Status:Closed    Asked:Jun 15, 2018 - 10:04 AM

Pewrt variable and decimal places

I'm confused about the variable Perwt. The description says that it is an 8-digit variable with two decimal places "This is a 8-digit numeric variable with 2 implied decimal places." and "but there is no actual decimal in the data.".

But actually, some censuses as Brazil 2010 or Colombia 2005, have decimal places in the variable Perwt. How can be this possible? Should I take just the first six numbers (before the decimal places)?

Many thanks in advance!

Do you have the same question? Follow this Question

Staff Answer


Jeff Bloem


This is a bit of a confusing detail. When the PERWT variable description mentions that there is no actual decimal in the data, this is referring to the data in the .dat file. Therefore, in order to use these weights properly, one needs to divide these values by 100. In most of the command files, however, this task is included automatically. So, when the data is read into you statistical software using an IPUMS command file, the weight values contain a decimal.


Jun 15, 2018 - 10:27 AM

Report it


In IPUMS Int'l, do the places of previous residence (e.g., MIGBR2) refer to c...
Why are not all the "not in the labour force" categories filled for the EMPST...
Contrary to documentation does EARNWEEK have two implied decimal points?
IRECODE1 Variable
Login   |   Register

Recently Active Members

View More »

Share |