Skip to contents

This function is meant to be used to clean up San Mateo cities in a dataframe. It will only look for San Mateo County cities - it doesn't look for other cities. Non-San Mateo cities will return an NA in the city_clean column.

Usage

smc_city_clean(data,
      city_col = "city",
      new_col = "city_clean")

Arguments

data

This is the name of the dataframe with the city variable in it that you'd like to clean

city_col

(optional): This is a string that specifies the name of the city variable you want to clean. By default, the function assumes the variable is called city.

new_col

(optional): This is a string to specify the name of the variable with the cleaned city names. By default the cleaned cities will be saved in a variable called city_clean

Value

a dataset with a new variable for the cleaned city values

Examples

data <- data.frame(
 city_dirty = c("Burligame", "Fost City", "San Mato", "Daily Cit", "S S Francisco",
                "South San Fransico", "SoSan Franc", "San Francisco")
 )

data %>%
    smc_city_clean(city_col = "city_dirty",
                   new_col = "city_smc")
#>           city_dirty            city_smc
#> 1          Burligame          Burlingame
#> 2          Fost City         Foster City
#> 3           San Mato           San Mateo
#> 4          Daily Cit           Daly City
#> 5      S S Francisco South San Francisco
#> 6 South San Fransico South San Francisco
#> 7        SoSan Franc South San Francisco
#> 8      San Francisco                <NA>