UNAIDS Data Accessibility

Project Purpose

Process, clean, and host tidy UNAIDS’ annual HIV Country Estimates data for analyts’ ease of access and use.

Overview

• Munged the messy spreadsheet UNAIDS posted on their site to tidy it into an analytic and viz ready dataset
• Added functions for standard graphical views
• Created a way for public data to be easy hosted and accessed

Outcomes

• Standardize output of an untidy dataset
• Added features to the dataset (e.g. PEPFAR affliation, goal reach, etc) to minimize analysts reinventing the wheel
• Allowed for easy access of data through an R package

Lessons Learned

• Leveraged GitHub Releases to host the dataset
• Designed unit tests
• Worked with stakeholder to understand data needs and explain limitations
• Learned the importance of hiding developer only functions from end users
• Optimized data storage/access as well as function parameters
• Improved naming conventions for functions
• Collaborated with UNAIDS to gain access to their database to establish a query