<p>As I was looking at Stats Profiles here, I came to a thought that top schools may differ in their preference in character of SAT scores. Difference may come from other reasons but I observe there are colleges accepting more balanced scorers and others accepting more contrasting scorers.
Is there more clear notion which colleges prefer which type of students? If two applicants with MATH 750, CR 750 and with MATH 800, CR 700 apply, do majority of colleges accept the former?</p>
<p>I doubt you can come up with any meaningful generalization that would override all the other factors, other than that an MIT or CalTech presumably would favor perfect math over balance.</p>
<p>I don’t have any sense that 800/700 is very different from 750/750, but I do have an impression that 800/640 is meaningfully better than 720/720, all other things being equal (and the student’s interests being clearly weighted towards the stronger area).</p>
<p>If they’re applying to a math/science program, a stronger math score will probably be desired. If they’re doing a liberal arts degree, then there won’t be as much emphasis on math.</p>
<p>For Harvey Mudd, MIT or Caltech 800 in math is pretty much the norm (though 800 on CR will help you “stand out” somewhat). For other types of schools I am pretty sure that if your scores are >700 the exact breakdown of the numbers matters very little.</p>
<p>
</p>
<p>I disagree strongly. In general, 700+ seems to be the first cut for all highly selective schools; other than tech-types. Strengths in “areas” can be demonstrated on Subject Test scores. For engineering types, it’s a 800 on Math 2.</p>
<p>There is little difference between 750 and 800 (one question). There is a more material difference between 700 and 800. So, 750/750 is not materially different from 800/800. But, there is a material difference between 800/700 vs. 800/800.</p>
<p>However, although it may be statistically significant, it is not meaningfully different.</p>
<p>I think it’s important at top colleges to be above the median on each section if you don’t have a hook. Looking at common data sets with good breakdowns, at top schools the difference in admissions rates goes up significantly between 700 and 750 and between 750 and 800 and you can see how few get in with a below median stat on any section when you back out the hooked applicants.</p>
<p>hmom5: At least half the class everywhere gets in with a below-median stat on any section, and probably a good deal more than that at colleges with normal yields (i.e., not Harvard).</p>
<p>bluebayou: I have seen several 800/640-type kids get into highly selective schools, including (rarely) the most highly selective schools. (Two patterns: talented humanities person with no interest in math, talented ESL math/science student.) I have not seen 720/720 kids get into highly selective universities without significant hooks (highly selective LACs may be a different story). I’m sure that does happen, but I haven’t seen it, hence my impression.</p>
<p>JHS, when you consider 40% of the class has a hook and that group probably has the vast majority of the under median scorers, that does not leave much room for the unhooked to have below median scores.</p>
<p>JHS:</p>
<p>I don’t dispute your experiences. But look at it from the adcom’s perspective: high math scorers are everywhere, but high math AND high CR scorers are much more rare. If you are an adcom at MIT, which also has an english, lingusitics and history programs, who is more attractive, the dual high scorer or the single high scorer?</p>
<p>Sorry, hmom5, but I don’t buy the 40% number, unless you are talking about Amherst or something similarly sized, or unless your definition of “hook” includes lots of students with above-median scores. (E.g. legacies; I have yet to see a legacy gain admission to my alma mater with below-median scores in any dimension.) </p>
<p>You are also assuming that there is near-total overlap among the below-median populations in each dimension. I doubt that. While certainly there are some students who are below median in everything, there are many who are below median in one or two dimensions and above it in others. Way more than half of the accepted students are going to be below median in something.</p>
<p>bluebayou: In general, yes, a dual high-scorer probably beats a single high-scorer, although I believe less frequently than you imagine (and MIT turns them away in droves). My proposition was that a single high scorer beats a dual not-quite-high scorer many places.</p>