Once you get your doctorate, the logical step for most STEM graduates is a postdoctoral position. There aren't many of these at all, most departments have doub

Once you get your doctorate, the logical step for most STEM graduates is a postdoctoral position. There aren't many of these at all, most departments have double the number of PhD students than they do postdocs. So right off the bat, half the students will be finding jobs elsewhere. Industry (hardware/software/analytics/whatever) pays more, is often more fulfilling and although your work isn't completely open it is usually more obviously relevant than your PhD. So 60% will find tenure track jobs? Well that's obviously true. If there are only about half the postdoc positions compared to the number of new PhDs then only half the PhD students will be getting tenure track jobs. Most postdoc positions are 2-3 years and are filled by newly minted PhDs. The logical evolution for most people is to do a postdoc (or two) and wait until their research is renowned enough for the university to offer a lectureship. You do that for another few years until they offer you a Readership, wait a bit longer and you get your coveted Professorship. If you're really lucky and you're top of your field you can hit assistant professor by 30-35 in the UK. Until you get the all-important lectureship offer you're a contract worker and there is always the prospect that funding will be cut and you won't be able to renew your position. Remember that once you've done your postdoc there are even fewer positions to be promoted into. Doing a PhD has largely put me off the idea of staying in academia, I think this is becoming more common. Being a professor would be nice, good pay (£60k min) lots of travel, but it's a very stressful job, you have no time to do any real research beyond project management and by that time you're thoroughly institutionalised. Should point out that my experience is from physics, ymmv. On the plus side I don't know anyone who has struggled to find a job from my department; plenty go into research in non university groups like Microsoft, the British Antarctic Survey, the Met Office, etc.