在R中基于指定地址获取Block Group普查数据的问题求助

我在使用R语言提取Block Group（区块组）的普查数据时遇到问题。手里有一批具体地址，需要先通过地址匹配到对应的Block Group，再提取该区块组的白人/黑人/亚裔/西班牙裔/两种族/其他种族占比等普查数据，但代码无法正常运行，求帮忙解决。

报错信息 #

Error in get_acs(...) : unused argument (blockgroup = block_group)

原代码 #

library(httr)
library(tidycensus)
library(purrr)

get_geocode_data <- function(address) {
  # The URL for the geocoding request
  url <- "https://geocoding.geo.census.gov/geocoder/geographies/onelineaddress"

  # The parameters for the request
  params <- list(
    address = address,
    format = "json",
    benchmark = "Public_AR_Current",
    vintage = "Current_Current"
  )

  # Make the request
  response <- GET(url, query = params)

  # Parse the response
  content <- content(response, "parsed", "application/json")

  # Extract the geographies data if it exists
  if (length(content$result$addressMatches) > 0) {
    geographies <- content$result$addressMatches[[1]]$geographies
  } else {
    geographies <- NULL
  }

  return(geographies)
}

addresses <- c(
  "Administration Building, 400 Bizzell St, College Station, TX 77843",
  "1210 Varsity Dr., E. Carroll Joyner Visitor Center, Raleigh, NC 27606",
  "1331 Cir Park Dr, Knoxville, TN 37916"
)

geocoding_data <- map(addresses, get_geocode_data)

# Define census API key
census_api_key("94d96c1dfb04afe285d0db958f78ded8a4b197f7")

get_population <- function(geographies) {
  # If geographies is NULL, return NA
  if (is.null(geographies)) {
    return(NA)
  }

  # Extract the state, county, tract, and block group
  state <- geographies$`2010 Census Blocks`[[1]]$STATE
  county <- geographies$`2010 Census Blocks`[[1]]$COUNTY
  tract <- geographies$`2010 Census Blocks`[[1]]$TRACT
  block_group <- geographies$`2010 Census Blocks`[[1]]$BLKGRP

  # Print out the geography parameters
  print(paste("State:", state))
  print(paste("County:", county))
  print(paste("Tract:", tract))
  print(paste("Block group:", block_group))

  # Get the total population for the block group
  population <- get_acs(
    geography = "block group",
    variables = "B01003_001",
    state = state,
    county = county,
    tract = tract,
    blockgroup = block_group
  )

  return(population$estimate)
}

# Get the total population for each block group
populations <- map(geocoding_data, get_population)

问题原因 #

报错直接原因是get_acs函数没有blockgroup这个参数，正确参数名是block_group（带下划线）。另外原代码只获取了总人口，没覆盖你需要的种族占比数据，下面是修正后的完整代码。

修正后的代码 #

library(httr)
library(tidycensus)
library(purrr)
library(dplyr)
library(tidyr)

# 地理编码函数：直接获取地址对应的Block Group信息
get_geocode_data <- function(address) {
  url <- "https://geocoding.geo.census.gov/geocoder/geographies/onelineaddress"
  
  params <- list(
    address = address,
    format = "json",
    benchmark = "Public_AR_Current",
    vintage = "Current_Current"
  )
  
  response <- GET(url, query = params)
  content <- content(response, "parsed", "application/json")
  
  if (length(content$result$addressMatches) > 0) {
    # 直接提取区块组层级数据，不用从Block层级转
    geographies <- content$result$addressMatches[[1]]$geographies$`Census Block Groups`[[1]]
  } else {
    geographies <- NULL
  }
  
  return(geographies)
}

# 待处理地址列表
addresses <- c(
  "Administration Building, 400 Bizzell St, College Station, TX 77843",
  "1210 Varsity Dr., E. Carroll Joyner Visitor Center, Raleigh, NC 27606",
  "1331 Cir Park Dr, Knoxville, TN 37916"
)

# 获取每个地址的地理编码数据
geocoding_data <- map(addresses, get_geocode_data)

# 设置Census API Key（建议执行一次install=TRUE，之后不用重复写）
census_api_key("94d96c1dfb04afe285d0db958f78ded8a4b197f7", install = TRUE)

# 提取区块组的种族人口及占比数据
get_race_data <- function(geographies) {
  if (is.null(geographies)) {
    return(tibble(
      address = NA,
      total_pop = NA,
      white_pct = NA,
      black_pct = NA,
      asian_pct = NA,
      hispanic_pct = NA,
      two_or_more_pct = NA,
      other_race_pct = NA
    ))
  }
  
  # 提取地理参数
  state <- geographies$STATE
  county <- geographies$COUNTY
  tract <- geographies$TRACT
  block_group <- geographies$BLKGRP
  
  # 定义需要的种族/族裔变量（ACS 5年估计，2021年为例）
  race_vars <- c(
    total_pop = "B01003_001",    # 总人口
    white = "B02001_002",        # 非西班牙裔白人
    black = "B02001_003",        # 非西班牙裔黑人
    asian = "B02001_005",        # 非西班牙裔亚裔
    hispanic = "B03003_003",     # 西班牙裔（族裔，非种族）
    two_or_more = "B02001_008",  # 两种或以上种族
    other_race = "B02001_004"    # 其他单一种族
  )
  
  # 获取ACS数据，修正block_group参数名
  acs_data <- get_acs(
    geography = "block group",
    variables = race_vars,
    state = state,
    county = county,
    tract = tract,
    block_group = block_group,
    year = 2021
  )
  
  # 整理数据为宽格式并计算占比
  result <- acs_data %>%
    select(variable, estimate) %>%
    pivot_wider(names_from = variable, values_from = estimate) %>%
    mutate(
      white_pct = round(white / total_pop * 100, 2),
      black_pct = round(black / total_pop * 100, 2),
      asian_pct = round(asian / total_pop * 100, 2),
      hispanic_pct = round(hispanic / total_pop * 100, 2),
      two_or_more_pct = round(two_or_more / total_pop * 100, 2),
      other_race_pct = round(other_race / total_pop * 100, 2),
      address = addresses[which(map_lgl(geocoding_data, ~ identical(.x, geographies)))]
    ) %>%
    select(address, total_pop, ends_with("_pct"))
  
  return(result)
}

# 批量获取所有地址的种族数据
race_results <- map_dfr(geocoding_data, get_race_data)

# 查看最终结果
print(race_results)

额外说明 #

1. 地理编码优化：直接提取Census Block Groups层级数据，避免从Block层级转取参数的冗余和错误。
2. 参数修正：将blockgroup改为block_group，匹配get_acs官方参数要求。
3. 变量补充：覆盖了你需要的所有种族/族裔变量，注意西班牙裔是独立的族裔分类，不属于种族变量范畴。
4. 数据整理：自动计算各群体占比并整理为易读的宽格式。
5. API安全：建议执行census_api_key("你的密钥", install = TRUE)一次，密钥会被保存到本地配置文件，后续脚本无需重复编写。

内容的提问来源于stack exchange，提问作者Fox_Summer