<p>I will also note that I don’t think the US News rankings are very good other than as an extremely crude metric. Revealed preferences give pretty strong evidence that HYPSM are the top 5 and I think are much more methodically sound way of ranking colleges than the arbitrary factors and weights of US News… </p>
<p>
</p>
<p>“Similar” and “much weaker” are both subjective, and I’m not sure if you’re using “statistically” rigorously. Objectively, MIT students’ scores are higher. Whether you consider the difference to be similar enough or rather large is up to you.</p>
<p>
</p>
<p>US News is not my primary source for determining what I think of schools.</p>
<p>
</p>
<p>Compared to MIT, I think they are probably stronger in physics, chemistry, and maybe math, where I think MIT is stronger in engineering disciplines, biology, and humanities.</p>
<p>If we estimate the median SAT score of both schools by summing the averages of the 25th and 75th percentiles for each section: MIT’s median is 2260 while Berkeley’s is 2035. Given that the SAT is out of 2400 these hardly seem “statistically similar”.</p>
<p>Mmmmm but…I don’t think that is how medians work.</p>
<p>@gravitas2 If a student were to get accepted to Berkeley’s EECS program, they would NOT be choosing MIT over Cal, at least not if he were smart.</p>
<p>@lidusha I’m aware that it’s only a estimate but I think it’s a pretty reasonable one and to the extent it’s systematically biased it should be biased for both schools. Why do you think it’s a poor estimate?</p>
<p>@"UMTYMP student" Couple of things for you to consider in regard to your SAT scores.<br>
- They’re for all undergraduate students and not Engineering students. I don’t have the stats but Berkeley’s engineering program is universally recognized as one of the top programs in the world and their students reflect this.
- MIT has less than 4,500 students while Berkeley has 26,000. Which do you think is easier to boast higher average SAT scores?
- MIT superscores SATs while Berkeley does not. Do you suppose that this may have some impact on the final numbers when one school cherry picks the best numbers while the other does not?<br>
- I suspect Berkeley students aren’t losing any sleep if couple of MIT students think they’re more than all that what’s merited. They might get a good chuckle out of it though…
- I think we can all agree that US News is not a good indicator. I mentioned it as a reference of ‘objective’ reasoning since there seems to be a void here. Of course what you bring up as your source of indicator is a mere popularity contest which seem to be perpetuated by your favorite news rag - which you claim to be not a fan of. Or was that @PiperXP?</p>
<p>1) Many of Berkeley’s program are recognized as very good in terms of research yet the level of undergrads at Berkeley is not as high as other schools. I’m willing to believe engineering students at Berkeley are better than the average students at Berkeley but not by a huge margin unless someone has evidence showing this.
2)The claim is that the average student at Berkeley is much weaker which is perhaps largely to Berkeley admitting so many students.
3)I’m not sure that would affect individual SAT sections although I admit it’s possible. I think the effects of superscoring are small compared to the difference in SAT scores between MIT and Berkeley though.
5)In what way are US News rankings ‘objective’? I don’t think US News is perpetrating the revealed preference rankings if that’s what you’re referring to. I’m aware that some people don’t give much weight to revealed preferences although I think it does a good job of aggregating large amounts of information about schools weighted by what students actually making decisions deem important.</p>
<p>
This is actually not what the data shows – very few students choose UCB over MIT each year. </p>
<p>The year for which I have data, the number is actually 4. This is despite the very large number of California residents who are accepted to MIT each year – I think it’s safe to assume that many more than four students were accepted both to MIT and to UCB EECS. More students chose Michigan and Georgia Tech over MIT that year than chose UCB, and I have no reason to suspect that year’s data was unusual.</p>
<p>@mollie This is sorta tangential but can you share the data on where students who declined MIT went that year? Like how many went to Harvard, Yale, Stanford, etc?</p>
<p>I can dig out my backup drive tonight and see if I can find the original file. I did post [url=<a href=“Prestige/Expensive vs. Good/Cheap - #190 by molliebatmit - Parents Forum - College Confidential Forums”>Prestige/Expensive vs. Good/Cheap - #190 by molliebatmit - Parents Forum - College Confidential Forums]this[/url</a>], a few months after I did the original analysis:
</p>
<p>EDIT: And from that thread, it appears I misspoke above – Berkeley did not get four MIT admits that year, they got three.</p>
<p>“If a student were to get accepted to Berkeley’s EECS program, they would NOT be choosing MIT over Cal, at least not if he were smart.”</p>
<p>Cadave: why do you say this? My son got into Berkeley’s EECS and is planning on turning it down for either MIT or Stanford. And we are in-state. (This is not a rhetorical question; I’m truly interested in why you think so.) Although I’m confident that the curriculum and employment recruitment is probably comparable at both schools, we think that the educational experience, enrichment opportunities, and ease of access to faculty/research, etc. will differ greatly due to the public vs. private nature of the schools. </p>
<p>@"UMTYMP student" ,
If sat is your yardstick,
at mit 25% score is = 670+740+680=2090
Ucb 75% score is = 720+770+750=2240
Mit students at 25% = 800*.75=600
Ucb stduent @ 75% = 4000 * .25 = 1000
So ucb has 1000 students having more than 2240 sat score
And mit has 600 students more than 2090
So, do your math and figure which has high sat score students </p>
<p>Calbearsmom, ucb and mit are both great universities but it all depends on how rich one is or how much aid one gets. </p>
<p>@moshot - Where on earth are you getting your stats from? I’m using the actual webpages for UCB and MIT.</p>
<p><a href=“Admissions statistics | MIT Admissions”>http://mitadmissions.org/apply/process/stats</a>
<a href=“Student Profile - Office of Undergraduate Admissions”>http://admissions.berkeley.edu/studentprofile</a></p>
<p>MIT 25% - 750+690+700 = 2140
UCB 25% - 660+630+640 = 1930</p>
<p>“So, do your math and figure which has high sat score students”</p>
<p><a href=“MIT Institutional Research”>MIT Institutional Research;
<a href=“http://opa.berkeley.edu/sites/default/files/2012-2013.pdf”>http://opa.berkeley.edu/sites/default/files/2012-2013.pdf</a>
As reported in CDS.
Also need to look at 25% of mit and 75% of ucb </p>
<p>^@moshot is comparing the bottom quartile of MIT (25%) with the top quartile of Berkeley (75%). I’m not sure what point he is trying to make.</p>
<p>Berkeley has more students overall (apparently 5X as many), so this is why they may have more students above a certain SAT score thresshold.</p>
<p>It’s like saying China has better living conditions than the U.S. because they have more people making $50,000 a year. Well, China has more than a billiion people and the U.S. has 250 million, so that skews the results.</p>
<p>@moshot - Yeah, I don’t see the point in comparing the bottom of MIT with the top of Berkeley. There’s obviously overlap in SAT scores, but MIT SATs as a whole is still clearly well above Berkeley.</p>
<p>Another major fallacy in moshot’s post is that the 25th/75th percentiles for composite scores is not the sum of the 25th/75th percentiles for each section. The 75th percentile of SAT scores at MIT is almost certainly less than 2380 and similarly I would expect the 25th percentile to be more than 2140 (using the more recent MIT admissions data). MIT also has considerably more than 800 students in each class.</p>
<p>The perhaps more serious problem with moshot’s analysis is that given the revealed preference data it is pretty clear that there is something very different between MIT students with ~2100 SAT scores and Berkeley students with ~2100 SAT scores. MIT students with low SAT scores for MIT will have some distinguishing feature or they wouldn’t be admitted.</p>
<p>“Calbearsmom, ucb and mit are both great universities but it all depends on how rich one is or how much aid one gets.” Yes, I totally get this. </p>
<p>But cadave doesn’t mention $$ at all in his/her assertion that “if a student were to get accepted to Berkeley’s EECS program, they would NOT be choosing MIT over Cal, at least not if he were smart” – so I assumed that he was referring to other non-financial reasons (especially since in-state vs. OOS wasn’t addressed either).</p>