The hidden labor cost of inherited data

Although inherited data (a.k.a. secondary data) are cheaper to get than primary data, they're more expensive in terms of data scientist labor-hours. One reason is that you'll have to do lots of extra documentation and you'll also need to detective work to figure out the real story about how the inherited dataset was created. This video walks you through the extra documentation you'll be expected to produce.
At a bare minimum, if you're working with inherited data, I recommend adding this to all your project documents:
“Since our project team did not participate in planning the study or data collection, it is possible that we are missing crucial context which renders our conclusions invalid.”
Learn more on my blog: bit.ly/quaesita_notyours
Don't forget to hit subscribe+notify! If you found this useful or enjoyable (amuseful?), the best way to say thank you is by sharing it.

[@jediyoda7338]

Listing assumptions and caveats is must. Many thanks for these daily tips ….

[@vikashkumar994]

Again great advice.

[@parrotraiser6541]

Was it Socrates who said that his wisdom consisted of realising how much he didn't know, or words to that effect?

[@brain_respect_and_freedom]

👍

[@greensock4089]

I hope nobody uses that disclaimer as a get out of jail free card for being data dangerous

[@azarel7]

C.Y.A !

[@greensock4089]

You're a bit quiet in this video