3 ms·

Try Googling "diversity performance teams." "Why Diverse Teams Are Smarter" - HBR [1] "Diversity and Work Group Performance" - Stanford [2] "New Research: Di

Try Googling "diversity performance teams." "Why Diverse Teams Are Smarter" - HBR [1] "Diversity and Work Group Performance" - Stanford [2] "New Research: Diversity + Inclusion = Better Decision Making At Work" - Forbes [3] There's a drumbeat of this stuff, both old and new, that comes out constantly. [1] https://hbr.org/2016/11/why-diverse-teams-are-smarter https://hbr.org/2016/11/why-diverse-teams-are-smarter [2] https://www.gsb.stanford.edu/insights/diversity-work-group-performance https://www.gsb.stanford.edu/insights/diversity-work-group-p... [3] https://www.forbes.com/sites/eriklarson/2017/09/21/new-research-diversity-inclusion-better-decision-making-at-work/ https://www.forbes.com/sites/eriklarson/2017/09/21/new-resea...

jl2718 8y ago
These are interesting evidence, but they don’t explore the question of why. The idea that mere color of skin would be causal to performance in anything other than hide-and-seek is kind of anaethema, right? I’m certain that there is a legitimate reason for the effect, but people are right to be skepical without any exploration of the cause. Are diverse teams more likely to disagree? Do they exhibit complementary recall of facts? Different modes of creativity? I think a more complete explanation of causality would result in a much more concerted effort to increase diversity and a lot less resistance. As it is, it feels like the diversity ‘lobby’ is actually trying to create controversy instead of proving value.
- colemannugent 8y ago
  It's sociology, they can't tell you the reason why because they don't know. Instead they rely on people wanting it to be true, so they build a "common sense" argument that lacks evidence. When these people push for diversity they often mean "have every skin tone represented in your team photos". While this feels nice and would certainly repel any accusations of not being "diverse enough" it is the wrong kind of diversity. Now here is my "common sense" argument: What people actually want is a diversity of opinions. This you can imagine having a positive effect on performance as each member of a team would have a slightly different take on things that may allow the group to discover a better solution. This would explain why the diversity of skin tone type approach somewhat works, as different races usually have different experiences, which usually means more diversity of thought when compared to a completely homogeneous group. The obvious drawback to diversity of thought is that it's harder to quantify and optimize for. Anyone can look at a team photo and tell if a race is over/underrepresented because we have evolved incredibly advanced pattern recognition abilities. You can't easily tell what someone's opinions are without talking to them for a little while. Management already has trouble hiring qualified candidates since they don't know what qualified candidates look like, so adding diverse thinking to the list of criteria just exacerbates the hiring problem. Plus if you actually could select thought diverse candidates you run the risk of not including all races, opening yourself up to accusations of bias that can be devastating in today's politically charged climate. Social science can't accurately model single humans well, let alone groups of humans as they attempt to solve problems. Even my analysis that we really ought to be concerned with diversity of thought is based on intuition rather than experimentation. It's a hard problem domain, and the solutions may not be intuitive.
  - foolrush 8y ago
    This is pure, uneducated rubbish. The reason that social science can't model (aka overfit) many phenomena, is that it involves strictly nonlinear data where cultures and humans are involved. Do some reading. Educate yourself. You may change your vantage.
    - colemannugent 8y ago
      >The reason that social science can't model (aka overfit) many phenomena, is that it involves strictly nonlinear data where cultures and humans are involved. Right, they currently can't accurately model the phenomena and thus their models have little to no predictive power. We are in agreement then. Also, it seems to me that social sciences tend to produce models that are underfit rather than overfit. So why then do we accept conclusions from scientists who have weak models with little predictive power? This is what my comment above was talking about. >This is pure, uneducated rubbish. It's definitely not uneducated and I don't think it's entirely rubbish either.
      - foolrush 8y ago
        Uneducated as in it is painfully clear you have never read anything by any sociologists. Have you tried doing linear algebra on nonlinear phenomena? The examples of successful sociological science are too numerous to name, and your laziness isn't going to result in someone Googling the easily found work for you. Go do your homework, or sit down and shut up.
- sonnyblarney 8y ago
  "they don’t explore the question of why" More importantly, they don't get into causality. Most diversity studies show correlation. The problem is - as companies get very big and stable, they can start to worry about diversity, and have such hiring programs. Early on, nobody has time to worry about anything but just getting the best people they can at the time, possibly in a panic. So, looking for diversity is often a case of sampling the winners. I'm not so cynical about diversity, but I am about the voices promoting it, who are often heavily biased. Diversity is polarizing topic, and like with all such things ... it's hard to get good data points.
orwin 8y ago
BUT diversity as "diversity of background" is never tested against diversity as "diversity of races" on those group performance studies. I think it should, because i'm pretty sure that if your "culture" is the same as your coworkers (same neighborhood, same school, same type of family, faith etc...); skin differences, except if you are subject to negative stereotyping in the field you're working on, should amount to almost nothing.

jl2718 8y ago

These are interesting evidence, but they don’t explore the question of why. The idea that mere color of skin would be causal to performance in anything other than hide-and-seek is kind of anaethema, right? I’m certain that there is a legitimate reason for the effect, but people are right to be skepical without any exploration of the cause. Are diverse teams more likely to disagree? Do they exhibit complementary recall of facts? Different modes of creativity? I think a more complete explanation of causality would result in a much more concerted effort to increase diversity and a lot less resistance. As it is, it feels like the diversity ‘lobby’ is actually trying to create controversy instead of proving value.

colemannugent 8y ago
It's sociology, they can't tell you the reason why because they don't know. Instead they rely on people wanting it to be true, so they build a "common sense" argument that lacks evidence. When these people push for diversity they often mean "have every skin tone represented in your team photos". While this feels nice and would certainly repel any accusations of not being "diverse enough" it is the wrong kind of diversity. Now here is my "common sense" argument: What people actually want is a diversity of opinions. This you can imagine having a positive effect on performance as each member of a team would have a slightly different take on things that may allow the group to discover a better solution. This would explain why the diversity of skin tone type approach somewhat works, as different races usually have different experiences, which usually means more diversity of thought when compared to a completely homogeneous group. The obvious drawback to diversity of thought is that it's harder to quantify and optimize for. Anyone can look at a team photo and tell if a race is over/underrepresented because we have evolved incredibly advanced pattern recognition abilities. You can't easily tell what someone's opinions are without talking to them for a little while. Management already has trouble hiring qualified candidates since they don't know what qualified candidates look like, so adding diverse thinking to the list of criteria just exacerbates the hiring problem. Plus if you actually could select thought diverse candidates you run the risk of not including all races, opening yourself up to accusations of bias that can be devastating in today's politically charged climate. Social science can't accurately model single humans well, let alone groups of humans as they attempt to solve problems. Even my analysis that we really ought to be concerned with diversity of thought is based on intuition rather than experimentation. It's a hard problem domain, and the solutions may not be intuitive.
- foolrush 8y ago
  This is pure, uneducated rubbish. The reason that social science can't model (aka overfit) many phenomena, is that it involves strictly nonlinear data where cultures and humans are involved. Do some reading. Educate yourself. You may change your vantage.
  - colemannugent 8y ago
    >The reason that social science can't model (aka overfit) many phenomena, is that it involves strictly nonlinear data where cultures and humans are involved. Right, they currently can't accurately model the phenomena and thus their models have little to no predictive power. We are in agreement then. Also, it seems to me that social sciences tend to produce models that are underfit rather than overfit. So why then do we accept conclusions from scientists who have weak models with little predictive power? This is what my comment above was talking about. >This is pure, uneducated rubbish. It's definitely not uneducated and I don't think it's entirely rubbish either.
    - foolrush 8y ago
      Uneducated as in it is painfully clear you have never read anything by any sociologists. Have you tried doing linear algebra on nonlinear phenomena? The examples of successful sociological science are too numerous to name, and your laziness isn't going to result in someone Googling the easily found work for you. Go do your homework, or sit down and shut up.
sonnyblarney 8y ago
"they don’t explore the question of why" More importantly, they don't get into causality. Most diversity studies show correlation. The problem is - as companies get very big and stable, they can start to worry about diversity, and have such hiring programs. Early on, nobody has time to worry about anything but just getting the best people they can at the time, possibly in a panic. So, looking for diversity is often a case of sampling the winners. I'm not so cynical about diversity, but I am about the voices promoting it, who are often heavily biased. Diversity is polarizing topic, and like with all such things ... it's hard to get good data points.

orwin 8y ago

BUT diversity as "diversity of background" is never tested against diversity as "diversity of races" on those group performance studies. I think it should, because i'm pretty sure that if your "culture" is the same as your coworkers (same neighborhood, same school, same type of family, faith etc...); skin differences, except if you are subject to negative stereotyping in the field you're working on, should amount to almost nothing.