Understanding Browser Usage Share Data
StatCounter Global Stats Fact Sheet
Sample size
When compiling statistics, larger sample sizes generally lead to more reliable results. Our stats are based on over 5 billion page views per month recorded across more than 1.5 million websites. We are not aware of any other publicly available data source with a larger sample size.
Browser Usage Stats based on Page Views
StatCounter measures internet usage trends. We track which browsers are actually used most. To accurately measure browser usage, we base our stats on page views (and not unique visitors). This means we take account of how frequently browsers are used and we also track multi-browser usage by individuals.
Geo weighting
Geo weighting involves artificially modifying data for individual countries when compiling stats on a worldwide basis.
The term "geo weighting" may appear impressive at first glance, however the reality can be quite different. Geo weighting data using inappropriate weights can erode the integrity of the initial information such that the output becomes meaningless.
We do not inflate any individual country stats when calculating overall worldwide stats. This is a deliberate decision due to the absence of any up-to-date, precise and consistent data source* for internet usage with which to accurately weight our data. We will not distort our stats by the application of unsuitable and out-of-date weights.
Furthermore our stats are based on a geographically diverse sample of over 5 billion page views per month across more than 1.5 million global websites. This means we have large enough sample sizes in the majority of countries to be confident that our stats are broadly representative of real world figures.
For the avoidance of doubt, geo weighting, if used, should only be applied to worldwide (or similar) data where stats for several countries are aggregated. Geo weighting is not relevant for individual country stats.
* The CIA Factbook (an oft-cited source for internet user data) conflates mobile internet users with desktop users. This renders the data wholly inappropriate for application to desktop browser stats.
Google Chrome Prerendering
From 1 May 2012, prerendered pages (which are not actually viewed) are not included in our stats. We have not adjusted any statistics published prior to this date.
Note that prerendering had no discernible impact on our stats when introduced in June 2011 or when removed from our stats in May 2012.
Independent
No conflicts of interest. StatCounter Global Stats does not receive payment from any of the organizations that we report on.
Transparent
We publish full information about our stats and methodology. Check our detection. Examine our individual country sample sizes. Ask us anything – we'll provide a comprehensive answer.
Free
No commercial interests influence our stats. All stats are available free of charge.
Primary Source
A primary source provides original data. Our stats are not subject to any third party influence or modification. Instead, our stats are collected, reviewed and published directly by us.