Silver Linings: How a Data Harmonisation Mishap Turned into a PhD Thesis

Date

March 7, 2024

Format

5min (Virtual)

Presenter

Cynthia A Huang

Venue

Submitted Talk, Data Mishaps Night 2024

 

Submission Notes

Title Ideas:

  • The story of the data mishap that gave birth to my PhD.
  • How losing and finding data …
  • A ‘simple data integration’ complex mishaps…
  • Building a simple tool to avoid a data mishap turned into a whole chapter of my PhD

TAKEAWAY: The right abstraction can help us avoid mistakes

Pitch:

How a “straight-forward” data harmonisation task turned into a month of detective work to figure out why my data didn’t match reference data prepared and published by other researchers, which let me to try and build a “simple” tool for implementing and validating many-to-many categorical recoding (not just one-to-one)… which has turned into a whole chapter in my thesis on mathematical abstractions of harmonisation

Slides