There have been many cases to where carbon dating has not been accurate, which is why we do the test many times. If a rock is dated to be 5 billions years old 5 times and two times is some crazy number, they are not going to take the average of all the numbmers, but the most accurate ones. Its just a matter of how you look at it.
This is off-topic, but I can't help replying since you've made significant errors.
First, your first sentence has no relationship to the second two. C-14 is not used to date rocks, not is it used to date anything more than about 50,000 years old.
Second, you allegation of scientists discarding data is unsupported, insulting, and just plain false. If a rock is dated to be 5 billion years old 5 times and two times is some crazy number, the first thing scientists do is try to figure out which numbers are wrong and why. Whether or not they can figure out what's going on they will report
all the numbers, and highlight the conflict if there still is one. You should read "The Age of the Earth", G.Brent Dalrymple, Stanford University Press, 1991. Not only is it an excellent exposition of the evidence for the age of the Earth, it contains
many examples of exactly what you claim doesn't happen; scientific publications of anomalous results, and the story of how the anomalies were resolved. When scientists come up with anamolous results, they put them out publically for the entire scientific community to work on. They will not (except for a very few dishonest ones) discard data they don't like. As Isaac Asimov said:
quote:
The most exciting phrase to hear in science, the one that heralds new discoveries, is not 'Eureka!' but 'That's funny...'.