Geek Feminism Wiki
Register

Overview[]

Although hiring discrimination on the basis of sex/gender is illegal in most developed countries, women often suffer overt or subtle discrimination when seeking work in technical fields.

Examples[]

One of the most replicated studies in the social sciences is the "résumé test." In this experiment, a collection of people are given résumés and asked to write their opinions of the applicant. The reviewers are given résumés that are identical except for the name at the top of the résumé - e.g., one will be "Jane Smith" and one will be "John Smith." Half the reviewers get the female version, and half get the male version. Their ratings are compared.

If the applicant is perceived to be female, she is rated lower for jobs that are perceived as male-appropriate or gender neutral than the identical résumé with a male name.[1][2] (This experiment has also been replicated with names differing in perceived race and other common parameters of conscious or unconscious hiring discrimination.) The gender of the reviewer does not change the result: Women rate women lower and men rate women lower.

A fascinating twist demonstrates the reviewers' conscious rationalization of an unconscious bias. When asked why they rated applicants as they did, reviewers will use the same data on the résumé to argue for the fitness of the male applicant and the unfitness of the female applicant. Reviewing the female applicant, they'll say "She only published 4 papers in peer-reviewed journals," where as the male applicant gets "He published 4 papers in a peer-reviewed journal, that's impressive!" Another form of rationalization takes the form of cherry-picking whatever job criterion matches the experience of the male applicant and presenting that as the most important qualification. If the male applicant has more technical experience than the female but less management experience, technical experience will be rated more important. If a second reviewer receives the same résumés but with names reversed, management experience will become more important.

Solutions[]

See also Finding women

When you have a job opening:

  • Explicitly state in your job advertisements that you welcome applications from members of minority groups and that you do not discriminate on the basis of sex, sexual orientation, religion, disabilities etc. (Even if such discrimination is illegal in your area anyway, stating it outright makes diverse applicants feel more welcome.)
  • Emphasize objective, measurable, and relevant qualifications over "cultural" or male associated personality traits.
  • Call out specific skills that women are socialized to be comfortable with associating with themselves: collaborative working style, interpersonal skills, time management.
  • Consider whether telecommuting and part-time opportunities can be offered.
  • Consider the implications carefully of hiring exclusively out of your geeky volunteer community. This is good for that community's health, but means you will inherit any diversity problems it has.
  • Evaluate your hiring criteria and procedures carefully to make sure they are not emphasising "cultural fit" qualities that actually mean "very like us".
  • Head hunt senior candidates as well as soliciting applications through advertisement.
  • Advertise in women's geek groups (where allowed, many have job ad facilities).
  • Actively monitor women's geek groups and consider their most active volunteers for roles.
  • Offer small additional referral bonuses for diversity candidates.
  • Since women geeks frequently have geek partners, consider how intra-office relationships can work in your organisation and establish fair, clear policy around it.
  • Avoid making snap assumptions if the candidate's resume includes long gaps. It may take a woman considerably longer to get employed in an industry known to discriminate. If anything, the fact you received her application despite the long gaps would indicate determination in the face of adversity.

Academia Pipeline[]

Colleges and universities should not be exempt from attention on this subject, since they form the pipelines into workplaces. In the scientific work setting, gender bias has been demonstrated using the resume test with the hiring of a laboratory manager as a test case. The bias was shown for both male and female evaluators of the resumes.  In many cases, such discrimination is not overt or even conscious on the part of the evaluator, but is subtle and due to socialized biases such as the perception of women as less competent.  See also the Scientific American blog post on this study.

Biases may also present themselves in the grading and evaluation of student coursework. For instance, academic settings take aggressive measures against plagiarism. A charge of plagiarism applies whether or not the act was "deliberate or accidental." This clause in the official policy leaves students with little recourse to defend themselves if their work closely resembles that of a classmate. Plagiarism charges can lead to an automatic fail, so an individual can be cleanly cut from the program.

It is possible for a programming student to receive a low grade even if a lab or assignment compiles and runs. Less objective ways of evaluating students may influence the grade, especially if a professor is consistently vague in his or her instructions.

Within the same course, it is difficult to compare the evaluation of a male student's work with that of a female student's while trying to avoid charges of plagiarism at the same time. It is easier to pinpoint a problem when a student consistently achieves high marks in other courses and then suddenly achieves drastically lower marks with one individual. It would also be easier to compare if the programming language used in a high grade course is very similar to that used in the considerably lower grade course.

The plagiarism threat also discourages students from helping other with course material, particularly if no tutors are available. This makes them fully dependent on the professor, who may have limited resources and time to answer individual questions, and in some cases, refuse to help or is selective in helping.

Ways to improve objectivity in academia:

  • Instructor staffing can be made to more accurately reflect the diversity of the population.
  • Avoid situations as much as possible where one professor teaches more than one course to the same student.
    • It may be a convenient thing to do if the course material is closely related, but objectivity trumps convenience in the pursuit of proper student evaluation.
    • Having different individuals lecturing and conducting labs (teaching assistants) may also help improve objectivity.
    • There might be something wrong if just two or three individuals control a student's entire semester's GPA, when it could be ten, and they are all from the same demographic profile, ie. A situation where the same professors are teaching different courses more than once, while also being in charge of all labs for those courses.
  • Provide more incentives in tutoring to encourage more to apply for the job, if students in the same course are discouraged from helping each other. People are unlikely to apply for a tutoring job if the wage is just above minimum wage.
  • Assign textbooks that closely follow the course material, rather than as a mere supplement, if students in the same course are discouraged from helping each other.
  • Do not penalize students for withdrawing from courses when they are not objectively evaluated in what should be an objective subject. Co-op programs that demand students be full-time and on-cycle reduce flexibility in this regard, since dropping any core course to re-take it later, with another professor, would disqualify them from entering the workplace through a work placement (See: Leaky Pipeline).

Geek Women with Disabilities in the Academia Pipeline[]

Some professors allow students to submit multiple copies of their assignments and grade the most recently copy as long as it is submitted before the announced deadline. Some do not, so the rule is somewhat arbitrary. A single-submission rule is hostile to students with Attention Deficit Disorder, who may underperform in academic situations not due to lack of skill or intelligence but because of their symptoms. Geek women with ADD that find themselves having to prove themselves over and over again may find things more difficult with their symptoms.

  • A single-submission rule, combined with a plagiarism rule with a "deliberate or accidental" clause, makes it more likely for students to accidentally plagiarise when they are openly invited to use a professor's code samples and then forget to credit it in their first submission. The student may remember 30 seconds later, but find him- or herself unable to correct the error. A student may also accidentally upload the wrong file. The chances of that occurring increases when the same professor teaches more than one course to the same student.
    • The penalty can lead to an automatic fail for the assignment or for the course by the very same professor that invited students to use his or her code samples.
    • This may lead to a last-minute Leaky Pipeline issues if the course credit is essential to achieving a work placement in a co-op or internship program.
    • A student with a documented disability may be granted accomodations, such as extra time for exams, but they may not necessarily be granted accomodations for all policies including accidental plagiarism.
    • It is also risky for visible minority students to get a diagnosis of Attention Deficit Disorder when they are often misdiagnosed by health care professionals in America as well as Europe. Canada is considerably less transparent by keeping mental health diagnosis statistics by race unpublished. For women of colour, a simple disagreement towards a misdiagnosis is to be labeled "argumentative" despite a calm tone.

The simplest thing to do to avoid discriminating is to allow multiple submissions for all students.

Resources[]

References[]