[ylem]

This is a challenging issue. I see a few reasons that companies want to increase demographic diversity. One would be that some companies are trying to address more demographically diverse markets and having more diverse teams may reveal more opportunities, or at least help to prevent blunders. For example, if you have people from many skin tones on your team, it's natural for then to say attempt to use a product on themselves and not fix it. If you don't have someone on the team (or at least in a test group) from those backgrounds, you may not think to test broadly--not that anyone was actively trying to be discriminatory--just that it's pretty natural to test things on yourself and within your network. Another reason is to have a stronger work force. A natural human thing to do is to recruit within networks--who did I go to school with? Work with? Where did I go to school? But, if I restrict myself to that pool, maybe I miss out on excellent candidates who went elsewhere. For example, I once found an amazing intern from a tiny college that I never heard of--but he was amazing (now staff at Carnegie Mellon). Finally, there are questions of social justice--organizations should reach out to populations that are underrepresented and ask if their recruiting/hiring processes have blind spots. On the recruiting side, are we reaching outside of our normal channels to recruit? For example, one of my coworkers took on an intern from a community college to do some physics/software project. The intern did well and transferred to the local state school and kept working--as a result, they got several job offers based on that work. On the hiring side, does the current methodology result in good hires? Are relevant characteristics being considered? For example, we sometimes bring on postdocs and as part of the review someone screens them based on their undergraduate grades. To me, this is dumb--no idea where it came from, I care about what they did during their PhD, not if they got a B in calculus as a teenager. It rates right up there with how fast they can run a 400 m race. Not terribly predictive of their ability to do research and probably results in the loss of some good candidates (for example, I met a young woman who went to a crap high school, spent college recovering from crap high school--she went to a small university for her masters and did incredibly well, and now she's at Harvard doing publishing good papers during her PhD--under the current system, she would probably not make it pass that screen). What happens once you have people in the institution is a longer discussion--basically, are people promoted fairly (based on performance)? Are these evenly applied?