Almost everyone who goes to grad school in physics does so thinking that they will one become a tenured professor at a large university. And anyone who has been around a physics graduate program for a while knows that for most of us that is simply not going to happen. A recent book by Paula Stephan entitled "How Economics Shapes Science" shows that 23% of physics PhD's hold tenure-track appointments 6 years after their PhD, which means that less than one-quarter of those that survive grad school will get to be a professor in the way they imagined when they started.
That's a dismal way to look at grad school, but I've made a strong assumption in the preceding paragraph that some of you probably already noticed. I assumed that every grad student wants to have a tenure-track position at a large research university. It turns out that what grad students want is far more diverse than that, and that it changes over the course of the average student's grad school experience. A recent study by a pair of management experts looked at exactly those questions and the results are fascinating. I recommend reading the entire paper as it's very well-written and accessible, but here at the two points that I found most interesting.
First, they showed that even when asked to disregard the likelihood of actually getting a job in one of six areas, only 37% of beginning grad students in physics rated a tenure-track faculty position at a research university as "highly desirable" and that the percentage of students with that opinion didn't change over the course of grad school. Note that the percentages can add up to more than 100% because respondents could indicate multiple areas as "highly desirable".
This indicates that new physics PhD's are not facing 1-in-4 odds of getting a tenure-track position, but rather that the odds are more like 1-in-2, assuming that there was little overlap between those that liked the "faculty-research" and "faculty-teaching" options.
The second highlight is the way that students' opinions of the six career paths change over the course of grad school. They tracked what percentage of students rated each career path at the end of their graduate careers versus their ratings when they entered grad school.
I find it very encouraging that most grad students realize that there are good things to do with a PhD in physics other than become your adviser, and that grad school actually does help open minds to other options.
Sauermann, H., & Roach, M. (2012). Science PhD Career Preferences: Levels, Changes, and Advisor Encouragement PLoS ONE, 7 (5) DOI: 10.1371/journal.pone.0036307