Code review Q2 2020 : Data publication

[luizaandrade]

Data cleaning code review checklist

Data source/survey round

Date

List of files to be checked [Add names or links]

• Master script

• Clean dataset(s)

• Cleaning scripts

Identifiers

• Deidentified data does not contain identifying variable

Reproducibility

• All scripts run from the master after adding the correct folder path to line(s) X (and XX)
• The master script is organized in a way that allows you to understand the general tasks being performed in the code
• The master script tracks which scripts create and use which files
• The data sets created by the reviewer are exactly the same as those shared by the coder

Code organization and readability

• Code names are informative
• It is clear in the code why tasks are being executed
• The code structure facilitates understanding of the tasks
• Code uses white space to improve readability
• There is extensive use of comments to explain the code
• The code is efficient (tasks are executed in the simplest way possible, loops are used when needed rather than repeating lines, pre-defined functions are used)
• Common tasks are abstracted and automated (e.g. using functions or macros)

Clean data set checks (pre-publication)

• The data does not include direct identifiers
• The data set has a clearly labeled, uniquely and fully identifying ID variable
• The level of observation of the data set is clear from the dataset name, ID variables and documentation
• Variables have informative labels or an acompanying dictionary
• Categorical variables have clear and informative value labels
• No modification is made from the raw to the clean data other then correcting problems
• No raw variables are processed (winsorized, for example)
• Variables can be easily traced back to the original questionnaire

Data cleaning tasks

• Are new variables being created in the cleaning do-files?
• Are any changes being made to observations values in the cleaning do-files?
• Check merges: Are any observations dropped? If so, is there a clear justification for that? If any observations didn't match, is that explained in the comments?
• Are missing values coded consistently? Are extended missing values used?