Attrition rates for PhD programs?

[newpsyche]

Hi, all! A rather specific question that is surprisingly not answered (or at least not clearly) in other forums: What would you say is that typical attrition rate for PhD programs? Specifically, in social sciences. I heard recently about a grad student who is leaving, and this startled me--I knew it was possible, but didn't know it actually happened. Now I'm unsure how common it is.

[fuzzylogician]

Attrition rates at my program are particularly low. The last time someone quit was before I joined the program (three different people from the incoming class of 2005 left during/after their first year, from what I hear). But I think it's not usual for there to be many more people who don't graduate, or who leave with a Masters at some point. I'll leave it up to more educated people than myself to give you numbers and links.

[Eigen]

I've always heard about 50%. Obviously higher and lower for different programs, but somewhere around there. This combines people failing out early, people leaving with masters, and people just drifting off ABD.

[Max Power]

You can pretty much calculate your program's attrition rate by counting the average incoming cohort size and counting how many people graduate each year. All the people on different time frames across cohorts should reasonably cancel each other out.

As Eigen said, there are several kinds of attrition. As programs get more selective, the failing out becomes a smaller part of it. They have high quality students coming in who are less likely to fail. There's no accounting for why people may decide to leave at other points or drift off ABD. Some people leave because they have a change of heart about academia. A lot of people arrive in grad school with only a vague idea of what an academic career is about. Some drift off ABD because they never learned how to be self directed in tasks and don't do well without having faculty members hand down assignments and prompts.

[TakeruK]

When I was visiting grad school, this stat was usually discussed. Often, it was the director/head of the program/department that told us average incoming class size, averaging graduating class size, and attrition rate, etc. However, almost all the schools quoted a "finishing rate" of something like 90-95%, which was suspiciously high. Upon further questioning, we then realised that they really meant "90-95% of the people who WANTED to finish their PhDs actually did" -- so basically only 5-10% of people actually fail their orals/quals/comps the max. number of times (usually twice) and were forced to leave. They admitted that a larger number (although didn't give stats) of people decided to not try the exam again a second time and left with either nothing or a masters! So, if you hear numbers from a school, make sure you know what is actually being counted!

My current school is in Canada and we all enroll in masters programs (2 years) specifically then apply for PhD programs (3-4 more years) at the end of our masters. Out of my cohort, ~40% of them have already left graduate studies, ~40% have concrete plans to continue onto a PhD, and the remaining ~20% are not yet sure what to do after a masters. In Canada, it's not unusual at all for people to decide that a masters is good enough -- a MSc opens a lot of job opportunities not available to BSc grads, but a PhD doesn't really open up more jobs than a MSc (except for tenure-tracked positions of course). It is unusual for someone to leave without getting a masters though. Overall, these numbers seem to also agree with Eigen's estimate of ~50% attrition.

[newpsyche]

Yep, I definitely think it depends on the program. Of mine (in the last five or six years), two have left after their masters and the rest have stuck with it.

[Xanthan]

http://www.phdcompletion.org/index.asp

"The Ph.D. Completion Project is a seven-year, grant-funded project that addresses the issues surrounding Ph.D. completion and attrition. The Council of Graduate Schools (CGS), with generous support from Pfizer Inc and the Ford Foundation, has provided funding to 29 major U.S. and Canadian research universities to create intervention strategies and pilot projects, and to evaluate the impact of these projects on doctoral completion rates and attrition patterns. An additional 25 partner universities are currently participating in various aspects of this project. The Ph.D. Completion Project aims to produce the most comprehensive and useful data on attrition from doctoral study and completion of Ph.D. programs yet available." (from their "Overview" page)

The following is from their attrition data, table 6: 10 Year Cumulative Attrition, Completion, and Continuing Rates by Broad Field.

(I can't get the image to work in the post preview, so here is a link: http://postimage.org/image/dlwt847it/)

Thus, after 10 years, in the life sciences (my area), 26.2% will have dropped out, 62.9% will have graduated, and 10.9% will still be continuing. I think that last number is the scary part. Getting a funded PhD is a good deal, but the longer it takes, the less good of a deal it becomes.

[Eigen]

Interesting to see how old that data is. It's based on conorts starting in 1992-93 and 1995-96. Also interesting is that in most fields, there was a significant drop in the 10 year completion rate (1-3%) going from the former to the latter.