Scaling down a proportion

41 Views Asked by At

Is it statistically incorrect to scale down a proportion? For example, if I have data stating that the average household in the US has two cars, would it be wrong to assume that each household in a state, say, Virginia, has two cars on average? If not, how can I scale down? Obviously, I could look up the population of the US and the state if it would be helpful, but I am not sure if using there is a correct way to scale down.

1

There are 1 best solutions below

0
On BEST ANSWER

I am fairly certain it is statistically incorrect to scale these statistics down, because Virginia might only have one car on average and their lack of cars is being compensated by the excess of cars in a state like California where there are maybe three cars on average per household. (California would be overcompensating here since there are a lot more households in California, but the point stays the same.)

As for looking up information about the population of the US and Virginia I don't think it would help much. You'd need extra data specifically related to the number of cars per household in Virginia.

When averaging you fundamentally lose information and there isn't really a way to reverse engineer the data from looking at averages.

I hope this answered your question.