The Long Tail of Google News Sources...

I saw this a few days ago on Waxy.org's link page. I took the information from the url linked to the title of this post and created an Excel spreadsheet to look at the frequency distribution for these data.

Just since I threw it into a spreadsheet, I realized I would love to share a simple frequency histogram and throw out some descriptive stats (arbitrarily chosen by the author):

These numbers are as of 8PM on 3/28/2005... they are constantly changing (bastards):

1172 total sources cited
7770 total stories

Mode of 1
Median of 2
Mean of 6.629692833
Standard Deviation of 17.8597173

The top thirty sources have a minimum of 50 stories cited from their pages and make up 2.56% of the total sources cited, yet have 35.44% of the stories (2,754) cited on Google News by volume.

The top 164 sources have a minimum of 10 stories cited, making up 13.99% of the total number of sources, yet have 69.14% of the stories (5,372) cited by volume.

The bottom 506 sources have one story cited by Google News, making up 43.17% of the total number of sources, but only having 6.51% of the total stories cited by volume.

Here is a snazzy graph showing the long tail and the distribution skewed to the left with a long tail to the right...



Fun With Scraped Data... 

google news sources citations