Women are not under-represented in science, they dominate in some fields and lag in others, part of the natural variation in humanity - but in academia at the highest levels, they are under-represented.

The reason is not sexism; women are actually over-hired, when they apply for the jobs. Instead, it is factors like that we don't engage in age-ism and penalize older researchers, the same reason that the average age of R01 grants has risen as older scientists have worked longer, so men were not fired to make room for women. The other big reason is that, by choice or by default, women care for kids more. Unlike corporations, the academic world demands a certain zealotry until you get tenure, so policies are far less-friendly, and there may be implicit bias in principal investigators who hire young researchers and wonder if a woman might soon want a baby. 

What can't be denied, and what has nothing to do with the retirement age of faculty members, is the number of women that are invited to speak at conferences and the percentage of grants awarded to women scientists. 

Writing in Cell Stem Cell, the Initiative on Women in Science and Engineering Working Group present seven strategies to advance women in academic science, engineering, and medicine the same way their careers have taken off in the private sector.

Some of the recommendations are a little too discriminatory to be implemented but the remainder are both workable and would make a difference.

1. Implement flexible family care spending

Make grants gender neutral by permitting grantees to use a certain percentage of grant award funds to pay for childcare, eldercare, or family-related expenses. This provides more freedom for grantees to focus on professional development and participate in the scientific community.

2. Provide "extra hands" awards

Dedicate funds for newly independent young investigators who are also primary caregivers to hire technicians, administrative assistants, or postdoctoral fellows.

3. Recruit gender-balanced review and speaker selection committees

Adopt policies that ensure that peer review committees are conscious of gender and are made up of a sufficient number of women.

4. Incorporate implicit bias statements

For any initiative that undergoes external peer review, include a statement that describes the concept of implicit bias to reviewers and reiterates the organization's commitment to equality and diversity.

5. Focus on education as a tool

Academic institutions and grant makers must educate their constituents and grantees on the issues women face in science and medicine. For example, gender awareness training should be a standard component of orientation programs.

6. Create an institutional report card for gender equality

Define quantifiable criteria that can be used to evaluate gender equality in institutions on an annual basis. For instance, these report cards may ask for updates about the male to female ratio of an academic department or the organization's policy regarding female representation on academic or corporate committees.

7. Partner to expand upon existing searchable databases of women in science, medicine, and engineering

Create or contribute to databases that identify women scientists for positions and activities that are critical components for career advancement.

2 and 3 were clearly created by someone in sociology with no experience in the real world. Number two is basically advocating a 'kiddie pool' segment of STEM, which isn't helping anyone, while number three says we shouldn't pick the best scientists, we should pick people of each race and gender who apply, regardless of their competence.  The rest would seem to be possible and helpful and already done, to a large extent, since universities love committee meetings.

The 2015 IWISE Working Group Meeting was supported by a grant from the Doris Duke Charitable Foundation. The initiative is also supported by The New York Stem Cell Foundation.

Cell Stem Cell, Smith et al.: "Seven Actionable Strategies for Advancing Women in Science, Engineering, and Medicine"