First she claimed the popular vote...
Clinton says she leads in popular vote - CNN.com
She could only be ahead if no one in Michigan voted for Obama. Uh, yeah right. She was the only one on the ballot and the ballots with anyone "written in" were actually thrown away uncounted! The only other choice was uncommited and 40 some % were uncommited.
Now she (or her staffers) is claiming a double digit win in PA? Well, I as an engineer never round of significant figures from the raw data before the result. I learned this in elementary school!
CNN Political Ticker: All politics, all the time Blog Archive - Final returns in Pennsylvania: did Clinton nab a double-digit win? « - Blogs from CNN.com
CNN calls this a more "traditional" way to calculate the score. I'm calling bullshit. Either way you calculate it, it lands at 9.28717837%.
They have the exact number of votes for each candidate. Rounding to the nearest significant figure is used when there is a question about the resolution of the data (you wouldn't use a ruler to measure a thousandth of a inch for example, you'd round to the nearest 1/32nd). They have the numbers down to one voter. If CNN doesn't want to type 9.28717837%, they should round it to 9.3%. Don't round off actual voters before the calculation and call it 10%! Are they saying there are insignificant voters out there?