search

dalejn / cleanBib

Probabilistically assign gender and race proportions of first/last authors pairs in bibliography entries
MIT License
149 stars 31 forks source link

Testing #6

Open dalejn opened 4 years ago

dalejn commented 4 years ago

No specific technical skills are needed to test the Binder. Please follow the instructions provided https://github.com/dalejn/cleanBib#instructions

While running through them, note that the instructions are designed for users with no coding experience. If you get stuck, we can probably do something about it to prevent it from happening to future users, either by clarifying the instructions or changing the code. So, we would love to hear any experiences of friction / sticking points!

jessicabartley commented 4 years ago

I have some suggested text for the instructions intro. It might be good to start with a quick framing and explanation of terms. When I first read the intro paragraph ("The goal of the coding notebook.... will be calculated.") I was a bit unsure about the .bib file (I use a Mendeley Word plugin to generate my references sections, not a .bib file) and the sentence organization made me think I needed to have already generated a .bib file in order to start the tutorial. I also thought it might be helpful to front-load the first author/last author way conceptualizing the gender representation breakdown.

I'm not sure how to suggest this edit as a (what I'm assuming is a) github.io webpage pull request, so I'll just add the suggested intro/framing text here :)

CleanBib* is intended as a tool to help identify and reduce bias in citation practices. The goal of this notebook is to walk through how to use cleanBib to generate a summary of gender representations across first and last author name pairs in your manuscripts references section.

In this instruction we will go over how to 1) generate a .bib file from your references, 2) clean the .bib file so it contains only references cited in your manuscript, 3) use the cleaned .bib file to generate a data table of author names that will query the probabilistic gender classifier, Gender API, and 4) calculate proportions of the predicted gender for the first and last author pairs (man/man, man/woman, woman/man, and woman/woman) from your references section.

*Insert the tool name if I've got it wrong

cleanthis-michael commented 4 years ago

I was also confused about having/generating the .bib file, and it took me a little time to figure out how to generate it since I've also been using the Mendeley Work plugin in the past.

After testing the code, the authors names of the papers below are extracted as first initials:

  1. Efficient computation and cue integration with noisy population code (Deneve, Latham, & Pouget, 2001).
  2. Sequence of changes in properties of neurons of superior colliculus of the kitten during maturation (Stein, Labos, & Kruger, 1973).
  3. The infinite regress illusion reveals faulty integration of local and global motion signals (Tse & Hsieh, 2006).
  4. Physical limits to sensation and perception (Bialek, 1987).
  5. Human cortical magnification factor and its relation to visual acuity (Cowey & Rolls, 1974).