<p>OK, I'm not sure I'm following your point but it seems to show cal and usc have similar sat scores.</p>
<p>another poster made the "Nobody at Cal matches..." statement.</p>
<p>i did say that HM and cal tech students are smarter than cal. i'll stick by that. take the average take the top take any segment of the sb at HM or tech and you'll find a smater (iq) student than the same student at cal.</p>
<p>75% of the sb at cal tech where their hs valD's.</p>
<p>
[quote]
Oh, and the SAT data is a bit more interesting than you give it credit...</p>
<p>Cal
Reading: 580-710
Math: 620-740
Writing: 590-710</p>
<p>USC</p>
<p>Middle 50% SAT CR 640 — 730
Middle 50% SAT Writing 650 — 730
Middle 50% SAT Math 670 — 760</p>
<p>Now, while it's impossible to argue that the USC numbers are higher, they're not even a full standard deviation higher in aggregate.</p>
<p>Plus, USC superscores, Cal does not. This almost certainly adds on points (I've read as much as 30 points a section.)</p>
<p>So unless we can control for the variable for superscoring (let's just say 30 points for the sake of argument), we don't get an accurate picture. Controlling for superscoring, the averages are...</p>
<p>Cal
Reading: 580-710
Math: 620-740
Writing: 590-710</p>
<p>USC
Reading 610 - 700
Math: 620 - 700
Writing: 640- 730</p>
<p>This provides an interesting picture. USC's numbers are a bit more tightly knit (less variation from the mean). However, Cal's numbers are not that far off.</p>
<p>What publics tend to have is much more wild variation from the mean. This is a problem, but may not necessarily reflect much in the classroom.
[/quote]
</p>
<p>UCLAri, you are looking at the wrong data. Those scores for USC are scores of Admits, not scores of those who matriculated. The numbers you gave for Berkeley are for its freshman class (those who matriculated), so we should compare likes with likes.</p>
<p>Berkeley
Reading: 580-710
Math: 620-740
Writing: 590-710</p>
<p>USC</p>
<p>Middle 50% SAT CR 630 — 720
Middle 50% SAT Writing 640 — 720
Middle 50% SAT Math 650 — 740</p>
<p>Superscoring accounted it's...</p>
<p>Berkeley
Reading: 580-710
Math: 620-740
Writing: 590-710</p>
<p>Overall: 1790 - 2160</p>
<p>USC
Reading: 600 - 690
Math: 590 - 690
Writing: 620 - 710</p>
<p>Overall: 1810 - 2090</p>
<p>So USC's averages are actually LOWER than those of Berkeley's, despite the fact that Berkeley has a much larger undergrad population.</p>
<p>so where's your proof that USC's superscoring inflates the scores by 30? </p>
<p>i read the medS stats and you're right cal is more successful than ucsd. </p>
<p>but i re-looked at sat scores for stanford, hm, tech and all three are higher than cals's so i'm not sure what "evidence" you're pointing to that shows cal sat's higher than those schools</p>
<p>appstressin,</p>
<p>I've only seen articles, but I can't find any right now.</p>
<p>Either way, we MUST account for superscoring when comparing SAT scores. Otherwise, we cannot account for the change in the y value without understanding that there will be some error involved.</p>
<p>
[quote]
i did say that HM and cal tech students are smarter than cal. i'll stick by that. take the average take the top take any segment of the sb at HM or tech and you'll find a smater (iq) student than the same student at cal.
[/quote]
</p>
<p>You're assuming that SAT is an accurate indicator of IQ. Which it may or may not be...</p>
<p>troll... right.
here's a little fact:
i would put more time into defending myself through exceptional data scouring and regressional modeling...but hey, i'm a very busy person. i have work to do and i don't have time to argue with some ftards about SAT scores and junk.</p>
<p>note to HS peeps: make up your mind by yourselves. don't listen to anyone here (including me). go out and do some research... look at the numbers and 3rd party reviews.</p>
<p>i obviously must be out of my mind...</p>
<p>...back to flywheels and induction motors...</p>
<p>vicissitudes and UCLAri,</p>
<p>So now it's become 30 more points for each individual subject caused by superscoring? Desperate and laughable. I didn't take SAT officially, so just a question here: You guys did all 3 parts in one sitting?</p>
<p>eastcoastbound,
It's clear you are still an undergraduate student and doesn't have much research lab experience. I know double digit Berkeley Professors. Some stories are appalling (again). A Berkeley associate Professor told me their department has a lab office available recently because a guy retired. There are more than 5 faculty members applied for the room. One of them was waiting in line for almost 10 years. It may be a bit exaggerated, or some departments are better than others. But it's simply not true that Berkeley has much superior research facillity. In the end, the guy sighed, "USC really has space!".</p>
<p>QW553,</p>
<p>In regression analysis when we have multivariate regressions, we have to do something to control for one of the variables when dealing with another one. 30 points is just an average increase per section that seems to be floated around as the average benefit that superscoring allows.</p>
<p>Unfortunately, we sometimes have the problem of lacking the data we need to correctly control for the average score increase that superscoring allows for.</p>
<p>
[quote]
So now it's become 30 more points for each individual subject caused by superscoring? Desperate and laughable. I didn't take SAT officially, so just a question here: You guys did all 3 parts in one sitting?
[/quote]
</p>
<p>How else do you propose that we control for the added variable? It's well-known and agreed upon that superscoring raises averages. </p>
<p>I took the old SAT, so it wasn't three sections. It is now.</p>
<p>"Some stories are appalling (again). A Berkeley associate Professor told me their department has a lab office available recently because a guy retired. There are more than 5 faculty members applied for the room. One of them was waiting in line for almost 10 years. It may be a bit exaggerated, or some departments are better than others. But it's simply not true that Berkeley has much superior research facillity. In the end, the guy sighed, "USC really has space!"."</p>
<p>What are you trying to prove? That Berkeley has a terrible research facility or that Berkeley's research programs are so prestigious that faculty members would endure 10 years without a proper workspace just to be a part of the research programs?</p>
<p>It is agreed upon that superscoring raises averages, but until I see data, 90 points total sounds reallllly high. 30 points for all three sections combined sounds much more in the ballpark.</p>
<p>Superscoring accounted for, my version (until you can provide data that it is worth 90(!) points)</p>
<p>Berkeley
Reading: 580-710
Math: 620-740
Writing: 590-710</p>
<p>Overall: 1790 - 2160</p>
<p>USC
Reading: 620 - 710
Math: 610 - 710
Writing: 640 - 730</p>
<p>Overall: 1870 - 2150</p>
<p>Even then, what we see is that the top of the 50th percentile of students at Cal are clearly as strong as those at USC.</p>
<p>This furthers my point that the mean is similar, but we just see more variation.</p>
<p>I just checked my old SAT score reports on CollegeBoard.
For each subject, Collegeboard gives you a range of scores that best represents the specific score that you received.</p>
<p>For my math score of 800, my score is best represented in the range of 770-800. For my CR score of 730, the range is 700-760. For my writing score of 750, the range is 730-770. From what I see, the range is (+/-) 30 in math and CR and (+/-) 20 in writing.</p>
<p>Math +/- 30
CR +/- 30
Writing +/- 20
That adds up to 80 points, so I don't think an adjustment of 90 points is outrageous.</p>
<p>QW553, lol your post is hilarious. only if i havent worked at a lab for almost 3 years, my name is not on two PNAS paper and one more (nature, science, or cell, depending where my PI will submit it) to come this fall the latest, or MIT and other number of schools spend hundreds of dollars wanting me to go to their grad school, yes i would consider myself a worthless undergrad without any research experience. i've been a lot of job talks of my department and my PI is currently on the search committee (in fact, i just went to one by a gentleman named Robert Best), i have to tell you, unless you are very well connected and doing very good research, there is little hope u will ever receive a faculty position at berkeley.</p>
<p>i cant edit anymore? i just want to delete my records because i dont want to start a ****ing contest again. i'm not going to involve any argument in this thread. can people just state their facts, then argue the implication of them in a logical way. plus, if u really have to argue SAT scores, please include means, standard deviation, confidence interval and a proper hypothesis analysis. and i suggest P<.05. and i am not going to answer any more ad hominem attack.</p>
<p>
[quote]
Then why did I show that the top 25% students at Berkeley, which is over 5 times the entire size of Harvey Mudd's undergrad size, completely crushes Harvey Mudd students when it comes to high school GPA/rank and SAT scores? It seems to me that there are at least five times as many people at Berkeley who are "smarter" than the undergraduates at Harvey Mudd. I used SAT scores and class rank data collected and published by the respective schools to make such an assertion. What evidence have you given at all for anything you are saying besides your own belief in them?
[/quote]
</p>
<p>"Completely crushes"?!?!?! haha who's the one making false, all-encompassing claims?</p>
<p>From what I saw, the SAT scores of the average Mudder are still higher than the SAT scores from a 75th percentile Berkeley The average Mudder's scores are even higher than a 87.5 percentile Berkeley student. Like I mentioned before, our math scores don't look as good because we hit the top of the scale at 75th percentile with 800. </p>
<p>You're using sketchy statistical analysis. We all know that SAT scores at colleges follow a bell curve model. So you are cutting the top fourth of that bell curve off and failing to mention that a majority of the SAT scores are more lower than higher. The lower scores scores in the lower 700s occur more frequently. So you compare that with the mid 50% of Mudd students, which means that at our 75th percentile the bell curve height is still pretty high. I think I'm going to make graphs and I'll get back to you.</p>
<p>With the whole GPA thing, obviously something is going on there. Do you really think 99% of Berkeley students were in the top 10%? By your logic that would make Berkeley more exclusive than HYP. Even though Mudd only has ~90% of students in the top 10%, I think it's obvious that the AVERAGE Mudder performed at a higher level than the AVERAGE Berkeley student.</p>
<p>I keep saying AVERAGE because you guys always go "you can't say EVERY student!!" when I keep saying AVERAGE every post.</p>
<p>Hmm, I feel like the constant warnings of the burdens of being an out of stater at a UC seem to fall on deaf ears everytime. Take it from me, an out of stater at Cal - if you can get a grant/merit aid from a private school you'll have a much easier time. USC is may be a bit more generous in terms of grant aid, if great stats are in the mix. But just to be clear, the UC system is by design for California residents, so the truth is that you won't get help with regards to finances.</p>
<p>Atomic: I trust you didn't take AP Stats, or you would realize that the SAT scores of MIT, CalTech or Mudd would definitely not be a bell curve, but would be skewed left since so many those kids easily score an 800, and would score higher if CB made the test harder.</p>
<p>apps:</p>
<p>In reviewing some other threads, I was surprised to see that you are a mom (kid at 'SC). The reason it was surprising is that your arguments are about as logical and well-crafted as many of the high schoolers who write and post.</p>
<p>
[quote]
The volume and intensity of Cal people to promote their school....showing that Cal is ... best in a lot of areas.
[/quote]
</p>
<p>WHERE, has anyone of this thread, ANYONE, said such a thing? Even CalX did not say Cal was the best at everything.</p>
<p>Methinks the lady doth protest too much! :rolleyes" </p>
<p>Perhaps the shoe fits, TrojanMom?</p>
<p>
[quote]
....actually makes them look insecure or have some type of chip on their shoulder....
[/quote]
</p>
<p>"Atomic: I trust you didn't take AP Stats, or you would realize that the SAT scores of MIT, CalTech or Mudd would definitely not be a bell curve, but would be skewed left since so many those kids easily score an 800, and would score higher if CB made the test harder."</p>
<p>Had you taken any calculus-based stats you would know that the sample variance distribution of any randomly sample (with N>40) is a bell curve. You don't scare me.</p>
<p>Here's what I was saying before put into a graph:
<a href="http://img490.imageshack.us/img490/2276/curvesdv4.jpg%5B/url%5D">http://img490.imageshack.us/img490/2276/curvesdv4.jpg</a></p>
<p>Well considering we are talking about the middle 50% at HMC, it does basically follow the bell curve (if I give you guys the edge of 75th percentile =800, even though as you were implying, it is probably somewhere in the 60s percentile since the SAT scale is maxed out).</p>
<p>Look, I'm trying to present a simplified version of the SAT scores just to show a point. Obviously it won't fit the exact model, but it shows my point at least.</p>
