Hi IPUMS folks,
I’ve been on a quest to obtain state-level poverty stats on 4 year-olds, and I’ve been using the SDA online analysis tool to create a table showing:
row: POVERTY
column: STATEFIP
selection filters: age(4), year(2000), poverty(10)
weight: sdawt
It appears Minnesota (statefip==27) is missing when I run the table with the above specification. However, when I don’t filter by poverty==10, Minnesota does appear (and the weighted N is reported is 0). Just checking if this is right – that there was a small enough number of 4 year-olds living below 100% of the federal poverty level in 2000 in Minnesota to report as 0? Thanks in advance for any light shed on this!
I just did a similar table that should identify the same people. This table shows the number of 4-year-olds in each poverty status in each year from 1998-2002 in Minnesota. These are unweighted counts, so they give the number of people in the sample (not the estimated number in the population). As you can see in the screenshot, the number of 4-year-olds in the sample under the poverty line is very small for every year, and is actually 0 in 2000. So that is why you are not seeing them show up in your tabulation.
More generally, a single age group in a single state in a single year will have a very small sample in the CPS. I recommend looking at IPUMS USA instead for this type of analysis. For the year 2000, you can use the 5% census sample, and for later years you can use the ACS, which is a 1% sample of the population each year. These have substantially larger sample sizes than the ASEC, and you can also use the SDA tool with IPUMS USA data. Note that POVERTY is defined differently in IPUMS USA (giving the family income as a fraction of the poverty line), so you’ll want to add an option to recode the poverty variable, for example:
poverty(r: 0-100 "Below" ; 101-* "Above")
1 Like
This is helpful. Thanks Matthew!
