isn't it interesting how all these meanings are true?
The number of murders in the United States can be verified with Wikipedia (although their number per 100,000 is indicated there), the popularity of IE can be found in the annual reports of W3Schools . Interest on W3Schools was slightly less than on the original schedule, and to determine the population, I had to visit the US Census Bureau.
Below are two graphs, one for 2003-2011, and the other for 2007-2011, which uses an approximation of the data of the original graph. As you can see, the original numbers are very accurate.
However, as was noted at the beginning, the comparison was initially incorrect, so a couple more graphs were made. In the first case, data were taken for 2003-2011, in the second - for 2007-2011.
As you probably already noticed - unfortunately, despite the fact that both broken lines have a similar tendency to decrease, the rate of change is still noticeably different.
According to the author of the chart: “Of course, I know that correlation does not mean a causal relationship, but it still seems to me that this is definitely something ...”
In fact, of course, we have before us an example of a funny (and provocative) correlation. His amusement is how cleverly the correlation between two completely unrelated meanings was noticed; the trouble is that he is trying to compare percentages and absolute values.
But isn't it interesting how all these meanings are true?
The number of murders in the United States can be verified with Wikipedia (although their number per 100,000 is indicated there), the popularity of IE can be found in the annual reports of W3Schools . Interest on W3Schools was slightly less than on the original schedule, and to determine the population, I had to visit the US Census Bureau.
New Charts
Below are two graphs, one for 2003-2011, and the other for 2007-2011, which uses an approximation of the data of the original graph. As you can see, the original numbers are very accurate.
In percents
However, as was noted at the beginning, the comparison was initially incorrect, so a couple more graphs were made. In the first case, data were taken for 2003-2011, in the second - for 2007-2011.
As you probably already noticed - unfortunately, despite the fact that both broken lines have a similar tendency to decrease, the rate of change is still noticeably different.
