The most interesting thing about statistics

Suppose you want to poll people in a country where 1 million people will vote, and then do the same in a country where 1 billion people will vote. How many more people do you have to poll in the second country to get the same polling accuracy as you do in the first?

The counterintuitive answer: you need to poll the same number in each country to get the same accuracy. Polling error is a function of sample size but independent of the size of the whole (voting) population.

**

The intuition behind this is that there is some “true” underlying probability you are trying to sample that is independent of population size. So suppose you are trying to figure out the odds that humans on average have boy versus girl babies. You take a random sample of 1,000 people to estimate it. Does it matter that the sample is out of a population on earth of 1 billion people on earth versus 6 billion people earth? The earth’s population is just an instantiation of the true underlying probability you are trying to determine, hence irrelevant.

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: