3 Getting clean and usable data

Cleaning your data will ensure that you always have an up-to-date database that can be analyzed without the risk of misleading those using information from the analysis. Database cleaning is primarily a logical process, including analyzing the consistency of data and performing triangulation with other information that may be available.

Some detailed resources specifically dedicated to data cleaning can be reviewed for in-depth, further details:

• Best practices for data cleaning, MDC Toolbox, CartONG -TDH, 2017

• Technical Brief on Data Cleaning, ACAPS, 2016
  • The Annex 01 provides a practical checklist for data cleaning.

If you are using excel, you can also directly review the section on data cleaning in the Toolbox dedicated to Excel (available in French) to learn how you can technically run these steps.

This section is divided into 4 sub-sections:

• 3.1 How to start data cleaning? A few steps to follow,
• 3.2 Export and format data,
• 3.3 Dealing with duplicates, errors, outliers, and missing data,
• 3.4 Recoding variables.