Synopsis

This program contains two do-files: var_label_scan.do and var_labeler.do.

The first (var_label_scan.do) scans all Stata (.dta) files found in a specified directory for variables that are not labeled. The output is a .csv file (var_labels.csv) which contains the following information:

• file: path of the .dta file
• var: variable name
• varlabel: variable label
• samp1-sampN: data sample for the given variable

After the user adds labels to the variables by editing the "varlabel" field of var_labels.csv, the second do-file (var_labeler.do) reads in the .csv file and labels the variables accordingly.

Instructions

1. In the file var_label_scan.do, edit the global variable directory_to_scan with the folder containing the data to scan
2. Specify where the output file should be saved by editing the cd path
3. Specify desired options (see below) and run var_label_scan.do
4. Edit the "varlabel" field of var_labels.csv and save the file
5. In the file var_labeler.do, edit the global path_to_csv with the path to the file var_labels.csv
6. Run var_labeler.do

Options for var_label_scan.do

• allvars: output all variables, including those with labels (default is to only output variables without labels)
• notunique: the data samples do not need to be unique (can speed up the program)
• decode_var: decode variables with value labels before writing samples to output
• samples(#): number of samples to output to excel (default is 5)