In btskinner/crosswalkr: Rename and Encode Data Frames Using External Crosswalk Files

crosswalkr

options(width = 100)

Overview

This package offers a pair of functions, renamefrom() and encodefrom(), for renaming and encoding data frames using external crosswalk files. It is especially useful when constructing master data sets from multiple smaller data sets that do not name or encode variables consistently across files. Based on renamefrom and encodefrom Stata commands written by Sally Hudson and team.

Installation

Install the latest release version from CRAN with

install.packages('crosswalkr')

Install the latest development version from Github with

devtools::install_github('btskinner/crosswalkr')

Usage

library(crosswalkr)
library(dplyr)
library(haven)

## starting data frame
df <- data.frame(state = c('Kentucky','Tennessee','Virginia'),
                 fips = c(21,47,51),
                 region = c('South','South','South'))
df

## crosswalk with which to convert old names to new names with labels
cw <- data.frame(old_name = c('state','fips'),
                 new_name = c('stname','stfips'),
                 label = c('Full state name', 'FIPS code'))
cw

Renaming

Convert old variable names to new names and add labels from crosswalk.

df1 <- renamefrom(df, cw_file = cw, raw = old_name, clean = new_name, label = label)
df1

Convert old variable names to new names using old names as labels (ignoring labels in crosswalk).

df2 <- renamefrom(df, cw_file = cw, raw = old_name, clean = new_name, name_label = TRUE)
df2

Convert old variable names to new names, but keep unmatched old names in the data frame.

df3 <- renamefrom(df, cw_file = cw, raw = old_name, clean = new_name, drop_extra = FALSE)
df3 

Encoding

## starting data frame
df <- data.frame(state = c('Kentucky','Tennessee','Virginia'),
                 stfips = c(21,47,51),
                 cenregnm = c('South','South','South'))
df

## use state crosswalk data file from package
cw <- get(data(stcrosswalk))
cw

Create a new column with factor-encoded values

df$state2 <- encodefrom(df, var = state, cw_file = cw, raw = stname, clean = stfips, label = stabbr)
df

Create a new column with labelled values.

## convert to tbl_df
df <- tibble::as_tibble(df)
df$state3 <- encodefrom(df, var = state, cw_file = cw, raw = stname, clean = stfips, label = stabbr)

Create new column with factor-encoded values (ignores the fact that df is a tibble)

df$state4 <- encodefrom(df, var = state, cw_file = cw, raw = stname, clean = stfips, label = stabbr, ignore_tibble = TRUE)

Show factors with labels:

as_factor(df)

Show factors without labels:

zap_labels(df)

btskinner/crosswalkr documentation built on March 15, 2024, 8:25 a.m.