Well, as the saying goes.. there's Lies, Damn Lies, and Statistics. But some are more self selecting than others. Some of these reports make an effort to create a non-biased pool of sites to gain data from, other just take whoever wants to join their list.. and those sites tend to be very tech heavy, which favors more non-IE results.
Let's take, for instance:
This one shows Safari with 15.2%. Wow. Safari is only available for MacOS if it's only counting desktop browsers (which it seems to be), and MacOS only accounts for about 7% of the market share. How, exactly does MacOS then have more than 2x the browser share than they have computers?
If the numbers include mobile, then that would include iOS and MacOS, but then those numbers would also include Android. And MacOS + iOS is far less than than 15% of the installed base of Android alone, much less Android + everything else that might be running Chrome (or some version of it). So those numbers, just based on the Safari stats are highly suspect.
First.. w3schools? Seriously? Apart from the fact that they're the worst joke on the internet... ( see
W3Fools ? Better web education for all ), w3schools is again very tech heavy... mostly visited by web developers which are going to be very chrome and firefox heavy. The bias is well known. In fact, look at the huge difference between safari usage in the two sites.. 15.2 vs 3.6, and chrome is almost 60% vs 40%.
Now, Statcounter is a completely apples to oranges comparison. Statcounters stats are based on page views, not on market share. In other words, they base their stats on which browsers view the most pages and do not remove duplicate ip's or try to find unique visitors.
In theory, users of chrome should view just as many pages as users of IE, but in practice we really don't know if that's the case.. since chrome users tend to be more technical, they may spend much more time surfing... or they may just visit a lot more pages than other people on a daily basis.. it's hard to tell. Their justification is that they count which browsers are the most USED, not the ones that have the most market share, and they also count users who use more than one browser.
Regardless, I've seen the browser statistics for a lot of sites, some of them quite large (Millions of dollars in revenue per day), and I've never once seen one in which chrome or any other browser came even close to surpassing IE in usage on any site I've monitored. So, simply put.. My own personal experience agrees with the net applications data, and not with the others. NetMarketShare data is also similar...
Market share for mobile, browsers, operating systems and search engines | NetMarketShare