34 variables, 2020

Here’s a list of the 34 variables used in the analysis of socioeconomic data for Chicago-area 2020 census tracts. The first seven variables plus the age variables (10-14) consist of data from the 2020 Census. The other variables come from the 2018-2022 American Community Survey (ACS). All the data were downloaded from IPUMS-NHGIS.

DENS  Population per square kilometer of land area
WHITE  Percent non-Hispanic white (alone)
AFAM  Percent non-Hispanic African-American (alone)
ASPI  Percent non-Hispanic Asian or Pacific Islander (alone)
HISP  Percent Hispanic/Latino
MARCHH  Percent of households including a married couple
SAMESX   Percent of households with same-sex spouses or partners
PERSHH  Persons per household
OWNOCC  Percent of occupied housing units owner-occupied
PCLT18  Percent of population aged 0-17
PC1829  Percent of population aged 18-29
PC30-49  Percent of population 30-49
PC5064  Percent of population aged 50-64
PC65UP  Percent of population aged 65+
FEMHH  Percent of households with female householder
PCAPINC  Mean per capita income
COLLDG  Percent of population 25+ with college degree
PUBTRN  Percent of workers 16+ using public transportation to get to work
MANPRF  Percent of workers 16+ with managerial or professional jobs
NOCAR  Percent of housing units with no vehicle available
HOUINC  Median household income
ITAL  Percent of population with Italian ancestry
POLISH  Percent of population with Polish ancestry
RUSS  Percent of population with Russian or Ukrainian ancestry
RENT  Median monthly rent
NONFAM  Percent of households non-families
HOUVAL  Median value of owner-occupied housing units
FORBRN  Percent of population foreign-born
UNEMP  Percent of workers 16+ unemployed
LINGIS  Percent of population 5+ unable to speak English well
POOR  Percent of families below poverty line
SINFAM  Percent of occupied housing units that are single-family homes
TENPLU  Percent of occupied housing units in buildings with ten or more housing units
YEAR  Median year housing built

These variables are very nearly the same as those used in the 2010 analysis reported on this blog, but there are some small differences, caused either by changes in the questions asked by the Census Bureau or by, well, events (or both):

[1] SAMESX. The data now include both married and partnered same-sex couples. The numbers (not surprisingly), while still generally low, are higher (and perhaps more reliable) than in 2010.

[2] FEMHH. The Census Bureau has (quite reasonably) done away with the concept “head of household” and substituted the less suspect term “householder,” which refers to the name of the person in whose name a housing unit is owned or rented. There is no longer the implication that he or she is head of anything. I don’t know how this change has affected the numbers.

[3] RUSS. This variable in 2020 includes both those who said they had Russian ancestry and those who said they had Ukrainian ancestry. I assume (but can’t prove) that many people whose ancestry was in the Russian Empire might now be more inclined to claim Ukrainian ancestry. The variable is a very imperfect surrogate for Jewish ancestry (the Census Bureau does not ask questions about religion).

[4] LINGIS. This variable is now used to report the proportion of those 5 and above who could not speak English very well. I don’t know how the change from “linguistically isolated” has affected the numbers.

[5] POOR. Formerly this slot was used for percent of families on public assistance. I couldn’t locate the identical question in the 2018-2022 ACS and substituted percent of families below the poverty line. This is the most important change in the variable list. I suspect but can’t prove that percent of families on public assistance (not defined with great precision) is closely correlated with percent of families below the poverty line.

 

 

This entry was posted in Urban. Bookmark the permalink.

Comments are welcome