Report Accuses Facebook of Gender Bias Against Female Developers

A newly published report has accused Facebook of rejecting female developer’s code 35 percent more often than code written by male developers.

The Wall Street Journal reports that an engineer at Facebook gathered data last year showing that code written by women at the company was more frequently rejected than code written by male developers. The report apparently set off a debate within Facebook internally about possible gender bias within the Silicon Valley tech company. CEO Mark Zuckerberg was even asked about the report during a weekly town hall meeting.

Senior Facebook officials conducted their own review of the Facebook engineer’s report following the outcry. Facebook’s Head of Infrastructure, Jay Parikh, published an internal post a month after the review which attributed the rejection rate gap to the engineer’s rank and level of work rather than their gender. This was seen by many employees as evidence that female developers were not rising through the ranks of Google at the same rate as their male counterparts.

A Facebook spokesperson told the Wall Street Journal that the initial analysis by the Facebook engineer was, “incomplete and inaccurate — performed by a former Facebook engineer with an incomplete data set.” However, the spokesperson did confirm Mr. Parikh’s analysis, which was based on data unavailable to the majority of employees. The spokesperson also stated that there aren’t enough women at senior engineering levels at Facebook and across many Silicon Valley tech companies.

Tracy Chou, a diversity advocate and former software engineer at Pinterest said, “especially for mass consumer products, there are just going to be blind spots.” The engineer that conducted the report who has reportedly been at the company for several years stated that she conducted the analysis, “so that we can have an insight into how the review process impacts people in various groups.” The engineer stated that she analyzed five years of data that she pulled from Facebook’s own open repository of code-review data.