<p>Ranking of institutions on CC, and indeed in a lot of other cases, tend to focus on selectivity of undergrads. Harvard has arguably the highest selectivity and is number 1. There are a lot of people that would hold though that other universities -- and indeed LACs are better at training undergrads. This undergrad selectivity is paramount. People point to a vague prestige factor or one study of cross-admit selection to justify a certain ranking of schools.</p>
<p>On top of this selectivity is a second measure: the academic productivity of the institutions. The best proxy for this is the strength of graduate schools.</p>
<p>Normally these two things mesh fairly well; in many fields, though not all, Harvard is tops and is also still the most selective undergrad.</p>
<p>The USNWR ranking tries to squeeze both these basic criteria together. There are big problems with this. 1) Schools like Dartmouth or some LACs may be under-rated because they don't have huge academic productivity, but are very selective. 2) Public schools are almost by definition less selective (on average; there are exceptions -- for example, engineering schools), though not across the board; indeed as the OP points out the top echelon of students at top publics may outperform their peers at many selective private schools. But they can have a core academic richness far surpassing schools with higher undergrad selectivity.</p>
<p>A unidimensional ranking is pointless. Berkeley is one of the best universities in the world, bar none. It is strong across the board -- arts, sciences, engineering and if you count UCSF (let the debates begin) medicine. This refers to its core academic productivity. In terms of undergrad selectivity overall, it doesn't match up to some of its academic productivity peers or academic lessers. </p>
<p>We need two axes. On one axes is breadth and quality of academic achievement. The highest ranking universities in terms of across-the-board academic breadth and quality are:</p>
<p>Stanford (full-service powerhouse)
Berkeley (full-service powerhouse)
Harvard
MIT</p>
<p>In terms of selectivity, there are plenty of posts on CC that parse this out, and the usual suspects would be:</p>
<p>HYPSM (most selective schools)</p>
<p>It's sort of pointless to argue Berkeley's position on a unidimensional ranking. Those who hold undergrad selectivity key will demerit Berkeley, even though its one of the best universities in the world. Those who argue against the proponents of a selectivity-focused ranking are arguing for essential academic productivity, and that is a fair argument, IMO.</p>
<p>The USNWR peer assessment is a measure used to determine, by proxy, academic achievement and broad academic excellence. Those who argue against its use, again, are essentially saying overall undergrad selectivity is paramount. They say it slants things favorably toward Berkeley as if core academic productivity is somehow illegitimate or a sleight of hand; this strikes me as kind of ridiculous since I am sure they would not make the same arguments in other cases that their pre-conceptions. Would they say the same thing about other things that clearly slant the rankings in favor of private institutions, things such as endowment sizes? Berkeley gets a lot of state funding, funding that in some senses provides an equivalency with a large endowment. That, however, is not counted. Also, Berkeley and UCSF are among the largest recipients of a wide variety of government and private grants.</p>
<p>Having gone both public and private, I'd say the privates are much more efficient generally in purveying a brand image. They are often over-rated. Berkeley overall is under-rated on the unidimensional scale, IMO. And those who argue its over-rated are arguing against its demonstrated primacy in core academic productivity, as I said.</p>
<p>In the end, who cares? The upper segments of students at Berkeley, at the very least, will do just fine. They are, after all, being taught by the leaders in their respective fields.</p>