This is a document written by graduate students 1 for everyone involved in academic research. The goal of this document is to identify practical actions that are accessible now to fight against sexual harassment and toxicity in academic research.
We welcome corrections, additions, and reframings. Please make a pull request or contact us if you would like to suggest a change.
- This document should be understood as a series of “patches” rather than a single overarching solution to sexual harassment. There are immediate steps that we can take as a community to help people, even if the larger solutions are still unclear.
- We created this document in response to several prominent cases of harassment and assault within our fields of computing research, frustrated by the lack of tangible actions.
- This document is not about other gender-based struggles in tech and academia for equal pay, promotions, hiring, etc.
- This document is not trying to convince anyone who doesn’t believe that sexual harassment is a problem. This document is targeted toward sympathetic allies. Likewise, none of our suggestions are aimed at perpetrators because we don’t know of any research that this would work (happy to be proven wrong, please let us know if you have evidence).
- This document is not focused on undergraduate students (e.g., it doesn’t discuss important aspects of the undergraduate experience like fraternities, dorms, and underage drinking).
- This document uses the term “victim” throughout, but different people might prefer “victim” or “survivor” depending on the recency of the experience, the framing of the discussion, and other factors. See here for a brief discussion of these terms.
- Additions and edits are welcome and necessary! Please feel free to make pull requests and/or contact us by email.
-
“Keeping the peace” or maintaining the status quo allows harassment and abuse to continue and increase. Silence, secrecy, and closed doors are harassment’s best friends. Taking this strategy, you might stop hearing about problems, but that’s likely because victims no longer trust you.
-
Being an active ally can feel uncomfortable. There might not be one “correct” action to take, and it can be difficult to know how to best support victims.
-
Sexual harassment is not a new problem. There are many resources, research studies, and experts available. Educate yourself and others, and try not to lose hope or treat this problem as unsolvable.
-
Harassers might be nice to you but drive other people out of the field. Be aware of your own biases, and don’t overweigh the harasser’s potential scientific contributions in contrast to their victims’ potential contributions.
-
Beware the “missing stair” — ”a person within a social group who many people know is untrustworthy or otherwise has to be ‘managed’, but whom the group chooses to work around, by trying to quietly warn others of their behavior, rather than deal with them and their behavior openly.” The metaphor of the broken staircase emphasizes the absurdity of creating elaborate workarounds for the “missing stair” rather than directly “fixing the staircase.”
-
Privilege can protect people. If there are ways to convey privilege to others, try to do that. Money and decision-making power are practical ways of shifting privilege.
-
When people have stable financial support (salary, healthcare, not waiting on reimbursements, etc.), they also have more power to make their own decisions.
- A Story of a F*** Off Fund by Paulette Perhach
- The Joy of F***-You Money by Richard Meadows
-
Annual reviews of research advisors by graduate students.
- Professors are reviewed for teaching but not typically not advising—we should add review processes that are more frequent and open to more students.
- Currently, the only review for advising is for tenure/promotion review, which only happens twice and only with a select number of invited student reviews.
- Students often have no outlet for feedback (e.g., a student might know something about another faculty member, not their advisor, but have no formal way to make this knowledge available to promotion/hiring/review committees).
- When evaluating tenure, solicit more letters from students who aren’t direct-reports of the professor up for tenure, particularly including students who previously reported to the professor.
- Put transparent, reliable protections in place for students who are sharing feedback such that they can be protected from persecution for their comments about a faculty member
- Professors are reviewed for teaching but not typically not advising—we should add review processes that are more frequent and open to more students.
-
Hold exit interviews with leaving students (graduating, “mastering out”, or otherwise); these are standard in industry.
-
Maintain records on student retention rates per professor.
- There is some nuance here, since sometimes the safer mentor is the one who supports a student leaving rather than coercing them to stay and continue producing work for the mentor. But someone should be noticing whether there are consistent patterns related to specific advisors.
- Framework for Accountability in Academic Research and Mentoring
-
Provide transitional funding for switching advisors and escaping unhealthy environments.
- This funding can make it possible for students to finish their degree, who otherwise would have dropped out of their PhD programs.
- Examples
- Computing Connections Fellowship (this is a wonderful example!)
- UCSD
- Examples of multiple institutions providing transitional funding
-
Make switching advisors easier, especially for international students.
- Possible improvements include: giving students transitional funding (see above); have faculty members beyond the Director of Graduate Studies who can talk through advisor switches, maybe per-area; keep data on whether certain advisors have students frequently quit/leave their group
- Make resources transparent and easy to find; include info about switching advisors available early on (e.g., in orientation).
- Consider making department requirements more fluid (committee chair and members can be from other departments). For students studying less common topics (often the most vulnerable), there simply may not be someone in the department who fits well enough to be a chair.
-
Formally include diversity and inclusion service as a factor in hiring and promotion, on an equal footing to other metrics. You might have to fight for this (and there are open questions about how to enforce this once implemented), but it can be done!
-
Recognize how intersecting identities (e.g., race, disability, gender) can make some people more vulnerable to harassment and less likely to be believed when they seek help.
- How Title IX has failed to protect survivors of color
- Race, Gender, and Sexual Harassment by Kimberlé W. Crenshaw
-
In general, workplace sexual harassment trainings don’t work and can actually increase sexual harassment rates. We recommend either providing trainings about available resources for victims or skipping trainings altogether.
- The promise and peril of sexual harassment programs by Frank Dobbin and Alexandra Kalev
-
Consider the negative effects of trainings on imposter syndrome, teaching people with marginalized gender identities to talk or act like men, implicit bias, etc.
- How Diversity Training Hurts Diversity by Rachel Thomas
- This Twitter thread on physical self-defense trainings by Nicole Bedera
- Stop Telling Women They Have Imposter Syndrome by by Ruchika Tulshyan and Jodi-Ann Burey
- All Career Advice for Women is a Form of Gaslighting by Ephrat Livni
-
PIs might want to explain to their students what to expect at conferences and what is acceptable and not acceptable behavior. Make sure students know where to report incidents in different scenarios (at conferences, while TAing, at department social events, etc).
-
Fund and promote bystander training so that harmful behavior is immediately called out.
- Here’s one nice method to employ as a bystander: “We don’t do that here.”
-
Consider the framing, timing, and required audience for trainings.
- Requiring sexual harassment training for everyone can be upsetting to victims, who can feel gaslit by these trainings after negative experiences seeking protection or recourse from the same institutions giving the training.
- One option (e.g., used at the University of Washington) is to provide an opt-out for most of the modules of the training for anyone who emails asking for accommodations (no explanation necessary, but you do have to make the extra effort to email Title IX).
-
Siloes, silence, and secrets allow harassers to thrive. They often are very skilled at isolating their victims and portraying themselves differently towards different people.
-
Create spaces for people to talk and feel safe comparing notes:
- General spaces (seminars, colloquium, reading groups) to build community and cross-lab connections.
- Focused spaces for minoritized groups to talk to each other. Relying on whisper networks can be harmful (since you have to be a member of the network to benefit) but they can also be powerful (connect people with similar experiences of the same person so that they can identify patterns and build a combined case).
- Educating spaces for allies to learn and prepare to protect others.
- Physical spaces that foster interactions between different groups (private meeting rooms, community lunch rooms and gathering spaces, comfortable offices).
-
Because of graduation timelines, cohorts of PhD students have to relearn much of the info in this document; master’s students have even shorter timespans and are thus even more vulnerable. One way to combat this is to create and support groups that pass down learnings from one cohort to the next. It can also be a good reason for such groups to have faculty or outside (e.g., lawyer) liaisons who persist beyond PhD timelines.
-
Cultivate a culture of respect instead of a culture that prizes aggression and rudeness, or that forces people to endure a “gauntlet of abuse” before being accepted into the community.
- Use science to stop sexual harassment in higher education by Kathryn B. H. Clancy et al.
-
In general: normalize calling people out immediately. Foster the sense that “we’re a team that needs to keep each other safe and that means calling each other out.” We all make mistakes sometimes, and calling each other out also means accepting each other’s fallibility.
- Community Agreements from the Intergroup Dialog Project at Cornell University
- Calling In and Calling Out Guide from Harvard University DEI
-
Compelled reporting (where someone reports what a victim has told them, even when the victim asks them not to share) can lead to retaliation, re-traumatizing by the investigation process (see below about Title IX), and other harms. See this Twitter thread and the following resource for further explanation about why compelled reporting is harmful: Compelled Disclosure: The Problem with "Mandatory Reporting" of Sexual Violence on College Campuses
-
Mandatory reporting policies, which are common at universities, often require compelled reporting. Until these policies change, you should make explicit who is a mandatory reporter and make explicit that confidentiality matters (as victims might not be aware of the harms that can arise from mandatory reporting and not know that they need to confirm confidentiality). Provide options who are not reporters. For example, faculty might want to list their mandatory reporter status in their email signatures or in their course syllabi.
-
As of August 2024, all U.S. faculty and staff are mandatory reporters thanks to Biden’s new Title IX regulation.
-
See the Academic Alliance for Survivor Choice in Reporting Policies for more information and resources related to mandatory or compelled reporting.
-
Some practical advice from this Twitter thread by Nicole Badera for mandatory reporters:
- "If a survivor is asking you for your help reporting, you should initiate a report. Always."
- Put a warning statement in your syllabus and on your website. This is a great example of such a statement: Mandatory Reporting Syllabus Statement
- Prepare for how you would handle reporting a disclosure: "...the short answer is to be a 'bad' mandatory reporter. You want to report as little as possible to reduce the likelihood that an investigation can take off without the survivors' consent. Don't pry for details. Don't name a perpetrator. Don't fish for contact information. Ideally, your report should include nothing more than the survivor's name and email. Then, it will be up to them to decide what—if anything—to share with Title IX. Be transparent about the mandatory reporting process. Tell the survivor what you will and will not share. Let them sit alongside you as you report if they want. As much as possible, prioritize their autonomy. Answer their questions. And connect them to a confidential victim advocate. Make sure they know they can have someone in their corner who won’t share everything with Title IX. In today’s Title IX terrain, that might require contacting an off-campus rape crisis center. Don’t forget to validate the survivor. Thank them for sharing their story. Apologize for your school's mandatory reporting policy. Promise them that you will prioritize their autonomy, even if their school won't. Let them know about other ways you can help if they need it."
- Reporting processes can be confusing or even harmful. A knowledgeable guide can help victims set expectations and get what they need from the process.
- See the doula metaphor from this Twitter thread by Sarita Schoenebeck
- Make sure these people are reliable (not law student volunteers), independent (not as helpful if they are employed by the university), and not mandatory reporters.
-
Prepare students: explain code of conduct, tell them what is inappropriate, connect them with senior students who can guide them, remind them that they do not need to drink/attend every social event if they do not want to.
-
Have an enforceable Code of Conduct (see below).
-
Do not pressure or force students to share hotel rooms.
-
Carefully decide whether and how to serve alcohol.
- Many perspectives and pros and cons in this Twitter thread and summarized here by Casey Fiesler
- Alcohol can be difficult at work events for many reasons, including gender discrimination (assuming pregnancy when not drinking), religious discrimination, health discrimination (alcoholism and other conditions).
- Arrange alternative activities that don’t involve alcohol (museum trips; exercise groups; academic interest brown-bags; non-alcoholic food and drinks)
- If you decide to serve alcohol:
- Also serve alternatives of equal quality (cost, taste)
- Use drink tickets
- Consider requiring people to pay for their own drinks
-
Understand the history of Title IX; it’s a bureaucratic, legalistic structure from a time when students were desperate for any protection.
-
Understand that Title IX can be more likely to hurt than help victims
- Examples and sources:
- Title IX: Falling short at 50: Despite men’s rights claims, colleges expel few sexual misconduct offenders while survivors suffer
- In survivors’ words: How colleges should better respond to sexual misconduct
- How University Title IX Offices Fail Survivors (“I chose certainty of safety over the possibility of justice.”)
- I Can Protect His Future, but She Can’t Be Helped: Himpathy and Hysteria in Administrator Rationalizations of Institutional Betrayal
- Examples and sources:
-
Understand that all such structures are built to protect the university as a money-making institution, not the victims
-
Request copies of all transcripts and records. Know that harassers might also have access to all transcripts and records.
-
It can still sometimes be useful to go to Title IX, but carefully weigh the pros and cons and take an advocate or lawyer with you (ideally a lawyer, but at the very least, take a supportive, powerful ally like a professor, who can witness everything and if nothing else, validate your experience). As an ally, offer to act as a witness or “support person” or help identify such a person. As a “support person,” you probably won’t be able to talk during the meetings, but you can still offer important emotional support to the victim.
In whatever space (your home department, conferences, etc.) remove plausible deniability and make bad behavior clear and identifiable
-
Have an enforceable Code of Conduct
-
Dating is fine! No one is trying to outlaw dating between people who don’t have power over each other! What’s not ok is: dating someone you have power over; asking out every single person in a space; asking the same person out repeatedly; touching without affirmative consent (this includes hugs and touching shoulders/backs/arms); not responding to complaints by changing behavior; etc.
-
Example Codes of Conduct:
Be careful about “solving” problems by passing them to other labs, departments, advisors, universities
-
In cases with a known serial harasser, “passing the harasser” might protect the current victims but provides the harasser with fresh victims who are probably outside the whisper network of the harasser’s previous community.
-
This can perversely help the harasser, whose CV now includes multiple collaborations and affiliations.
-
Sometimes, you may witness a case of passing the harasser that you have no power to stop. In this case, if you can safely do so, consider sharing information with people in the new community.
-
Most importantly: institute policies around hiring, recommendations, etc. that allow for this kind of feedback. Require job applicants to release information from their previous employer about sexual and other misconduct.
- See these recommendations from the Association of American Universities.
- See this legislation in Washington state.
- Datafication turns human experiences into data. Data can be useful to make arguments and track improvement, but datafication can also take resources away from solving problems that are already well understood. There's a place for surveys and documentation (e.g., tracking graduation rates, asking students about their comfort in the department), but these should be weighed carefully against other efforts. Don't immediately prioritize demands for evidence that sexual harassment is a problem in research; your energy might be spent more effectively elsewhere.
- Viral Justice by Ruha Benjamin
- Data Feminism by Catherine D'Ignazio and Lauren F. Klein
-
Universities are caught in the bureaucratic trap of Title IX, so official university procedures are unlikely to result in protection or justice. For example, some universities interpret Title IX as preventing individual university departments from having their own enforceable Codes of Conduct that meaningfully address sexual harassment and assault (though other universities may have different interpretations).
-
Conferences are not bound by Title IX! Conferences can have their own Codes of Conduct and justice/protection systems.
-
You can also look for other people with power in the situation, e.g., the harasser’s committee, the chair of the harasser’s department, hiring committees and workshop committees working with the harasser. (Each of these should be approached with caution as no individual can be assumed to be an ally.)
-
If you feel safe doing so, speaking publicly (social media, news outlets) is another alternative.
-
You can completely circumvent Title IX and your university by going directly to the U.S. Office of Civil Rights.
- You can file a complaint here.
-
Understand that when you prioritize giving a harasser a second chance, there will likely be a large, expanding cost to their past and future victims, and to the general trust in your community. Consider instead giving a “second chance” to the victims.
- This isn’t meant to be retributive, and this doesn’t mean you should never give someone a second chance. It means considering how many chances this person has already had, how many chances their victims have had, and how your community can both protect victims and create an environment where harassment is unlikely (the harasser has a chance to not be a harasser).
- If tempted to protect a serial harasser because of their research, consider that their research output, no matter how brilliant, does not outweigh the lost output of multiple other people being forced from the field.
-
Your name as a researcher carries a ton of weight and respect (even awe for junior students). Use it wisely.
- For serious cases of harassment or other abuse:
- Think very carefully about collaborating with someone you know has harassed, abused, or assaulted others. Think about what this signals to other people in your community who are aware of the perpetrator. Think about what this signals to the victims. Think about how those not in the know (e.g., more junior students) might interpret your association as an endorsement, leading to fresh victims for the perpetrator.
- Similarly, think carefully before promoting their work (e.g., on social media or via emails) or sharing or endorsing their hiring advertisements.
- Citing and teaching a harasser’s work can itself carry harms, but not citing or teaching their work can also harm their (possibly junior) coauthors. Consider how to cite and teach while giving context to your audience.
- For serious cases of harassment or other abuse:
-
You might make mistakes. Harassers are often very good at disguising their actions, and they might behave very, very nicely to you and others who aren’t their victims. But at least do some research and ask around, and try to do better going forward.
- Harassers are nice to me, and probably to you by Sarah Milstein
-
Talk to your students one-on-one about TA assignments. Check that they’re comfortable with the professor and other TAs. If you know a professor or TA will cause problems, shield your student by letting them TA for a different course.
-
When hiring, speak with multiple references/contacts by phone if possible.
-
As an advisor, if your students aren’t bringing you problems, they probably don’t trust you. Grad students have all kinds of problems (money and reimbursements, family, physical and mental health, visa issues, research, sexual harassment, racial harassment, etc.) so if you haven’t heard anything, it’s unlikely to be because they’re not experiencing problems — they’re just not telling you.
-
Speaking up or calling people out can be scary. It takes practice! Consider taking bystander training to prepare yourself.
-
See above about mandatory reporting and how you can protect your students while being a mandatory reporter. All U.S. faculty and staff are now mandatory reporters.
-
There are no right answers here, and there are costs and benefits to each path forward.
-
Prioritize your health and safety. Your survival is a triumph in itself.
-
Find allies, both powerful (professors) and supportive (other students who have gone through similar experiences). Allies who are independent from your institution can be particularly valuable (e.g., you might have a local advocacy center not affiliated with your institution).
-
Carefully research (or ask your allies to help you carefully research) your reporting options. Each option has costs and benefits.
-
Always ask about confidentiality before sharing your story. Many academic resources are not confidential and may include mandatory reporting.
-
Document everything. Record everything, if possible. Write things down in a dated journal. Tell other people you trust (if safe) so that you have witnesses. Screenshot/save digital communication. Try and get communication with departments/people in power in writing. This can help you both make your case to any arbitrator and can validate you later, if you are gaslit or begin to doubt your own experiences.
-
Proceed slowly and cautiously with all reporting processes (e.g., Title IX, HR). Consider going through an official reporting process (e.g., Title IX) only if you have the energy and strength for a fight. The advantage is that there will at least be some kind of official report and evidence, removing deniability for the university if more problems happen. But there are many downsides (see above). Importantly, your harasser might have access to anything you say.
-
Other options often include: community groups (e.g., women’s resource centers, student organizations), community and university police, going public (social media, news outlets) with your experience or asking someone with more power to relay your experience publicly.
-
Unfortunately, the process of seeking support can itself take a toll on your mental health. Don’t discount this. Lean on your support network. Seek mental health support or ask your supporters to help you find mental health support.
-
There are no right answers here, and there are costs and benefits to each path forward.
-
Center the victims and what they tell you they need. Do not do anything without the victim’s explicit permission. Here is a great guide to being an ally.
-
Offer to accompany them to appointments and meetings as a witness and support person.
-
Help them through the bureaucratic process, which can be exhausting and overwhelming. Help them research their options, identify powerful allies, navigate Title IX and other systems, organize documents, help them navigate the healthcare system to receive mental health support.
-
Listen without judgment and depending on the situation, perhaps without offering advice or solutions. Be a witness and believe their story. Guard the victim from their harasser. If you are at an event where both are present, stick with the victim. Never force the victim and harasser into conversation or interaction, no matter how mundane or innocent it may seem to you or how much time has passed.
-
Are there other evidence-based training programs that can help keep researchers safe?
-
How can students collaborate across universities to improve research culture? How can students collaborate with lawyers, activists, and other people external to academia?
-
This document arose most directly in response to cases of gender-based harassment—how should these actions change based on other aspects of marginalization and identity?
-
It can be very hard to implement the changes recommended in this document, taking time away from other research activities. Efforts to improve safety should be tangibly rewarded by managers and review committees.
-
It’s awful to consider the collective amount of time and energy wasted on fighting these creeps. They indirectly hurt our research and scientific advances in so many ways, in addition to all the direct costs of their harassment and abuse.
-
Related: Gaslighting from universities and departments compounds this wasted energy. Institutions with power can help or hurt; anyone with power should consider which they are doing.
- I Can Protect His Future, but She Can’t Be Helped: Himpathy and Hysteria in Administrator Rationalizations of Institutional Betrayal by Nicole Bedera
- The promise and peril of sexual harassment programs by Frank Dobbin and Alexandra Kalev
- Compelled Disclosure: The Problem with "Mandatory Reporting" of Sexual Violence on College Campuses by Jennifer J. Freyd
- Beyond Trigger Warnings: A Survivor-Centered Approach to Teaching on Sexual Violence and Avoiding Institutional Betrayal by Nicole Bedera
- Race, Gender, and Sexual Harassment by Kimberlé W. Crenshaw
- Down Girl: The Logic of Misogyny by Kate Manne
- The Real Reason Women Quit Tech (and How to Address It) by Rachel Thomas
- Feature, Not A Bug by Mar Hicks
- Harassers are nice to me, and probably to you by Sarah Milstein
- How Diversity Training Hurts Diversity by Rachel Thomas
- Allies’ Guide (Callisto)
- Title IX: Falling short at 50: Despite men’s rights claims, colleges expel few sexual misconduct offenders while survivors suffer
- In survivors’ words: How colleges should better respond to sexual misconduct
- How University Title IX Offices Fail Survivors
- The Mobilization of Title IX across U.S. Colleges and Universities, 1994-2014 by Celene Reynolds
- More than 70 lab heads removed from NIH grants after harassment findings
- Complaint by Sara Ahmed
- On the Wrong Side: How Universities Protect Perpetrators and Betray Survivors of Sexual Violence by Nicole Bedera
- Sexual Justice: Supporting Victims, Ensuring Due Process, and Resisting the Conservative Backlash by Alexandra Brodsky
These are researchers focused on sexual harassment in academia whose work we've learned from. They were not involved in the creation of this document.
- Dr. Nicole Bedera
- Dr. Kathryn Holland
- Dr. Celene Reynolds
- Dr. Kate Clancy
- Dr. Sarah Peitzmeier
- Dr. Charlene Y. Senn
- Dr. Kate Manne
- Dr. Frank Dobbin
If you or your department would like to hire someone to help design a safer workplace, you might want to check out Beyond Compliance Consulting which is run by experts and follows evidence-based approaches.
Footnotes
-
Maria Antoniak and Alexa VanHattum, beginning May of 2021, with significant input from other graduate students and faculty from Cornell and other universities. This document is written from the perspective of graduate student researchers in Information Science and Computer Science at a North American university, as white women with a lot of privilege. This document is very much a draft written by non-experts. We don’t research sexual harassment, though we’ve thought about it a lot, read about it a lot, and talked about it a lot with victims and allies. We welcome corrections, additions, and reframings. ↩