Invisible Women: Kickoff

Today’s targets

• Discussions and input: Where are you coming from?
• Demo: How did I put what I imagined into something concrete?
• Housekeeping

From the announcement

• Data curation activity
• Invisible Women by Caroline Criado Pérez
• We will find and prepare referenced data for public use.
• If you are inquisitive about the topic “data (collection) bias”, “user research and design” or the book itself, you are welcome to join this group.

Discussions and input

When you signed up for the activity:

• What were your priors about this activity?
• What were you hoping to get or achieve?
• What level of commitment do you see yourself giving?
• What kind of credit are you expecting?

How I imagined the activity

• A distributed form of partial fact-checking

• Workflow:

  • Extract “publicly-digestable” claim
  • Find the “shop talk” version of the claim
  • Point to the data source
  • Write a journal entry or a “Captain’s Log”

• Examples for textbooks and for after-school exploration

  • No excuses anymore, they are free, someone already did the footwork!
  • Research project suggestions for further exploration which may be suitable at any level (there is even a book called Communism for Kids by MIT Press!)

• Studying people and scientists (meta-stuff?)

  • The way people, whether “common” or “trained”, look at / process evidence
  • The time costs of doing distributed open science or team projects

Any similarity is purely coincidental.

• Nancy Reid’s 2023 JSM talk about “The Importance of Foundations of Statistics”

  • Pushes further looking into details of the analysis, the results, and the theory

• Framework for Open and Reproducible Research Training (FORRT) Replications and Reversals Project

  • One of my contributions was the “stakes effect”.

Why imagine it this way?

• “Grassroots” approach vs “in-your-face” approach to change
• Simplest way to onboard is to do archival work: data curation
• Distributed resource building to create community and opportunities to change one’s mind
• “Higher” goals have a more solid and concrete base: “Should the project’s mission be to provide guidelines to include women and other minorities in research, data collection, and policy rather than data curation?”

Tool to be used

• Use Stroll: notetaking tool based on TiddlyWiki with Roam-like capabilities
• Tiddlers -> linked notes -> “card catalog” in wiki format
• Readily distributable without a lot of fuss: only 1 html file is needed
• Localized to your own work style and preferences
• Meant to be casual: let your mind “roam free”
• Aggregate all later for an official version

Demo time

• What to look for (not exhaustive)

  • Claims which are statistics-adjacent: some are of the “descriptive” nature while some are more “domain-specific” stats
  • Biases which you think may be corrected with enough time or hard to correct even with enough time
  • Tidbits involved in the conduct of research, e.g. sex attribution

• A rough “protocol”

  • Claim as extracted from the book should be self-contained.
  • Claim can then be processed by hunting down the main source and the data linked to that source.
  • Broken links may occur: so need to go down a “rabbit hole”. Document the search path.
  • Data may exist but not in the form it originally appears. Document your findings.
  • Write a “Captain’s Log”.

• Processed outputs

  • A zettelkasten of sorts (visually and functionally)
  • Ready-to-use examples and possibly exercises (like writing prompts)
  • And possibly more depending on your interests

Possible questions for your “Captain’s Log”

• Thoughts about the claim, the measurements, etc…
• What search engines were you using?
• How (How far) did you go down the “rabbit hole”?
• Were you logged in? Which browser?
• How long did it take you? Time yourself.
• And more …

Housekeeping

• Meetings:

  • via Microsoft Teams (may change depending on the circumstances and the hiccups)
  • first Monday of every month, 1900-2000 Eastern Time

• The key resource: Invisible Women

  • Fair use issues?