Equal Opportunity = rigging the rules?
bat
today I had a conversation with our HR Manager, and we discussed the idea of
giving prospective hirees an IQ test. This already had been discussed
before, and now she said that she was unable to find an IQ test that would
be certified for equal opportunity. I became curious, why is this necessary,
and she explained that giving any quantative test at job interview is risky
because it can create grounds for a lawsuit based on discrimination. I asked
how this is possible. She gave an example.
According to the friend of her family, a firefighter, the city must maintain
certain quote in different races working for the city; becoming a
firefighter involves certain exam. Since black candidates are known to pass
this exam worse than white, the passing grade for them is 70 while for
whites it's 90. This is fully official policy, based on interpretation of
the equal opportunity as adjusting the criteria in such a way that all
candidates have equal opportunity regardless of their skill.
So now she is afraid that if we offer IQ test, the same logic will result in
losing potential lawsuit - at the trial, she imagined, we would be required
to prove that the criteria for the IQ test was absolutely fairly balanced
and adjusted for different races. Hence there are some providers that
"certify" various employment tests.
>From my point of view, all this "adjusting" is 100% racism, but on the other
hand I don't think she exaggerated about the firefighters. She also gave
another example - in her previous work for private company, they had to
maintain certain ratio of races among the employees because this was
required by their biggest customer Costco. Why was Costco concerned? because
Costco believes in equal opportunity, and requires it from all their
vendors.
So I'm wondering, is there indeed a practical risk in running job candidates
through IQ test? maybe not so much from the strictly law perspective, but
basing on the established practices in this politically correct nightmare.
This is all in Illinois.
Regards
Timothy
>
> So I'm wondering, is there indeed a practical risk in running job candidates
> through IQ test? maybe not so much from the strictly law perspective, but
> basing on the established practices in this politically correct nightmare.
> This is all in Illinois.
All an IQ test does is tell you how well the person scores on an IQ
test. Those tests are fairly meaningless, since the specific skills
needed to ace such a test are not directly relevant to any real-world
job.
That said, you definitely can get away with administering standardized
tests to applicants. Even the government does it, and they are of
necessity the most "politically correct" institution around. Your
main problem is a practical one: the tests cost money and take time to
administer and they don't really tell you anything useful.
Mike Jacobs
> today I had a conversation with our HR Manager,
<snip>
> and now she said that she was unable to find an IQ test that would
> be certified for equal opportunity.
I don't doubt that. IQ tests are culturally biased and, in fact, the
whole idea of IQ testing is scientifically suspect. What, exactly,
does it measure? Ordinary foax put WAY too much faith in standardized
testing to actually prove ANYTHING close to what it purports to test.
What an IQ test really measures is your ability to take an IQ test.
And, any contention that any such test simply measures innate, genetic
aptitude rather than learned, acquired cultural behaviors is
laughable. Ask any competent social scientist, or research these
issues yourself in Google or Wiki.
There are other, more useful measures of ability to perform the actual
job duties than such a test. Since in fact IQ test results ARE
skewed in favor of those who have a certain cultural background - a
family that encourages reading and scientific curiosity, a high
quality Anglo-style education, familiarity with common cultural tropes
of American culture - it is no surprise that middle- and upper-class
"white" test-takers do better on such tests.
> I became curious, why is this necessary,
> and she explained that giving any quantative test at job interview is risky
> because it can create grounds for a lawsuit based on discrimination.
Yes, it can. Have you been asleep through the entire confirmation
hearing for Justice Sotomayor a few months ago? Have you never heard
of Ricci v. DeStefano, the New Haven white firefighters' suit against
the city to _restore_ their promotion test results that the City had
thrown out because almost no Blacks or Hispanics had passed the test?
And, have you not heard that SCOTUS recently _overturned_ that result,
required the City to reconsider, and that even more recently, the
white firefighters in New Haven actually _did_ get the promotions they
had been tested for? Where have you been? Certainly not paying
attention on Earth, or in cyberspace either, because it's been all
over the place.
> I asked
> how this is possible. She gave an example.
>
> According to the friend of her family, a firefighter, the city must maintain
> certain quote in different races working for the city;
NO, the law does NOT impose or require any quota system. At least,
it is not SUPPOSED to be a quota system. That's what the New Haven
case was all about. The recent SCOTUS opinion was scorchingly clear
that the City must pick a racially-neutral means of choosing whom to
hire or promote and that the chosen method must actually be related to
some job-required skill set, not simply an abstract and general
"intelligence" or "aptitude" test. If the test _does_ meet those
requirements, the City (or other hiring entity) may use it to sort out
qualified candidates.
It's perfectly OK, frex, to test airline pilots on their ability to
fly in bad weather, their familiarity with radio procedures, their
knowledge of airspace restrictions and anti-hijacking protocols, etc.,
since those are directly related to the needed job skills, and no one
wants to have incompetent airline pilots flying around checking their
Myspace pages on their laptops while their autopilot takes them 200
miles past their destination (those guys lost their jobs pronto,
BTW). Likewise, it would make sense for firefighters up for promotion
to be tested on the kinds of skill sets they would actually have to
USE to do their job. There is nothing wrong with that - it is a Bona
Fide Occupational Qualification (BFOQ) and therefore by law NOT
discriminatory even if it does impact different racial or ethnic
groups to a different degree.
What is NOT permissible is using some general, culturally-biased
"aptitude test" or "IQ test" that is _known_ to be biased,
statistically, in favor of "white" applicants (or at least those who
were raised in uppercrust Anglo culture) which has no real value as a
predictor of job abilities or future success in performing those
duties.
> becoming a firefighter involves certain exam. Since black candidates are known to pass
> this exam worse than white, the passing grade for them is 70 while for
> whites it's 90. This is fully official policy, based on interpretation of
> the equal opportunity as adjusting the criteria in such a way that all
> candidates have equal opportunity regardless of their skill.
That is so totally wrong. In fact it sounds like this agency you work
for is setting itself up for a suit almost identical to the one that
was brought against the City of New Haven. Your employers are MIS-
INTERPRETING the anti-discrimination laws in so many ways I don't know
where to begin. Do they HAVE A LAWYER on the City's staff advising
them about this, or is this HR person simply a loose cannon on deck?
> So now she is afraid that if we offer IQ test, the same logic will result in
> losing potential lawsuit - at the trial, she imagined, we would be required
> to prove that the criteria for the IQ test was absolutely fairly balanced
> and adjusted for different races.
Which can't be done, because the IQ tests you are talking about are
inherently biased and worthless to show actual job skills. Therefore,
they are not a permissible winnowing tool to narrow the field of
applicants.
Perhaps someday someone _will_ develop a test that actually tests
something we can call "intelligence" that is not culturally biased.
As a trained social scientist before becoming a lawyer, I doubt it.
But even if they did, it would still NOT HAVE ANY RELATION to the
actual set of job-required skills and would STILL not be a pemissible
BFOQ. So one way or the other, using an IQ test to weed out
candidates in this situation is a Very Bad Idea. Your HR rep is
right about _that_ aspect.
> Hence there are some providers that "certify" various employment tests.
Correct. There ARE tests that are appropriate to be given in
particular job-hiring and promotional situations, because they ARE
reasonably related to required skills and are NOT culturally biased
and so they ARE a BFOQ. The "providers" you mention have social
scientists on staff who crunch the statistics and also look at the
actual content of the tests and, basically, "test the tests" to be
able to certify them as appropriate for use as a BFOQ.
> >From my point of view, all this "adjusting" is 100% racism,
The way you describe it, it may well be - did you expect an argument
from the liberals here? Sliding-scale passing grades for different
racial groups, and/or explicit racial quotas are, on the one hand, a
SLOPPY, LAZY way to try to ensure "racial equality" and, OTOH, are
just as bad as quotas that used to ban or limit the number of minority
applicants who would be accepted for a position (often, a big fat zed,
making it real easy) instead of having RACIALLY NEUTRAL, COLOR-BLIND
hiring practices which is, in fact, the GOAL of EEO legislation and
regulation. Read the New Haven opinion which contains a fairly good
background discussion of these issues. You can find it online in many
places, including http://www.supremecourtus.gov/opinions/08pdf/07-1428.pdf
> but on the other hand
> I don't think she exaggerated about the firefighters.
I don't know what you mean by that. If you intended to refer back to
her factual description of the procedures the City currently uses to
"adjust" the scores of white and black applicants, I guess you will
have to take her word for it unless you have some reason to think she
was pulling your leg. It's not the way the law says things are
SUPPOSED to be done, but I wouldn't put it past a wrongheaded City
administration to actually do things that way. I've certainly heard
worse.
> She also gave
> another example - in her previous work for private company, they had to
> maintain certain ratio of races among the employees because this was
> required by their biggest customer Costco. Why was Costco concerned? because
> Costco believes in equal opportunity, and requires it from all their
> vendors.
That anecdote is far too general to make any sensible reply regarding
its compliance with, or possible mis-interpretation (by you, the HR
lady, the "private company," or Costco, I'm not sure which) of, the
requirements of the Equal Employment Opportunity laws.
Certainly, one lesson we can learn from both the City firefighters,
and the Costco examples, is that EMPLOYERS SHOULD CONSULT LEGAL
COUNSEL before setting up any cockamamie plan that they THINK is going
to comply with EEO law based on some 25-words-or-less, sound-bite
version of what they THINK the law requires. Would these people take
out their own appendices? NO, they would hire a trained and qualified
surgeon. SO why do some managers think that a do-it-yourself EEO
plan makes any more sense? Potential multi-million dollar lawsuits
against their company are at stake if their naive, crude, home-made
attempts at achieving "racial balance" backfire in their faces. Tell
the bosses to TALK TO A LAWYER for Pete's sake. Employment lawyers
spend YEARS learning all the ins and outs of what is permissible and
what is not, and massive forests have been turned to pulp to write
books about it. EEO COMPLIANCE IS NOT A SOUND-BITE, DO IT YOURSELF
PROJECT for an employer, folks.
> So I'm wondering, is there indeed a practical risk in running job candidates
> through IQ test?
Absolutely yes, there is a risk. See above discussion.
> maybe not so much from the strictly law perspective,
Yes, it IS a risk from a strictly law perspective.
> but
> basing on the established practices
The "established practices" you refer to, IF they exist, are probably
ALREADY out of compliance with the requirements of the law. Your
City managers need to bring the City's attorneys on board, PRONTO, and
re-evaluate what they are doing in light of the _Ricci_ decision.
> in this politically correct nightmare.
"Politically correct" has nothing to do with it. IMO even the _use_
of that term is one more symptom of the right-wing "sound bite"
mentality that tries to reduce these complicated, socially-valued
goals to an absurdity so that they can be castigated as "liberal
foolishness" or worse. I don't know where you stand on the political
spectrum, but if you _believe_ that EEO is nothing more than
wrongheaded attempts at being "political correct" and a "nightmare"
then you stand somewhere to the right of Rush Limbaugh and Sarah
Palin. Such reductionist polemicization is not conducive to any
actual THOUGHT on this issue or any other issue. Ironically, the
right wing that is constantly shouting "politically correct" as their
knee-jerk reaction against any liberal idea are themselves the ones
who are being "politically correct" by their own, far-right concept of
what is "correct" - slavishly invoking ideological totems without
regard to practical needs - even while they accuse the left
(incorrectly) of doing the same thing.
> This is all in Illinois.
EEO is a matter of Federal law (although each state is free to enact
their own EEO laws if they do not conflict with the requirements of
Federal law). The SCOTUS opinion in _Ricci_ is binding on all the
states and every employer in USA who has to face EEO issues in testing
applicants for hiring or promotion. Somebody at your City who knows
what they are doing ought to read it.
> Regards
Yes, cheers. Let's hope for a good New Year, and good New Decade.
The Oughts were pretty Oughful, all things considered.
--
This posting is for discussion purposes, not professional advice.
Anything you post on this Newsgroup is public information.
I am not your lawyer, and you are not my client in any specific legal
matter.
For confidential professional advice, consult your own lawyer in a
private communication.
Mike Jacobs
LAW OFFICE OF W. MICHAEL JACOBS
10440 Little Patuxent Pkwy #300
Columbia, MD 21044
(tel) 410-740-5685 (fax) 410-740-4300
Cy Pres
[Different test scores considered a "pass" for firefighters depending
on what race they are.]
>So I'm wondering, is there indeed a practical risk in running job candidates
>through IQ test? maybe not so much from the strictly law perspective, but
>basing on the established practices in this politically correct nightmare.
>This is all in Illinois.
You might want to take a look at the recent Supreme Court decision
Ricci v. DeStefano. 129 S.Ct. 2658, 106 Fair Empl.Prac.Cas. (BNA)
929, 92 Empl. Prac. Dec. P 43,602, 174 L.Ed.2d 490, 77 USLW 4639, 09
Cal. Daily Op. Serv. 8212, 2009 Daily Journal D.A.R. 9567, 21 Fla. L.
Weekly Fed. S 1049, U.S., June 29, 2009 (NO. 07-1428, 08-328).
This case had very similar facts, even being about testing
firefighters in a manner which favored minorities.
>From the syllabus to the case:
"New Haven, Conn. (City), uses objective examinations to identify
those firefighters best qualified for promotion. When the results of
such an exam to fill vacant lieutenant and captain positions showed
that white candidates had outperformed minority candidates, a
rancorous public debate ensued. Confronted with arguments both for and
against certifying the test results-and threats of a lawsuit either
way-the City threw out the results based on the statistical racial
disparity. Petitioners, white and Hispanic firefighters who passed the
exams but were denied a chance at promotions by the City's refusal to
certify the test results, sued the City and respondent officials,
alleging that discarding the test results discriminated against them
based on their race in violation of, inter alia, Title VII of the
Civil Rights Act of 1964. The defendants responded that had they
certified the test results, they could have faced Title VII liability
for adopting a practice having a disparate impact on minority
firefighters. The District Court granted summary judgment for the
defendants, and the Second Circuit affirmed.
"Held: The City's action in discarding the tests violated Title VII.
Pp. 2672 - 2682."
More at:
http://www.scotuswiki.com/index.php?title=Ricci%2C_et_al._v._DeStefano%2C_et_al.
So yes, I'd say there's a very good chance the policy you discuss is
problematic.
Deadrat
> Hello,
>
> today I had a conversation with our HR Manager,
The implication here is that "bat" is in some position of authority in
his company and had a serious conversation with HR about giving IQ tests
to applicaants.
I call bullshit.
1. IQ tests are not only thoroughly discredited as a reliable tool for
measuring mental acuity (See SJ Gould, _The Mismeasure of Man_.), but no
company is looking to employ people who are just "smart." If that were
the case, where would they get an HR department? Companies want
employees who are among other things, reliable, responsible, honest,
trainable, collegial, and (often) experienced. An IQ score tells them
nothing about these qualities, and an HR manager in a company large
enough to have an HR manager would know that.
2. "bat" seems completely ignorant about employment law. He's surprised
that "quantitative" tests might be discriminatory; he thinks racial
quotas are legal; he relies on FOAF evidence about firefighters but seems
unaware of the Ricci case.
3. "bat"'s grievances loom so large. He knows that black candidates
can't compete against white candidates: they "pass this exam worse than
white," in his words. He believes that equal opportunity means adjusting
hiring criteria "regardless of ... skill." He uses scare quotes when he
describes companies that certify employment tests. He thinks
"adjusting" scores, for which he has no evidence, is "100% racism." (I'm
guessing he's conflating university admissions criteria with employment.)
Based on third-hand anecdotal evidence, he claims to live in a
"politically correct nightmare" that trumps the law, or as he puts it,
"the strictly law perspective."
<snip/>
> Regards
Instead of regards, I'd like some regarding, bat. Like what your post is
really about, because I'm having a hard time believing it actually has to
do with employment policy.
And, Mike, nice reply. But pace yourself, my man. A new year is
starting up.
bat
MJ> On Dec 31, 7:52 am, "bat" <b...@bats.com> wrote:
??>> today I had a conversation with our HR Manager,
MJ> <snip>
??>> and now she said that she was unable to find an IQ test that would
??>> be certified for equal opportunity.
MJ> I don't doubt that. IQ tests are culturally biased and, in fact, the
MJ> whole idea of IQ testing is scientifically suspect. What, exactly,
MJ> does it measure?
The ability to think logically and make conclusions. We of course are not
going to run iq test on workflow applicants (our business is
manufacturing), but the whole topic of iq testing came up as finding the
solution to the most prevalent complaint on the office employees - "they are
idiots". I can easily show parallels between the puzzles offered by usual iq
tests and problems faced by document control employee deciding what
departments need to approve specific engineering change, or quality engineer
investigating the root cause of the failure. Not to mention the managers.
MJ> the chosen method must actually be related to some job-required skill
MJ> set, not simply an abstract and general "intelligence" or "aptitude"
MJ> test. If the test _does_ meet those requirements, the City (or other
MJ> hiring entity) may use it to sort out qualified candidates.
MJ> It's perfectly OK, frex, to test airline pilots on their ability to
MJ> fly in bad weather, their familiarity with radio procedures, their
MJ> knowledge of airspace restrictions and anti-hijacking protocols, etc.,
MJ> since those are directly related to the needed job skills, and no one
MJ> wants to have incompetent airline pilots flying around checking their
MJ> Myspace pages on their laptops while their autopilot takes them 200
MJ> miles past their destination (those guys lost their jobs pronto,
MJ> BTW).
I think this exact example demonstrates why testing these "specific" skills
does not make much sense. These pilots were not incompetent, they were
irresponsible. They certainly were certified, and they did pass the very
exams that demonstrated their familiarity with radio procedures etc. If
there's a test that could measure person's ability to disregard his duties,
I'm sure it's much closer to "abstract and general" than to "job-related
skill set". Or a test that would measure person's tendency to look for the
means to solve the problem vs. looking for an excuse not to solve it and
then blame someone else. Specific job-related skills can be acquired quickly
when the attitude is right; it's not the skill that is in deficit, it's the
right attitude. I'm not saying that iq test is the full solution to such
benchmark, but it's not too far from it. If someone could suggest a test
relevant to these tendencies, I'm all ears.
Regards
Deadrat
> Hello,
>
> MJ> On Dec 31, 7:52 am, "bat" <b...@bats.com> wrote:
>
> ??>> today I had a conversation with our HR Manager,
> MJ> <snip>
> ??>> and now she said that she was unable to find an IQ test that
> would ??>> be certified for equal opportunity.
>
> MJ> I don't doubt that. IQ tests are culturally biased and, in
> fact, the MJ> whole idea of IQ testing is scientifically suspect.
> What, exactly, MJ> does it measure?
>
> The ability to think logically and make conclusions.
Please support this contention.
Let me warn you, though that any attempt is a fool's errand. In fact,
you will fail at the attempt to rigorously define "[t]he ability to think
logically." And logical things aren't always right. Logically (or at
least inductively), I can conclude that things in motion tend to come to
rest. But that's wrong.
> We of course are
> not going to run iq test on workflow applicants (our business is
> manufacturing), but the whole topic of iq testing came up as finding
> the solution to the most prevalent complaint on the office employees -
> "they are idiots".
And if you ask the office employees, I'm sure you'll find that the most
prevalent complaint about the office managers is that "they are idiots."
> I can easily show parallels between the puzzles
> offered by usual iq tests and problems faced by document control
> employee deciding what departments need to approve specific
> engineering change,
No, you can't. But please show that I'm wrong by demonstrating these
"parallels."
If your company can't get its document control employees to route
documents to the correct departments, then your company's training is
lacking. Look to that before you attempt to raise the average IQ of your
employees.
> or quality engineer investigating the root cause of the failure.
I've done root cause analysis of software. Critical to this task is
experience in the language, the application, and the operating
environment. It also helps if you like solving puzzles. But just
because someone can solve the puzzle questions on an IQ test doesn't
mean he'll be good at root cause analysis. And I defy you to show a
correlation.
Which is exactly why the you're going to run into legal problems by
using IQ tests as an employment sieve.
> Not to mention the managers.
Unless your company is very unusual, your first-level managers are a
collection of sneaky little shits who look out only for themselves by
spending their work hours taking credit for the accomplishments of others
and shifting the blame for their own failures. Their deviousness no
doubt often rises to the level of genius, but that's not detectable on on
IQ test.
<snipped: careless pilots example >
> I think this exact example demonstrates why testing these "specific"
> skills does not make much sense. These pilots were not incompetent,
> they were irresponsible. They certainly were certified, and they did
> pass the very exams that demonstrated their familiarity with radio
> procedures etc.
Do you really not understand how little sense this makes? You're
proposing that testing for competence is nonsensical because competent
people might be irresponsible. This is just another reason why I'm
having a hard time believing you have anything to do with managing a
manufacturing company. Initially, an employer gauges responsibility by
interviewing a prospective employees about their work experience and
education. Thereafter, managers monitor responsibility by conducting
performance reviews. These things may be done badly, of course, but
they're better than administering IQ tests.
> If there's a test that could measure person's ability
> to disregard his duties, I'm sure it's much closer to "abstract and
> general" than to "job-related skill set".
You're sure of a lot of things that aren't true. Please tell us why you
think responsibility is better measured by tests of abstract thinking
than by tests of specific skills or aptitudes. Be detailed. Show your
work.
> Or a test that would measure
> person's tendency to look for the means to solve the problem vs.
> looking for an excuse not to solve it and then blame someone else.
> Specific job-related skills can be acquired quickly when the attitude
> is right; it's not the skill that is in deficit, it's the right
> attitude.
And you think an IQ test can measure an employee's attitude? Really?
> I'm not saying that iq test is the full solution to such
> benchmark, but it's not too far from it.
Please support this contention. Start with the definition of your
metric, so we can agree on what "far" means in this context.
> If someone could suggest a test relevant to these tendencies, I'm all
> ears.
There isn't one. An IQ test, however, will be the perfect disaster. Not
only will it fail to identify those competent quality control engineers
who are responsible and have the right attitude, but when you're charged
with discrimination, that failure will be used to show that the test
isn't related to the job.
> Regards
Whatever.
Seth
> MJ> I don't doubt that. IQ tests are culturally biased and, in fact, the
> MJ> whole idea of IQ testing is scientifically suspect. What, exactly,
> MJ> does it measure?
>
>The ability to think logically and make conclusions.
Very few tests measure that (alone); they assume some "basic
underlying knowledge" which is where the cultural bias enters.
> We of course are not going to run iq test on workflow applicants
>(our business is manufacturing), but the whole topic of iq testing
>came up as finding the solution to the most prevalent complaint on
>the office employees - "they are idiots". I can easily show parallels
>between the puzzles offered by usual iq tests and problems faced by
>document control employee
But you aren't enough of an expert on testing that your claim will be
convincing (to a court, if someone sues).
> deciding what departments need to approve specific engineering
>change, or quality engineer investigating the root cause of the
>failure. Not to mention the managers.
On the other hand, if you wrote a test that involved giving candidates
the same problems that people had to solve in the past few years, and
asking what they'd do (or how they'd go about solving them), it would
be hard to argue that you aren't testing relevant skills.
> MJ> the chosen method must actually be related to some job-required skill
> MJ> set, not simply an abstract and general "intelligence" or "aptitude"
> MJ> test. If the test _does_ meet those requirements, the City (or other
> MJ> hiring entity) may use it to sort out qualified candidates.
Unless the lawyers get involved. Some years ago, I was in a position
where my department was hiring APL programmers (as fast as we could
find good ones). We had a programming test that had been written
in-house, and modified as we went along, by some of the world's best
APL programmers. (It was a very hard test, but we were looking only
for extremely good programmers.) Then the lawyers told us we couldn't
give the test to applicants unless some outside authority had designed
it and certified that it was unbiased. (How I'd go about putting
cultural bias into an APL test I was never able to figure out. Maybe
if somebody didn't know there were 7 days in a week or how leap years
worked or something, but we were perfectly willing to give a candidate
any relevant information for doing the test. Approximately nobody
knows precisely how 30/360 day count works, without the description in
front of them.)
> MJ> It's perfectly OK, frex, to test airline pilots on their ability to
> MJ> fly in bad weather, their familiarity with radio procedures, their
> MJ> knowledge of airspace restrictions and anti-hijacking protocols, etc.,
> MJ> since those are directly related to the needed job skills, and no one
> MJ> wants to have incompetent airline pilots flying around checking their
> MJ> Myspace pages on their laptops while their autopilot takes them 200
> MJ> miles past their destination (those guys lost their jobs pronto,
> MJ> BTW).
>
>I think this exact example demonstrates why testing these "specific" skills
>does not make much sense.
It makes a lot of sense. I don't want to fly with a pilot who can't
handle it when the plane hits weather.
> These pilots were not incompetent, they were irresponsible.
Unfortunately, there's no test for "not irresponsible". But I'd
rather have a competent irresponsible pilot than an incompetent one.
> If there's a test that could measure person's ability to disregard
>his duties, I'm sure it's much closer to "abstract and general" than
>to "job-related skill set".
But you'd have to prove that any such test actually does measure a
person's likelihood of disregarding his duties.
> Specific job-related skills can be acquired quickly when the
>attitude is right;
In some cases, yes; in other cases, if your brain isn't wired the
right way, you aren't ever going to gain those skills no matter how
good your attitude. (Consider a job that required someone to see
things and argue the way I do in this newsgroup; how many people here
think they could acquire such a skill?)
Seth
bat
CP> Pp. 2672 - 2682."
CP> More at:
CP> http://www.scotuswiki.com/index.php?title=Ricci%2C_et_al._v._DeStefano%2C_et_al.
CP> So yes, I'd say there's a very good chance the policy you discuss is
CP> problematic.
Thanks; but I think, there's a difference. The New Haven case woul be good
argument against the practices of the city of that firefighter. But there
was quite defined passing grade, so the hiring decision was quantative:
passed=hired, did not pass=not hired. In our case, the idea is to offer the
test just as yet another phase of the interview, not as the "passing"
factor; in the end, if we hire the individual, it would be not just because
he passed the test (we aren't even going to define the passing number), but
basing on the whole experience from the interview, including the test. If
unseccessful applicant wanted to sue, he would only have the mere fact that
the test was offered - but then, there were many other questions that also
were offered, including the most standard HR questions - so from my own
understanding, the merits of suing for IQ test would not be any better than
suing for the question "tell me about your weakensses", for instance. With
the full understanding that "anybody can sue for anything", still, do you
think that this possibility is real?
thanks
Seth
Deadrat <a...@b.com> wrote:
>I call bullshit.
>
>1. IQ tests are not only thoroughly discredited as a reliable tool for
>measuring mental acuity (See SJ Gould, _The Mismeasure of Man_.), but no
>company is looking to employ people who are just "smart."
Nobody said they were; the IQ test would be _part_ of the
interview/hiring process, not the definitive answer. If there were a
way to determine who is "smart", wouldn't a smart employee be better
than a non-smart one, other qualifications being equal?
> If that were the case, where would they get an HR department?
Most companies hire them. Some outsource them.
> Companies want employees who are among other things,
Bingo! "among other things" Isn't _one_ of the things they want
"smart"?
> reliable, responsible, honest, trainable, collegial, and (often)
>experienced. An IQ score tells them nothing about these qualities,
But you _started_ by admitting those are only _some_ of the qualities
wanted.
(I'll note that you didn't include "competent" or "skillful" in your
list of qualities. Some companies really do consider those quite
important in hiring decisions.)
> and an HR manager in a company large enough to have an HR manager
> would know that.
Would know that there are a lot of qualities the ideal employee has?
Sure.
Seth
Mike Jacobs
> In article <hhmc06$97...@speranza.aioe.org>, bat <b...@bats.com> wrote:
> > MJ> I don't doubt that. �ｿｽ IQ tests are culturally biased and, in fact, the
> > MJ> whole idea of IQ testing is scientifically suspect. �ｿｽ What, exactly,
> > MJ> does it measure?
<snip>
> > MJ> the chosen method must actually be related to some job-required skill
> > MJ> set, not simply an abstract and general "intelligence" or "aptitude"
> > MJ> hiring entity) may use it to sort out qualified candidates.
>
> Unless the lawyers get involved.
<example of lily-livered lawyers banning a worthwhile test because it
wasn't "officially certified" snipped>
Well, that gets us back to the issue of how much risk one is willing
to accept vs. the expected benefit, as well as the issue of which side
of the problem you're standing on, and thus what perspective you have
on the issue.
As any MLM regular knows, in USA anyone can sue anyone else at any
time for any reason. So, even what Seth's company's lawyers did
would not guarantee that _no_ employee or prospective employee passed
over for this APL programming job would be able to sue for
discrimination. However, the lawyer's advice _did_ minimize that
risk, and also minimized the risk of a _successful_ such suit, since
the "seal of approval" from a recognized certifying body is an easier
and more likely way to prove that one's test is not culturally
biased. However, consider whether that level of security against
successful suit had an unacceptable downside, i.e. that the company
then had _no_ good way to weed out the mediocre APL programmers from
the top-notch ones, and wanted the latter so badly that they were
willing to take the risk of being sued by the mediocre ones, and then
having to defend the basis on which they made that decision.
Plus, the company (knowing its goals, and its risks) is likely to see
things from a different pespective than will the disgruntled
prospective employee who is contemplating whether to sue or not. Not
only does the burden-of-proof issue normally make the employee's
threshold of unacceptable, suable bias somewhat _higher_ (on whatever
metric is used) than the company's threshold of acceptable, doable
testing, even if they are both acting in good faith; but the
perspective of the advice offered by each sides' lawyer is also going
to be different depending on what his client is trying to accomplish.
If the company WANTS to be able to test for only high-level candidates
for APL programmers, the company lawyer is going to be looking for
ways that the company can do that (if he has any cojones, coupled with
any actual understanding of why such a test is almost certainly NOT
culturally biased even if it is not "officially certified" as such),
while the prospective plaintiff's lawyer is going to be looking for
ways to poke holes in the company's testing scheme to show bias, and
will salivate over the fact that they were using an in-house, not-
independently-certified test. What that means is, the company may
well decide to go forward with the non-certified test if they want the
result badly enough and need it quickly enough (so they don't have
time to submit their in-house-developed test for independent vetting),
but both sides will realize that their doing so is more likely to lead
to a lawsuit and thus increase the transaction costs all around.
> > Specific job-related skills can be acquired quickly when the
> >attitude is right;
>
> In some cases, yes; in other cases, if your brain isn't wired the
> right way, you aren't ever going to gain those skills no matter how
> good your attitude. �ｿｽ(Consider a job that required someone to see
> things and argue the way I do in this newsgroup; how many people here
> think they could acquire such a skill?)
IMO your skill set is unique, Seth. Someone else could argue _their_
way, sure, but they would never do precisely _your_ way, with your
voice, your syntax, your perspective. Which fact, tongue-in-cheek as
it sounds, proves your point (and mine). Some jobs, like actors,
musicians, artists, professors, columnists, troubleshooters of
whatever profession, are so keyed to the skills of one particular
person that no one else could fill the bill the employer wants.
Deadrat
> In article <GOudnZ_WwdJQDaPW...@giganews.com>,
> Deadrat <a...@b.com> wrote:
>
>>I call bullshit.
>>
>>1. IQ tests are not only thoroughly discredited as a reliable tool for
>>measuring mental acuity (See SJ Gould, _The Mismeasure of Man_.), but
>>no company is looking to employ people who are just "smart."
>
> Nobody said they were; the IQ test would be _part_ of the
> interview/hiring process, not the definitive answer. If there were a
> way to determine who is "smart",
There's no way to determine (or even realiably define) who's "smart."
Although both "bat" and I think we know intuitively who belongs to the
set of smart people (and, of course, that we belong there as well), we
don't have a reliable written test to tell us.
> wouldn't a smart employee be better
> than a non-smart one, other qualifications being equal?
Not in New London, CT, where in 1996 they rejected an applicant for the
police department because he was too "smart."
>> If that were the case, where would they get an HR department?
>
> Most companies hire them. Some outsource them.
Sit back, take a deep breath, and ask yourself, "Did I really take that
as a serious question?"
NOw let out your breath slowly. Get it now?
>> Companies want employees who are among other things,
>
> Bingo! "among other things" Isn't _one_ of the things they want
> "smart"?
Sure, but hardly the point. The topic is that companies want a simple,
reliable, unbiased test to determine which applicants are smart.
I want to be young, rich, and handsome. That isn't gonna happen either.
<snip/>
> Seth
Cy Pres
>"bat" <b...@bats.com> wrote in news:hhmc06$97h$1...@speranza.aioe.org:
>> If someone could suggest a test relevant to these tendencies, I'm all
>> ears.
>There isn't one. An IQ test, however, will be the perfect disaster. Not
>only will it fail to identify those competent quality control engineers
>who are responsible and have the right attitude, but when you're charged
>with discrimination, that failure will be used to show that the test
>isn't related to the job.
Though I agree with you in general about the scientific merit of IQ
tests, courts have generally upheld the use of IQ and other types of
aptitude of personality test, even when the effects of the test tend
to be racially discriminatory, specifically in the line of cases
following Washington v. Davis, 426 U.S. 229 (1976) (test of verbal
skills in which white applicants tended to score better not invidious
discrimination under equal protection clause).
Note that if the purpose of adopting the test *was* in order to
discriminate racially, there would be a violation of equal protection.
I find this a ridiculous conclusion, since in both cases, the
plaintiff has suffered exactly the same harm, and it invites the use
of "fig leaf" tests presumably to serve some neutral job-related
purpose but in actuality *intended* to be racially discriminatory.
After all, nobody will admit to a discriminatory intent if they are
not quite foolish.
Cy Pres
>In our case, the idea is to offer the
>test just as yet another phase of the interview, not as the "passing"
>factor; in the end, if we hire the individual, it would be not just because
>he passed the test (we aren't even going to define the passing number), but
>basing on the whole experience from the interview, including the test.
This introduces a number of wrinkles. I would still urge that any
entity adopting such a policy talk to lawyer(s) first, beause "putting
the thumb on the scale" by adding points based on race, even if it
doesn't dispose of the hiring situation by itself, is still going to
get anyone *not* hired under such a regime (whether or not a minority)
considering themselves to be discriminated against. It's not really
the kind of thing that should be done shooting from the hip.
Also, on top of the federal laws against discrimination, there is also
generally a thicket of state regulations and equal opportunity agency
bodies with administrative processes of their own. You generally
don't want to be involved with any of them if you can avoid it.
Deadrat
news:nq02k5h3le385geil...@4ax.com:
> On Sat, 02 Jan 2010 14:32:15 -0600, Deadrat <a...@b.com> wrote:
>
>>"bat" <b...@bats.com> wrote in news:hhmc06$97h$1...@speranza.aioe.org:
>
>>> If someone could suggest a test relevant to these tendencies, I'm
>>> all ears.
>
>>There isn't one. An IQ test, however, will be the perfect disaster.
>>Not only will it fail to identify those competent quality control
>>engineers who are responsible and have the right attitude, but when
>>you're charged with discrimination, that failure will be used to show
>>that the test isn't related to the job.
>
> Though I agree with you in general about the scientific merit of IQ
> tests, courts have generally upheld the use of IQ and other types of
> aptitude of personality test, even when the effects of the test tend
> to be racially discriminatory, specifically in the line of cases
> following Washington v. Davis, 426 U.S. 229 (1976) (test of verbal
> skills in which white applicants tended to score better not invidious
> discrimination under equal protection clause).
Washington v Davis says that a test that has a rational purpose does not
violate due process even if the test has a discriminatory effect. The
claim of a violation of due process arose because the employer was
Washington, DC, which is specifically exempted from the 1964 civil rights
act.
We've been using the term "IQ test" without defining it. IQ tests are
supposed to measure innate intelligence, and although verbal skills are
part of that, IQ tests aren't simply tests of verbal skills. The latter
may certainly be defended if verbal skills are an important part of the
job requirements.
But bat's company doesn't enjoy the exemptions of the federal government,
and bat wants his company to employ "smart" people, not just those with
an extensive vocabulary.
> Note that if the purpose of adopting the test *was* in order to
> discriminate racially, there would be a violation of equal protection.
<snip/>
Due process, I think.
I should make it clear that I think bat's company could mount a
successful defense to a charge of discrimination. It would (naturally)
depend on the test, the internal company discussions, and the rest of the
facts.
But as an IQ test would be worthless, I stil don't think it's worth the
risk. Bat's MMV.
bat
http://hbswk.hbs.edu/archive/5124.html
Harvard Business Review, Vol. 83, No. 11, November 2005.
"Although IQ tests were not originally intended for use in business, studies
have shown that these instruments predict work performance at least as well
as competency interviews do (the most common assessment tool used today for
hiring and promotion) and about ten times better than personality tests do.
(...)
The skills that IQ tests assess represent a fraction of a person's existing
cognitive abilities. Some of the skills measured�such as vocabulary,
arithmetic, and spatial reasoning�have almost no relevance to managerial
work. Moreover, the topics tested would seem academic and elementary�indeed,
almost insulting�to people with extensive professional experience. The
format is also ill suited to business. Executives rarely if ever confront
problems that have just one right answer; nor do they have the option of
picking one answer from several choices listed. IQ test questions don't
assess the practical, on-your-feet thinking skills needed in business.
What's more, these tests have been repeatedly accused of racial and gender
bias.
Yet, despite these very real shortcomings, IQ tests are still a better
predictor of managerial success than any other assessment tool. "
Deadrat
Great. HBR says that a seemingly unsuitable test that doesn't measure
anything actually predicts a quantity that can't be measured. Or is at
least subject to selection bias.
In other words, the people who use IQ tests report that the tests predict
their managers' "success." I'm assuming that the companies didn't allow
HBR to independently assess how well the managers did.
The article notes that IQ tests are ten times better than personality
tests. Ten times nothing is still nothing. Although I'm still trying to
figure out what "ten times" could mean in this context. Ten times as
many managers with "high" IQ tests got promoted than those with "good"
personality scores?
Supposedly IQ tests work as well as competency interviews. Ten times a
fraction is still small. Interviews are difficult to do well, so the bar
is very low.
But I suppose I should withdraw my claim that IQ tests are worthless in
this context. If your company doesn't want to take the time, effort, and
expense to train interviewers or to develop or buy unbaised tests of
relevant skills, then you might be able to save money by simply
administering IQ tests to applicants. Then, if HBR is right, you won't
be any worse off.
And then, if you're unlucky, you'll be sued for discriminatory practices,
and you'll be able to sit and listen to the plaintiff's read the second
paragraph you've quoted. These suits aren't easy for plaintiffs to win,
so I'm not saying it's a slam dunk for them.
Seth
>> Harvard Business Review, Vol. 83, No. 11, November 2005.
. . .
>> Yet, despite these very real shortcomings, IQ tests are still a better
>> predictor of managerial success than any other assessment tool. "
>
>Great. HBR says that a seemingly unsuitable test that doesn't measure
>anything actually predicts a quantity that can't be measured.
"Managerial success" can be measured, at least indirectly: raises,
promotions, etc. If a "seemingly unsuitable" test actually predicts
how well a candidate will do (if hired), then I say the "seeming" was
wrong and the test is quite suitable.
It doesn't matter whether or not I can explain why something works
(has predictive power), only whether or not it actually does. The
latter can easily be tested statistically.
>In other words, the people who use IQ tests report that the tests predict
>their managers' "success." I'm assuming that the companies didn't allow
>HBR to independently assess how well the managers did.
What would you consider an "independent assessment"?
>The article notes that IQ tests are ten times better than personality
>tests. Ten times nothing is still nothing.
What is your evidence that personality tests are "nothing"? The
predictability value of a test is a measurable statistic.
> Although I'm still trying to
>figure out what "ten times" could mean in this context. Ten times as
>many managers with "high" IQ tests got promoted than those with "good"
>personality scores?
What were the correlation coefficients? I'd say that 98% correlation
is ten times as good as 80% correlation (error of 2% vs. 20%).
>Supposedly IQ tests work as well as competency interviews. Ten times a
>fraction is still small. Interviews are difficult to do well, so the bar
>is very low.
So what do you suggest that's better? If all the existing tools are
bad, then determining which are best and how well each combination
works is beneficial.
>But I suppose I should withdraw my claim that IQ tests are worthless
>in this context. If your company doesn't want to take the time,
>effort, and expense to train interviewers or to develop or buy
>unbaised tests of relevant skills,
Why do you believe this is even possible, much less not prohibitively
expensive?
> then you might be able to save money by simply administering IQ
>tests to applicants. Then, if HBR is right, you won't be any worse
>off.
That is, you'll be better off.
>And then, if you're unlucky, you'll be sued for discriminatory practices,
>and you'll be able to sit and listen to the plaintiff's read the second
>paragraph you've quoted.
The bit about despite the tests measuring stuff that doesn't seem
relevant, they work anyway? Or the statements that they've been
*accused* of being biased?
Seth