Census Data Using an API on R studio

Question

So, I am new to using R, so sorry if the questions seem a little basic!

But my work is asking me to look through census data using an API and identify some variables in each tract, then create a csv file they can look at. The code is fully written for me, I believe, but I need to change the variables to:

S2602_C01_023E - black / his 

S2602_C01_081E - unemployment rate 

S2602_C01_070E - not US citizen (divide by total population) 

S0101_C01_030E - # over 65 (divide by total pop)

S1603_C01_009E - # below poverty (divide by total pop)

S1251_C01_010E - # child under 18 (divide by # households)

S2503_C01_013E - median income 

S0101_C01_001E - total population 

S2602_C01_078E - in labor force

And, I need to divide some of the variables, like I have written, and export all of this into a CSV file. I just don't really know what to do with the code..like I am just lost because I have never used R. I try changing the variables to the ones I need, but an error comes up. Any help would be greatly appreciated!

library(tidycensus)
library(tidyverse)
library(stringr)
library(haven)
library(profvis)

#list of variables possible
v18 <- load_variables(year = 2018,
                  dataset = "acs5",
                  cache = TRUE)
#function to get variables for all states. Year, variables can be 
easily edited.
get_census_data <- function(st) {
  Sys.sleep(5)
  df <- get_acs(year = 2018, 
            variables = c(totpop = "B01003_001",
                          male = "B01001_002",
                          female = "B01001_026",
                          white_alone = "B02001_002",
                          black_alone = "B02001_003",
                          americanindian_alone = "B02001_004",
                          asian_alone = "B02001_005",
                          nativehaw_alone = "B02001_006",
                          other_alone = "B02001_007",
                          twoormore = "B02001_008",
                          nh = "B03003_002",
                          his = "B03003_003",
                          noncit = "B05001_006",
                          povstatus = "B17001_002",
                          num_households = "B19058_001",
                          SNAP_households = "B19058_002",
                          medhhi = "B19013_001", 
                          hsdiploma_25plus = "B15003_017",
                          bachelors_25plus = "B15003_022",
                          greater25 = "B15003_001",
                          inlaborforce = "B23025_002",
                          notinlaborforce = "B23025_007",
                          greater16 = "B23025_001",
                          civnoninstitutional = "B27010_001",
                          withmedicare_male_0to19  = "C27006_004",
                          withmedicare_male_19to64 = "C27006_007",
                          withmedicare_male_65plus = "C27006_010",
                          withmedicare_female_0to19  = "C27006_014",
                          withmedicare_female_19to64 = "C27006_017",
                          withmedicare_female_65plus = "C27006_020",
                          withmedicaid_male_0to19  = "C27007_004",
                          withmedicaid_male_19to64 = "C27007_007",
                          withmedicaid_male_65plus = "C27007_010",
                          withmedicaid_female_0to19  = "C27007_014",
                          withmedicaid_female_19to64 = "C27007_017",
                          withmedicaid_female_65plus ="C27007_020"), 
            geography = "tract",
            state = st )
  return(df)
} 

#loops over all states
df_list <- setNames(lapply(states, get_census_data), states)

##if you want to keep margin of error, remove everything after %>% 
in next two lines
final_df <- bind_rows(df_list) %>%
  select(-moe)

colnames(final_df)[3] <- "varname"

#cleaning up final data, making it wide instead of long
final_df_wide <- final_df %>%
             gather(variable, value, -(GEOID:varname)) %>%
              unite(temp, varname, variable) %>%
              spread(temp, value)

#exporting to csv file, adjust your path

write.csv(final_df,"C:\Users\NAME\Documents\acs_2018_tractlevel_dat.
a.csv" )

Hello and welcome to SO. It is much easier to help you if you provide a good, reproducible example with sample data. See also: https://stackoverflow.com/questions/5963269/how-to-make-a-great-r-reproducible-example — Julian_Hn, Jul 17 '20 at 20:35
@Julian_Hn I tried to clean it up a bit, but thank you for that advice! — onedayatatime, Jul 17 '20 at 20:46
You just posted a census API key in clear text. I have edited your question to remove it. Never post secret keys publicly! — Julian_Hn, Jul 17 '20 at 20:51

score 1 · Answer 1 · answered Jul 17 '20 at 21:03

Since you can't really give an reproducible example without revealing your API key, I'll try my best to figure out what could work here:

Let's first edit the function that pulls data from the API:

get_census_data <- function(st) {
  Sys.sleep(5)
  df <- get_acs(year = 2018, 
            variables = c(blackHis= "S2602_C01_023E",
                          unEmployRate = "S2602_C01_081E",
                          notUSCit = "S2602_C01_070E")
            geography = "tract",
            state = st )
  return(df)
}

I've just put in two of the variables, but you should get the point.

Try if this works for you. And returns the data that is stored in the respective variables.

thank you so much!! i tried this and ran it, but for some reason, i get the data for the old variables? like i don't get this new blackHis, unemployment rate, etc data, just the old total population, population of females, etc — onedayatatime, Jul 17 '20 at 21:13

Census Data Using an API on R studio

1 Answers1