Minority Groups "Special Rights"

[LSSBB]

The problem is the under representation of minorities in your hypothetical company.

As far as I know, that makes it a problem in that diverse companies outperform companies that aren't diverse. Once you achieve optimal diversity for your business (whatever that is), then it would no longer be a problem.

 

[Belz...]

The problem is the under representation of minorities in your hypothetical company.

Again, you haven't established that there's a problem in the first place.

 

[Archie Gemmill Goal]

Again, you haven't established that there's a problem in the first place.

Minorities are under represented. Thats a problem. We live in a society where we areare trying to eliminate under representation.

What you are arguing is that this problem might have an explanation. Im fine with that. After you have done the work to demonstrate it. Not before.

 

[Roboramma]

Minorities are under represented. Thats a problem.

No it isn't.

 

[Belz...]

Minorities are under represented. Thats a problem.

I'm asking you why. See below.

We live in a society where we areare trying to eliminate under representation.

No, we're trying to eliminate discrimination.

What you are arguing is that this problem might have an explanation. Im fine with that. After you have done the work to demonstrate it. Not before.

I think yours is the positive claim, Archie.

 

[Archie Gemmill Goal]

No it isn't.

[pantoseason] Oh yes it is [/pantoseason]

 

[Roboramma]

[pantoseason] Oh yes it is [/pantoseason]

Okay, I'll expand, though I really can't see why it's necessary: There's no inherent problem with underrepresentation. There's a problem with discrimination, which is one possible explanation for underrepresentation, but underrepresentation itself is neither good nor bad, it's just a fact.

 

[Archie Gemmill Goal]

I'm asking you why. See below.

Because whenever we see under-representation it is a sign of...

No, we're trying to eliminate discrimination.

Bingo.

I think yours is the positive claim, Archie.

Your way just wouldn't work practically. Anyone could simply excuse their workplace imbalance by saying 'well that's just the way it is' and unless you had the time and resources to investigate every single business and their hiring processes in depth then all but the most egregious examples of discrimination are going to be fine.

Let's take another example. Minority turnout at elections is in many cases lower than the average. Your approach would be to say 'that's not a problem. minorities are just less inclined to vote.' My approach says 'something is wrong there. What can be done to get more minorities to vote?'

If you want to give an easy out to discriminatory employers then fine but I set the bar higher than that personally. You don't create change by justifying the status quo. There needs to be pressure to force it.

 

[marplots]

We live in a society where we areare trying to eliminate under representation.

No, we're trying to eliminate discrimination.

Yes, but under-representation is our proxy measure for discrimination - because we can't see into anyone's head to tell what their motivations are. We are stuck looking at behavior instead.

I can't tell how someone feels about black people, but I can count how many homes they've sold to black families.

 

[Roboramma]

Yes, but under-representation is our proxy measure for discrimination - because we can't see into anyone's head to tell what their motivations are. We are stuck looking at behavior instead.

I can't tell how someone feels about black people, but I can count how many homes they've sold to black families.

It's not a good proxy when the data is so messy. In your example, is there actually discrimination or are black people just less likely to be able to afford the homes that this realtor is selling? Are black families just less likely to want to own homes?

Archie makes the argument that this is special pleading: yes there will be cases where the underrepresentation is not caused by discrimination, but generally discrimination will be the, or at least a cause. I just don't think the world is that simple, and don't think that from underrepresentation we can assume discrimination.

I think that it's entirely possible to make a good case based on statistics, but not naively.

 

[Belz...]

Because whenever we see under-representation it is a sign of... [discrimination]

No, we're talking about people not applying for jobs in field X. That's not discrimination. If you go to discrimination as a first explanation for every discrepancy or difference, no wonder you think it's all problematic.

If you always think underrepresentation implies discrimination, you're actually doing a disservice to women and minorities. And majorities, too.

Let's take another example. Minority turnout at elections is in many cases lower than the average. Your approach would be to say 'that's not a problem. minorities are just less inclined to vote.' My approach says 'something is wrong there. What can be done to get more minorities to vote?'

The analogy doesn't work because A) this is public sector stuff and B) it's still not about discrimination.

If you want to give an easy out to discriminatory employers then fine but I set the bar higher than that personally.

Nice appeal to shame. That kind of stuff doesn't work on me, principally because it's based on a strawman.

 

[marplots]

It's not a good proxy when the data is so messy. In your example, is there actually discrimination or are black people just less likely to be able to afford the homes that this realtor is selling? Are black families just less likely to want to own homes?

Archie makes the argument that this is special pleading: yes there will be cases where the underrepresentation is not caused by discrimination, but generally discrimination will be the, or at least a cause. I just don't think the world is that simple, and don't think that from underrepresentation we can assume discrimination.

I think that it's entirely possible to make a good case based on statistics, but not naively.

I agree. I think a good parallel is in the police touchstone of "specific and articulable facts" - the principle that, before the cops stop and roust you, they have to have a good reason, a reason they can state clearly, with facts they can point to. I want the same clarity in policy-making. Statistics fill this role, as would testimony from a victim of discrimination.

So, I'm going to look at the data and see what is worth looking into further and what isn't. The numbers do constitute evidence, but shouldn't be the entire corpus of my case.

At issue then is whether the burden of proof now shifts to the possible offender. Whether they have to respond to my concerns. One response would be to show other reasons for the discrepancy, unrelated to discrimination based on a protected category.

 

[Roboramma]

When I say that I think it's possible to make a good case based on statistics, that doesn't mean that you can make that case with only the information that a particular group is underrepresented.

For instance, take your realtor. Let's say that we go further than showing that he sells fewer homes to black families than their representation in the population at large: we also show that he is selling fewer homes to black families than are other realtors working in the same market, and that this difference in statistically significant. In that case I think you would have a good case to, as you say, shift the burden of proof*. In that case it makes sense to say to him "how is it that you are selling so many fewer homes to black families than your competitors are?" That doesn't mean no response is possible, but we have eliminated a lot of the noise in the data.

My claim, though, is that the data is in general very noisy, and I don't think it's a good idea to just assume we're seeing a signal because we see some particular trend away from some naively expected number.

As I said earlier I simply don't see any reason to expect any particular representation. As such not finding that particular representation doesn't tell me anything.

*I also think that in general this is how the law works and it seems pretty valid to me.

 

[marplots]

(some snipped)
My claim, though, is that the data is in general very noisy, and I don't think it's a good idea to just assume we're seeing a signal because we see some particular trend away from some naively expected number.

As I said earlier I simply don't see any reason to expect any particular representation. As such not finding that particular representation doesn't tell me anything.

I agree this might be an issue, but it's an issue well known and understood by statisticians. I would expect them to handle the data appropriately, both to measure inherent noise and to set some reasonable trigger level for a closer look.

My assumption though, is that mere statistics wouldn't be the most common path to investigation or enforcement measures. I'd expect that individual complaints/allegations of discrimination would rise to the top. But that's just a guess on my part.

 

[Archie Gemmill Goal]

It's not a good proxy when the data is so messy. In your example, is there actually discrimination or are black people just less likely to be able to afford the homes that this realtor is selling? Are black families just less likely to want to own homes?

Archie makes the argument that this is special pleading: yes there will be cases where the underrepresentation is not caused by discrimination, but generally discrimination will be the, or at least a cause. I just don't think the world is that simple, and don't think that from underrepresentation we can assume discrimination.

I think that it's entirely possible to make a good case based on statistics, but not naively.

But my case is that when we have seen things like estate agents having a suspiciously low number of black clients and then force them to do something about it we generally see that the number of black clients increases. So its not that black people don't want to buy houses.

Now maybe the next time the problem will be underrepresentation of gay couples and maybe it won't work and maybe there will be reasons for it. But it probably won't be based on past experience and if it turns out that it is the case then that's fine too. Just do the work first.

 

[Belz...]

But my case is that when we have seen things like estate agents having a suspiciously low number of black clients and then force them to do something about it we generally see that the number of black clients increases. So its not that black people don't want to buy houses.

I don't want to play devil's advocate here but under that scenario would it be possible that they allow more black people to buy houses who normally couldn't afford it just to meet the government requirements?

 

[Archie Gemmill Goal]

No, we're talking about people not applying for jobs in field X. That's not discrimination. If you go to discrimination as a first explanation for every discrepancy or difference, no wonder you think it's all problematic.

It's discrimination to me if you know that your company has an under-representation of minorities and do nothing about trying to improve that but just assume that those groups don't want to work for you because of their own reasons. If you have a better word for it then fine but let's not get bogged down in the semantics of what we call it.

If you always think underrepresentation implies discrimination, you're actually doing a disservice to women and minorities. And majorities, too.

How so?

The analogy doesn't work because A) this is public sector stuff and B) it's still not about discrimination.

I don't see how the sector matters. If you want to call it something else then it's still a problem. And I think the analogy deserves a bit more depth of response than this. Minorities are under-represented in the ballot. Senator Cletus P Racehater says 'that's because them ******** are too damn lazy to vote'

Should we just accept that there is no problem there?

Nice appeal to shame. That kind of stuff doesn't work on me, principally because it's based on a strawman.

No its based on the practical consequence of your position. If 'it might not be discrimination' is an excuse then anything except documented clear and egregious discrimination goes.

 

[Belz...]

It's discrimination to me if you know that your company has an under-representation of minorities and do nothing about trying to improve that but just assume that those groups don't want to work for you because of their own reasons. If you have a better word for it then fine but let's not get bogged down in the semantics of what we call it.

Archie, discrimination has to be a deliberate action. Not doing anything to correct something which may not have anything to do with you in the first place isn't discrimination. Furthermore, as I said we're talking about a situation where the candidates don't follow the general population's proportion, not a matter of selection by the employer.

How so?

First, because you cry wolf and risk alienating potential allies in the fight against discrimination, so that when real cases crop up they aren't willing to help out. Second, because if the solution to discrimination is government intervention, and your claim of discrimination is false but believed, then you are making an intervention where none is needed, potentially harming people along the way.

I don't see how the sector matters.

No one's selecting the voters, Archie. If there's a problem, it's one of convincing people to get off of their asses and vote.

Should we just accept that there is no problem there?

No one's talking about making assumptions, here. If you want to check it out to determine if there's a problem to be solved, fine. I'm saying that you shouldn't assume that a problem exists.

No its based on the practical consequence of your position.

Irrelevant. It's an appeal to emotion, and is thus discarded.

 

[Archie Gemmill Goal]

I don't want to play devil's advocate here but under that scenario would it be possible that they allow more black people to buy houses who normally couldn't afford it just to meet the government requirements?

Anything is possible but unless estate agents work differently where you live they don't actually do the finances.

If your general point is that the people they find are somehow less qualified or capable then as someone mentioned earlier businesses with greater diversity tend to perform better so that would be strong evidence against such a general claim.

Of course it may be that your assessment criteria of who is an isn't qualified actually need to be reviewed also. Quite often employers create BS lists of qualifications and experience to screen candidates that have little or nothing to do with their ability to do the job.

So for example requiring a degree in sanitation engineering for a toilet cleaner might in and of itself be discrimination if the population of degree qualified sanitation engineers is 99% white. If you remove that requirement then you will get a lot of 'unqualified' candidates being hired to the new job and they will be used as evidence of how AA is a scam to favour incapable minorities over highly qualified white folks.

 

[Belz...]

Anything is possible but unless estate agents work differently where you live they don't actually do the finances.

If your general point is that the people they find are somehow less qualified or capable then as someone mentioned earlier businesses with greater diversity tend to perform better so that would be strong evidence against such a general claim.

Of course it may be that your assessment criteria of who is an isn't qualified actually need to be reviewed also. Quite often employers create BS lists of qualifications and experience to screen candidates that have little or nothing to do with their ability to do the job.

So for example requiring a degree in sanitation engineering for a toilet cleaner might in and of itself be discrimination if the population of degree qualified sanitation engineers is 99% white. If you remove that requirement then you will get a lot of 'unqualified' candidates being hired to the new job and they will be used as evidence of how AA is a scam to favour incapable minorities over highly qualified white folks.

Ok thanks for the explanation.