It's really a weird, strange kind of twilight zone. Alleged experts are just spouting totally unfounded (and even contradicted by evidence) assertions and stating them as fact. Some of them really hurtful and hateful even. And few dare to call them out, demand evidence for their unproven claims, or demand metrics and results of improvements they claim their policies would result in. It's like a nutty cult that's being ruled by fear.
In my opinion it is also profoundly damaging to the cause of actually improving inequality and reducing discrimination and reducing division.
> Intersectionality is a theoretical framework rooted in the premise that human experience is jointly shaped by multiple social positions (e.g. race, gender), and cannot be adequately understood by considering social positions independently.
Is it deeper than that?
What is relevant is that the general discourse focuses more on systemic differences in opportunity to disadvantaged groups. In this case, we're talking about women as the ostensibly disadvantaged group.
I think what people fail to understand though, is that marginalization can take many forms. If women are less likely to continue interviewing after rejection, this doesn't mean that they have the same opportunity; instead, it suggests that systemic factors play into this. The "intersectional" take on this is really focused on how different components of identity work together to shape an individual's experience, in subtle ways that compound. Hence, just growing up in a society where young women and men have even mildly different attitudes impressed upon them from a young age, can have more noticeable effects on their outcomes much later in life
No it doesn't. This is a false dichotomy. It can be any number of things, including something systemic, but only one explanation of many is eliminated. All that it suggests is that interviewers don't discriminate (against women).
That's the reason today there's a lot of focus on black trans women because activists are trying not to repeat this mistake.
> any disparity between any demographic groups is wholly attributable to bias
Unless you're literally arguing that some groups are genetically superior or inferior this statement is true by definition -- that's what bias means.
If you're a member of two different minority groups, your experience is not described by taking a union of the bad things from each. Eg. If being black results in 3% less <> and being a woman results in 5% less <>, then being a black woman does not result in 8% less <>.
This is not people interested in truth, this is people who want to control how you think.
No different to the racists of yesteryear, especially since the current wokesters are the grandchildren of klan members.
uh, what?
The common thread is the belief in that whites are naturally superior to everyone. Which is why they don't support income based affirmative action and fetishize the idea of race.
Put it another way: anyone who thinks Obama is an African American is part of the problem.
He's had none of the experiences an African American (defined as the black skinned descendants of US slaves) has had and there was no one in either of his families who would have.
Calling that African American is a bit like calling a chihuahua a wolf.
Those experiences of the descendants of slaves you mentioned are the reasons why race is so important when it comes to building wealth and prosperity in your community.
Just reel yourself back in right there... Even after achieving high levels of political status he was mocked and harrassed by millions... solely based on his skin color. anyone can look them up.
https://thehill.com/policy/energy-environment/410625-new-epa...
https://www.alternet.org/2009/09/10_horrifying_racist_attack...
https://www.cbsnews.com/news/west-york-pennsylvania-republic...
https://www.ebony.com/news/pro-trump-officer-racist-michelle...
https://www.independent.co.uk/news/world/americas/mayor-char...
https://www.salon.com/2014/10/01/7_disgustingly_racist_anti_...
https://hiphopwired.com/870840/tennessee-exec-obama-lynching...
Or the racists of currentyear.
>the current wokesters are the grandchildren of klan members
No they're not. I'm definitely not.
This is because their principal theory of demographics — that all group differences are entirely due to one group's use of power to undermine the other — is wrong. There are many factors involved in group differences, and they are all being completley ignored because of political correctness. Even to say that sexism sometimes only indirectly causes disparity in enmplpyment - as an example, teachers/mentors investing more time in boys - is becoming problematic because it posits that interviewers themselves may possibly not be overtly or implicitly sexist. The idea that not only the "system," but each member in the powerful group is out to get all minorities/women, is ridiculous and, as you say, hateful.
It is, and you see it in this person's comment. It was actually surprising to them that they did not observe discrimination. They have been indoctrinated to believe without evidence that the society they live in is a sea of hateful and/or ignorant bigots who oppress and discriminate. The oppressed classes are indoctrinated to believe they are under constant attack and oppression by their neighbors and friends and colleagues. And the oppressor classes believe they hold some responsibility and shame for these real or imagined atrocities. It's just a hell of a way to go through life, and I can't see anything much good coming from it.
I'm not saying for a minute that discrimination does not exist, that it should not be addressed and reduced, or that it can not concentrate in organizations and positions of power, before people start accusing and denouncing me.
The example frequently given is the proverbial workplace that in order to hire enough black people and women, hires many black men on the factory floor and white women in the offices while black women don’t get hired and despite this people declare that equity has been accomplished. People paid attention to women’s issues and black issues, but not the INTERSECTION of those issues.
My criticism is that intersectionality is that it’s blindingly obvious and there is limited value in literally deviding up identity groups into an exponentially expanding amount of subgroups. Once you start exploring the intersection of 3 identities you’ve already kind of lost the plot and may as well start talking about issues on the personal/individual level.
I have no idea what people THINK intersectionality means. I think people who know what that word means keep using it without explaining it and people get really confused and think it’s just a general term for woke ideology and never stop to think what the “intersection” they’re referring to is.
How do trans immigrant women feel about the gender wage gap? How do male Irish steelworkers feel about it? So on and so forth… and don’t you dare tell anybody their identity group and their unique intersection of those identities isn’t important.
It’s an ideology I would try to convince my enemies to adopt to encumber them with endless amounts of petty political bickering so they never got anything done.
This whole intersectionality business is spreading way too fast, being too prominent in the media and front and center everywhere to not be at least encouraged by an hostile nation.
At the limit, intersectionality approaches individualism.
When you know the agendas and world views of those same people, everything starts to make sense. It's quite unfortunate to say the least, but hopefully more people wake up.
https://www.wbur.org/hereandnow/2021/08/18/name-discriminati...
The claims basically amount to any inequity between groups is due to discrimination, except in the cases where the inequity "favors" the oppressed class (and where oppressed and oppressor classes have no real objective criteria or definitions, aside from what intersectionality experts claim them to be).
And that institutional policies of discrimination are an the acceptable and effective way to deal with this inequity, but also you must not refer to it as systemic or institutional discrimination.
I bet the normal variation in responses for the same group submitted to separate groups of companies exceeds 3%. That is, I could send 1,000 resumes to one batch of companies and 10.0% would get interviews. I could send the same 1,000 resumes to a different batch, and 10.3% would get interviews. Boom 3% difference for the same candidates.
A result does not have "power". An experiment has power -- the ability to detect a a given effect size a certain percentage of the time -- but a result is either statistically significant, or it is not.
As for "noise", statistical significance takes random noise into account. That is the point of the calculation -- it asks if a given result exceeds the threshold of what you'd expect to find at random some percentage of the time. If it does, the result is deemed significant.
A 3% difference could be enormous, or it could be miniscule. We can't say anything based on this information alone, and certainly can't say it's "likely a tiny effect". On a sample of thousands, a 3% difference is big. On a sample of tens, a 3% difference is small.
Not really. Only if it is many, many thousands. Assuming a totally random acceptance rate of 1/5:
a = 0;
b = 0;
for (c of Array(1000)) {
if (Math.random() > .8)
a++;
if (Math.random() > .8)
b++;
}
console.log(`a=${a}, b=${b}, a is ${(a/b - 1)*100}% more likely than b`)
> a=209, b=201, a is 3.9800995024875663% more likely than b
literally the first run. And even in absolute terms, I got this on the third run: >a=192, b=219, a is -12.328767123287676% more likely than b
That's an absolute difference of 2.7%. Again, 100% random data.I think I get what you're going for here -- you're trying to simulate a coin flip? -- but what you've actually done is made successive draws from a uniform random number generator. The software is designed to return numbers that fall along the interval [0,1) with equal probability. Thresholding the numbers and dividing their counts is not a meaningful transformation; the result is still just a uniformly distributed random number. It's like...the ratio of heads in two identical, unfair coins or something.
If all "random numbers" were uniform like this, then no, we wouldn't expect an X% difference to be any more or less likely based on the magnitude of the underlying sample. But when we're talking about something like a a population mean, then the behavior of the errors on estimates is very different indeed, and most estimates cluster around the true (aka population) value:
https://online.stat.psu.edu/stat415/lesson/9/9.4
As the sample size for an experiment of this sort gets larger, the bell curve of expected errors gets sharper and sharper, and it becomes increasingly less likely to see errors >= X, for any value X. In the limit of large N, the distribution of sample errors around a known mean approach a normal distribution:
https://www.jmp.com/en_us/statistics-knowledge-portal/t-test...
For what it's worth, the expected proportion of N heads in M coin flips is modeled using the binomial distribution, which is also bell-shaped and illustrates the same idea:
This model is built on the assumptions that if candidates are actually totally equally likely to be picked (the null hypothesis for the experiment above), any given candidate has a p=.2 chance of being hired (given an arbitrary but reasonable hire vs interview ratio of 1:5). Which is just a weighted coin flip. This is indeed a binomial distribution, and my point is that results ±3% of the mean (p*M), even at M=1000, are still fairly probable. When comparing two such results, it's almost expected.
The part where you did this in a loop, with two calls per iteration, and then divided the counts and called it a percentage is wrong. It's certainly not a Binomial distribution. It's just the ratio of two binomial random variables.
This is wrong. That is a very meaningful transformation. It is the standard way (https://stats.stackexchange.com/questions/240338) to turn a uniform distribution into a Bernoulli distribution.
Getting a single value with Bernoulli distribution is called a Bernoulli trial (https://en.wikipedia.org/wiki/Bernoulli_trial). Repeating this gives you a Binomial distribution (see your own wikipedia link).
Long story short: GPs code is a perfectly valid way of sampling the Bernoulli distribution. It is inefficient because it needs so many random values, but it mimics the actual process happening in real life making it easier to understand than generating a Binomial sample from the Binomial distribution's CDF.
The OP didn't do what was described in the SO post. They did something else.
So this Wikipedia article is a fever dream? https://en.wikipedia.org/wiki/Bernoulli_distribution
> they calculated the ratio of two binomial random variables, and presented that as a percentage.
Ok, now I'm confused. I 100% agree with that statement. I thought your whole point was that OPs code was not a valid way to sample from a binomial distribution?
But then what is your criticism? Are you arguing that the Binomial distribution does not model the original experiment correctly?
> I thought your whole point was that OPs code was not a valid way to sample from a binomial distribution?
The code OP posted was just taking the ratio of two binomial random variables. It's not "sampling from a binomial", except (perhaps) in the sense that each of those random variables was the result of independent coin flips.
We really need to be more precise in our terminology here. "Sampling from a distribution" can mean a lot of things. Based on the sibling comments, it seems like they were trying (?) to sample from the binomial CDF.
Setting this aside, my high-level point was OPs calculation doesn't have anything to do with error distributions.
I know it to mean only one thing: Generating a value in such a way, so that if that process were to be repeated, the generated values follow the given distribution. How exactly this is done is irrelevant, as long as the distribution is correct. See also https://en.wikipedia.org/wiki/Pseudo-random_number_sampling.
> [...] it seems like they were trying (?) to sample from the binomial CDF.
This is technically unclear usage of terminology. You can not actually sample from a CDF. But it is clear that you are referring to Inverse transform sampling (https://en.wikipedia.org/wiki/Inverse_transform_sampling) – where you sample from a Uniform distribution and use that sample to generate a sample form a non-uniform distribution using that distribution's CDF.
> The code OP posted was just taking the ratio of two binomial random variables. It's not "sampling from a binomial", except (perhaps) in the sense that each of those random variables was the result of independent coin flips.
Once again: Since the Binomial distribution is the distribution of a series of independent coin flips, doing a series of independent coin flips is a perfectly valid way of sampling the binomial distribution.
> Based on the sibling comments, it seems like they were trying (?) to sample from the binomial CDF.
As they explain in the sibling comment, they generate two samples from the binomial distribution and compare them to each other the same way the original authors did. What they achieve by this is sampling from the same random variable that the original authors were implicitly sampling from. They then took multiple samples from that variable in order to get a feel for its distribution, to confirm their original point: That 3% is not an uncommonly big value under that distribution.
> Setting this aside, my high-level point was OPs calculation doesn't have anything to do with error distributions.
So I don't quite now what you mean by "error distribution". I assume you mean the distribution of the random variable that had the value of 3% in the article? If so, then OPs calculation does – as explained – have a lot to do with that distribution. It does not calculate that distribution, but it samples from it, which is a useful way to get a feel for a distribution without having to do any fancy mathematics or research.
If you have an expectation that someone won't interview well and they surprise you, you'll be more likely to note their positive characteristics and move them forward.
Likewise, if you expect someone to interview well and they disappoint you, you'll be more likely to note their negative characteristics and won't move them forward.
This will be true even if the interviewees give nearly identical interviews.
Does that explain this? If a woman finds herself being compared to an interviewer's expectations for men instead of the interviewer's expectations for women, she does worse, and vice versa.
Another class of extreme examples would be public political figures. There is only one person who has successfully implemented a city-wide smoking ban. There is only one person who was a minister of finance, and then worked on in-game economics.
"Blind recruitment means recruiters cannot tell the gender of candidates because those details are removed from applications."
"The trial found assigning a male name to a candidate made them 3.2 per cent less likely to get a job interview.
Adding a woman's name to a CV made the candidate 2.9 per cent more likely to get a foot in the door."
https://www.abc.net.au/news/2017-06-30/bilnd-recruitment-tri...