Question

Status:Closed    Asked:Dec 07, 2017 - 11:06 AM

How do I pull Boston data for 1980 by NECTAs?

I want to pull data for Boston in years 1980, 1990, 2000, 2010 and 2015 keeping the boundaries constant. For other cities, such as Chicago, I have used the counties that comprised the MSA in year 1980 and pulled data based on county FIPS indicators for subsequent years. However, since the 1980 census used towns and cities as the geographic unit in New England rather than counties, I can't apply the same methodology. However, I can't find a geographic variable that identifies NECTAs. What do people do when pulling 1980 Boston MSA data?


 
Do you have the same question? Follow this Question
 

Staff Answer

Suffolk County, which is the county where Boston is located, is actually identifiable in the 1980 5% and 1% samples, however, it is not identifiable in the 1990 or 2000 samples. This spreadsheet lists the counties which are identifiable in 1950-onward data.

If you are interested in a geographic variable with consistent boundaries over time, then I would suggest looking into using CONSPUMA and CPUMA0010. For 1980, the lowest-level geographic codes are for county groups (CNTYGP98), which for Massachusetts are comprised of the cities and towns. This page contains maps and detailed compositions by state for 1980 county groups.

 

Dec 08, 2017 - 04:13 PM

0
0
Report it

As I understand, you'd ultimately like to get data for 1980, 1990, 2000, 2010 and 2015 for the 1980 Boston MSA definition, which was built from towns and cities rather than counties.


The METAREA variable identifies the Boston MSA using each sample's "contemporary" MSA definitions. So the METAREA code for the 1980 sample would identify the 1980 MSA, but the 1990 and later METAREA codes would not. An additional problem: the 1980 "county groups" (not really groups of counties--more like PUMAs) that the METAREA codes are based on don't match up perfectly with the 1980 MSA, so METAREA's 1980 Boston code actually omits 15% of the MSA's population. (This page gives the omitted proportion of each MSA's population in each sample.)


In short, I'm quite sure that what you're proposing to do is not possible with public census microdata.


However, you could obtain census summary tables for towns and cities from 1980 through 2015 from IPUMS NHGIS. To obtain New England town and city data through NHGIS, enter the NHGIS Data Finder, click on the Geographic Level filter, and select the "County Subdivision" filter. You can then explore which tables are available, select and download them, and then, if you have a record of which towns and cities comprise the Boston MSA, you could sum data for all of those towns and cities separately for each census source year. NHGIS also supplies some time series tables for county subdivisions that already link together 1980, 1990, 2000, 2010, and 2008-2012 ACS data.

 

Dec 13, 2017 - 05:06 PM

0
0
Report it

Answers

As I understand, you'd ultimately like to get data for 1980, 1990, 2000, 2010 and 2015 for the 1980 Boston MSA definition, which was built from towns and cities rather than counties.


The METAREA variable identifies the Boston MSA using each sample's "contemporary" MSA definitions. So the METAREA code for the 1980 sample would identify the 1980 MSA, but the 1990 and later METAREA codes would not. An additional problem: the 1980 "county groups" (not really groups of counties--more like PUMAs) that the METAREA codes are based on don't match up perfectly with the 1980 MSA, so METAREA's 1980 Boston code actually omits 15% of the MSA's population. (This page gives the omitted proportion of each MSA's population in each sample.)


In short, I'm quite sure that what you're proposing to do is not possible with public census microdata.


However, you could obtain census summary tables for towns and cities from 1980 through 2015 from IPUMS NHGIS. To obtain New England town and city data through NHGIS, enter the NHGIS Data Finder, click on the Geographic Level filter, and select the "County Subdivision" filter. You can then explore which tables are available, select and download them, and then, if you have a record of which towns and cities comprise the Boston MSA, you could sum data for all of those towns and cities separately for each census source year. NHGIS also supplies some time series tables for county subdivisions that already link together 1980, 1990, 2000, 2010, and 2008-2012 ACS data.

 

Dec 13, 2017 - 05:06 PM

0
0
Report it

OTHER QUESTIONS NEEDING ANSWERS

Can I merge 1980 data with 2005 by conspuma?
How can I generate MSA-level data with consistent MSA boundaries in 1990, 200...
Foreign-born population by county in 1970, 1980 and 1990.
How to calculate the population of a specific year (for example 1974) with de...
Login   |   Register

Recently Active Members

View More »

Share |