10 March 2008

Death rates

How many people in the world do you think will die in the next day?

See here for a psuedo-hint, which is the somewhat easier related question which inspired this.

See here for the answer, from the U. S. Census Bureau.

4 comments:

.mau. said...

my own estimation (400,000 deaths) was wrong by a factor of nearly three.
The strange things is that I supposed that the world has 8 billion people, and a person would live for 55 years, that is 20,000 days. Of course they are rough estimates, especially in thinking that there is no distribution at all in life expectancy, but the error seems large. Any idea where I have been wrong?

Anonymous said...

Neat, based on the pseudo-hint, I took the approximate world population (6B), divided it by 100, and then divided it by 300, and got 200k for my rough estimate, and then pulled out a calculator and divided by 365 for a more refined estimate, and turned out to only be out about 10%. Not bad for an order of magnitude answer :)

Joshua Zucker said...

mau,
The world only has a little over 6 billion people, and more of them are young than old (population is still growing, after all) and in China and India (which is all that matters :) the life expectancy is well over 55 years (more like 60-70: and when you average in the next biggest country, the US, you start to get close to 70).

So I think you had two smallish factors because of overestimating world population and underestimating life expectancy, and then a larger factor because of neglecting the age distribution of the current population. All three of those factors point the same direction, which is why your error was surprisingly large.

.mau. said...

Joshua,
the U.S. Census Bureau ( here) estimates the present world population at 6.65 billion.
Current world like expectancy, according to UN (here) is 67.2 years, that is 24544 days.
Indeed with those data my estimate would have become ~270000 deaths per day. I still find it strange that the distribution of ages of death may account for a factor of nearly 2!