4 ms·

> In particular, as part of Google's mandatory diversity training, you're presented with Google's internal research showing that diversity hiring yields both be

by bostik 9y ago

> In particular, as part of Google's mandatory diversity training, you're presented with Google's internal research showing that diversity hiring yields both better individual and group performance. Are these numbers available to public somewhere? I'm asking because I would dearly want to find comprehensive data. When I've done some of my own research, I have run into sets of studies, which all agree on one thing: more diverse groups get better results. The snag is that these studies disagree - wildly - with the magnitude of the effect. Some claim maybe 2% improvement. Some others claim up to ~30%. Yet others somewhere inbetween. The figures also fluctuate quite a bit over time. Study from 2008/2009 shows numbers that are almost entirely unrelated to a study from 2014/2015. Some studies claim that gender diversity alone bears very little benefit[ß], and other studies from the same time come out with results that sort of imply you will need cultural diversity for notable results. But by far the worst piece of news is that the measured and/or recorded numbers appear to be geographically distinct. Results from US studies are different from those in UK, which in turn differ from mainland Europe studies. That only raises more questions. Such as... - are the numbers consistently different across regions? - ... across continents? - ... across cultures? - what is the measured effect between "just" gender diverse and culturally diverse groups? - do age groups matter? ... and so on. I genuinely would like to know. (For the record, and you can check this from my earlier posts as well: I am firmly in the camp who believe that diversity is beneficial.) N.B. The most recent study I know of is the McKinsey report from 2015.[0] --- ß: I can dig out the research articles if there's interest. I have some of them sort of easily accessible. 0: http://www.mckinsey.com/business-functions/organization/our-insights/why-diversity-matters http://www.mckinsey.com/business-functions/organization/our-...

gurkendoktor 9y ago
Are you aware of a study that is focused on (software) engineers? The McKinsey study looks at companies' top management for its diversity metrics, but I don't really trust studies across professions. Most anecdotes I've read about the benefits of diversity boil down to "if only we'd had a <minority> on the team, we would have noticed <problem affecting minority>". But as a developer, my requirements are set by project managers (50% women in my experience), the design and marketing is done by a diverse team, testing is done by other people (or end-users, sigh); so developers are "only" concerned with the actual implementation. Consequently, I don't find it intuitive that more diversity (of gender/ethnicity) would help my team. Do developers in other companies or countries (Germany here) have wider responsibilities, and thus, diversity is more important? Can engineers at Facebook or Google write user-facing features on a whim? (The terrible mic drop Gmail gimmick comes to mind...) Could that explain the difference in how much teams benefit from diversity?
- bostik 9y ago
  > Are you aware of a study that is focused on (software) engineers? Sadly no :( - I'm fully aware that the studies in this field mostly focus on financial outcomes, and because they are commissioned by consulting/recruiting agencies, they are of course targeting the management group. That's one of the reasons I would love to see the numbers from Google's internal research. As for this: > Most anecdotes I've read about the benefits of diversity boil down to "if only we'd had a <minority> on the team [...] I am not sure whether I should be offended or merely aghast. From the tone you used, it feels like you have witnessed something more akin box-ticking than strategy, cargo-culting over rationality. I know my reasons are different but I suspect they could be equally selfish. 1. In my experience, women who stay in tech tend to be pretty damn good. Even if we ignored all other factors, it would be rational from purely talent-acquisition point of view to target female engineers in hiring. In addition, women often have greater capacity for empathy and thus can become a force multiplier in practically any team. 2. Constructive arguments from different viewpoints can only make us stronger. Things I am blind to can be spotted by someone else. Things my usual peer group is blind to can be spotted by persons from different peer group(s). But most importantly - things that are obvious to people who share my background can be pointed out and questioned by people whose backgrounds are different. The thing I, as an engineer, worry about is myopia. > as a developer, my requirements are set by project managers (50% women in my experience) That is an interesting observation. Do you think that is due to self-selection or external factors? Also: if PM roles can have a 50/50 gender split, does that mean that women are more likely to seek promotions that lead to PM type positions? > Do developers in other companies or countries (Germany here) have wider responsibilities, and thus, diversity is more important? I think it depends less on the country and more on the company [or national?] culture. In engineering driven organisations, definitely. In a rigid hierarchy with nothing-but-top-down approach, probably less so. But then again, in the light of what I know, the latter category should benefit from diversity in their upper management. Where else would the alternate viewpoints come from? And finally, I'm going to pick this one up separately: > Consequently, I don't find it intuitive that more diversity (of gender/ethnicity) would help my team. I'm sorry for you. But also, I am glad that you brought this up. You have just highlighted that MY experience and view of diversity can be irrational or plain odd to someone from a different cultural environment. For that - thank you.

by bostik 9y ago

gurkendoktor 9y ago

Are you aware of a study that is focused on (software) engineers? The McKinsey study looks at companies' top management for its diversity metrics, but I don't really trust studies across professions. Most anecdotes I've read about the benefits of diversity boil down to "if only we'd had a <minority> on the team, we would have noticed <problem affecting minority>". But as a developer, my requirements are set by project managers (50% women in my experience), the design and marketing is done by a diverse team, testing is done by other people (or end-users, sigh); so developers are "only" concerned with the actual implementation. Consequently, I don't find it intuitive that more diversity (of gender/ethnicity) would help my team. Do developers in other companies or countries (Germany here) have wider responsibilities, and thus, diversity is more important? Can engineers at Facebook or Google write user-facing features on a whim? (The terrible mic drop Gmail gimmick comes to mind...) Could that explain the difference in how much teams benefit from diversity?

bostik 9y ago
> Are you aware of a study that is focused on (software) engineers? Sadly no :( - I'm fully aware that the studies in this field mostly focus on financial outcomes, and because they are commissioned by consulting/recruiting agencies, they are of course targeting the management group. That's one of the reasons I would love to see the numbers from Google's internal research. As for this: > Most anecdotes I've read about the benefits of diversity boil down to "if only we'd had a <minority> on the team [...] I am not sure whether I should be offended or merely aghast. From the tone you used, it feels like you have witnessed something more akin box-ticking than strategy, cargo-culting over rationality. I know my reasons are different but I suspect they could be equally selfish. 1. In my experience, women who stay in tech tend to be pretty damn good. Even if we ignored all other factors, it would be rational from purely talent-acquisition point of view to target female engineers in hiring. In addition, women often have greater capacity for empathy and thus can become a force multiplier in practically any team. 2. Constructive arguments from different viewpoints can only make us stronger. Things I am blind to can be spotted by someone else. Things my usual peer group is blind to can be spotted by persons from different peer group(s). But most importantly - things that are obvious to people who share my background can be pointed out and questioned by people whose backgrounds are different. The thing I, as an engineer, worry about is myopia. > as a developer, my requirements are set by project managers (50% women in my experience) That is an interesting observation. Do you think that is due to self-selection or external factors? Also: if PM roles can have a 50/50 gender split, does that mean that women are more likely to seek promotions that lead to PM type positions? > Do developers in other companies or countries (Germany here) have wider responsibilities, and thus, diversity is more important? I think it depends less on the country and more on the company [or national?] culture. In engineering driven organisations, definitely. In a rigid hierarchy with nothing-but-top-down approach, probably less so. But then again, in the light of what I know, the latter category should benefit from diversity in their upper management. Where else would the alternate viewpoints come from? And finally, I'm going to pick this one up separately: > Consequently, I don't find it intuitive that more diversity (of gender/ethnicity) would help my team. I'm sorry for you. But also, I am glad that you brought this up. You have just highlighted that MY experience and view of diversity can be irrational or plain odd to someone from a different cultural environment. For that - thank you.