Added checklist item for downstream bias mitigation

[glipstein]

Context:
This PR focuses on a missing piece in A. Data Collection about considering ways to include data that will facilitate downstream bias measurement and mitigation, for instance by including demographic data for protected groups (e.g. race, gender). This seems to be a huge existing barrier to being able to trust ML systems not to codify or perpetuate existing inequities. It's also in line with Ibram Kendi's quote on "ignoring race" that "Terminating racial categories is potentially the last, not the first, step in the antiracist struggle."

Considerations for review:

• Updated checklist in yml and docs
• Updated examples in yml and docs
  • added apple pay example from list in issue Process for contributing to Examples / Resources docs #98
  • moved up facial recognition example from fairness across groups
  • moved amazon recruiting bias example from proxy discrimination to fairness across groups (since it seemed a better fit there, and the list otherwise had almost all google examples)
• I also had come across this article "How Algorithms Discriminate Based on Data They Lack: Challenges, Solutions, and Policy Implications." Not sure there's a place for things like this, but dropping here for reference https://www.jstor.org/stable/10.5325/jinfopoli.8.2018.0078
• If this is merged, will need to create an issue for the translated files to reflect the updates

Also

• As a bonus, added misleading representation example from Example for C.3 Honest representation: Georgia COVID-19 Viz #104 while in the examples files

[codecov]

Codecov Report

Merging #119 (dd9a8b9) into master (60b7f8d) will not change coverage.
The diff coverage is n/a.

@@           Coverage Diff           @@
##           master     #119   +/-   ##
=======================================
  Coverage   96.80%   96.80%           
=======================================
  Files           6        6           
  Lines         188      188           
=======================================
  Hits          182      182           
  Misses          6        6           

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 60b7f8d...dd9a8b9. Read the comment docs.

[ejm714]

Some suggested rewordings and different locations for examples. @glipstein take a look at let me know your thoughts

[jayqi]

Looks good to me!