<p>Another example of drawing conclusions from non-representative samples. This is why newspapers were printed with headlines of DEWEY WINS BY A LANDSLIDE, when Truman eventually won the election. They hadn’t done proper sampling in polls.</p>
<p>No one from Payscale has ever called me up to ask what I earn, and I’ve never filled out one of their forms. Most of those who fill out their forms are looking for jobs, and that is by no means going to give them a representative sample of Princeton, Harvard or Stevens grads. Not to mention that the data needs to be sliced and diced by career choice and countless other factors to be relevant.</p>
<p>Throw out the Payscale data – it’s meaningless, just as the StudentReviews.com data that says that 60% (or whatever) of the generally miserable, whining students who decide to post on their web site wouldn’t return to a particular school if they had to do it again is meaningless. Data obtained without using rigorous sampling and other surveying techniques to ensure that the data represents the full population and like populations is worthless. </p>
<p>The closest thing I’ve seen to a real salary survey of Princeton graduates was my 25th reunion survey. That indicated that 18% were earning over $500K. That’s probably a more representative sample (because the entire class was sent a survey form), but we still don’t know if it was truly representative of the entire class.</p>
<p>I would guess that Harvard, Yale, Brown and other Ivy grads have similar earnings as do the Princeton grads, and even more so, that the student who chooses to go to Dartmouth instead of Princeton will earn just as much in his or her career. The differences in the caliber of education amongst these schools is slim, and there are more opportunities at all of them than any student can take advantage of.</p>