In Big-Life-Lab/recodeflow: Contains functions to interface with variable details sheets, including recoding variables and converting them to PMML

knitr::opts_chunk$set(echo = TRUE, collapse = TRUE, comment = "#>")
source("../R/test_data_generator.R")

What is the function rec_with_table?

rec_with_table is the function used to recode variables within your dataset into harmonized variables. rec_with_table reads:

• variable for a list of variables to recode, and
• variable_details for instructions on how to recode the variable(s)

library(recodeflow)
status <- rec_with_table(data = tester1, 
                         variables = "status", 
                         variable_details = variable_details,
                         log = TRUE,
                         var_labels = c(status = "Status"))

Output - Warning messages

Warning messages will appear when your dataset is missing variables in your variable worksheet or rec_with_table() call.

Output - Notes

By default, rec_with_table() prints notes to the console.

Notes can give important details on what and how variables are harmonized. For example, notes can include details on:

• which datasets include a variable,
• how a variable changes across datasets, and
• if the recoding rules for a variable are dependent on the source dataset

You should read and consider notes before deciding whether to harmonize a variable.

The notes output is from the column notes in variable_details

Arguments - Required

• data A dataframe(s) containing data to be recoded.

• variables Either a string of variables to recode or a dataframe containing the variables (e.g., variables)

• variable_details A dataframe containing the specifications for recoding.

#To recode only a few variables
age_sex_1 <- rec_with_table(data = tester1, 
                         variables = c("age","sex"),
                         variable_details = variable_details)


#To recode all variables listed on the variable worksheet
recoded1 <- rec_with_table(data = tester1,
                          variables = variables,
                          variable_details = variable_details)

Arguments - Default

With rec_with_table() the following arguments are only called if you want to modify them.

• database_name defaults to NULL. As a result, rec_with_table() uses the database indicated in the data argument.

• else_value defaults to NA. Values out of range set to NA.

• append_to_data defaults to FALSE. Recoded variables will not be appended to the original dataset.

• log defaults to FALSE. Logs of the variable recoding will not be displayed.

• notes defaults to TRUE. Information in the Notes column from variable_details worksheet will be displayed

• var_labels is set to NULL. This argument can be used to add labels to a recoded dataset that only contains a subset of variables from variables worksheet. The format to add labels to a subset of variables is c(variable = "Label").

• attach_data_name is set to FALSE. This argument can be used to append a column that denotes the name of the dataset each recoded row came from.

Big-Life-Lab/recodeflow documentation built on Jan. 19, 2024, 10:41 p.m.