Skip to main content
This guide covers common issues you may encounter when using the GSS StatsBank API with R and how to resolve them.

​
Common Issues

Problem: Your query returns no data or results that don’t match expectations.Solutions:
  1. Verify variable names - Variable codes are case-sensitive. Check the table metadata to confirm exact spelling:
metadata = table_req |> 
  req_perform() |> 
  resp_body_json()

# List all variable codes
purrr::map_chr(metadata$variables, "code")
  1. Verify value names - Values must match exactly as they appear in the metadata:
# Get available values for a specific variable
variable_info = metadata$variables |> 
  purrr::keep(~ .x$code == "WaterDisposal") |> 
  purrr::flatten()

tibble(Code = unlist(variable_info$values))
  1. Include all required variables - Some tables require selections for all variables. If in doubt, use the "all" filter:
list(
  code = "Geographic_Area",
  selection = list(filter = "all", values = list("*"))
)
Problem: Requests fail with timeout or connection errors.Solutions:
  1. Increase timeout - Large tables may take longer to process:
table_req |> 
  req_body_json(query_list) |> 
  req_timeout(300) |>  # 300 seconds = 5 minutes
  req_perform()
  1. Check API availability - Verify the API is accessible:
request("https://statsbank.statsghana.gov.gh:443/api/v1/en/") |> 
  req_perform() |> 
  resp_status()
  1. Reduce query size - Request smaller subsets of data instead of entire tables:
# Instead of filter = "all", select specific items
list(
  code = "Geographic_Area",
  selection = list(
    filter = "item",
    values = list("Ghana", "Greater Accra", "Ashanti")
  )
)
Problem: API returns an error about invalid JSON format.Solutions:
  1. Verify JSON structure - Ensure your query list follows the correct format:
query_list = list(
  query = list(
    list(
      code = "VariableName",
      selection = list(
        filter = "item",
        values = list("Value1", "Value2")
      )
    )
  ),
  response = list(
    format = "csv"
  )
)
  1. Check for special characters - Some variable values may contain special characters. Use exact strings from metadata.
  2. Validate with jsonlite - Test your JSON structure: 
library(jsonlite)
toJSON(query_list, auto_unbox = TRUE, pretty = TRUE)
Problem: API returns a 404 error when trying to access a table.Solutions:
  1. Verify the path - Double-check each level of the hierarchy:
# Start from the database level
build_url(URL, "PHC 2021 StatsBank")

# Then navigate to the topic
build_url(URL, "PHC 2021 StatsBank", "Water and Sanitation")

# Finally to the table
build_url(URL, "PHC 2021 StatsBank", "Water and Sanitation", "waterDisposal_table.px")
  1. Check spelling and spacing - Folder and table names are case-sensitive and must match exactly.
  2. Confirm table extension - Tables must end in .px
Problem: read_csv() fails to parse the API response correctly.Solutions:
  1. Check response status - Verify you received data:
response = table_req |> 
  req_body_json(query_list) |> 
  req_perform()

# Should return 200
resp_status(response)

# Check content type
resp_content_type(response)
  1. Inspect raw response - Look at the actual data returned:
raw_data = resp_body_raw(response)
rawToChar(raw_data)
  1. Adjust CSV parsing options - Some tables may need different settings:
data = response |> 
  resp_body_raw() |> 
  readr::read_csv(
    locale = locale(encoding = "UTF-8"),
    na = c("", "NA", "..")
  )
Problem: Required R packages are not installed.Solutions:Install required packages:
# Core packages
install.packages("httr2")
install.packages("tidyverse")

# Additional helpful packages
install.packages("jsonlite")
install.packages("readr")
Load packages at the start of your script:
library(httr2)
library(tidyverse)
Problem: The build_url() helper function returns errors or unexpected results.Solutions:
  1. Verify function definition - Ensure the complete function is defined:
build_url = function(URL, ...) {
  path = list(...)
  req = request(URL)
  
  full_req = purrr::reduce(path, req_url_path_append, .init = req)
  
  is_table = FALSE
  if (length(path) > 0) {
    if (grepl("\\.px$", path[[length(path)]], ignore.case = TRUE)) {
      is_table = TRUE
    }
  }
  
  response = req_perform(full_req)
  body = resp_body_json(response)
  
  if (is_table) {
    message("Endpoint reached: Table found.")
    message("Available variables:")
    print(map_chr(body$variables, "code"))
  } else {
    key = if(length(path) == 0) "dbid" else "id"
    print(map_chr(body, key))
  }
  
  return(full_req)
}
  1. Check purrr is loaded - The function requires purrr from tidyverse:
library(purrr)

​
Getting Additional Help

PxWeb Documentation

Review the official PxWeb API specification

Available Databases

Check the list of available databases and their structure

​
Debugging Tips

1

Enable verbose output

Use req_verbose() to see full HTTP request and response details:
table_req |> 
  req_body_json(query_list) |> 
  req_verbose() |> 
  req_perform()
2

Test incrementally

Build your query step by step, testing each component:
# Test 1: Can you reach the database?
build_url(URL, "PHC 2021 StatsBank")

# Test 2: Can you reach the topic?
build_url(URL, "PHC 2021 StatsBank", "Water and Sanitation")

# Test 3: Can you reach the table?
table_req = build_url(URL, "PHC 2021 StatsBank", "Water and Sanitation", "waterDisposal_table.px")
3

Validate metadata

Always check metadata before building queries:
metadata = table_req |> req_perform() |> resp_body_json()
str(metadata, max.level = 2)
If you continue experiencing issues after trying these solutions, the API may be undergoing maintenance or experiencing technical difficulties. Try again later or contact the Ghana Statistical Service for support.

Build docs developers (and LLMs) love

Get started for freeTalk to us