@paranoia destroys ya:
"Colbert made some observations last night on terrorism. More Americans are killed by falling furniture."
I think the claim was first made on AllGov, but it has been picked up by the blogosphere and widely repeated so it is difficult to identify the source.
Unlike most of the blogosphere, AllGov at least gives their sources. From which -- if you are one of the few people who actually checks sources -- you can see that the claim is substantially misleading, in four main ways.
Firstly, nearly all of the victims of toppling appliances and furniture are children under the age of 8. You might argue that this makes it worse (more years of lost life, etc.), but it thoroughly distorts the *perception* of the statistic. We are meant to think "gee, killed by falling television, that must be extremely rare" -- and for adults, IT IS. But adult deaths by toppling appliance / furniture are not even in the ballpark with adult deaths by terrorism.
Secondly, their figure for Americans killed by terrorist attacks is based on “private citizens” only. This figure omits not only all military personnel, and all government employees, but even employees of other companies engaged on government business.
Thirdly, the two periods being compared are not the same. Although the data were readily available to report the same periods, "for some reason" they chose two slightly different periods such that they omitted the worst year for terrorism. Now it would be reasonable to claim that 9/11 was a special case. You might claim it was a "once in a hundred years" event, and amortize it over the whole century (but won't the next “once in a hundred years” event be even worse?) Or you might claim it could not have occurred before the development of wide-bodied jets, and amortize it over 30 years. Or you could claim that such a case is just too hard to handle, and make a note that you are omitting it, which would be honest but rather prick the sails of your argument. What you cannot do is just subtly omit it without telling anyone; that is academic dishonesty.
On just these 3 points, you could easily turn this spin right back around into “adult Americans were nearly a hundred times more likely to be killed by terrorists than by falling heavy objects in the home and workplace”, but that just doesn't sound like a very remarkable statistic, and it certainly isn't as funny. But the fourth point is by far the most remarkable:
Their source for terrorism fatalities clearly shows that annual deaths due to terrorism declined sharply almost every year since the “War on Terrorism” was begun!! (I really didn't expect that!) The per capita, per annum risk of a US citizen being killed by terrorists is now – but formerly was not – 35 times lower than the world average. You may or may not think that was worthwhile, but it is certainly a remarkable piece of sophistry to turn that source data into this joke.