Althouse: End unpaid internships, drop the SAT for college admissions, and stop requiring a college degree for jobs where it's not needed.

March 9, 2012

End unpaid internships, drop the SAT for college admissions, and stop requiring a college degree for jobs where it's not needed.

Charles Murray has some ideas about narrowing class differences in America.

132 comments:

traditionalguy said...: College degrees and SATs that get college admissions have had a major value: the graduates are literate. Reading and writing still has tremendous value within society.

Should we replace the College degree with essay tests like GED tests?

Ask the Chinese Tiger Moms.; 3/9/12, 10:36 AM
Chip S. said...: Give everybody the Wonderlic test. The NFL seems to think it's a useful partial predictor of job performance.

Jobs that require specific knowledge, as opposed to simply requiring at least trace amounts of That Indefinable Something That Can Never Be Mentioned But Seems Nevertheless To Be Measurable and Correlated With Desirable Outcomes, can require advanced training in those areas.; 3/9/12, 10:42 AM
Methadras said...: This argument basically describes how to fill job numbers by grabbing the low hanging fruit. The idea that merit has some merit for getting into good schools or good jobs is being sidelined by these suggestions. Paying ones dues has a place in society. It teaches something. I agree that certain jobs requiring a degree can be a little stupid, but the rest is pap.; 3/9/12, 10:42 AM
Methadras said...: This comment has been removed by the author.; 3/9/12, 10:43 AM
Bryan C said...: I'm not interested in telling people who they can and cannot work for, or that they can't work for free if they want to. That's not any kind of "reform". If we want more young adults working real jobs, cut minimum wage so that they can compete with illegals and other under-the-table employees.; 3/9/12, 10:45 AM
cubanbob said...: get rid of griggs vs duke power and the problem is solved.ea; 3/9/12, 10:46 AM
MnMark said...: There seems to be a root cultural aassumption that has taken hold that economic differences are bad, and that the situation will only be good when there are no significant numbers of people who are richer than anyone else. Marxism truly has won a stealth victory in the West. The culture of envy is thoroughly entrenched.

We should be thrilled that modern technologies and knowledge allow people to leverage their creativity to produce such phenomenal amounts of value, which the free market then rewards them for with billions of dollars. They deserve it.

Cave man society was very egalitarian by our standards. Everyone was equally dirt poor, and there was no possibility of becoming very well-off.

Even if they managed to steal enough from productive people so that the richest person had no more than, say, twice what the poorest person had, then the envy level would simply be adjusted so that the idea of someone having twice as much was immoral. If the richest person only had two cents more than the poorest, they'd argues that those two cents more were immoral.; 3/9/12, 10:52 AM
rcocean said...: I'd settle for getting rid of the SS payroll cap, increasing the capital gains rate, and increasing the top rate back to 50 percent.

And capping the mortgage deduction at $50,000.; 3/9/12, 10:52 AM
IM-MTO said...: I agree with you 100%. College Degrees to be a retail store manager or police officer for instance is about as stupid a thing as I've ever heard of. We need to emphasize skills and abilities again. Getting a real job getting your hands dirty has become something beneath so many people, because they have all been told how special they are. Especially if they went to college, even if all they learned was how to use a condom and do a beer bong.; 3/9/12, 10:55 AM
hawkeyedjb said...: "I'd settle for getting rid of the SS payroll cap, increasing the capital gains rate, and increasing the top rate back to 50 percent.

And capping the mortgage deduction at $50,000."

The country has no problems that can't be solved by a trillion dollar tax increase.; 3/9/12, 10:57 AM
Revenant said...: I wasn't aware that "intelligent" was a social class.; 3/9/12, 10:57 AM
Herb said...: class mobility in the US is better than almost any other country in the world. why is this even worth a discussion.; 3/9/12, 10:57 AM
bagoh20 said...: I'm a big believer in testing ability rather than credentials or any other indirect determiner.

No degree should be required for anyone who can pass something like the bar exam, or a similar test that is thorough, often including actual demonstration of ability.

Many jobs have no business asking for a college degree, it's got little to do with the work.

Even a job in something like engineering would be better filled by an exhaustive test that could easily prove if you can do engineering, rather evaluating some institution's ability to fail the uneducated, which they rarely do.

This could be expensive, but so is hiring people who have little more than a piece of paper proving payment and attendance.

I believe driving tests for licence or DUI arrest should be the same. Can you drive well, period, regardless of age, or current happiness index..; 3/9/12, 10:59 AM
edutcher said...: The last one is the biggie. A lot of jobs require the sheepskin, or expect it if you want to advance, but it's really not needed for the job.

And a lot of trades pay as well as the professions, so those need to be encouraged, as well.; 3/9/12, 11:00 AM
Sorun said...: Let's experiment with airline pilot jobs first.; 3/9/12, 11:05 AM
Eric Palmer, PeopleSoft Consultant said...: This comment has been removed by the author.; 3/9/12, 11:06 AM
Eric Palmer, PeopleSoft Consultant said...: Even in professional level jobs (analyst, accountant, etc), what's required isn't a college degree, but the skills we assume a college degree brings. The more accurate and defensible way to measure those skills is to do so directly: through a test. But most people don't know how to develop a reliable, valid test. Interviews are much easier, but you can only interview a few people. So employers use artificial requirements as a tool for winnowing the applicants to a small number. They don't consider the fact that artificial requirements can also weed out your best candidates.; 3/9/12, 11:08 AM
MadisonMan said...: I don't see Colleges embracing an idea that suppresses demand for College degrees.

I'm all for ending unpaid internships however. I think a strong economy would help that.; 3/9/12, 11:17 AM
Eric Palmer, PeopleSoft Consultant said...: This comment has been removed by the author.; 3/9/12, 11:17 AM
cubanbob said...: since the duke power ruling eliminates the ability for employers to give iq and skills tests for th most part, sat testscand college degrees are defacto substitues for them. besides colleges are for the most part the new high school.

rcocean i would even happier is social security was privitized along with medicare and the rest of the entitlement state was severly rolled back.; 3/9/12, 11:19 AM
Eric Palmer, PeopleSoft Consultant said...: I did a major study of professional level accounting positions for a public agency once. They were hiring based on mathematics and accounting knowledge. They were not having long term success with their hires. What the study showed was that their most successful hires had excellent writing skills, and could draft a well written, well reasoned argument. I developed an assessment system for them similar to an MCAT or an LSAT: a pencil and paper exam that measures mathematics, reading comprehension, and critical thinking, combined with a critical reasoning essay. The organizational management were so happy with the first crop of hires that I practically got marriage offers.; 3/9/12, 11:20 AM
Eric Palmer, PeopleSoft Consultant said...: Cubanbob: the Duke ruling doesn't eliminate employers testing for skills and abilities, it simply requires a study that demonstrates a clear correlation between the skill or ability being measured, and success on the job. A test that is based on a demonstrated correlation is a valid test. The case law is very supportive of such examinations, provided the study adheres to established professional standards.; 3/9/12, 11:25 AM
Anonymous said...: class mobility in the US is better than almost any other country in the world. why is this even worth a discussion.

No, it isn't:

Several studies have been made comparing social mobility between developed countries. One such study (“Do Poor Children Become Poor Adults?") found that of nine developed countries, the United States and United Kingdom had the lowest intergenerational vertical social mobility with about half of the advantages of having a parent with a high income passed on to the next generation. The four countries with the lowest "intergenerational income elasticity", i.e. the highest social mobility, were Denmark, Norway, Finland, and Canada with less than 20% of advantages of having a high income parent passed on to their children.

Like I told Pogo on another thread. Just because you desperately want something to be true, doesn't make it so.; 3/9/12, 11:28 AM
cubanbob said...: Yideshe as a practical matter what large employer would gamble on this when suing for discrimination to do, the bar of proof for the plaintiff is low and the risk of a damage award for the employer so high? no hr manager ever got fired for playing it safe.; 3/9/12, 11:30 AM
Peter said...: " College Degrees to be a retail store manager or police officer for instance is about as stupid a thing as I've ever heard of. We need to emphasize skills and abilities again."

OK, so how do you do that without testing for the skills and abilities you want?

Whether the tests are objective or subjective, they're sure to have disparate impact. And disparate impact leaves one open to massive liability.

Therefore, tests remain the exception and not the rule. And they will remain so, so long as disparate impact hangs as a Sword of Damocles over anyone who dares to test.

Which leaves us with proxies for desired skills and abilities. College degrees may be low quality proxies, but what do you have that's better?; 3/9/12, 11:31 AM
Anonymous said...: There seems to be a root cultural aassumption that has taken hold that economic differences are bad, and that the situation will only be good when there are no significant numbers of people who are richer than anyone else.

Can you provide any evidence that this "root cultural assumption" represents even a significant minority view in this country?

Just because many on the right mus-characterize the concern about the lack of social mobility and concentration of wealth in this country, doesn't mean your assertion is true.; 3/9/12, 11:31 AM
Anonymous said...: Even a job in something like engineering would be better filled by an exhaustive test that could easily prove if you can do engineering, rather evaluating some institution's ability to fail the uneducated, which they rarely do.

You don't know the first thing about what it takes to get an engineering degree, do you? Grade inflation and the inability to fail the uneducated might be a problem in some disciplines, but you certainly won't find it in the hard sciences or engineering.; 3/9/12, 11:34 AM
cubanbob said...: freder what you told pogo is far more applicable to you.
now as for the countries you mentioned you forgot thevrest of the story; they are far smaller making an export driven economy easier, they are far more homongenous racially and culturally than the US. If Canda had the US mix it wouldn't be doing so great, if Norway had no oil it would what it was before, poor and if Denmark wasn't where it is with easy entre to the european markets it would't be doing so well.; 3/9/12, 11:37 AM
Anonymous said...: Even a job in something like engineering would be better filled by an exhaustive test that could easily prove if you can do engineering

Ever heard of the FE and the PE exam? (and btw, in some but not all states, practical experience can substitute for a formal engineering degree); 3/9/12, 11:39 AM
Anonymous said...: cubanbob--None of what you said changes the fact that Herb's original statement is simply not true. Class mobility in this country is not "better than almost any other country in the world."; 3/9/12, 11:42 AM
KCFleming said...: Freder, you were wrong in the other thread, too.; 3/9/12, 11:47 AM
Econophile said...: Freder said: "Just because you desperately want something to be true, doesn't make it so."

Income mobility just happens to be something I've studied a while and you're looking at some very misleading statistics just to make a point.

So let me modify your comment:
Just because you desperately want something to be true doesn't mean every "study" cited on Wikipedia that you think proves your point is done correctly. In other words, you must acknowledge your own confirmation bias.; 3/9/12, 11:50 AM
Eric Palmer, PeopleSoft Consultant said...: Cubanbob: every requirement for a job can be held to these standards. The same law applies to a degree requirement, a GED requirement, a requirement to have worked one year in a similar job, etc. No matter how you are screening, including interviews, whatever criteria you use to select IN and select OUT, is an assessment. And it is required to be based on a study demonstrating a clear nexus to the job. If you are an employer, and you are hiring someone, you are in this boat. The only difference is that these formal tests were hit early on. These other approaches will have their day.; 3/9/12, 11:52 AM
KCFleming said...: Don't confuse him with facts, Econophile.

He's got a message to sell.; 3/9/12, 11:52 AM
Balfegor said...: Re: Bagoh20:

No degree should be required for anyone who can pass something like the bar exam, or a similar test that is thorough, often including actual demonstration of ability.

That sounds nice and all, but you'll drown in disparate impact racial discrimination lawsuits. I don't think they've devised a police or firefighting exam yet that hasn't been challenged because of disparate impact. Frankly, I think it's a miracle Bar associations aren't getting sued under some disparate impact theory -- for first time takers in California Black passage rates are around 60% of White passage rates, which is well below the arbitrary 80% usually used for disparate impact lawsuits.; 3/9/12, 11:53 AM
cubanbob said...: Freder Frederson said...
cubanbob--None of what you said changes the fact that Herb's original statement is simply not true. Class mobility in this country is not "better than almost any other country in the world."

3/9/12 11:42 AM

Freder you have just proved that your argument holds no water. Most of us not being communist of one variety or another do not consider income disparities and class mobility to be problems to be solved by government unless those results are caused by government action or legislation.; 3/9/12, 11:58 AM
Carol said...: Murray restricted his latest book to a cmoparison only among whites. But his suggestions would increase the white/black gap even more, based on his own research on IQ in the Bell Curve.

Certifiable butts in seats through easy HS and college courses at least nominally qualifies the Oppressed for AA hiring.; 3/9/12, 11:59 AM
Revenant said...: Balfegor,

The reason the bar exam doesn't get nailed for disparate impact is that all the people ruling on the issue have person experience with the bar exam. They know it isn't racist.

It is only those OTHER tests, that they aren't familiar with, that must be racist.; 3/9/12, 12:01 PM
Anonymous said...: Income mobility just happens to be something I've studied a while and you're looking at some very misleading statistics just to make a point.

Do you have a link to a study that contradicts my claim and shows that the statistics are "misleading"? Excuse me if I don't take an anonymous commenter's word for it.; 3/9/12, 12:01 PM
Anonymous said...: Don't confuse him with facts

Like I said, provide the facts. All I see is unsupported assertions.; 3/9/12, 12:04 PM
cubanbob said...: Yiddishe Bloyger said...
Cubanbob: every requirement for a job can be held to these standards. The same law applies to a degree requirement, a GED requirement, a requirement to have worked one year in a similar job, etc. No matter how you are screening, including interviews, whatever criteria you use to select IN and select OUT, is an assessment. And it is required to be based on a study demonstrating a clear nexus to the job. If you are an employer, and you are hiring someone, you are in this boat. The only difference is that these formal tests were hit early on. These other approaches will have their day.

3/9/12 11:52 AM

Yiddeshe you are making my point: look at the hurdles and hassles an employer has to go through to try to avoid be sued. It's just easier to hire a college graduate. With enough thrust even a brick could be made to fly but you wouldn't want to be a passenger on that plane.; 3/9/12, 12:04 PM
Eric Palmer, PeopleSoft Consultant said...: There is a misunderstanding in a number of these posts about the difference, legally, between formal tests, and things like interviews and application requirements (a BA, etc). The misunderstanding is that one is a legal risk (formal tests), while the other is safe.

EVERY method that you use for selecting among applicants is a test. Period. And all tests are subject to the same basic standards. Every one of those approaches is required to meet the 80% rule for adverse impact, just like a formal test is.

The reason we have so much established case law concerning formal exams, but not interviews or a degree requirement, is mostly accidental: formal exams were targeted early because they are often used by employers hiring large numbers of people, and because they often apply to jobs that happen to be unionized. At one time those tests seemed completely safe, just like BA requirements seem safe now. And just like formal exams, the BA requirement will eventually get noticed, and then employment processes will start dropping like dominos.; 3/9/12, 12:06 PM
stressdoc said...: Freder - Please name even one state that still allows engineering registration without a degree. Furthermore, the vast majority of engineering graduates never register as PE's. Your assumption that all graduates from even accredited engineering schools are qualified to perform engineering duties is completely wrong. While it may be somewhat less pronounced in engineering than in some of the liberal arts areas, grade inflation is alive and well in many (certainly not all) engineering schools.; 3/9/12, 12:09 PM
Eric Palmer, PeopleSoft Consultant said...: Hiring a college grad because he or she has a college degree is only less risky than giving them an exam because the law hasn't paid much attention to it yet. But the same law applies in both cases.

And you can't just hire someone because they have a degree. Not generally. If you advertise your position, you could get 200 applications (or more). ANY method you use for selecting between them is a test.; 3/9/12, 12:10 PM
Hagar said...: I do not know what the above arguments have to do with the title heading, but as for hiring people, I have always felt that the people responsible for getting whatever the job is, done, should also be allowed to hire the people to do it.

The current system of avoiding responsibility by turning the hiring over to a "Human Resources Dept.", or some such euphemism, the personnel of which know nothing about the work to be done and so have to rely on resumes submitted listing "credentials" deemed valuable by other "credentialed" people, is worthless.; 3/9/12, 12:12 PM
Econophile said...: I really enjoy reading Murray, but when he says "... we should get rid of..." his premise is already bad.

What or whom is this "we" he refers to? There is no collective body making these decisions beyond private organizations and individuals themselves. Is he suggesting there should be?

Colleges and business will always require something in the spirit of the SAT and college degrees for obtaining information on large numbers of candidates. To say that these things do not convey useful information and so these organizations are miscalculating is one thing (if so, they'll figure that out); to suggest ending their existence is another.; 3/9/12, 12:13 PM
cubanbob said...: stressdoc said... what you said is absolutely correct, getting a degree and being able to pass are two very different things but fredder is invulnerable to the facts that don't agree with his view of how the world should be.

yidesshe perhaps one day, perhaps one day soon you point about a BA degree will be proven right. until then any employer, especially a large employer need but to look at their EPL policies to see what is and what isn't covered and have a talk with their labor lawyer. it's just easier and safer to go the BA route unless you do have a very specific need that only tests can weed out the the unqualified. however when the time comes that the BA becomes as much a problem as the exams then God help us as the outsourcing will be far worse than it is today.; 3/9/12, 12:18 PM
rcocean said...: Get rid of the law school monopoly. If you can pass the Bar exam - you get a law degree.

No need for anyone to go to "Law School" - Lawyers aren't like doctors. There are no super 'life or death' technical matters being decided.; 3/9/12, 12:18 PM
cubanbob said...: Hagar said...

unless you have been the defendant in such a lawsuit then you don't know just how much the law is stacked against an employer, the employers best option is usually a settlement to avoid a huge damage award. that's life in the big city today.; 3/9/12, 12:21 PM
Mark O said...: To hell with merit. Go with bloodlines and geography. Maybe first names.

Let anyone teach Con Law. Feelings are important. It's a living document, right.

Go back to the old slogan: The only power Duke Power will have is electric power.

Who needs freedom?; 3/9/12, 12:24 PM
Blue@9 said...: I really enjoy reading Murray, but when he says "... we should get rid of..." his premise is already bad.

What or whom is this "we" he refers to? There is no collective body making these decisions beyond private organizations and individuals themselves. Is he suggesting there should be?

^This.

These are fun little theories, but we may as well argue about angels dancing on the head of a pin.

Why is it these "solutions" always require some kind of central-command economy and bureaucracy to effect?

Murray now wants to outlaw private employers from requiring college degrees for certain positions... who decides which jobs should require degrees? Oh yeah, not the actual business owner, but a judge or bureaucrat in Washington.

What about banning unpaid internships... how do you even enforce that? They're already not on the payroll, so are you going to have internship police to make sure that no unpaid volunteers are at the workplace?; 3/9/12, 12:27 PM
Balfegor said...: Re: Revanant:

The reason the bar exam doesn't get nailed for disparate impact is that all the people ruling on the issue have person experience with the bar exam. They know it isn't racist.

Well sure. But they've also taken the bar exam themselves -- do they honestly think that it's got much to do with the actual practice of law?

Re: Yiddishe Bloyer:

The reason we have so much established case law concerning formal exams, but not interviews or a degree requirement, is mostly accidental

I think it's a little more than accidental -- if hiring is a black box without clearly articulated standards, it's hard to identify the practice that is clearly racist. It's the flip-side of the reason why outright racist quotas have been outlawed in college admissions, but mushy black-box "affirmative action" has been permitted, even though we all know it's employed to achieve the same result. An exam with a fixed cut-off is, in contrast, a pretty easy thing to get results and statistics for, to establish the prima facie case.

With respect to college degrees, you're absolutely right that they don't insulate employers from liability (in fact, I think Duke Power concerned high school degree requirements), but as you move up the income scale, I think courts have been more willing to accept that college degrees are a valid requirement. That said, I think most of the reason they haven't been challenged as heavily is that most people who would be losing out in the hiring process, and would therefore be bringing these suits against employers with college degree requirements, already have college degrees themselves, so it doesn't really matter. And once you're deciding between college graduates, it seems like it would be much harder to show that things like GPA or class rank don't matter, once the employer rebuts the prima facie case.; 3/9/12, 12:28 PM
Balfegor said...: Re: Mark O

To hell with merit. Go with bloodlines and geography. Maybe first names.

The obvious solution is nepotism.; 3/9/12, 12:30 PM
Peter said...: Surely Murray's least defensible assertion in Murray's NYT Op-Ed is where he wants to stop using the SAT- not because of its flaws, but because people perceive that it is biased?

If people's perception is faulty, why not work to correct that perception instead of just moving on to a different test (which will surely be subject to equally strong attacks)?

Murray says, "We can also drop the SAT in college admissions decisions. The test has become a symbol of new-upper-class privilege, as people assume (albeit wrongly) that high scores are purchased through the resources of private schools and expensive test preparation programs.; 3/9/12, 12:31 PM
CatherineM said...: I agree with the college degree for jobs where it's not needed. You need a BA to get a job as an exec asst in NYC - the did away with secretarial schools. Yet, still, most are not prepared for the job like they were 20 years ago coming out of Katharine Gibbs.

Additionally, I have friends with 20 years of experience working as Exec assts for Fortune 500 CEOs and are told by headhunters/HR people that they aren't qualified for the job they are MASTERS at because they don't have a BA (or don't get found when resumes are "searched" in recruting databases for applicants with BA).

It's absurd!; 3/9/12, 12:32 PM
Dan in Philly said...: The problem with the higher education bubble group is that there is currently no viable alternative for employers to ensure their employees are able to be good employees. If you are not able to put yourself through college, you will likely not make a good employee in a white collar job, even if technically your job doesn't call for a college education.

So the ones who say the college degree is just a screening device are certainly correct, and alas there's really no better one we have yet devised.; 3/9/12, 12:44 PM
cubanbob said...: Balfegor said.. when it comes to black boxes coupled with disparate outcomes the formula equals get ready to write a big check and pray the insurance policy language is sufficiently vague that the ambiguity favors the policyholder because God is definitely not on your side in such a case.; 3/9/12, 12:48 PM
Hagar said...: The reason for requiring a college degree as one of the prerequisites for registation as a Professional Engineer, is that the school is supposed to have taught you at least a smattering of knowledge across the entire field of your branch of engineering. Thus, when you start working, you are supposed to be able to tell when you are getting in over your head, or into unfamiliar territory, and get assistance from someone familiar with that kind of work. You had better, because you will be held responsible for doing so and can be sued for damages and lose your license to practice if there is a failure connected to your work.; 3/9/12, 12:52 PM
Balfegor said...: Re: cubanbob:

when it comes to black boxes coupled with disparate outcomes the formula equals get ready to write a big check and pray the insurance policy language is sufficiently vague that the ambiguity favors the policyholder because God is definitely not on your side in such a case.

Really? You look at white collar jobs, the hiring process is in many cases almost totally opaque -- it's all subjective interviews -- and there's huge and obvious racial disparities in employment. Maybe that's because the source pool of interested persons also reflects those same racial disparities, but I'd honestly be kind of surprised if that were the case.; 3/9/12, 12:53 PM
Anonymous said...: Please name even one state that still allows engineering registration without a degree.

California; 3/9/12, 12:56 PM
cubanbob said...: Balfegor said...
Re: cubanbob:

when it comes to black boxes coupled with disparate outcomes the formula equals get ready to write a big check and pray the insurance policy language is sufficiently vague that the ambiguity favors the policyholder because God is definitely not on your side in such a case.

Really? You look at white collar jobs, the hiring process is in many cases almost totally opaque -- it's all subjective interviews -- and there's huge and obvious racial disparities in employment. Maybe that's because the source pool of interested persons also reflects those same racial disparities, but I'd honestly be kind of surprised if that were the case.

3/9/12 12:53 PM

First you have to pass the post: how many applications are even considered if a degree isn't in the resume? Now for large companies that have exposure you seriously believe that mix of hires doesn't have a rough correlation to the qualified applications received and the interviews don't reflect that?; 3/9/12, 1:08 PM
Balfegor said...: Re: cubanbob:

First you have to pass the post: how many applications are even considered if a degree isn't in the resume? Now for large companies that have exposure you seriously believe that mix of hires doesn't have a rough correlation to the qualified applications received and the interviews don't reflect that?

Isn't the whole problem of disparate impact how you define "qualified applications?" I'm sure the mix of hires roughly matches how the company has defined "qualified applications," -- much less certain that the mix of qualified applications matches the mix of applicants.; 3/9/12, 1:11 PM
Hagar said...: I am surprised that California still allows that, and I think it must be due to the heavily unionized California State government.

However, California has, or at least used to have, one of the stiffest EIT exams in the country, recognized by all other states, except New York, also unionized, and I believe I have heard that their P.E. exam is quite rigorous.
So I think that the practice in California must be to also take a look to see just how your P.E. registration was obtained.

(How come I can't edit in this new format?); 3/9/12, 1:15 PM
Balfegor said...: Sorry, I'm mixing and matching interviews and the resume filtration process. Interviews are total black box, and the one I was talking about, and I expect there's some disparity there, but it may be defensibly small. I'd still be surprised if there's proportionately so few African-Americans with "qualified" applciations for some positions, as the number I see who have gone through into white-collar positions is extremely low.

Resume filtration is a different black box (other than obvious things like "college degree in relevant major").; 3/9/12, 1:15 PM
cubanbob said...: Balfegor said...
Re: cubanbob:

First you have to pass the post: how many applications are even considered if a degree isn't in the resume? Now for large companies that have exposure you seriously believe that mix of hires doesn't have a rough correlation to the qualified applications received and the interviews don't reflect that?

Isn't the whole problem of disparate impact how you define "qualified applications?" I'm sure the mix of hires roughly matches how the company has defined "qualified applications," -- much less certain that the mix of qualified applications matches the mix of applicants.

3/9/12 1:11 PM

Like Stalin said its not the vote that counts but who counts the votes. In employment cases the HR dept. consults with the legal consultant whose advise is play it safe because the ultimate decision is the on whose application is good is the plaintiff's lawyer and the law is on his side, so unless the applicants mix is so terribly awful the SOP is go the safe way to go. That is how the system has been set.; 3/9/12, 1:21 PM
cubanbob said...: Balfegor said...
Sorry, I'm mixing and matching interviews and the resume filtration process. Interviews are total black box, and the one I was talking about, and I expect there's some disparity there, but it may be defensibly small. I'd still be surprised if there's proportionately so few African-Americans with "qualified" applciations for some positions, as the number I see who have gone through into white-collar positions is extremely low.

Resume filtration is a different black box (other than obvious things like "college degree in relevant major").

3/9/12 1:15 PM

It depends on the field you are in. Without a back and forth about racial preferences and abilities or cultural inclinations for different fields if you are interviewing for a position that for whatever reason attracts very few minority applicants then you have a point about disparate outcomes. If not, then you have a problem if your hires aren't roughly proportional and you are already on the labor lawyer network radar unless you have massive documentation to back you up (which itself is rather costly). A recent famous case was the CT fireman case about disparate outcomes. Lightning isn't harmful to you until you get zapped but it would foolish to stand out in the rain with a wire tied to a kite.; 3/9/12, 1:29 PM
Geoff Matthews said...: The problem with doing away with the SAT is that it is the best predictor of college performance and success. Couple this with grade inflation in high school, and I'd like to see some objective tests around.; 3/9/12, 1:34 PM
Old RPM Daddy said...: I agree with CatherineM and others: Why require a degree when the job doesn't require that level of education. But it seems misguided to force the issue somehow. Better that individual firms have the guts to quit asking for it, and find another way to sort through candidates. I suppose it would be a more difficult process, but might save money in the long run, as non-degreed hires might accept lower pay (having presumably less debt to service).

As it is, I don't think anyone would argue that the value of a college degree is decreasing substantially in this country. We're on our way to becoming a country like the Philippines, my wife's homeland. Visiting a mall a few years back, my wife told me that most of the checkout girls had college degrees. People with only high school diplomas were out sweeping the streets.; 3/9/12, 1:34 PM
Luke Lea said...: But the biggest problem Murray identified in his new book is that of working-class families falling apart. He doesn't address that issue.

As for affirmative action, as a native son of the South I'd prefer to see it extended to cover all the major ethnic groups and geographical regions of the country, including rural and small-town areas as well as the big cities. Our multi-cultural democracy needs a governing elite that is representative of the ethnic and geographic diversity of America.

There are more than enough talented high school seniors out there to stock the the Ivy League. You'd have student bodies composed of the best and the brightest members of every population group in the country (including the absolutely brightest I might add) instead of just the best and brightest minorities. Much healthier.; 3/9/12, 1:45 PM
jimbino said...: Murray is so right.

Seeing that:

California allows professional engineers to practice without state certification;

Wisconsin allows lawyers to practice without taking the bar exam;

Wisconsin and New Hampshire allow drivers to drive without liability insurance;

Bill Gates, Michael Dell, Steve Jobs and Mark Zuckerberg became billionaires without finishing college;

Isn't it about time to institute Milton Friedman's recommendation to put an end to all certification?; 3/9/12, 1:54 PM
Bruce Hayden said...: The unpaid internships operate to provide signaling, which is, to some extent, anti-democratic. Nevertheless, it is something that is going to happen regardless. All you have to do is look at where someone went to undergraduate and high school, and you have most of the information required.

And, even paid work in the area of interest may provide signaling. I know of several undergrads whose parents got them summer lab jobs. And, others who got good summer jobs through family connections in other areas.

Also, let me give a shout out for REUs (Research Experiences for Undergraduates), available in a lot of STEM areas. Mostly paid for by the NSF, but there are some funded by NASA too. The undergraduates get to do research in cutting edge areas during their summers, and they actually get paid for it (though some of that pay sometimes has to go for living expenses). For a STEM undergraduate, I would think that these REUs would be more useful than an unpaid internship, esp. when applying to graduate school. They are limited though, so are competitive.; 3/9/12, 2:14 PM
Alex said...: But the biggest problem Murray identified in his new book is that of working-class families falling apart. He doesn't address that issue.

Show me a time when working class families weren't falling apart. I'm sure there is some golden period right?; 3/9/12, 2:22 PM
bagoh20 said...: "Let's experiment with airline pilot jobs first."

As a licensed pilot, I got no problem with a pilot that is qualified in every other way but has no college degree. What possible help would an undergraduate or graduate degree be as compared to those years getting experience flying instead? Answer: College would make the average 30 year old a less qualified pilot. And that same analysis applies to nearly every occupation except becoming a college professor.

Put simply, college is a very inefficient path to learn most or occupations when compared to a more direct alternative. That we blindly accept it as something more is foolish, and most people don't internally in their own mind about their own profession.; 3/9/12, 2:46 PM
bagoh20 said...: And the reason we have all this disparate impact legal bullshit in the first place is because lawyers spend too much time in school and don't have real work to do. You don't see the carpenters getting into the electricians' business and mucking it up. No, because they have work to get done - building to do.

Law should be considered a subset of Demolition Studies and lawyers renamed Dream Crushers. But, I kid...a little.; 3/9/12, 2:59 PM
Joe said...: My local university (which used to be a community college) has an open admissions policy. If you have a high school diploma or a GED (or are a high school student and want to to concurrent enrollment) they will admit you. If you haven't take the SAT or ACT, they have a test you can take. It determines whether you need to take remedial courses.

Not requiring college degrees for many jobs would be a simple first start, though there are already several that don't do this. Many require certificates and where I work, if you know C++, we don't care if you ever went to school at all.

(I've never been asked about my college degree except out of curiosity--my degree is in film, I'm a software engineer. More than half the software engineers I know don't have computer science degrees. Most have some sort of engineering, but several have no degree or BAs in some silly major like me.); 3/9/12, 3:05 PM
Joe said...: Alex, June 3, 1956 at 10 am.; 3/9/12, 3:06 PM
Blue@9 said...: Okay everyone, here's the solution to your social engineering problem:

1) Start your own business.
2) Don't require college degrees when hiring.; 3/9/12, 3:34 PM
Chuck66 said...: I like the idea of ending requirement of a college degree that is just on-the-job training. Right now, its either:

A) You spend a lot of money getting a 4 year degree that does little or nothing for your career. Why not just start working at age 18.

B) If you didn't go to college, you are eliminated from many jobs that have no need to hire college graduates.; 3/9/12, 4:08 PM
Chuck66 said...: This right winger could agree to some things....such as capping the mortgage interest deduction. I have no problem with a wealthy person buying a very expensive house. But I could easily live with capping his deduction at a lower amount. It is not taking money from a successful person. It instead it eliminates part of a subsidy to purchase an expensive house.; 3/9/12, 4:13 PM
Unknown said...: And make credit reports just for obtaining credit only. Nothing else.
And make the internet truely anonymous and owned by the indiviual nics like Jay and Carol Herman; 3/9/12, 4:30 PM
Unknown said...: I once applied for a job at a McDonalds and the manager wanted a college degreed person. I doubt that is a company policy but...

I interpreted as a selective excuse not to hire me, he just raised the bar when it was convenient to do so.; 3/9/12, 5:54 PM
Unknown said...: What someone said here.

A well written business letter to a bank that negotiates the solution to a problem. That is worth a 4 year college degree. Do you need college to learn that?; 3/9/12, 5:59 PM
Unknown said...: cubanbob

Most of us not being communist of one variety or another do not consider income disparities and class mobility to be problems to be solved by government unless those results are caused by government action or legislation.

Probably 100 percent of your fortune 500 compannes and 100 percent of the large local governments and school systems are now following racial justice programs. They all retain industry consultants. All the industry consultants have a strong class oppressor/victim background that comes right out of Russia. It's been coated with a lot of honey to be digestible and addictive to people pleasing leaders. But it is still unproductive crap.

But tell that to a leader and he will be shocked, shocked I tell you and then pursue a new justice law to get you some shock treatment also.

Less government.; 3/9/12, 6:05 PM
wild chicken said...: "Show me a time when working class families weren't falling apart."

Maybe 5 years ago, when couples fought over who would get the house.; 3/9/12, 6:10 PM
Seven Machos said...: The problem with doing away with the SAT is that it is the best predictor of college performance and success.

You have absolutely no idea what you are talking about. You are completely and totally full of shit. You are making things up.

The SAT is not a predictor of college success, and makes no claims to be. And it can't because multiple studies have shown that the SAT is not a predictor of performance and success.

The SAT predicts how rich you are quite well, and nothing else.; 3/9/12, 6:55 PM
Blue@9 said...: The SAT is not a predictor of college success, and makes no claims to be. And it can't because multiple studies have shown that the SAT is not a predictor of performance and success.

Except for the studies that do show it. Go ahead, google "SAT predict college success." Results are all over the place.; 3/9/12, 7:11 PM
Blue@9 said...: The SAT predicts how rich you are quite well, and nothing else.

I actually did quite well on the SAT and was dirt poor at the time. I know, one data point...; 3/9/12, 7:12 PM
Cincinnatus said...: stressdoc, only a small minority of engineering work requires a PE cert.; 3/9/12, 7:15 PM
Seven Machos said...: Blue -- Show me the studies that say that the SAT correlates with college performance and success. After all, if they are all over the Internet, they must be easy to find.

Show me further where the creators of the SAT make such claims that you are assigning to their product.

Tell us also what SAT stands for.

I'll wait.; 3/9/12, 7:16 PM
Blue@9 said...: Blue -- Show me the studies that say that the SAT correlates with college performance and success. After all, if they are all over the Internet, they must be easy to find.

Yes they are. I even told you the google search. You're on a computer. If you have any intellectual integrity you'll acknowledge what I've told you.

Show me further where the creators of the SAT make such claims that you are assigning to their product.

I didn't argue this point.

Tell us also what SAT stands for.

Now you're just being dumb, and pretending like it's clever.; 3/9/12, 7:21 PM
Blue@9 said...: btw, when I said "Results are all over the place," I should have clarified that I meant the studies go both ways when it comes to establishing any correlation.

I actually have no clue it does, but I wanted to point out that it's not really a settled debate.; 3/9/12, 7:25 PM
Seven Machos said...: Blue -- The SAT itself will not make the claims about its own product that you are making for it. Its creators cannot, because the data don't support it.

What the data supports is a modest correlation between success in the first year of college and scores. Beyond that -- beyond survey courses that are very often fill-in-the-bubble tests just like the SAT -- there is nothing.

I know a great, great deal about this topic. It is my area of expertise in the world. It is very likely not yours, and I find it pointless to debate with people on the topic because they don't know what they are talking about. They spout bullshit, like Geoff up there.

The SAT is the Wizard of Oz. It's a bad test. It used to be a worse test. You need only look at how much it has morphed itself into what the ACT is to understand how poorly the SAT has performed in its reason for existence.; 3/9/12, 7:28 PM
Blue@9 said...: Blue -- The SAT itself will not make the claims about its own product that you are making for it.

It doesn't matter what the SAT claims. And I'm not making any claims about the SAT test-- are you even reading my posts?

What the data supports is a modest correlation between success in the first year of college and scores.

Okay, but right up above you say this:

The SAT is not a predictor of college success, and makes no claims to be. And it can't because multiple studies have shown that the SAT is not a predictor of performance and success.

The SAT predicts how rich you are quite well, and nothing else.

Your statements are rather contradictory.

I know a great, great deal about this topic. It is my area of expertise in the world. It is very likely not yours, and I find it pointless to debate with people on the topic because they don't know what they are talking about.

This is just a "Respect my Authority!" argument. Considering that you just contradicted yourself, I'm dubious about your credentials.; 3/9/12, 7:36 PM
Seven Machos said...: Blue -- College performance the first year is not college performance. The implication, which I am sorry to have to spell out to you, is that there is not a correlation beyond the first year, and so the correlation is pointless.

As far as argument from authority, yeah, that's it. I know more than you do about this topic. I know the SAT. I know what's on it, and what used to be on it. I know the history of test, and the body of research.

Finally, I think it's objectively true to say that you are making claims about the SAT when you say that studies show something about it. Were those claims actually about poodles in microwave ovens? The way that advertisers use sex to sell except dating sites, who strangely don't?; 3/9/12, 8:03 PM
Blue@9 said...: Blue -- College performance the first year is not college performance. The implication, which I am sorry to have to spell out to you, is that there is not a correlation beyond the first year, and so the correlation is pointless.

Good grief, you're getting more ridiculous with each post. In a four year program, one year is 25% of your college success. It's like saying "Yeah, this tweak improves your performance in the first 50 laps of the Indy 500, but that's only a quarter of the race, so really there's no performance improvement." The only way your statement could make sense is if those high-SAT, fast-out-the-gate freshman deteriorate and do worse in the following three years.

First year is usually the toughest for students too. I really should have no argument with you, because as I noted, I have no particular stance as to whether the SAT correlates with college success. However, your contradictory statements are starting to make me think there is something to it.

As far as argument from authority, yeah, that's it. I know more than you do about this topic. I know the SAT. I know what's on it, and what used to be on it. I know the history of test, and the body of research.

That's great. Have a cookie. Next time don't sink your own credibility by making absolute statements and then backtracking.

Finally, I think it's objectively true to say that you are making claims about the SAT when you say that studies show something about it.

Now you're being a dumbass and thinking that you're clever.

You: SAT correlates to NOTHING but wealth and the studies show it!

Me: Wait, a quick internet search seems to indicate that it's not settled. Some studies agree with you and others don't.

You: Oh yeah? Why are you defending the SAT? Look at my huge tail feathers!

I know the SAT.

I'm the most interesting man in the world and I drink Dos Equis.; 3/9/12, 8:16 PM
Revenant said...: Seven,

The reason SAT scores weakly predict college success is that colleges filter out the low scores before even letting the students *in*.

Take UCSD for example -- a solid state university here in San Diego, but by no means an Ivy. The 25th percentile (the middle of the lowest-scoring half) of freshmen had composite scores of 1710. The 75 percentile was 2080. These correspond to national percentile rankings of 73 and 94.

I'm sure there is only a weak correlation between SAT and success among those freshmen, for the same reason I would expect a weak correlation between high school GPA and college performance among a group of students whose GPAs ranged from 3.7 to 3.9. Add in the C, D, and F students -- then you'll see a strong correlation. :); 3/9/12, 8:50 PM
david7134 said...: Basically the SAT and similar exams are a joke. They can easily be gamed. Someone that makes a mediocre scores can take a course and score high on the second round. The course only gives you the game rules and tricks. The essays are a joke. Again it requires gaming.

The Chinese started this about 2000 years ago and the sum result was pathetic bureaucracy, like now. With real smarts being passed over.

I never could take the exams in the 60's. Finished high school with a C. The next year in college, I was on the Dean's list and entered medial school in 3 years. Then got an MD with honor. From there, very successful.

The fact is that now I would only get a job as a ditch digger.

Now, college is too much overemphasized. People can make a better living without it.; 3/9/12, 9:19 PM
Penny said...: Charles Murray wrote a couple of books that seem to get people talking about shit.

Come on!

It's not like he invented the internet.

Al Gore is going to be REALLY pissed!; 3/9/12, 10:11 PM
Seven Machos said...: Blue -- It's touching that find me the most interesting man in the world because I know a ton about the SAT. I wish more women felt the same way.

It's also touching that you say that you aren't really that interested in the topic, but that doesn't keep you from spouting your opinions that you formed after a cursory Google search, or chastising me for brushing off your obvious lack of knowledge.

As for your analogy to car races, it's utterly flawed. To begin with, the SAT doesn't improve anything. It's a test to used for admissions, so you would need to focus on qualifying for car races for your analogy to have any hope. Car race qualification is very standardized and involves speeding around a track, which is exactly what you do in a car race. The SAT is a qualification that does not test anything that you will be doing in college. You will not be doing any of the remedial math on the SAT in college. You will not be reading little stories and answering questions about them in college. You will not be filling in missing words in sentences in college. You will be writing essays, but admissions staffers in college care the least by a country mile about the essay score on the SAT.

The SAT was until recently purposefully not academic in nature; it involved nothing that anyone learned past eighth grade. (Now, a small part of the test is geared toward some slightly higher level math, and it would be intelligent to ask yourself why the creators made these changes.) The SAT does not test what people learned academically, or their ability to learn in college. This is widely accepted.

Now, tell me again how you don't really care, but say it in eight paragraphs. Include a horrific analogy.; 3/9/12, 10:32 PM
Seven Machos said...: Rev -- Your points are stronger than Blue's, but it begs the question: why have the test? When you get to the scores these kids getting into remotely elite colleges have, you are talking about missing a few questions or getting them right.

The difference is not brilliance or not. If you are brilliant, you will get a high score on the SAT because brilliant people tend to do well on tests of remedial math and reading. And if you are brilliant, you are only going to UCSD by choice or for financial reasons.

The difference between a 1200 and a 1300 is for the non-brilliant is learning how to plug in numbers for letters on math questions and simply learning how to spend your time wisely. That's what any SAT teacher will drill into your head.

Had Murray dropped the "albeit wrongly" from his piece, he would be exactly right.

The ACT, incidentally, is harder to coach because it is more like a comprehensive school exam. It's not a comprehensive school exam -- not at all -- but it more closely resembles it. It's a better test.; 3/9/12, 10:39 PM
Penny said...: "Rev -- Your points are stronger than Blue's, but it begs the question: why have the test?"

You asked this question seriously, so I will answer it seriously.

We are a "discriminating" audience, provided we have "skin" in the outcome of the game.; 3/9/12, 11:16 PM
Seven Machos said...: Penny -- Here is the argument that people make about the SAT.

1. It is good to have an objectively-graded and meaningful, substantive test that the same everyone who wants to get into college.

2. The SAT purports to be such a test.

3. People believe that the SAT is such a test.

4. Therefore, the SAT is good.

The first premise is absolutely riddled with factual and conceptual errors, thus rendering the rest of the rest of the premises a farce. And that's a big social problem.; 3/9/12, 11:24 PM
Seven Machos said...: Penny -- You make the same error that everyone else makes: you assume that the SAT is a meaningful, substantive test.

It's not. It's a piece of shit. Shit is shit, despite its ubiquity.; 3/9/12, 11:36 PM
Chip S. said...: This research established the relationship between SAT and g, as well as the appropriateness of the SAT as a measure of g, and examined the SAT as a premorbid measure of intelligence. In Study 1, we used the National Longitudinal Survey of Youth 1979. Measures of g were extracted from the Armed Services Vocational Aptitude Battery and correlated with SAT scores of 917 participants. The resulting correlation was .82 (.86 corrected for nonlinearity). Study 2 investigated the correlation between revised and recentered SAT scores and scores on the Raven's Advanced Progressive Matrices among 104 undergraduates. The resulting correlation was .483 (.72 corrected for restricted range). These studies indicate that the SAT is mainly a test of g.

--Meredith C. Frey and Douglas K. Detterman, "Scholastic Assessment or g?The Relationship between the Scholastic Assessment Test and General Cognitive Ability." Psychological Science, June 2004.; 3/10/12, 12:08 AM
Seven Machos said...: Chip -- Are you really suggesting from a cursory search of the internet that the SAT is a test of intelligence? Really? Nobody, and I mean nobody has taken such a claim seriously for decades. Psychometrics is a disgusting fringe science with a terrible racist past. Don't embarrass yourself.

Further, the SAT now is much, much different than the SAT was in 2004, and it was much different then than in 1994, or 1984. No serious researcher would discount these differences.

This conversation will soon drift into IQ. It's a holy grail among a small segment of academics, and it's entirely too commonly bandied about. So let's get this out there: there is no such thing as IQ. There is no number that says how smart you are. It is simply bullshit.; 3/10/12, 12:31 AM
Seven Machos said...: Chip -- Are you really suggesting from a cursory search of the internet that the SAT is a test of intelligence? Really? Nobody, and I mean nobody has taken such a claim seriously for decades. Psychometrics is a disgusting fringe science with a terrible racist past. Don't embarrass yourself.

Further, the SAT now is much, much different than the SAT was in 2004, and it was much different then than in 1994, or 1984. No serious researcher would discount these differences.

This conversation will soon drift into IQ. It's a holy grail among a small segment of academics, and it's entirely too commonly bandied about. So let's get this out there: there is no such thing as IQ. There is no number that says how smart you are. It is simply bullshit.; 3/10/12, 12:34 AM
Chip S. said...: There are two very different propositions being conflated here. One is whether the SAT is a good predictor of "success" in college. The other is whether the SAT is a good measure of anything at all, except parental income.

My prior comment provided some evidence that the SAT is not a meaningless test. And it clearly refutes your claim that "nobody" takes seriously the argument that the SAT measures general intelligence. If you've got links to the innumerable studies you keep alluding to, by all means provide them. I will simply point out that among the articles listed at my link that cited the paper in question, none were criticisms of it.

Look, that study finds a very high correlation b/w the SAT and the Armed Forces test. The latter test is used by the military precisely b/c it does a good job in measuring suitability for various jobs. It's widely--very widely--used and accepted as a good measure of innate ability. That's a fact.

Any test-taker's intelligence is rather highly correlated with his parents' intelligence, which in turn is correlated with the parents' income. So it's not terribly surprising that SAT scores are correlated with parents' income.

As for the SAT's predictive power for "college success," it's strange to discuss this topic without specifying clearly how "college success" is measured.

It can't simply be measured by a four-year grade-point average, since there are well-known systematic differences across disciplines in grade distributions.

Grades during the first year of college are a better measure because that's the time when there's the greatest degree of comparability across the courses students take. It's still not perfect, but it's better by far than junior or senior year, when English majors are doing their thing and Physics majors are doing something entirely different.

But let's stipulate that SAT could only partially predict "true" college performance if that were capable of being measured.

If, as seems likely, college success depends on both intelligence and effort, then even a perfect measure of intelligence can't perfectly predict success unless intelligence and effort are perfectly correlated. And if they're not perfectly correlated, then once a college has selected students on the basis of intelligence (as indicated by SAT scores), the differences in student performance will be due to the residual variation among them, particularly unobservable differences in effort or motivation.

It's pointless to indict the SAT for not measuring effort or motivation if it's primarily designed to measure intelligence. And it's baseless to claim that the test doesn't measure intelligence because it only weakly predicts college grades.; 3/10/12, 12:46 AM
Chip S. said...: Correction: One of the papers citing the article I quoted was a criticism--not of the reported correlations with the Armed Forces test, but of the authors' attempt to convert SAT scores into IQ scores, which I considered to be an extraneous part of their findings.; 3/10/12, 12:55 AM
Seven Machos said...: When did effort come into anything? The SAT sucks because it doesn't measure anything. The strongest correlation is wealth, but it doesn't measure wealth, and to suggest so is ridiculous. There is a modest correlation between the SAT and success as a first-year college student, but not anything beyond that.

The SAT tests how well you take the SAT. That's it. Which would be great if the SAT was some half-decent test. But the SAT is a shit test. Therefore, it produces shit results that can be gamed.

Finally, your mention of the military is a wonderful irony. The military is the last bastion of psychometrics. You should ask yourself why that is. Is it perhaps because the military must quickly and efficiently sort people constantly? And is it because psychometrics provides a very crude way to do that necessary thing?

Now, you should ask yourself: is our education system so much like the military that we must resort to similar, crude testing devices? I can understand that your answer is yes, in the name of efficiency. But the collective endowments of our great universities are more than the GDP of many countries and they with far fewer people. We can and should do better.

Finally, the thing you are utterly avoiding is that the entire premise of anything involving g is IQ, and IQ is bullshit. It doesn't exist. It is a fable.; 3/10/12, 1:01 AM
Chip S. said...: Now, you should ask yourself: is our education system so much like the military that we must resort to similar, crude testing devices?

Nowhere in what I've written will you find any support for the sole reliance on SAT scores for college admissions. That would be ludicrous. In fact, the most detailed study of college performance I've seen (one that controls pretty well for differences in courses taken) suggests that class rank works about as well as SAT scores in predicting grades.

This thread was originally about the advisability of screening job applicants on the basis of college credentials. The discussion naturally turned to the use of aptitude tests rather than sheepskin requirements. So the relevant comparison here is the predictive power of aptitude tests vs. college degrees for performance on various jobs. Both will be imperfect; that's not news to anybody. And even if a BA adds some predictive power to test results alone, is that added predictive power worth the cost of the degree? (By the way, Armed Forces test scores are used all the time by economists to predict individuals' earnings in civilian occupations. They are not simply some random sorting device that the military finds convenient.)

I agree completely that there's a history of abuse of intelligence testing by eugenics advocates, and it's crucial not to commit the fallacy of equating "g" with "fitness" or "worth" or anything else aside from the ability to perform certain classes of tasks. I share your revulsion at the potential misuse of "g" measurement. But it strikes me as silly on its face to deny that people are differentially capable of learning advanced mathematics or other intellectually challenging subjects.; 3/10/12, 1:23 AM
Seven Machos said...: I'm not saying that the SAT is imperfect. Lots of things are imperfect. We are humans. That's how it goes. I'm saying that the SAT is an utter piece of shit. It's so many degrees below imperfect that it makes imperfect look perfect.

Secondly, I have issued the same challenge at Althouse many times that I will issue you: what's on the SAT? How is it different than it was 10 years ago? Twenty years ago? You don't know, and I know you don't know. But for reasons that can only be foolish becausee they arise from ignorance and blind trust in a billion-dollar test-making company, you accept the test and its outcomes.

Finally, it never ceases to amaze me how people apparently cannot get their heads around the fact that people can be smarter or dumber than each other either in certain areas or across the board, yet IQ doesn't exist. Of course it's silly on its face to deny that people are differentially capable of learning advanced mathematics or other intellectually challenging subjects. But, dig it, man: the SAT won't answer this question, nor will any similarly superficial standardized test. And there is no number of how smart you are. You are just smart, in some ways but probably not all ways.; 3/10/12, 1:35 AM
Chip S. said...: We're left with this: You believe that your time at the CEEB gives you a unique insight into the Big Test. That may well be so. But you haven't laid claim to any inside info about the Armed Forces Qualification Test, and that correlated quite well with the SAT in most of the latter's incarnations. So I don't think I need to know every nuance of every iteration of the SAT to form a judgment about the overall utility of aptitude testing.

When people here refer to the "SAT" they probably aren't thinking of its current form, but rather the form in which they themselves took it. For a lot of Althousians, that may have been a long, long time ago, when the thing did more closely resemble an IQ test than a general-knowledge test. So where you see them as citing a very precise (and flawed) test, I see them as making a more general reference to aptitude tests in general. If you scroll up to the top of this thread, you'll see that I chose the Wonderlic test as an example of such tests, simply b/c of its use by a wide range of employers, most famously the NFL.

The ultimate idea behind "g" or "IQ" is that abstract thought has distinct properties that are independent of the specific subjects being thought about. To the extent that tasks differ in the degree of abstract thought that they demand, and to the extent that these differences across tasks can be discerned, "g" scores will be useful in predicting success at a variety of tasks. Those predictions will always be imperfect b/c there are other factors besides innate ability that determine success.; 3/10/12, 2:02 AM
Seven Machos said...: But you haven't laid claim to any inside info about the Armed Forces Qualification Test

I don't know the first Goddamn thing about the Armed Forces Qualification Test (and I certainly have spent no time with the College Board).

What I do know a lot about is the SAT, and I'm telling you that it is shit. It was shit and it remains shit. I also find it interesting that you managed to elide over the fact that you so obviously don't know dick about what's on the test now or what was on it before. I don't pontificate about things I don't know about. But that's just me.

Concering abstract thoughts, every single one of the changes to the SAT has been away, sometimes radically, from the kind of goofy abstract questions of the past. The reason its creators moved away from those things is to make the test harder to coach.

Now, Chip, you gotta ask yourself: how is it that there can be some innate intelligence number inside people that can be tested, yet it also can be that testing exactly that thing is much easier to game? How it can be that analogies and antonyms and stupid math questions involving x, y, and z can be both a test of your innate ability and ridiculously easy to manipuate with coaching techniques? And if those questions were so great, why do we now see essays and functions, which involve real mathematical skill, and have nothing to do with abstract ability?; 3/10/12, 2:17 AM
Seven Machos said...: But you haven't laid claim to any inside info about the Armed Forces Qualification Test

I don't know the first Goddamn thing about the Armed Forces Qualification Test (and I certainly have spent no time with the College Board).

What I do know a lot about is the SAT, and I know it is shit. It was shit and it remains shit. I find it interesting that you managed to elide over the fact that you so obviously don't know what's on the test now or what was on it before. I don't pontificate about things I don't know about. But that's just me.

Concering abstract thoughts, every single one of the changes to the SAT has been a move away, sometimes radically, from the kind of goofy abstract questions of the past. The reason its creators moved away from those things is to make the test harder to coach.

Now, Chip, you gotta ask yourself: how is it that there can be some innate intelligence number inside people that can be tested, yet it also can be that testing exactly that thing is much easier to game? How can it be that analogies and antonyms and stupid math questions involving x, y, and z can be both a test of your innate ability and ridiculously easy to manipuate with coaching techniques? And if those questions were so great, why are they gone? Why do we now see essays, which aren't abstract at all, and functions, which involve real mathematical skill and have nothing to do with abstract ability?; 3/10/12, 2:25 AM
Blue@9 said...: Rev -- Your points are stronger than Blue's

Really, which ones? The one where I point out you're a sloppy thinker and writer when you say the SAT doesn't correlate to anything other than wealth--and then later backtrack, admitting that it has a modest correlation to first year college success? Or the one where I showed you're full of shit when you said no study has found any such correlation? If you want to argue that those studies were flawed and/or superseded by subsequent research, be my guest, but don't flat out lie.

What is terribly sad is that you're a clueless buffoon. You may know some facts, but you beclown yourself by lying in order to bolster your arguments--lies that are easily uncovered by a simple Internet search. You probably do have a much better understanding of the SAT than anyone here, but you've destroyed your credibility by being such a dumbass.

but it begs the question: why have the test?

Because colleges continue to give it weight. Because they'd rather use something that exists and has mindshare than develop something more meaningful. Because they'd rather not spend the money to develop a new test or quadruple the admissions staff to delve really deeply into applications.

Any other questions?; 3/10/12, 2:46 AM
Revenant said...: Rev -- Your points are stronger than Blue's, but it begs the question: why have the test?

To identify overall academic and intellectual ability, basically. In theory high school GPA could be used for that, but in practice high school GPA is wildly inflated. A 4.0 doesn't mean what it used to.; 3/10/12, 2:48 AM
Revenant said...: And if you are brilliant, you are only going to UCSD by choice or for financial reasons.

I don't know what "brilliant" means in this context -- but like I pointed out, UCSD students scored far above average on the SAT. A quarter of them scored in the top 6% nationally.

On the other hand, if you only got a 1200 or 1300... maybe community college is a better choice. If you can't handle simple math and vocabulary problems, how are you expecting to survive in an event marginally challenging university?; 3/10/12, 2:53 AM
Seven Machos said...: Rev -- We can do better and we should do better. The ACT has been a good start, and the SAT slowly morphing into the ACT is cause for optimism. But both tests are far too narrow, and they are poor indicators of anything useful.

Money is not the issue. Making tests is not expensive and it's a volume business. The libertarian in you should favor a market of tests for different people with different skills, instead of a duopoly of shit and near-shit.

Blue -- I had a post responding to some of your comments, but Bloggers ate it. I decided not to bother because you just aren't worth it. You don't know anything about the test you are defending. The fact that your argument has now become that the SAT is good because of the inertia of the college admissions system is especially hysterical.; 3/10/12, 2:59 AM
Chip S. said...: Now, Chip, you gotta ask yourself: how is it that there can be some innate intelligence number inside people that can be tested, yet it also can be that testing exactly that thing is much easier to game?

Well, I'd hate for you to think I was pontificating, so I'll just take a wild stab at this one: (1) There's a finite number of ways to ask the same conceptual questions, so once outfits like Princeton Review started getting access to actual old test questions (I don't have to explain that history to an expert like you), they made it easier to prep for the questions. (2) There's an element of strategy in answering any multiple-choice test, and this is a teachable skill (no doubt you're familiar with "Joe Bloggs"). So in the olden days when people naively believed that no prep was possible, the SAT probably gave somewhat more accurate results.

But that just says that there were exploitable weaknesses in the historical SAT. It doesn't say that it was ever useless. Only you keep saying that it's complete shit, in the apparent belief that that constitutes a reasoned argument.

Now, don't get me wrong, you wouldn't be 7Machos if you didn't give us fun phrases like

stupid math questions involving x, y, and z

(the horror!) and

ridiculously easy to manipuate with coaching techniques.

I love that stuff. Really--not being snarky here. We all have our internet personas, and yours is distinctive and entertaining. But I think you really do understand that "ridiculously easy" isn't a quantitative measure of coachability. And I know you're aware of the fact that if the old SAT really had been utterly coachable, there wouldn't have been a need to rescale the scores to hide the decline.

But, again, you're the only person here with an SAT fixation. You're probably right that it's well past any pretense of being an IQ test. As to the question of why this evolution occurred, well, there are lots of stories about that. I'd say that your horror at the correlation b/w SAT scores and parental income suggests one explanation: a lack of ability or will to defend against the charge of class bias.; 3/10/12, 3:01 AM
Seven Machos said...: Rev -- To this day nobody cares about the new section on the SAT. A 1200 to me is 600 on what you would call math and verbal in the days or yore.

UCSD is the third-best school in a public school system in a state full of great private schools.; 3/10/12, 3:02 AM
Seven Machos said...: Chip -- Test prep costs money. In New York City, $400 per hour is common. Courses are expensive. Ask Roy Hale in Springfield Missouri if he is willing to part with $1500 for a test prep course for his kid. I'd love to be there for that.

The decline you speak of was not related to coaching.

But you fail to answer any of the questions I asked. I'll put it a different way. Say, you took the SAT back in, say, 1995, and you got an 1100. I took you by your tender hand, and I taught you a new way to do analogies, and I taught you how to plug in numbers for letters, which makes about 15 math questions a breeze for you, and I taught you how to manage your time. You take the SAT again, and you get a 1300.

Are you smarter? Because you just took a test of innate ability and all.

If you aren't smarter, how did you do better on a test of innate ability? Perhaps the word innate is giving you some problems. Merriam Webster has a website.; 3/10/12, 3:11 AM
Seven Machos said...: Blue -- I tried to respond to your comment but my comment got eaten. Then, I decided that you are a blowhard not worth my time.

Good luck out there detecting fallacies based on your paucity of knowledge, though, dude. You're gonna go far.; 3/10/12, 3:19 AM
Chip S. said...: The decline you speak of was not related to coaching.

No shit! That would be some pretty bad coaching.

You really do seem to have a fundamental misunderstanding of what numbers represent, to the point where you don't even understand that I've answered all your inane questions. But, for some crazy reason, I'll try one more time:

There is no contradiction at all in saying that the ability measured by a test that no one has advance knowledge of is innate, while at the same time any particular person's score on that test will be higher the greater his familiarity with the questions on the test. It's most easily stated this way:

S = f(A, F), where S represents a student's test score, A represents the student's innate ability, and F represents the student's familiarity with the test's content and style. You were apparently shocked and crestfallen to learn once upon a time that F has a positive effect on S. You seem to have concluded, therefore, that S is uncorrelated with A. That is an invalid conclusion.

You assert--at least implicitly--that the entire correlation of family income with S is due to the use of income to buy more F through expensive coaching. But you haven't presented any systematic evidence of that; and no, "data" is not the plural form of "anecdote". Please feel free to provide this evidence.

OTOH, maybe you'd find an anecdote more persuasive than that x,y, z shit. So here's one: A professor (in a relatively mathematical field) once told me that he considered a student's vocabulary to be an excellent indicator of intelligence. Does this mean that vocabulary can't be taught? If vocabulary can be taught, does that mean the professor was a deluded fool?; 3/10/12, 3:29 AM
Chip S. said...: 7M, It's v. late and I'm v. tired, so I hereby cede the rest of my time to Blue@9.

buenas noches; 3/10/12, 3:33 AM
Seven Machos said...: You assert--at least implicitly--that the entire correlation of family income with S is due to the use of income to buy more F through expensive coaching.

It could be coaching. More than likely, it's a correlation produced by the general fact that a wealthy household tends to be a magical success incubator for everything, including higher scores on crappy tests.

The reason I say that the decline in scores wasn't related to coaching is that you said: if the old SAT really had been utterly coachable, there wouldn't have been a need to rescale the scores to hide the decline. I have no idea why scores went down, but I do know that coaching of richs kids exacerbated the anti-coaching changes to the SAT because it was seen, quite correctly, as an unfair situation.

Your position that a test can test innate ability but you can do better if you have advanced knowledge is baldly illogical. The claim was that the SAT was uncoachable. You couldn't study for it. Just show up. Obviously, that was a lie. But you are apologizing for it. Why? Either there is innate ability which can be tested or there is not. Which is it?

A better solution would be to have a marketplace of tests. Tests are not particularly expensive to make. There is no reason to have the duopoly of crap and crap lite that we have now.

Finally, I do not think I am the only one here with an SAT fixation. You've got a little bit of one going yourself. Anyway, it's been stimulating. I mean that. Good night.; 3/10/12, 3:43 AM
Michael said...: Why would we want to eliminate "class" differences in the first place?
What makes Murray think that kids are equipped for college with or without high SAT scores?
The poor are fucked. Their liberal owners have no interest in doing the things needed to be done, not while they vote for more bread and more circus. The poor are done for and the "class" that works and strives will continue to rule.
If all,all, the money of the one percent were given,given,to the poor the once percent would have it back in twenty years. With interest. You cannot overcome stupidity with the elimination of "class".; 3/10/12, 7:18 AM
SGT Ted said...: since the duke power ruling eliminates the ability for employers to give iq and skills tests

When I did the Veterans Adminstration Vocational Rehab for my service disabilities, they gave me Cognitive and Physical skill based tests to see how they could best get me a job.; 3/10/12, 7:39 AM
SGT Ted said...: Employers should use the ASVAB. I think it has valid results in a general way.

If it's good enough for a color blind Armed Forces, good enough for the civilian wanna-be elites that seek to rule them.; 3/10/12, 7:57 AM
NotWhoIUsedtoBe said...: We overvalue verbal intelligence. We undervalue the ability to work.; 3/10/12, 8:04 AM
Steve said...: The last two people I hired for a marketing position and a purchasing position both had 'only' a high school diploma. At the final interview level they beat out three, four year degrees and two, two year degrees because I believe their work ethic and ability is better then the other folks.

Both have started out great. So I'm with you Blue@9! Now if I could keep some more of my money to grow my business faster I could hire more people like this. Instead I have to worry about how much cash is set aside for my quarterly tax payments that the elite in government will use to give lottery winners food stamps.; 3/10/12, 8:10 AM
NotWhoIUsedtoBe said...: Intelligence has become the new virtue, replacing work ethic.

People talk about how smart someone is, rather than how hard-working.

People we don't like are "stupid," or "retarded."

For a non-judgemental people we sure like to categorize based on smarts. Since smarts are largely inherited this isn't very fair.; 3/10/12, 8:18 AM
FuzzyFace said...: Companies require college degrees for a practical reason beyond the whole, "we think it is a reliable predictor of something important," thing. HR departments receive massive numbers of resumes, way too many to examine individually. They therefore need ways to filter them down to a manageable level, and presume that - all else equal, somebody with a degree is a better risk than somebody without.

I agree that a test can be better, but they are difficult to do in the general case. One employer did hold tests for programming positions. Anybody could show up and take it, and if you scored high enough, they would interview. The tests were directly related to programming knowledge. Once I joined the company, I took part in grading them, and was appalled at how low many people scored. It was a much better filter than examining resumes.

But how do you write a test for a sales or administrative position? Is it even possible?; 3/10/12, 11:00 PM