Status:Closed    Asked:Mar 24, 2018 - 01:45 PM

Puma Adjustments--Hurricane Katrina in Louisiana

On IPUMS website, it says "in the 2006-2011 ACS, persons living in Louisiana PUMAs 01801, 01802, 01905 were all coded as living in Louisiana PUMA 77777. Due to the effects of Hurrican Katrina, the original three PUMAs no longer had sufficient population to be separately identified."

However, I found 16184 observations with PUMA code 1801, 12344 obervations with PUMA code 1802, 0 observation with PUMA code 1905, and 3179 observations with PUMA code 77777 in 2008 3-yrs ACS.

I don't understand why PUMA 1801 and 1802 still exist but PUMA 1905 disappears. Should it be that all these three PUMAs are grouped into 77777? Does it mean the observations in 77777 all comes from 01905?

Many thanks for any clarifications!

Do you have the same question? Follow this Question

Staff Answer

The observations you are counting for PUMAs 01801 and 01802 are the number of observations for all PUMAs 01801 and 01802 in the US. PUMA is state-dependent. The codes must be read in combination with one of the STATE variables (STATEFIP or STATEICP). By not specifying the state of Louisiana (STATEFIP == 22), your count includes observations from Florida, Missouri, Pennsylvania and Washington. Once you analyze PUMA with one of the STATE variables, you will find that, as indicated in the documentation, persons living in Louisiana PUMAs 01801, 01802, and 01905 were all coded as living in Louisiana PUMA 77777.


Mar 26, 2018 - 09:19 AM

Report it


Number of PUMAs in ACS 2014 is greater than the number of PUMAs in the shapef...
Preselect MULTYEAR when selecting PUMA for multi-year ACS?
Correspondence between Puma 1990 and Puma 2000
PUMA codes assigned to all states [ACS 2012 and 2013 (1% sample)]
Login   |   Register

Recently Active Members

View More »

Share |