Introduction

Census data provides valuable resource for understanding and analyzing US demographic, social, and economic trends within a population. In this code through, we will explore the basic of handling census data in R with packages like ‘tidyverse’ for data wrangling and ‘ggplot2’ for creating informative visualizations.

Packages

Getting data

Step 1: Getting API Key

Census API key is a unique code that allows users to access and retrieve specific datasets or information from the census bureau’s database. You can get a Census API key here: https://api.census.gov/data/key_signup.html. Remember to keep it private as it is upnique for each person like your password.

Access census data in R with census_api_key. load your API key in ““. install = TRUE will save it for future use. That way there is no need to enter API key in”” every time you use it.

census_api_key("")
## To install your API key for use in future sessions, run this function with `install = TRUE`.

Step 2

Use tidycensus package to download information.

CenDF <- load_variables(2017, "acs5", cache = TRUE)     #load the data set, 2017 is the end of year. acs5 set the time frame from 2013-2017. 
CenDF$label <- toupper(CenDF$label) #Make valuables to upper case
head(CenDF)
## # A tibble: 6 × 4
##   name        label                                concept             geography
##   <chr>       <chr>                                <chr>               <chr>    
## 1 B00001_001  ESTIMATE!!TOTAL                      UNWEIGHTED SAMPLE … block gr…
## 2 B00002_001  ESTIMATE!!TOTAL                      UNWEIGHTED SAMPLE … block gr…
## 3 B01001A_001 ESTIMATE!!TOTAL                      SEX BY AGE (WHITE … tract    
## 4 B01001A_002 ESTIMATE!!TOTAL!!MALE                SEX BY AGE (WHITE … tract    
## 5 B01001A_003 ESTIMATE!!TOTAL!!MALE!!UNDER 5 YEARS SEX BY AGE (WHITE … tract    
## 6 B01001A_004 ESTIMATE!!TOTAL!!MALE!!5 TO 9 YEARS  SEX BY AGE (WHITE … tract

In this example we want to check the median house value growth. By searching “Median Value” in label in CenDF data set to find the B25077_001 name. The “get_acs” function is used to filter out the data set for the state or county you would like.

M_Home<- c(Median_House_Value = "B25077_001",
           HHIncome = "B06011_001") 

CenDF<- get_acs(geography = "state", year = 2017, survey = "acs5", variables = M_Home,geometry = T)
head(CenDF)

Data Wrangling

As it is shown above, the CenDF data set is showing two rows of information about each county. And we want just one record for each county.

CenDF <- CenDF %>% 
    select(-moe)%>%  #Remove moe column
    spread(variable, estimate)%>%   #spread data from long to wide. "variable" is the column name and "estimate" is the value you want for variable.
    mutate(HHInc_HousePrice_Ratio_2017 = round( HHIncome/Median_House_Value*100,2))   # "mutate" creates a new column
datatable(CenDF)  # create an interactive table
## Warning in instance$preRenderHook(instance): It seems your data is too big for
## client-side DataTables. You may consider server-side processing:
## https://rstudio.github.io/DT/server.html
head(CenDF)
## Simple feature collection with 6 features and 5 fields
## Geometry type: MULTIPOLYGON
## Dimension:     XY
## Bounding box:  xmin: -179.1489 ymin: 30.22333 xmax: 179.7785 ymax: 71.36516
## Geodetic CRS:  NAD83
##   GEOID       NAME HHIncome Median_House_Value                       geometry
## 1    01    Alabama    24476             132100 MULTIPOLYGON (((-88.05338 3...
## 2    02     Alaska    32940             261900 MULTIPOLYGON (((-166.5772 5...
## 3    04    Arizona    27517             193200 MULTIPOLYGON (((-114.8163 3...
## 4    05   Arkansas    23789             118500 MULTIPOLYGON (((-94.61783 3...
## 5    06 California    29454             443400 MULTIPOLYGON (((-118.6044 3...
## 6    08   Colorado    32401             286100 MULTIPOLYGON (((-109.0603 3...
##   HHInc_HousePrice_Ratio_2017
## 1                       18.53
## 2                       12.58
## 3                       14.24
## 4                       20.08
## 5                        6.64
## 6                       11.33
CenDF <- CenDF %>% mutate(state = c("AL", "AK","AZ","AR","CA","CO","CT","DE","CL","FL","GA","HI","ID","IL","IN","IA","KS","KY","LA","ME","MD","MA","MI","MN","MS","MO","MT","NE","NV","NH","NJ","NM", "NY","NC","ND","OH","OK","OR","PA","RI","SC","SD","TN","TX","UT","VT","VA","WA","WV","WI","WY", "PR"))

Data Virsalization

ggplot(CenDF, aes(fill = Median_House_Value/1000)) +
  geom_sf(color = "white") +
  coord_sf(xlim = c(-130, -60), ylim = c(25, 50)) +
  theme_void() +
  theme(panel.grid.major = element_line(colour = "transparent")) +
  scale_fill_distiller(palette = "Blues", direction = 1, name = "Estimate in thousand") +
  labs(title = "Median Household Value by State", caption = "Source: US Census/ACS5 2017")

We can also choose a state and see its county level status.

M_Home<- c(Median_House_Value = "B25077_001",
           HHIncome = "B06011_001") 
CenDF2<- get_acs(geography = "county", year = 2017, survey = "acs5", variables = M_Home,geometry = T) 
CenDF2 <- CenDF2[grep("California$", CenDF2$NAME), ] %>%
   select(-moe)%>%  
    spread(variable, estimate)%>%   
    mutate(HHInc_HousePrice_Ratio_2017 = round( HHIncome/Median_House_Value*100,2))   
head(CenDF2)
ggplot(CenDF2, aes(fill = Median_House_Value/1000)) +
  geom_sf(color = "white") +
  theme_void() +
  theme(
    plot.background = element_rect(fill = "white"),  # Set the plot background color
    panel.grid.major = element_line(colour = "transparent"),
    plot.margin = margin(20, 20, 20, 20, "pt")  # Adjust margins (top, right, bottom, left)
  ) +
  labs(title = "Median Household Value by State", caption = "Source: US Census/ACS5 2017") +
  scale_fill_distiller(palette = "Oranges", direction = 1, name = "Estimate in thousand")

ggplot(CenDF2, aes(x =HHInc_HousePrice_Ratio_2017, y =NAME )) +
  geom_bar(stat = "identity", fill = "darkblue", color = "white") +
  labs(x = "HHInc_HousePrice_Ratio_2017", y = "COUNTY") +
  theme(axis.text.x = element_text(angle = 45, hjust = 1))