3 ms·

A lot of "science research" are biased because of the researcher's own "unconscious bias". When we talk about scientific method, it involves coming up with a h

A lot of "science research" are biased because of the researcher's own "unconscious bias". When we talk about scientific method, it involves coming up with a hypothesis, and devising a controlled experiment that validates or invalidates the hypothesis. This is a hard problem even for objectively observable science such as physics, and I think it's extremely hard for social science. It's really hard to be NOT biased if you're a part of the hypothesis you're trying to prove. I think there should be a fundamentally different approach to doing these social science experiments. When I see these people parrot some "science research paper" to push their agenda, I shake my head, especially when the real world statistics is opposite of the research. For example, I see people say "research says diversity brings creativity, and if you're working on an early stage startup, you should make your team as diverse as possible, because the research says so." I won't go into my own opinion on this because I realize opinions are subjective. However I would like to ask, if you look at actual real world stats, how many extremely successful tech companies started out with a "diverse" team? Turns out most tech startups started out with like-minded people working tightly knitted together. (Again, I'm not making value judgment, this IS the reality). So as someone who's starting to work on a startup company, should you try so hard to create a diverse team because some lab research with controlled environment of 20 people sample size said so? Or would you rather follow the hard fact? I honestly think these diversity consultants are harmful for the society, NOT because of what they are claiming to try to do (I DO believe diversity is important in the long term, I'm just talking about how weak their "research" evidence is compared to what really happens in the real world), but because of what they are actually doing, without even realizing their own "unconscious bias".

Geekette 9y ago
" I think it's extremely hard for social science. It's really hard to be NOT biased if you're a part of the hypothesis you're trying to prove." That's why mechanisms such as control groups and blind analyses exist. It also bears saying that researchers consist of all types, not necessarily part of the demographic affected and some study results (including those on diversity) were invalidations of their hypotheses, with some studies having been repeated elsewhere with similar results. "However I would like to ask, if you look at actual real world stats..." The studies I've seen showing positive effects of diversity have also been based on actual statistics and many assessed several sectors, not just tech. That many startups in a specific region are similar based on demographics doesn't invalidate benefits of diversity.
- cocktailpeanuts 9y ago
  > The studies I've seen showing positive effects of diversity have also been based on actual statistics and many assessed several sectors, not just tech. That many startups in a specific region are similar based on demographics doesn't invalidate benefits of diversity. I was specifically talking about the startup context. I think it's simply wrong for these diversity consultants to abuse their PC power to spread false information. And going further, my point was that these people use controlled research results to make their point when the real world demonstrates plenty of evidence showing that it doesn't work that way. Think of any successful company (not just tech but historically) and see how many of them won over their competition in their early days because they worried about creating a diverse team when they were 5 people bootstrapping. You won't see many. Most businesses succeed by building or providing products/services people want. That's the most important thing as an entrepreneur. Nowadays I see entrepreneurs and VCs bragging on social media how their firm has "achieved diversity" because they have 50% women and 50% black, etc. That's the last thing you should be worried about during your early days in startup journey.

Geekette 9y ago

" I think it's extremely hard for social science. It's really hard to be NOT biased if you're a part of the hypothesis you're trying to prove." That's why mechanisms such as control groups and blind analyses exist. It also bears saying that researchers consist of all types, not necessarily part of the demographic affected and some study results (including those on diversity) were invalidations of their hypotheses, with some studies having been repeated elsewhere with similar results. "However I would like to ask, if you look at actual real world stats..." The studies I've seen showing positive effects of diversity have also been based on actual statistics and many assessed several sectors, not just tech. That many startups in a specific region are similar based on demographics doesn't invalidate benefits of diversity.

cocktailpeanuts 9y ago
> The studies I've seen showing positive effects of diversity have also been based on actual statistics and many assessed several sectors, not just tech. That many startups in a specific region are similar based on demographics doesn't invalidate benefits of diversity. I was specifically talking about the startup context. I think it's simply wrong for these diversity consultants to abuse their PC power to spread false information. And going further, my point was that these people use controlled research results to make their point when the real world demonstrates plenty of evidence showing that it doesn't work that way. Think of any successful company (not just tech but historically) and see how many of them won over their competition in their early days because they worried about creating a diverse team when they were 5 people bootstrapping. You won't see many. Most businesses succeed by building or providing products/services people want. That's the most important thing as an entrepreneur. Nowadays I see entrepreneurs and VCs bragging on social media how their firm has "achieved diversity" because they have 50% women and 50% black, etc. That's the last thing you should be worried about during your early days in startup journey.