College ain't for everyone--but neither are standardized tests

I think this WSJ article by one Charles Murray does a good job of explaining why college doesn't make sense for a lot of people:

Imagine that America had no system of post-secondary education, and you were a member of a task force assigned to create one from scratch. One of your colleagues submits this proposal:

First, we will set up a single goal to represent educational success, which will take four years to achieve no matter what is being taught. We will attach an economic reward to it that seldom has anything to do with what has been learned. We will urge large numbers of people who do not possess adequate ability to try to achieve the goal, wait until they have spent a lot of time and money, and then deny it to them. We will stigmatize everyone who doesn't meet the goal. We will call the goal a "BA."

You would conclude that your colleague was cruel, not to say insane. But that's the system we have in place.

Murray's proposed solution? Judge applicants on what they know, not how they came to know it:

The solution is not better degrees, but no degrees. Young people entering the job market should have a known, trusted measure of their qualifications they can carry into job interviews. That measure should express what they know, not where they learned it or how long it took them. They need a certification, not a degree.

In principle, this sounds great. What we should be interested in is the end result of a person's education, not the educational process that led to that result. If two people are equally qualified for a position, it should make no difference that one of them went to a prestigious university while the other one had no formal schooling at all. Where and how a person was educated does not necessarily correlate with that person's true value as an employee.

However, Murray goes one step too far by arguing from here that we should therefore ignore degrees and focus instead solely on standardized tests modeled on "the CPA exam that qualifies certified public accountants". He does not offer any evidence that standardized tests are any more reliable than degrees as indicators of an employee's fitness. Rather, he just seems to operate under the assumption that for any set of skills, there is a standardized test that will yield an accurate snapshot of a person's proficiency in those skills:

The merits of a CPA-like certification exam apply to any college major for which the BA is now used as a job qualification....

[...]

...Certification tests would provide all employers with valuable, trustworthy information about job applicants.

But this is really controversial. While the benefits of standardized tests are clear: universal standards, quantifiable results, etc.--they also introduce plenty of complications. A few I can think of from the comfort of my armchair:

• People who do poorly under pressure/time constraints are disadvantaged
• People unfamiliar with test-taking are disadvantaged
• A one-time test does not necessarily reflect a person's day-to-day productivity or work ethic
• Cramming is rewarded
• Test design can unintentionally favor some socioeconomic or ethnic groups
• Educators end up teaching only material that is likely to be covered by the test
• Not all skills--for example, paper writing--can be captured by a sit-down test

I think these drawbacks are pretty significant. If I'm looking for someone who is going to be working at a desk serving some public administration function, I'm not really interested in whether this person can perform under pressure--I'm more interested in how this person performs on a routine, regular basis. How dependable is this person? How consistent? How thorough? Of course, I'm interested in the person's analytical skills and such--and certainly, a high score on a standardized test can be sufficient condition for this. But a person with the skills I need won't necessarily do well on an administered test. That's why it's good to have other indicators--like for instance GPA, which is a measure of success and diligence over an accumulation of years rather than a one-time swing for the fences.

I sympathize with Murray's point: for too many people, four-year degrees are a waste of time, and their extreme expense prevents many from climbing up the social ladder. But I think it's wrong to conclude from this that standardized tests alone hold the key. While they work for some professions, for many they are of limited value. One approach may be to keep the "degree" concept but make earning them less expensive and less weighted down with superfluous requirements. For example, there could be a general degree for "Administration" that would guarantee a person's proficiency in language comprehension and communications skills. There could even be a standardized set of assignments for the course that would impartially enforce universal standards. Such a degree would avoid many of the difficulties of a one-shot certification exam while keeping many of the advantages of the GPA metric, and hopefully allow a JC student to compete on an even playing field with someone from a fancy Ivy League school.

In the end I think we should be wary of one-size-fits all solutions--but that doesn't mean we can't be doing more to make the education system more efficient, meritocratic, and accessible.