This workshop was prepared by Nanticha Lutt, Tiffany Lowe-Power, Dhruv Patel, and Michi Taga for the 2018 Plant and Microbial Biology Departmental retreat at UC Berkeley.

If you use these materials, we would appreciate it if you fill out the following form so we can track usage: Form Link. These metrics might help us justify resources for future efforts.

Recognizing implicit bias in recommendation letters reduces the impact of that bias on our judgement of applicants’ qualities.

• The goal was to advance diversity and equity by minimizing a gatekeeping barier to advancement in scientific careers. Implicit bias in letters of recommendation has been most studied in letters for male vs female applicants, so our data driven presentation of the problem/barrier focused on the types of implicit bias in letters that disadvantage women. However, we hope that letter evaluators and letter writers can use a similar lens with which to identify bias based on racial, class, gender identity or other categories.

• One goal of this workshop is to promote awareness of implicit bias. Once bias is recognized, the productive path forward is less clear, but some of our ideas are presented at the end (see last ppt slide). A further goal is for letter writers to recognize implicit bias in their own letters and revise them accordingly.

Timing: 1 hr.

• Brief presentation to present the scientific data on the manifestations of implicit bias in letters.
• Prepare the audience for a small group discussion where they read mash-ups of excerpts of anonymized letters of rec from the Trix and Psenka 2003 paper. Note: When we presented this the first time, the audience didn't understand the context of the source material, which distracted the small group discussions. Make sure they understand that the text of the letters originated in bona fide rec letters for medical faculty positions (although it has been anonymized in the original paper & aggregated into just a few letters here).
• Break out sessions: everyone reads the sample letters and in groups of 3-4, they identify and discuss the manifestations of bias in those letters.
• Round-up discussion: a Poll Everywhere was set up to have respondants answer questions in real time via their phones.

1. Source Material-Trix and Psenka, 2003
2. Slides
3. Break-Out Activity
4. Guide for Activity Facilitators
5. Handout - implicit bias and gender
6. Handout - implicit bias and race. Note that this isn't referenced in the workshop explicitly because I found this after the workshop was hosted in 2018.

Alternatively, All files are zipped here

This work is licensed under a Creative Commons Attribution 4.0 International License.

Any of the materials can be edited, improved, or tailored for a specific audience. Although we do not require it, we would appreciate the workshop being acknowledged as having been "originally designed by the UC Berkeley PMB department Diversity and Equity Committee".