Similar data from the 2000 census

Like the 2010 Census, data from the 2000 Census is directly accessible through American Factfinder. However, with each new ten-year Census, the questions get modified and the data is recompiled in different ways. In the 2000 Census there is no table that is an exact equivalent to the 2010 Table P5. Initially I thought table QT-P5 might be the same, but it does not distinguish Latin Americans separately. It turns out Table P004 is the closest equivalent.

First, set the search filters the same way to see if you can get data at the Tract level:

TableP004_01_search-filters_alltracts_CoCoData

Then filter the data by year:

TableP004_02_search-filters_year2000_CoCoData

Then search for tables that identify Latinos as a separate population:

TableP004_03_search-filters_L:atino+not_CoCoData

This leaves us with only one candidate table:

TableP004_04_table-P004-identified_CoCoData

Select the table and then review the contents to see if it has the right data:

TableP004_06_inspect-table-contents_CoCoData

This table does not disaggregate Latinos by race, as do columns D011-Do17 in the 2010 Table P5. But we are not using those variables either; so the “Hispanic or Latino” category suffices as an equivalent to Table P5’s column D010.

Prepare to download:

TableP004_07_choose-DL-format_CoCoData

Uncheck the “Include descriptive data element names” so that you don’t get a second row of descriptive data in your CSV. If you are going to import that CSV data into R (or other statistical software), only Row #1 should have descriptive information in it that will become the names for columns of data. If you do have a second row of ‘descriptive element names’, you will need to delete Row#2 before importing it into R for analysis.

However, on subsequent pages I also argue that you need to revise the Row#1 names because the default names that the Census provides are utterly cryptic, like HC01_VC05. I will ask you to refer to the ‘separate file of data and annotations’ referred to above, which is the ‘metadata.csv’ that gets downloaded with your main data file. But you have a choice: you could include that second row within your main file, use it as your guide for renaming the columns into short, readable names, and then delete Row #2 so that every row after Row #1 only contains data. So long as you know how to prep your CSV data by producing a clean, import-ready file, you can decide for yourself whether to keep that descriptive Row #2 in your download as the guide for revising Row #1, or use the separate medatata file as the guide.

 


Leave a Reply