Finding Value in Standardized Testing
College entrance tests (SAT and ACT) are the subject of some of the most heated and controversial debate in all of higher education. Trustees and administrators are not of one mind on the tests’ value or even whether they ought to be used. Critics of the tests argue that not all students are able to show their ability in a high stress “timed” environment and that the tests have built-in biases that disadvantage minority students. Colleges are reevaluating the use of these tests as accurate measures and considering if they want to continue to require students to take them for admission purposes.
While the debate about the tests is instructive, there are at least four important reasons to maintain the common use of SAT and ACT for prospective students. These rationales include: the availability of standardized data; an emphasis on skill preparation; attention to academic quality and better informed decision-making by all involved in the admissions process.

Colleges should do whatever they darn well please. If a college wants to specialize in teaching really intelligent students, that's its perogative. If a philanthropist wants to endow a college dedicated to providing educational opportunities to mediocre students, that's his own business, too.
A state's university system can dedicate itself to whatever legal mission the state legislature decides is in the best interest of the state (or, alternatively, in the best interest of its members' political careers).
If a college _wishes_ to favor more intelligent students, it needs a way to estimate students' intelligence. No estimate is perfect, but the S.A.T. is a pretty good one. If it uses it, however, fairness demands that it should allow applicants who undertest to offer alternate evidence of their intelligence.
I hail from the state of Washington, infamous for it's WASL (Washington Asessment of Student Learning) testing in grades 4, 7, 10, and nowadays pretty much all the others. Tests similar to the WASL are being used in place of standardized testing in classrooms across America. Tests like these, and their result have no correlation with academic value, seeing as they are corrected by teachers, rather than machines, amounting to varied scores for kids who put similar answers depending on who the corrector is. Also, these tests are commonly graded by a standard of "showing work" rather than getting the correct answer, and in certain portions, simply writing a longer answer is what gets you a better score. Things like this lead me to believe (yes this is a bit off topic, and feel free to label me a conspiracy theorist) that these tests are often used as a method of personal data collection derived from the views portrayed in one's answers on the test, carried out by state governments. Plus, teachers, students, and administrative staff HATE tests like the WASL. I don't know if this rule stands in other states with similar tests, but if a school scores low enough on the WASL, the government can come in, take over, start booting teachers, assigning different curriculums, etc. Decisions best left up to local school boards. In addition, these threats cause teachers to become nervous about their jobs and only teach kids what they need to know for the tests (known by local anti-WASL advocates as "teaching to the test") and neglect the things that our youth REALLY need to know.
I've been thrown through the BS that is the WASL for years now, and I really don't see the point in them. Can a few sets of questions really detail every last bit of knowledge a student has?