Hi, I was hoping to do an analysis of a particular combined metropolitan statistical area (CMSA) with ACS data. But, I see that in the MET2013 variable, the CMSA is split between two separate MET2013 codes. If I combine the data from these two codes - do I get the CMSA that I’m looking for?
Specifically, I wanted to do an analysis of the Greenville-Spartanburg-Anderson, SC CMSA. But, in the MET2013 codes, they are split between Spartanburg (43900) and Greenville-Anderson-Mauldin, SC (24860). If I combine the data under those two MET2013 codes, do I get the CMSA I’m aiming to analyze?
Finally, would that combination represent the Greenville-Spartanburg-Anderson, SC CMSA in the same way that the MET2013 Youngstown-Warren-Boardman, OH-PA (code 49660) represents it’s own CMSA?
Alternatively, I’ve thought about trying to recreate the the CMSA by combining county-level data? I also don’t think the PUMA geographies represent the CMSA’s in the way that I would hope - so I’m more hopeful that the MET2013 variable is the way to get at the geography I want.
Thanks for any help!
Anyone able to help me out with this?
The metro area you are referring to is not a Consolidated Metropolitan Statistical Area (CMSA), as those were discontinued in 2003. According to the 1999 delineation file, it was a Metropolitan Statistical Area (MSA) until 2013 at which point delineations changed and it became a Combined Statistical Area (CSA), which consist of adjacent MSAs. In 2013, the Office of Management and Budget (OMB) released new MSA delineations (more information on Delineation Files provided by the Census) and as a result, pre-2013 and post-2013 metro areas cannot be consistently compared (more discussion on this in this forum post). Metro area variables created by IPUMS USA are based on the relationship between Public Use Microdata Areas (PUMAs) and Metropolitan Statistical Areas (MSAs). METAREA is based on pre-2013 delineations whereas MET2013 is based on post-2013 delineations.
You may want to read about Omission/Comission errors at the bottom of the MET2013 definition page as you consider using values in this variable. Omission errors are cases where a resident of an MSA is not identified as a resident and Comission errors are the opposite, where a non-resident is identified as a resident. The MET2013 protocol allows for both types of errors, which can introduce inconsistencies, especially in smaller metropolitan areas where PUMA boundaries aren’t nested well within metro area boundaries. Therefore, it would not be advisable to compare Greenville-Spartanburg-Anderson from METAREA to Greenville-Anderson-Mauldin and Spartanburg from MET2013 without verifying that the underlying geographies match. PUMAs may be your best option to obtain geographic information that is comparable across time. Crosswalks to link 2000 and 2010 PUMAs can be found here.
1 Like