A PhD should also be about being honest in ones arguments in the popular press. PhD's in Education may take 13 years (is that the median?) because most student

A PhD should also be about being honest in ones arguments in the popular press. PhD's in Education may take 13 years (is that the median?) because most students are part time. In good graduate programs that provide guaranteed funding for students, 4-6 years is customary (and funding sometimes runs out at year 5). It's hard for me, as an academic, to trust an author who twists statistics to make his case. And what does it mean to improve society? Basic biomedical science rarely promises improvements, but has given us revolutionary technologies (the structure of DNA, the genetic code, cloning, transgenic animals, etc, etc). The essence of basic science is that you do not know what you are going to learn, or whether it will be useful. You cannot plan unimagined discoveries. Should graduate students start doing "research" sooner? That is certainly the standard in biomedical research, where very few courses are required. But the consequence is that students know a lot about what they have done, and very little about other disciplines. How does a graduate student in biology become a computational biologist without learning programming and algorithms? The comments about having research experience are spot-on. And many of the frustrations people describe could be reduced by learning more about one's advisor before joining the group.