<p>"the SAT doesn't measure intelligence...
if anything it measures socioeconomic status of the family. "</p>
<p>thats just one of the many excuses out there. how does economic status effect ur ability to do math or to analyze a passage?</p>
<p>See test scores by Family Income in Table 11 (page 4) in
[url=<a href="http://professionals.collegeboard.com/profdownload/Total_Group_Report.pdf%5DTotal">http://professionals.collegeboard.com/profdownload/Total_Group_Report.pdf]Total</a> Group Report: College-Bound Seniors 2008<a href="from%20%5Burl=http://professionals.collegeboard.com/data-reports-research/sat/cb-seniors-2008%5DCollege-Bound%20Seniors%202008%5B/url%5D">/url</a>.</p>
<p>[url=<a href="http://www.collegeboard.com/prod_downloads/highered/ra/sat/coaching.pdf%5DEffects">http://www.collegeboard.com/prod_downloads/highered/ra/sat/coaching.pdf]Effects</a> of Coaching on SAT Scores<a href="from%20%5Burl=http://professionals.collegeboard.com/data-reports-research/sat/data-tables%5DSAT%20Data%20Tables%5B/url%5D">/url</a>.</p>
<p>[url=<a href="http://professionals.collegeboard.com/profdownload/Avg_Scores_of_Repeat_Test_Takers.pdf%5DAverage">http://professionals.collegeboard.com/profdownload/Avg_Scores_of_Repeat_Test_Takers.pdf]Average</a> Scores of Repeat Test Takers<a href="from%20%5Burl=http://professionals.collegeboard.com/data-reports-research/sat/data-tables%5DSAT%20Data%20Tables%5B/url%5D">/url</a>.</p>
<p>gcf101, interesting results in Table 11 in the CB report. I think that the level of a student's preparation in school and the availability of books in the home have a big impact, and both are often correlated with family income. College Board has expanded the income range for which they give data (relative to previous years). But I think it would be interesting to have even more data, splitting the "$200,000+" group into strata. I'd like to know whether the increase of scores with income tails off (probably, at some point) and where that happens, in the income and score range.</p>
<p>@Supereagle10 - It may be in your eyes an "excuse" but it is unfortunate and cannot be ignored.
Kids coming from wealthier families logically live in more affluent neighborhoods.
More affluent neighborhoods have bigger school budgets and thus more resources and opportunities.
There are definitely brilliant people from poorer neighborhoods that don't do as well on tests like the SAT/in school because they a) can't afford fancy SAT prep courses and b) don't attend schools with top-notch teachers, attend schools that use outdated textbooks, perhaps have their ability to concentrate lessened because of home-related factors.</p>
<p>I agree with geologistsrock, the quality of the schools makes a significant difference in a student's preparation for the SAT; that's what I had in mind by "the level of a student's preparation in school" in post #83. In my opinion, the quality of schooling over an 11-12 year period outweighs SAT prep courses quite a bit.</p>
<p>To all of you who persist in criticizing the SAT -- on the "basis" that some "unfair" factor X (income, social class, private school) is the reason some applicants score better than others -- you need a remedial course on statistics. Upper income/class students no doubt score better on the SAT than low income /class students, but that doesn't mean that the SAT is unfair to, or "underscores" low income students. A CORRELATION, NO MATTER HOW STRONG, DOES NOT IMPLY CAUSATION. In addtion to attaining higher SAT scores, children from high income families also generally do better in college and earn more.</p>
<p>Here's another example, the number of cavities in elementary school children and vocabulary size have a strong positive correlation, but only an idiot would argue that vocabulary causes cavities. Instead the correlation exists because both respond to changes in an undisclosed variable, the child's age.</p>
<p>A lot of these posts disparaging the SAT are in the nature of vocabulary-causes- cavities arguments. Leave this to the professionals.. And, by professionals, I dont mean the self serving, PC hacks at UCB. Their "study" should have been entitled "lies, damn lies and statistics." I have never seen a more blatantly dishonest excuse for statistical analysis.</p>
<p>If everyone studied the same amount and in the same way then the SAT would measure intelligence. But since kids are taking all these prep classes/studying 15 hours a week then its a pretty bad indicator</p>
<p>I seriously doubt that anyone taking an SAT class and studying for it for fifteen hours per week is doing much to raise his score.</p>
<p>I kind of agree with you vicariousparent. A lot of schools say that they do not place that much importance on the SAT, but I definitely think that it plays a huge factor in the college process. Although demeaning it has become popular, especially for the SAT optional schools.</p>
<p>You all seem to ignore the elephant in the middle of the room. Kids from wealthier families score better on the tests not because their parents have more wealth but because their parents are smarter, at least on average. Wealth in our society is only rarely inherited, read The Millionaire Next Door. The same characteristics that made these people wealthier also helps their kids score better: intelligence and diligence. At least to some degree these kinds of things are inherited characteristics. This is of course not universal, some people, like myself, come from poor backgrounds and do exceptionally well on standardized tests. The reason these tests were invented was to discover those kids and give them a chance to develop their abilities.</p>
<p>I think the typical reader of CC needs to be a bit careful with the conclusions of this study.</p>
<p>This study is dealing with SAT scores of around 1000-1100 (which I assume is for the CR and Math portion only). An increase of about 5-6% in this range corresponds to about a 10% increase in percentile rank.</p>
<p>From what I gather from many of the 'chances' thread, the typical student reader of CC worries about SAT scores of around 2000-2100 (for the entire CR, Math and Writing). I often see comments that these scores are 'low' for admission to an elite school like HYPS. It seems that these readers feel that a score of 2200+ is needed. </p>
<p>Going from 2100 to 2200 is also about a 5% increase in score, but only about 5% increase in percentile rank, much less significant than the 5% increase in score for the 1000-1100 range.</p>
<p>I doubt that there is much difference in terms of graduation rates with students who have scores in the 2000-2100 range versus students with scores at 2200+, even at places like HYPS.</p>
<p>So I guess I am wondering if there is some 'ceiling' effect. At what point (if any) do further increases in average SAT score lead to very little increase in graduation rates, GPA or both. And if this point does exist, why do we need the SAT to make any finer distinctions above this mark? Or why should an admissions officer care about deciding between candidate A and candidate B on the basis of SAT scores if both are higher than this hypothetical point? At this point, other factors should come into play.</p>
<p>I am not a big fan of standardized tests, but from my experience, students who do score very low on these tests, even with a decent GPA, do struggle in college. I have taught students who have below the 50 percentile rank and college classes for them are very hard, even with an OK GPA.</p>
<p>However, I doubt that I could tell much of a difference between a student who scores in the 93 percentile rank from one who scores in the 98 percentile rank.</p>
<p>sk:</p>
<p>one other confounding factor is the arbitrary cap at 800 of each test, i.e., it doesn't test above 800. Some kids who earn that score get lucky on a test version (happen to be a poem that they did in class); others find that the 800 is "easy." Those in the latter category might easily score 1,000+ if the test was longer and normed that way. Perhaps you migh then find a difference, not in graduation rates (since kids tend to gravitate towards majors that they can graduate in), but in a difference in class performance.</p>