Question

Status:Closed    Asked:Aug 10, 2017 - 03:53 PM

I'm trying to run a query for EDUC for all the counties in state of Louisiana from the years 2001-2008

When I run the table, the counties don't show up correctly (just the countyfips number)so I have no idea whether this is giving me information on the entire United States or if this is a crosstab of the state of Louisiana with some random countyfips assigned. Is there a different way that I should be running this query? I'm using the ACS dataset. Sorry for the very basic question, I am new to using this database system and the codebook/getting started pages were no real help.



 
Do you have the same question? Follow this Question
 

Staff Answer

avatar

Jeff Bloem

Staff

Note that counties (either via COUNTY or COUNTYFIPS) are state-dependent. This means you need to control or filter for a particular state when analyzing IPUMS USA data at the county level. It sounds like you might be using the online analysis program. If so, you can generate a table by using the following:

Row: EDUC
Column: COUNTY
Selection Filters: STATEFIP(22)

I hope this helps.

 

Aug 10, 2017 - 04:04 PM

0
0
Report it

Voted Best Answer

Ah, yes, actually not all counties are identifiable in all years in public use data. This is due to confidentiality restrictions. You can see what counties are available in which samples via this spreadsheet. This spreadsheet, or this resource page, will also help in understanding what county each COUNTY code represents within specific states.

 

Aug 10, 2017 - 04:18 PM

1
0
Report it

Answers

Thank you for your assistance. The state of Louisiana has 64 parishes and the rumber of columns obtained by your suggestion only gives me 11 columns and they don't start with the leading state FIPS number (22---). Is there another function for this?

 

Aug 10, 2017 - 04:12 PM

0
0
Report it

That makes much more sense. As a follow up, is there a quick way to see which data sources collect EDUC so that I can add it to my table? I’m trying to develop a panel dataset with 37 parishes in Louisiana and won’t be able to run a panel regression model with that much missing data Since the PUMS data only provides information for 10 of the 37 that I need for my research.

 

Aug 10, 2017 - 04:32 PM

0
0
Report it

You can look at the Availability Tab of the EDUC varaiable, but this will only give you information about the availablity of EDUC within the IPUMS USA project, which only provides US Census data. Thus, analysis at the county level for EDUC will be limited.

Another option, if you are looking for aggregated data at the county level is the IPUMS NHGIS project, which provides aggregated tables of US Census data. Because this data is aggregated, rather than of individual records, the limiting confidentiality restrictions don't apply. I think you may find this resource very useful.

I hope this helps.

 

Aug 10, 2017 - 04:39 PM

1
0
Report it

OTHER QUESTIONS NEEDING ANSWERS

Does anyone have sample code for using svydesign function in R?
Which variables should I use in my data extract for determining father-son pa...
What "Psychiatric, Nursing, and Home health aid" occ 3600 encompasses?
Difference in Sample Size between ACS 2016 and Census Technical Documentation...
Incomplete alphabetic variable list and broken IPUMS group table in thee CPS.
What R package is needed for IPUMS USA for weighting? Is it "survey"? Do you ...
Login   |   Register

Recently Active Members

View More »

Share |