Fundraising databases often contain inconsistent data entry. fundr provides tools to normalize phone numbers, ZIP codes, and convert between naming conventions.
Phone Number Normalization
The normalize_phone() function handles the many ways
phone numbers get entered:
messy_phones <- c(
"(214) 555-1234",
"972.555.4567",
"817-555-8901",
"1-469-555-0123",
"+1 (214) 555-9876",
"2145557890",
"214/555-4321"
)
normalize_phone(messy_phones)
#> [1] "214-555-1234" "972-555-4567" "817-555-8901" "469-555-0123" "214-555-9876"
#> [6] "214-555-7890" "214-555-4321"Output Formats
Choose your preferred output format:
phone <- "(214) 555-1234"
# Dashed format (default)
normalize_phone(phone, format = "dash")
#> [1] "214-555-1234"
# Digits only
normalize_phone(phone, format = "digits")
#> [1] "2145551234"
# E.164 international format
normalize_phone(phone, format = "e164")
#> [1] "+12145551234"Handling Extensions
phones_with_ext <- c(
"214-555-1234 x100",
"972-555-4567 ext. 200",
"817-555-8901 extension 300"
)
# Drop extensions (default)
normalize_phone(phones_with_ext)
#> [1] "214-555-1234" "972-555-4567" "817-555-8901"
# Keep extensions
normalize_phone(phones_with_ext, keep_extension = TRUE)
#> [1] "214-555-1234 x100" "972-555-4567 x200" "817-555-8901 x300"Strict vs. Lenient Mode
By default, normalize_phone() requires exactly 10
digits. Use strict = FALSE to attempt salvaging numbers
with extra digits:
problematic <- c("12145551234", "0012145551234")
# Strict mode - returns NA for invalid
normalize_phone(problematic, strict = TRUE)
#> [1] "214-555-1234" NA
# Lenient mode - takes last 10 digits
normalize_phone(problematic, strict = FALSE)
#> [1] "214-555-1234" "214-555-1234"7-Digit Numbers
Some databases store local numbers without area codes:
local_numbers <- c("555-1234", "5554567")
# By default, 7-digit numbers are invalid
normalize_phone(local_numbers)
#> [1] NA NA
# Allow 7-digit numbers
normalize_phone(local_numbers, allow_7 = TRUE)
#> [1] "555-1234" "555-4567"ZIP Code Normalization
The normalize_zip() function standardizes US ZIP codes
and ZIP+4:
messy_zips <- c(
"75201",
"75201-1234",
"752011234",
"75201 5678",
" 75201 "
)
normalize_zip(messy_zips)
#> [1] "75201" "75201" "75201" "75201" "75201"Output Formats
zip <- "75201-1234"
# 5-digit only (default)
normalize_zip(zip, format = "zip5")
#> [1] "75201"
# ZIP+4 when available
normalize_zip(zip, format = "zip9")
#> [1] "75201-1234"
# Raw digits
normalize_zip(zip, format = "digits")
#> [1] "752011234"Strict vs. Lenient Mode
problematic_zips <- c("7520112345", "0075201")
# Strict mode
normalize_zip(problematic_zips, strict = TRUE)
#> [1] NA NA
# Lenient mode - salvages valid portions
normalize_zip(problematic_zips, strict = FALSE)
#> [1] "75201" "00752"Case Conversion
Convert strings between common naming conventions:
# Snake case (database columns)
to_snake_case("FirstName")
#> [1] "first_name"
to_snake_case("lastName")
#> [1] "last_name"
to_snake_case("Date of Birth")
#> [1] "date_of_birth"
# Title case (display labels)
to_title_case("first_name")
#> [1] "First Name"
to_title_case("date_of_birth")
#> [1] "Date Of Birth"
# Camel case (programming)
to_camel_case("first_name")
#> [1] "firstName"
to_camel_case("date of birth")
#> [1] "dateOfBirth"Converting Data Frame Column Names
Use convert_names() to transform all column names at
once:
df <- data.frame(
`First Name` = "John",
`Last Name` = "Smith",
`Date of Birth` = "1980-01-15",
check.names = FALSE
)
names(df)
#> [1] "First Name" "Last Name" "Date of Birth"
# Convert to snake_case
df_snake <- convert_names(df, case = "snake")
names(df_snake)
#> [1] "first_name" "last_name" "date_of_birth"
# Convert to title case
df_title <- convert_names(df_snake, case = "title")
names(df_title)
#> [1] "First Name" "Last Name" "Date Of Birth"Working with the Portfolio Dataset
The fundr_portfolio dataset intentionally includes messy
phone numbers and ZIP codes for practicing cleanup:
# Sample of messy phone numbers
sample_phones <- fundr_portfolio$phone_number[!is.na(fundr_portfolio$phone_number)][1:8]
sample_phones
#> [1] "512.375.2067" "(713) 891-0940" "+1 (361) 882-5883"
#> [4] "+1 (817) 622-3340" "361-4636998" "409/512-0230"
#> [7] "254/490-8249" "+1 (210) 891-3829"
# Clean them
normalize_phone(sample_phones)
#> [1] "512-375-2067" "713-891-0940" "361-882-5883" "817-622-3340" "361-463-6998"
#> [6] "409-512-0230" "254-490-8249" "210-891-3829"
# Sample of messy ZIP codes
sample_zips <- fundr_portfolio$zip[!is.na(fundr_portfolio$zip)][1:8]
sample_zips
#> [1] "75219-9418" "78701 4888" "77005-7098" "73102 7081" " 77005 "
#> [6] "75099-3876" " 75205 " "77005"
# Clean them
normalize_zip(sample_zips)
#> [1] "75219" "78701" "77005" "73102" "77005" "75099" "75205" "77005"Cleaning the Full Dataset
Here’s how you might clean the entire portfolio using dplyr:
library(dplyr)
portfolio_clean <- fundr_portfolio |>
mutate(
phone_clean = normalize_phone(phone_number),
zip_clean = normalize_zip(zip, format = "zip5")
)Or with base R:
# Create a copy
portfolio_clean <- fundr_portfolio
# Clean phone and ZIP
portfolio_clean$phone_clean <- normalize_phone(portfolio_clean$phone_number)
portfolio_clean$zip_clean <- normalize_zip(portfolio_clean$zip, format = "zip5")
# Check results
head(portfolio_clean[!is.na(portfolio_clean$phone_number),
c("phone_number", "phone_clean", "zip", "zip_clean")])
#> phone_number phone_clean zip zip_clean
#> 1 512.375.2067 512-375-2067 75219-9418 75219
#> 2 (713) 891-0940 713-891-0940 78701 4888 78701
#> 3 +1 (361) 882-5883 361-882-5883 77005-7098 77005
#> 4 +1 (817) 622-3340 817-622-3340 73102 7081 73102
#> 5 361-4636998 361-463-6998 77005 77005
#> 6 409/512-0230 409-512-0230 75099-3876 75099Handling Invalid Values
Both normalization functions return NA by default for
invalid values. You can customize this:
invalid_phones <- c("123", "not-a-phone", "")
# Default - returns NA
normalize_phone(invalid_phones)
#> [1] NA NA NA
# Custom invalid value
normalize_phone(invalid_phones, invalid = "INVALID")
#> [1] "INVALID" "INVALID" "INVALID"Summary
| Function | Purpose | Key Options |
|---|---|---|
normalize_phone() |
Standardize phone numbers |
format, strict, allow_7,
keep_extension
|
normalize_zip() |
Standardize ZIP codes |
format, strict
|
to_snake_case() |
Convert to snake_case | - |
to_title_case() |
Convert to Title Case | - |
to_camel_case() |
Convert to camelCase | - |
convert_names() |
Convert data frame column names | to |