It's called mathematics, you were supposed to learn it in school.
There are any number of ways to calculate DNA similarities among species and somebody who reports 98% DNA similarity between two species uses one set of assumptions, somebody else reporting, say, 80% is using different assumptions.
Both can be "correct", but the point of the number is to illustrate how closely related the species are and if we are talking about humans & chimpanzees, then what matters most is that no other living species is more closely related to humans than chimps.
In other words: 80% might be valid provided you don't somehow find that, say, elephants are 90% similar.
Then we'd know something is rotten in Denierland.
Kalamata: "Not even close, Joey.
Even when using the old, fabricated number of 98.8% genetic similarity, there are about 35 million differences in our genomes.
That’s 35 MILLION, Joey."
Sure, and there are millions of DNA differences among different human beings too.
But that's out of, what is it, 3.3 billion base pairs?
And when they compared human to ancient Neanderthal DNA they found Neanderthals were much closer to humans than to chimpanzees.
That makes perfect sense when you consider the timeline of evolution as revealed in thousands of fossils.
