WeWork ($WEWORK) is top-of mind right now, with the workspace company's IPO and fascinating S-1 among the top of the heap. In the shadows of pre-IPO deconstructions of WeWork, however, is controversy over the company's hiring and HR practices. A gender discrimination suit alleges the company pays men more for the same jobs at WeWork U.K. Meanwhile, recent headlines called out the company for having an all-male board.

But when it comes to the way the company is searching for new talent, it appears to be ahead of the game. That's because language bias in WeWork job listings has trended toward being more female-friendly over the past year.

We ran thousands of job listings published by WeWork through the same algorithm created by Kate Matfield that uses a list of gender-coded words from a Journal of Personality and Social Psychology research paper in order to quantify "gender-coded" language. It aims to uncover the job listings that may be discouraging to females or males, respectively.

By running WeWork's job listings through the gender decoder, we get a look at how WeWork is aiming to fill future positions via the language it uses in job listings. As you can see above, in 2017, WeWork coded job listings, on average, at a 46% male bias. These days, however, new job openings are written at a 41% male bias, favoring female applicants.

While that looks good for WeWork on the whole, as it suggests the company is aiming to balance out a male-skewed workplace, not all divisions at the company are coding their job listings equally.

The most male-biased job listings at WeWork are in its Legal division, at 50.62%. Finance comes in at 46.88% followed by Technology at 44.62%.

Category

% Masculine Bias (Average)

Legal

50.62%

Other

46.88%

Finance

46.35%

Technology

44.62%

Intern

43.65%

Education

43.32%

Community

42.72%

Information Technology

40.22%

Marketing

38.19%

People

38.00%

Building Operations

37.44%

Business Operations

36.77%

Building Design & Development

35.77%

Sales

35.64%

Public Affairs

34.76%

Real Estate

31.30%

Executive

15.31%

At the lowest end is job openings categorized as "Executive", where job listings strongly favor female applicants with a male bias of just 15.31%. This is likely because the number of listings for executives is much smaller than those of other divisions, but it also shows that the company is making a concerted effort to attract top female talent.

For instance, the most in-demand job cateogry at WeWork is for "Community", which includes professionals that run local offices and engage directly with WeWork tenants and customers. These job listings favored male applicants in early 2019 at a male bias of nearly 53%, but as of late have dropped to 41% on average.

For Technology job listings — WeWork's second most in-demand job type — male bias in job listing language has been on the rise since April 2019, approaching 50% at a current average of 46%.

Sales job listings at WeWork are fairly female-friendly, with a gender bias of just 36.5%. This after a early-2019 high of a still-female-friendly 38.6%.

The rest of WeWork's various divisions are on divergent paths when it comes to gender bias coded in to job listings. Of course, the fewer openings in these divisions makes averaging out gender bias a bit more erratic, but we include some outliers for comparison.

About the data and the gender decoder:

This gender decoder is based on the research paper "Evidence That Gendered Wording in Job Advertisements Exists and Sustains Gender Inequality" (Journal of Personality and Social Psychology, July 2011, Vol 101(1), p109-28), written by Danielle Gaucher, Justin Friesen, and Aaron C. Kay. In this paper, the three looked at language in job descriptions they found to be "feminine" or "masculine" in nature and whether or not men and women may be off-putted by sample job descriptions.

Thinknum tracks companies using information they post online - jobs, social and web traffic, product sales and app ratings - and creates data sets that measure factors like hiring, revenue and foot traffic. Data sets may not be fully comprehensive (they only account for what is available on the web), but they can be used to gauge performance factors like staffing and sales. 

Further reading: