|  | 
			
			
			
					
					
			
				
					
						
	Nor do you know the exact number of murders, so there.
		
			
			
				
					  Originally Posted by TheSpoonald   Yeah, but you can't do that unless you know how many 100,000's there are, which you don't.  So the rate is bunk. 
 So you estimate number of murders, then you estimate population, then you calculate murders / 100k.
 
 
 
 
	There are definitely people who don't show up on the census though, right? Are these more likely to be legals or illegals?
		
			
			
				
					  Originally Posted by TheSpoonald    Census data does include illegals.  It's currently illegal to ask citizenship data on the census.  So the illegals are included in total headcount.  We just don't know how to differentiate them from the native population.  The 330M remains constant. 
 
 
 
	Sure, and if there's more blue-eyed people in the country than you thought, the number of crimes committed by blue-eyed people is higher than you thought. Doesn't mean any given blue-eyed person is more dangerous than they were before you got better at counting them.
		
			
			
				
					  Originally Posted by TheSpoonald   And even if your math is right, which it isn't, my point to MMM would still be correct.  And that point was that the yale data means that the number of crimes committed by illegal immigrants is higher than you thought. |