What are some of the things that could happen if all colleges & universities in the US start admitting students only based on merit?

One thing that I think a lot of answers here miss is that the goal of a private university is not to admit the best students, it's goal is to admit the student body that will set up the university for long-term success. A school that starts admitting based on SATs alone will filter out a lot of future politicians and, civil leaders and entrepreneurs and be left with a lot of future professors and engineers. That's a great strategy for some schools, but for others it would mean a marked decline in reputation and donations. This fact can lead to good outcomes (like direct generous financial aid policies) and bad outcomes (misalignment of who gets admitted and who "deserves" to be admitted). But mostly it's just the way things are. I think it has bigger implications for state universities, whose missions should be "increase the human capital of our state as much as possible by admitting students to whom our education can provide the greatest combined benefit for the student and state", which is probably harder to pull off well especially given political constraints.

Very little change would be detectable. Leaving aside the measurement issue (which other answers have considered), keep in mind that so-called selective schools are in reality a very small percentage of all higher educational institutions in the US--maybe 200 to 250 versus an N that probably ranges from conservative estimates of about 3,000-4,000 to more inclusive estimates of upwards of 7,000. Here is Barron's list of selective schools, with distinctions made across four categories of selectiveness: http://www.nytimes.com/interactive/2013/04/04/business/economy/economix-selectivity-table.html. By the time you get to category 4, you are, in truth, not really that selective. Here is something else to keep in mind: there is considerable research in support of the notion that the college from which people graduate has little to no effect on their long-term prospects (see http://economix.blogs.nytimes.com/2013/04/04/what-makes-a-college-selective-and-why-it-matters/). The trick is not necessarily to get in to the very best school, it's to complete whatever program you start. In other words, the challenge for many, many students is not getting in, it's staying in--not access, but academic success. The advantage for selective schools is twofold: 1) in being selective they can choose those with a greater likelihood of academic success (read graduation) and 2) they have more resources than unselective schools and can use those resources to boost graduation rates. Recent research suggests that wealth confers considerable educational advantage prior to children's actual entrance into school. The educational system in the US can (and does) narrow that advantage somewhat, but not entirely.

Using a single standardized test score is a terrible idea. It measures such a limited portion of a student's skills and potentials. In fact, it probably wouldn't change the colleges much, because the rich kids have better education, more exposure to vocabulary, SAT tutors, etc. In fact, parents would pressure schools to focus all their efforts on teaching how to get good SAT scores, instead of real instruction.   On the other hand, devising a better way to measure merit is a great idea, but not an easy one. The only simple step colleges could take is to stop 'legacy' admissions, which would help, but also eliminate millions in alumni donations, requiring schools to either cut costs or raise tuition even higher.   There are more fundamental problems here. Poor children are not only less likely to do well in school and less likely to participate in extra-curriculars (for a variety of reasons) they're less likely to go to college even if they could get in, because they can't afford tuition, costs and living expenses for all the years it would take. A real meritocracy would require all university and living expenses to be paid by the State, which some countries do, but it would be a tough row to hoe in the US.   The real problem, to me, is that if you wait for college, you're already too late. Kids from the poorest families already have worse grades, worse test scores and worse prospects in life. If you want the country to be a true meritocracy, you have to start providing real support to kids much earlier. If we did that consistently, then we'd really have something.

Deciding a student's worth based only on the score of single standardized exam is what it is  wrong with Indian education system. It encourages students to mug up the things as opposed to learning the subject entirely. Student master the exam not the subject The overall profile of student including his/her academic grades, projects, research work, field experience define the student as a whole Great or poor performance in any single criterion can be misleading

Things that could happen if college entrance was based on standard exams: Innovation in the States would come to a crawl as creative minds were passed over for better test-takers The leadership greats who revolutionize how we motivate workers and manage business will cease to appear in our companies. Individuals with other valuable, but not easily tested, strengths will be missing from our workforce. Only the children of families who can support their education will succeed, relegating those from poor or troubled families largely to remain in that situation, leading to even greater class stratification. Those who make study materials and courses for such exams will make a great deal of money.

Assuming merit to be measured as it is now (GPA, SAT, AP, recommendations) one difference would be an increase in the admission of women -- currently universities try to keep a rough parity between men and women, but to do this, they have to manipulate the data. Women would become a majority -- and perhaps a dominant one. I taught for many years in Korea, where a single national exam -- taken just once each year by high school seniors -- determines entry to the top schools. The rich spend thousands of dollars (or much more) preparing to pass this one test. This assures that those who are test-wise will be advanced, and that some corruption will be rewarded. It also assures that education will be seen as only useful for passing a test -- it can be forgotten afterward. To illustrate how this works, I have seen a Korean student barely able to speak English who was able to score at the highest level in SAT writing -- by memorizing passages written by someone else. A skilled test-prep teacher can make this work surprisingly well. If real merit could be measured, it would require something far more sophisticated than the SAT -- after all, the ideal engineering trainee might be quite different from the ideal future business manager or the ideal future teacher or doctor or lawyer. Keep in mind that education is inherently conservative. When Francis Bacon was providing us with an approach to the modern scientific method, Oxford was still using Aristotle to teach science -- as if nothing had advanced in 2,000 years. We still lecture to students -- a technique created in the middle ages to allow pupils to copy books as they were read. The printing press, invented more than 500 years ago, didn't have much effect on this practice. Neither did audio recording, film, radio or TV, and even the Internet is at present more decorative than essential to formal education. We still labor as our distant ancestors did, and little seems likely to change that practice.

Well, merit. There's the rub. How do we determine merit? What aspects of a person should we measure, and how should we measure those attributes? Standardized tests are popular, but many institutions of higher learning use other things. They look at grades and extracurricular activities and service to the community and entrepreneurial activities and many other things that I can barely even think about. Private institutions can probably accept anyone they want for any reason they want. However public institutions have to follow the rules that the state legislatures come up with. So admissions get politicized. We haggle over what merit means, and everyone gets all righteous on each other. It is very clear to me that tests do not measure merit. It is also clear to me that vast majority of people think tests are a perfectly find way to measure merit. So what will happen? At first more Asians and fewer blacks will get into state schools. Later on, as people figure out this is not a good thing, they'll find different ways to measure merit. Maybe they'll even decide that working in a poverty stricken neighborhood gives one additional merit, and they might even decide that doing so well on tests without spending any time at all in needy communities is a demerit. It's all politics. And over time, it will change. People don't like affirmative action because they associate it with advancement due to the color of your skin, as opposed to for any meritorious reason. So if people figure out how to advance folks for meritorious reasons, even if they don't do as well on tests, we could achieve the goals of those who want to redress perceived historical injustices.

Depends entirely upon how one defines the term "merit". Although standardized test results and such are often used I would prefer it to include the philosophy that's best articulated by the phrase "success is not a measure of where you end up in life but rather how far you get from where you started". I would argue that if the concept of "merit" were somehow to include a measure of how far you have come to arrive at the score you achieved, rather then simply based upon the score's absolute value, we would witness more 'achievers' getting in and fewer of those who simply feel entitled.

This already happens in France. The exams are not easy, definitely far more difficult than the SAT or even the IIT JEE. Here are some admissions exams for École Polytechnique. http://www.sujets-de-concours.net/sujets/xens/2013/mp/maths1.pdf http://www.sujets-de-concours.net/sujets/xens/2013/mp/maths2.pdf http://www.sujets-de-concours.net/sujets/xens/2013/mp/phys.pdf http://www.sujets-de-concours.net/sujets/xens/2013/mp/info_option.pdf http://sujets-de-concours.net/sujets/xens/2013/commun/informatique.pdf These are from just round one to see who's worth talking to. Round two involves a battery of oral examinations. I think this is a wonderful system that truly fosters academic excellence. Contrary to some stereotypical claims that innovation and creativity will become stifled, the French have created some of the most impressive feats of engineering like this No words necessary the Millau Viaduct the Airbus A-380 the TGV (fastest bullet train in the world) the Minitel If the US copies the French grandes écoles system, no longer will Americans have to settle for expensive and academically inferior undergraduate universities like Harvard and MIT. The US higher education system is truly a grave sickness that I hope Western Civilization cures itself from.   Remerciez Dieu pour le système éducatif français!