SAT scores well in predicting college success

<p>
[quote]
In other words, the non-submitting students benefit from the fact that others are taking standardized tests

[/quote]
Not necessarily true, because colleges also have information about high schools that come from standardized measures of comparison OTHER than SAT scores. For example, one can easily find out that my daughter's high school has an California API of 9 on a 10 point scale. So a college might look at that, plus her GPA and class rank as a good indication that she was a strong student.... but definitely want more from a high school with a lower API. Most of the information they would need would be in the school's profile submitted by the g.c. -- but I use GreatSchools.net to find info about schools anywhere, and I'm sure that colleges could do the same. </p>

<p>Also, I think that in a completely test optional environment, students would still be free to submit alternative tests. For example, my daughter has a 4 or 5 on every AP exam she took; as far as I know, the AP exams all have an essay component and I think that enables her to better show her thought process and mastery of a subject -- she also did very well on SAT writing. So it would make sense for her to have submitted her AP English scores rather than SAT CR scores, for example. </p>

<p>I agree with MarathonMan that nonsubmitting students are assumed to have weaker scores, but I don't see a problem with that -- they are being evaluated on the strength of their records within the contexts of their schools, and in general the colleges have reasonably good information about the quality of their school. Obviously the SAT or ACT would be more important for a kid coming from a weak school .... but that is no reason to force all kids across the board to go through the competitive exercise of taking and retaking the test for the sake of improving their scores. I think that if all colleges were test optional, we would have saved money on needless retakes of exams for my daughter -- as it was, she looked at those tests as a barrier to consideration at many colleges, and it added a lot of unnecessary stress to the whole admissions process.</p>

<p>CA schools may have metrics that colleges can use instead of SATs; but that does not hold true for all schools. So a student from CA comes from a school with a CA API of 9/10. How would that compare with a student from another state? Should college engage in detective work?

[quote]
Also, I think that in a completely test optional environment, students would still be free to submit alternative tests.

[/quote]
.
This does not suggest a completely test-optional environment. It merely substitutes one national standardized test for another. And some of the tests, namely the APs, are even more expensive than the SATs and less available to a broad range of students. As well, how should adcoms compare AP scores with ACT or SAT scores?If the SAT test can be prepped, (read more affluent students can pay for tutoring) AP classes tend to be found in the more affluent districts.
For good or ill, the SAT is a metric that is national, not state-wide; and not limited to schools that have greater resources than others.<br>
Various posters have criticized the expenses of SATs and APs. It is a valid point. But it is the price Americans pay for having a fee for service, relatively low tax system. I did not have to pay a penny for taking the baccalaureat or any of the tests before it. But my parents paid far higher in taxres than an American family with the same income level.</p>

<p>Marite, colleges aren't interested in comparing schools to another with great detail --they just want to get a general idea of whether the school is rigorous or not. But any case the data to compare is there and readily accessible -- Greatschools.net assigns a ranking to all school, also on a 10-point scale, which is based solely on standardized testing data controlled for demographic factors. (My d's school is ranked "8" by Greatschools). Most states also have data that can be used -- the NCLB law pretty much mandates that the states be collecting the data. The testing data is public info, and if there was a market for it for colleges, it would be produced for them in whatever form they wanted.</p>

<p>And when I referred to substituting an AP exam, I did not mean that would be mandatory -- just that a true test optional school would offer the option of different tests, if the student wanted to submit them. No one has to submit any tests. Obviously kids with very strong test scores would find it advantageous to submit. As would kids coming from weak schools or kids whose grades weren't as strong as they could be.</p>

<p>My understanding is that a true test optional school would be school that did not require ANY standardized test. So far, I do not know of any. Just want to be clear about terminology.</p>

<p>In your scenario, kids from Exeter and Andover would submit umpteen AP scores, and kids from Holeinthe ground, TX would submit....what? By the way, Holeintheground, TX (obviously not the real name, but a real community of mostly first-generation immigant Hispanics), sent a kid to Harvard a couple of years ago.</p>

<p>As for the Greatschools.net, how many schools does it include? All? Some?
I am surprised you think that NCLB provides good metrics. First, as far as I know, most NLCB-mandated tests are 10th grade leve tests, hardly a gauge of what it would take to excel at Harvard or Barnard College. Second, they differ wildly across states. The NY Regents tests are a totally different beast from MCAS (MA), of the Texas tests, or the CA tests.
What you are proposing would entail adcoms trying to figure out how to compare applicants from totally different educational and testing systems. The fact is, the colleges do compare schools. That's why they ask for the school profile. They also have to wonder what an A from school x means as opposed to an A from school y. Or how to interpret the rec from teachers from school x who have never before sent a student their way.</p>

<p>The irony is that the US with its highly diverse educational system needs standardized tests more than countries with a far more uniform k-12 educational structure and curriculum.</p>

<p>Greatschools.net has no rating for most of the schools in our county. For our school it has a dozen or so parent and alumni comments. (Fairly accurate I'd say.) But it looks like ratings are solely based on poll numbers - not very scientific.</p>

<p>And is there a notsogreatschools.net? For the kind of schools for which the SAT is probably far more useful than for the Andover and Exeter or Scarsdale or Brookline, MA?</p>

<p>
[quote]
My understanding is that a true test optional school would be school that did not require ANY standardized test. So far, I do not know of any.

[/quote]
Fairtest.org lists around 700 - <a href="http://fairtest.org/optinit.htm%5B/url%5D"&gt;http://fairtest.org/optinit.htm&lt;/a&gt;&lt;/p>

<p>Some of the schools that are SAT optional actually require SATIIs. Applicants need to use the Fairtest list with a lot of caution.</p>

<p>EDIT: I looked up some names on the Fairtest list. There are indeed some schools that are totally test-optional.</p>

<p>Re "Greatschools.net" -- it lists ALL schools for which it can get data - the title of the site doesn't mean exclusivity. (It lists plenty of rotten, terrible schools too). </p>

<p>The site does have a place for parents and students to submit reviews, but the "rating" doesn't come from that source - it comes from standardized test data:
[quote]
These ratings provide a quick overview of how students
in this school performed on standardized tests.</p>

<p>A "10" means that the group is in the top 10% for test scores
when compared to students in the same group statewide.
A "1" means that the group is in the bottom 10%.

[/quote]
Basically they do a ranking of schools within each state by whichever metric the state uses and reports. It is true that they don't have data for every state, but they seem to have it for most. </p>

<p>I wasn't suggesting that the colleges rely on that web site in any event -- I was suggesting that if they wanted, they could rely on the same public information that the web site uses for its calculations, just as I can rely on information from the Common Data Set to evaluate colleges without necessarily subscribing to US News to get their take on it. I'm sure that if the colleges wanted that info, there would be a marketing or data collection firm quite willing to assemble it for them -- maybe the same firms they already use to target zip codes for their mailings and help develop profiles for enrollment management. </p>

<p>And again, there is no need to compare high schools from one state to another or to develop a ranking of high schools. Colleges don't CARE about that -- if they did, California students would be at a terrible disadvantage when applying to east coast schools because everyone knows our public school system sucks. But colleges want geographic diversity even if that means from drawing from all sorts of high schools in towns where no one has ever heard of IB and kids are lucky if their high schools offer 2 APs. </p>

<p>And they do NOT expect kids coming from those high schools to have perfect SAT scores in order to be considered, even if they require the scores --they look at scores in context of the school and the state where the student is coming from. </p>

<p>The argument has been raised that the scores can be used for "validation" of GPA -- and I am responding to that, saying that the colleges have other data to validate the GPA -- namely the school's own profile and the public data available.</p>

<p>Let me see if I have it right. High schoolers should not have to take standardized tests in order to apply to colleges. However, when reviewing their applications, colleges may rely on greatschools.net whose rankings rely on standardized test data.<br>
One way or the other, it looks to me like colleges need high schoolers to take standardized tests. If not a specific applicant, then a large proportion of students at that applicant's school. That's where the validation comes in. </p>

<p>I would be leery of relying too much on the greatschools data. I have no idea whether the two schools I am citing, ours and a neighboring one, are in the greatschools network. I read an article about a month ago, however, that the participation rates in SAT-taking for these two schools differed widely. Despite the fact that our school has a 40% F/R lunch population, 80% take the SAT. The other school, with and only marginally larger F/R population has a SAT-taking rate below 60%.</p>

<p>
[quote]
But colleges want geographic diversity even if that means from drawing from all sorts of high schools in towns where no one has ever heard of IB and kids are lucky if their high schools offer 2 APs.

[/quote]

Yes, indeed. And that is why a student from Holeintheground, TX, whose school offers only two APs and has never heard of IB should be taking the SAT.</p>

<p>marite: you took the thunder out of me. I was thinking of asking the same question about that site.</p>

<p>Not only that for TX schools it relies on TAKS data....According to my S, TAKS is a joke.</p>

<p>Simba: Thanks. I made the point earlier that different states have dfferent state tests. The TAKS vs. MCAS comparison reinforces this point.<br>
The MCAS is supposedly one of the toughest state tests. However, it is administered to 10th graders, not 11th and 12th graders. And its contents and scoring have been tinkered with ever since it was introduced. It was originally intended (or at least presented) as a diagnostic test rather than as a high states exit test. As a result, different student cohorts within the state cannot be meaningfully compared across the years.
While the CB has been criticized for its fees, the MCAS cost the state (i.e. the taxpayers) millions of dollars every year; and the test takes up a whole week instead of one Saturday morning. Another issue is that the MCAS proctoring and conditions under which the tests are administered vary hugely even within the same school.</p>

<p>I also agree with the point you made earlier that in many parts of the world the curriculum is much more standardized than US. Thus US needs a standardized test more than ever.</p>

<p>Even No Child left behind relies on tests to track improvements.</p>