Median Income with IPUMS

This script calculates the weighted median income of an arbitrary region (group of counties) using data from IPUMS-USA.

library(tidyverse)
library(ipumsr)
library(matrixStats)

Read the data files.

ddi <- read_ipums_ddi("usa_00006.xml")
data <- read_ipums_micro(ddi)

## Use of data from IPUMS-USA is subject to conditions including that users should
## cite the data appropriately. Use command `ipums_conditions()` for more details.

Remove irrelevant data, including wages listed as 99999 and observations under age 25. Then calculate the weighted median income using the weightedMedian() function from the matrixStats library.

tx <- data %>% filter(STATEFIP==48, INCWAGE != 0, INCWAGE < 999999, AGE >= 25)
weightedMedian(tx$INCWAGE, tx$PERWT)

## [1] 40000

The FIPS codes listed here represent the counties of the Comptroller’s Gulf Coast region. First calculate the total median income of the region, then calculate it by educational attainment: high school or equivalent, associate degree, bachelor’s, and master’s.

gulfcoast <- tx %>% filter(COUNTYFIP %in% c(15,39,71,89,157,167,201,291,321,339,471,473,481))
weightedMedian(gulfcoast$INCWAGE, gulfcoast$PERWT)

## [1] 42000

gchighsch <- filter(gulfcoast, EDUCD == 62 | EDUCD == 63 | EDUCD == 64)
weightedMedian(gchighsch$INCWAGE, gchighsch$PERWT)

## [1] 30000

gcass <- filter(gulfcoast, EDUCD == 81 | EDUCD == 82 | EDUCD == 83)
weightedMedian(gcass$INCWAGE, gcass$PERWT)

## [1] 42000

gcbach <- filter(gulfcoast, EDUCD == 101)
weightedMedian(gcbach$INCWAGE, gcbach$PERWT)

## [1] 59000

gcmasters <- filter(gulfcoast, EDUCD==114)
weightedMedian(gcmasters$INCWAGE, gcmasters$PERWT)

## [1] 72000