* 1.3 billion United States
* 570 million Brazil
* 280 million Turkey
* 260 million Germany
* 250 million Thailand
* 240 million China
* 240 million United Kingdom
* 180 million Indonesia
* 160 million Canada
* 140 million India
* 109 million Russia
StatCounter is a global service with members from every country in the world but it is more popular in certain countries. Therefore we collect more data from these countries. This means reports for Worldwide/Asia/Europe/North America/Oceania/South America can be skewed towards the countries where StatCounter is more popular. However, you can view each stat on a country level which negates any potential for a bias.
We could compensate for this bias in our reports for Worldwide/Asia/Europe/North America/Oceania/South America based on the total internet population size from each country. But this would mean manipulating the data, and we prefer to leave the data untouched. If other people want to build upon this data and manipulate it to account for any regional bias then they are very welcome to do so per our licensing agreement http://gs.statcounter.com/about.
We talk about our methodology here.
Our stats are based on aggregate data collected by StatCounter on a sample exceeding 4 billion pageviews per month collected from across the StatCounter network of more than 3 million websites. From this sample we analyze the sources of the referring traffic to compile our search engine reports.
Slashdot Mirror
Hi,
I work at StatCounter and I would just like to point out that we have a very diverse sample size from around the world.
As per http://gs.statcounter.com/faq#sample-size for July 2009 here was the breakdown of our sample pageviews for the month.
* 1.3 billion United States
* 570 million Brazil
* 280 million Turkey
* 260 million Germany
* 250 million Thailand
* 240 million China
* 240 million United Kingdom
* 180 million Indonesia
* 160 million Canada
* 140 million India
* 109 million Russia
StatCounter is a global service with members from every country in the world but it is more popular in certain countries. Therefore we collect more data from these countries. This means reports for Worldwide/Asia/Europe/North America/Oceania/South America can be skewed towards the countries where StatCounter is more popular. However, you can view each stat on a country level which negates any potential for a bias.
We could compensate for this bias in our reports for Worldwide/Asia/Europe/North America/Oceania/South America based on the total internet population size from each country. But this would mean manipulating the data, and we prefer to leave the data untouched. If other people want to build upon this data and manipulate it to account for any regional bias then they are very welcome to do so per our licensing agreement http://gs.statcounter.com/about.
We talk about our methodology here.
Our stats are based on aggregate data collected by StatCounter on a sample exceeding 4 billion pageviews per month collected from across the StatCounter network of more than 3 million websites. From this sample we analyze the sources of the referring traffic to compile our search engine reports.
I'm from StatCounter and I would just like to address your concern. The detection for baidu was added on the 5th March 2009 at 21.00 GMT. When a new detection is added it is noted on the visual graph (but not in the csv download).
Also if you look at the stats just for China you can easily see Baidu's dominance there.