ATLAS e-News

Statement on the gender bias article

27 May 2008

Comments on the May 13th feature, Study exposes gender bias in High Energy Physics:

The article about gender issues in the ATLAS e-News of 13 May 2008 has produced strong reactions, and we wish to add here some clarifications. Note that the purpose of the article was to inform the community about a publically available article written by Sherry Towers on her claims of gender bias in D0, and also most importantly to take this as a trigger to promote discussions as to what measures ATLAS may take to avoid possible biases. Let us not forget that the ATLAS e-News are meant to be an informal forum where also individual opinions can be expressed, the e-News are not an official communication channel for Collaboration policy matters.

With this note we want to state very clearly that the ATLAS e-News article does not imply at all that ATLAS as Collaboration would share or endorse the conclusions of the article by Towers, they remain her own responsibility. In particular neither the ATLAS Collaboration Board Chair nor the ATLAS Management have any doubts that the important gender issue is given due consideration in the D0 Collaboration. Furthermore, we wish to point out that we feel very committed to a fair treatment of all ATLAS members, women and men. We are certainly not aiming at any concepts like the ‘productivity’ as postulated in Towers’s article as important criteria of assigning talks for conferences.

Peter Jenni (ATLAS Spokesperson)

Kerstin Jon-And (ATLAS Collaboration Board Chair)

Study exposes gender bias in High Energy Physics

13 May 2008

Sherry Towers, the author of “A Case Study of Gender Bias at the Postdoctoral Level in Physics, and its Resulting Impact on the Academic Career Advancement of Females."

Last month, the HEP community was reeling from the news that a former Dzero scientist had uncovered damning evidence of gender discrimination on the Run II Dzero experiment. Sherry Towers claims that female postdoctoral researchers at DZero had to be three times as productive as their male counterparts in order to be awarded the same number of conference presentations.

Presentations are fundamentally important to the career progression of young physicists in the field. Since author lists on huge collaborative experiments such as Dzero often run into the hundreds – and are printed alphabetically – asking a researcher to present the group’s work indicates to the audience, including potential future employers, that they are considered to have had a significant input.

Towers’s study, which used detailed data from public online databases in order to avoid the possibility of survey bias, looked at 57 postdocs – 48 male; 9 female – working on the experiment between 1998 and 2006. It compared the number of internal papers produced by each researcher with the number of conference talks that they were offered by Dzero’s senior administration, and investigated how this affected the academic career advancement of the young scientists.

The females were involved with more of the internal papers than were 24 of the 48 males. Yet, based on this measure of "productivity," they were each only awarded an average of one third the number of presentations that the average male was. Even when productivity was not taken into account, males still received significantly more conference presentations than their female peers.Not only that, Towers also found conference allocation to be the most significant factor in determining which females went on to secure faculty positions.

The Fermilab Equity Office, to whom Towers submitted a formal complaint on the matter in 2006, declined to investigate, insisting that 10% of all Dzero physicists were female and, accordingly, 10% of conference presentations were given by females. But, says Towers, this did not take into account the relative output of the women in comparison to their male colleagues, for which they ought to have been rewarded justly, nor did it factor for the seniority of the presenters. She also contests Dzero’s baseline figures; her own detailed examination of the 2006 photo-library of researchers showed a female contingent of over 15%.

When Nature reported on this story just after the pre-print of the paper appeared online at arXiv.org, the reaction was astounding. The piece has remained the most commented news item on the website for over two weeks, although many of the comments have simply added weight to the theory that sexism is rampant in the physical sciences. One reader, Paul Kantorek, writes:

“Women in physics are generally harder working than male colleagues … they do not, however, contribute a great deal of original ideas and rigorous logical analysis to the research.”

Michael Pyshnov, a reader of an article appearing in the Chronicle of Higher Education, suggests that the women in the sample were only more prolific because they “just loved to make internal reports”. Towers has heard it all before. “Don’t these comments point to why there is a gender imbalance subtle biases like these are creeping into the decision making process?” she asks.

While the debate rages online between narrow-minded thinkers like Kantorek and Pyshnov, and others who have taken great offence to their and others’ comments, Towers herself has been inundated with emails from female graduate and undergraduate students keen to share their stories of being discriminated against.

“I’ve received only two emails from male particle physicists,” Towers points out, adding that the solution to the gender-bias problem is a “no-brainer”: simply make the conference allocation process more transparent and democratic; take the closed-door meeting and make it public, if there really is nothing to hide.

Transparency is exactly what Co-chair of the ATLAS Speakers Committee, Anna Di Ciaccio, accredits for the apparent success of ATLAS in avoiding a similar situation arising here. The Speakers Committee allocate talks from lists of names provided by the Physics Groups Conveners, taking into account remarks made by Institute Representatives about who deserves particular visibility or promotion. “Many people are involved in making the decision, so I think it makes the decision more justified, and transparency is better insured,” she says.

The prominence of female Physics Group Conveners at ATLAS also helps to ensure, albeit indirectly, that fairness is maintained at the highest level. But Pauline Gagnon of the ATLAS Women’s Network warns that this is no reason to become complacent:

“Even though ATLAS has been appointing women in many key positions, including in the Speakers Committee, it is nevertheless crucial to put in place safeguard systems to make sure no bias creeps in anywhere, be that in conference allocation or representation of women at any level,” she says. “Sexism and discrimination are prevalent in our society and ATLAS is not immune to that. Adding more safeguards will benefit everyone and ensure that no group or individual is unfairly treated.”

Female physicists currently make up around 15% of the total at ATLAS, and are responsible for roughly 22% of conference talks given in 2008, according to Anna. Although this appears to be a promising figure, no statistical data is currently available on the productivity level of anybody at ATLAS, or how this translates into presentation allocations. But, in light of the recent study, ATLAS is already making plans to ascertain these figures in order to promote fairness across the board.

“It’s a serious matter,” says Kerstin Jon-And, Chair of the ATLAS Collaboration Board. “ATLAS is well aware of the issue, and we try to appoint both men and women to the key committees. But, to my knowledge, we haven’t yet quantified the distribution of talks and papers, and I think it would probably be a good idea to do that.”

Following the release of Towers’s paper, discussions were held at ATLAS about how to improve the statistical information currently held about all aspects of work here, to ensure that rewards are given fairly and current measures which strive towards gender equality are proving effective. “We also need to make sure we don’t discriminate according to institute, nationality, or anything else,” said Anna. “This is a basic principle of justice, but to uphold it, we need statistics.”

Towers welcomes the positive moves from ATLAS: “The number of particle physicists in the world roughly matches the population of a small town, so the traditions of how experiments are run just keep getting passed down,” she explains. “That’s why it’s so heartening to have a response from ATLAS showing that they’re actually considering departing from tradition and trying to develop something that’s much more open and democratic.”