1/3 making a mistake when entering grad school?

[Loric]

Are they really the top applicants if 1/3 are not finishing? Wouldn't the top applicants nearly all finish?

 

Go ask the philosophy forum, they're convinced that admissions is a lottery.

[St Andrews Lynx]

As has been said already in this thread, a lot of the reasons that people don't finish have nothing to do with the applicant's skills or abilities.

 

It's also the case that if you are a "top applicant" it means (i) you're smart (ii) well-qualified (iii) have gathered work/research experience in your chosen field (iv) you've cultivated good LORs before entering grad school. What I've seen with top applicants like those in my field is that they realise they've actually got a good stab at the professional job market and decide they want to get on the career ladder now, rather than wait 5 years. 

[Loric]

But think there's no common ground between those who quit..? Nothing that could be found and used as an indicator? No precursor traits?

 

So i'm to believe that dropping out of a grad program is a spontaneous and uncontrollable/unavoidable event?

Edited January 4, 2014 by Loric

[Loric]

And specifically.. why should graduate programs take in people who are unwilling to wait and finish their degree before entering the job market?

 

Those types seem like they'd be the first target to eliminate from the applicant pool. Companies go out of their way to avoid hiring people who are going to quit. On boarding is one of the most expensive things related to filling job positions. Presumably procuring and vetting new graduates is just as costly for an institution.

 

Wouldn't getting rid of the "will quits" lower the costs across the board for everyone?

Edited January 4, 2014 by Loric

[Eigen]

Money doesnt just come out of nowhere though. If someone who has a 1 in 3 chance of dropping out is funded, that money is not given to someone else. It's a lot of wasted funding.

 

If we're talking about STEM fields, that money goes to people who are getting paid to do research off of grants, or are teaching. So it's not waste of funding, from the standpoint of the person paying. 

 

IE, if someone spends 5 years doing research and ends up with a MS, the person paying them for those 5 years probably doesn't really care. If they weren't a productive researcher and teacher, they wouldn't have been kept on that whole time. So they were a good employee during their 5 years, and it doesn't really have a negative connotation to end up with an MS, if that's where your life and career are going. 

[Loric]

But again, hiring someone new costs a ton of money in and of itself. If you can retain a current employee - or better yet, dont hire a quitter - you save copious amounts of money.

 

It is not a simple swap. You lose someone, you lose money. It is not free to replace someone.

[DerpTastic]

If it's illness, probably not. If they end up leaving for another job, you might be able to see that they're not 100% invested in grad school. However, no one would tell the school they're applying to that they aren't all that invested in school. Additionally, I think the amount of people willing to go through the stress and pressure of a PhD program are most likely pretty dedicated to it at the time they apply/begin. Many end up getting burnt out, but that could be years down the road.

 

There might be shared traits for people that end up dropping out, but do you think these tests would be extremely accurate? What if you end up not accepting people who would have finished that have stellar applications? 

 

I'm not sure if rates are low enough that I would really be worried.

[Loric]

And a note:

 

SHRM, the Society for Human Resource Management, estimated that it costs $3,500.00 to replace one $8.00 per hour employee when all costs — recruiting, interviewing, hiring, training, reduced productivity, et cetera, were considered. SHRM’s estimate was the lowest of 17 nationally respected companies who calculate this cost.

[Eigen]

But think there's no common ground between those who quit..? Nothing that could be found and used as an indicator? No precursor traits?

 

So i'm to believe that dropping out of a grad program is a spontaneous and uncontrollable/unavoidable event?

 

I wouldn't say spontaneous and uncontrollable, but I haven't seen a lot of common ground other than life issues. It's usually not people who aren't doing well in the program, or who wouldn't likely finish if they stuck it out. Of the people I know that have had to leave, it's been:

 

Having to go home and take care of sick parents

Cancer/Significant illness

Offer of a dream job position from a contact somewhere

Sick spouse/spouse getting a job across the country

Sick children

 

And then there are the people who work for a few years and find out that while they could complete the degree, it won't really lead to a career that they want or are interested in. A lot of those people end up leaving and starting up again in another field. This might be avoided, but it's usually something that requires a lot of academic socialization to see, or being exposed to academia. 

 

You also have people leaving due to advisors moving, grant funding running out, or finding out the program/advisor they are working for is anywhere from a really bad fit to toxic. 

 

I think you leaving one of your programs would fall under the last category, from what you've posted. 

[St Andrews Lynx]

I've seen personality traits that lead me to suspect somebody is going to quit their PhD program. But these traits may not come out in the applications or LORs (the traits might not be apparent until they're in a grad school environment).

 

I've also watched a very smart candidate (has research experience, a fellowship, does well in courses) make the mistake of trying to get into a PI's group to the extent of neglecting all other options...only to watch that PI's funding fall through, and find themselves placed in like their 3rd choice group. I've seen other talented "top applicants" have similar issues trying to get into their group of choice (the PI just liked someone better, there was really limited funding). I wonder if those students will still be here in 4 years' time...

 

I would say that's it is all well and good being a "top applicant"...but even the most fantastic, experienced grad student will require optimal conditions (say a hands-off supervisor, small research group, living in the same city as their partner). Put that fantastic grad student in sub-optimal conditions and they're going to be miserable and contemplate dropping out. Of course the admissions committee tries to match grad students to PIs in the admittance process, but the adcom can't control how many or which applicants eventually accept the school's offer. 

 

Side note: I'm a chemist, one of the first things you learn as a chemistry student in high school is about entropy (disorder). One of the Laws of Thermodynamics says that the entropy of the universe is constantly increasing. Most of what you see out there is just disorder. 

[DerpTastic]

If we're talking about STEM fields, that money goes to people who are getting paid to do research off of grants, or are teaching. So it's not waste of funding, from the standpoint of the person paying. 

 

IE, if someone spends 5 years doing research and ends up with a MS, the person paying them for those 5 years probably doesn't really care. If they weren't a productive researcher and teacher, they wouldn't have been kept on that whole time. So they were a good employee during their 5 years, and it doesn't really have a negative connotation to end up with an MS, if that's where your life and career are going. 

 

This is also a good point. 

 

 

But again, hiring someone new costs a ton of money in and of itself. If you can retain a current employee - or better yet, dont hire a quitter - you save copious amounts of money.

 

It is not a simple swap. You lose someone, you lose money. It is not free to replace someone.

 

I'm wondering, in what way do you mean they lose a ton of money swapping people out? In wasting time to train the next worker, or losing an experienced worker for a less experience one?

 

I could see if you had tons of people leaving after a year that it would be a burden. But if someone leaves after 4-5 years, well, they were most likely going to leave after 6 years with a PhD anyway. You don't keep these workers all that long anyway.

[Loric]

If it's illness, probably not. If they end up leaving for another job, you might be able to see that they're not 100% invested in grad school. However, no one would tell the school they're applying to that they aren't all that invested in school. Additionally, I think the amount of people willing to go through the stress and pressure of a PhD program are most likely pretty dedicated to it at the time they apply/begin. Many end up getting burnt out, but that could be years down the road.

 

There might be shared traits for people that end up dropping out, but do you think these tests would be extremely accurate? What if you end up not accepting people who would have finished that have stellar applications? 

 

I'm not sure if rates are low enough that I would really be worried.

 

Don't think already not accept hundreds of people who would likely do stellar?

 

If 1/3 of the pool is quitters and there's more than 2x as many applicants as positions, you're already coming out ahead in any attempt to weed out the quitters (if at all successful).

 

3 slots. 2x as many applicants means 6 applicants.. 1/3 quitters.. you have 4 people who would stay on and finish. You have 3 slots, so if you weed out the 2 quitters you can have 3 slots who don't quit from the 4 applicants.

[Loric]

This is also a good point. 

 

 

 

I'm wondering, in what way do you mean they lose a ton of money swapping people out? In wasting time to train the next worker, or losing an experienced worker for a less experience one?

 

I could see if you had tons of people leaving after a year that it would be a burden. But if someone leaves after 4-5 years, well, they were most likely going to leave after 6 years with a PhD anyway. You don't keep these workers all that long anyway.

 

Replacing a worker costs 30-50% of the annual salary of entry-level employees, 150% of middle level employees, and up to 400% for specialized, high level employees

[DerpTastic]

Replacing a worker costs 30-50% of the annual salary of entry-level employees, 150% of middle level employees, and up to 400% for specialized, high level employees

 

What are these stats for though? I'd be tempted to say a TA/RA in a PhD program doesn't follow these stats, as it works a little differently, but I can't really comment. Generally, they are constantly replacing people as other graduate. It's not really the same as other jobs, which tend to hire people intending to keep them on long term.

[St Andrews Lynx]

And specifically.. why should graduate programs take in people who are unwilling to wait and finish their degree before entering the job market?

Tell me how the adcom would be able to predict that. 

[Loric]

There is the cost of a rolling program but every time you replace someone before the end of the program you're increasing the cost.

[VioletAyame]

First I think we need to distinguish between the cost for an individual to attend and complete grad school and the cost of any given program to admit said individual. It seems that the argument has been made against "not completing one's degree is wasteful for oneself" and now we're arguing the cost for the institution.

From what I just read in this thread, it looks like there are to types of reasons for non-completes; let's call them external and internal.

External reasons are things like advisors leaving, losing funding, family issues, etc. In short it is not the intention nor the fault of the grad students to drop out but instead the circumstances did not work out. I don't see any plausible way to foresee this, any indicator which adcoms can use to flag these candidates. They are, like you said, uncontrollable events.

Internal reasons are when students voluntarily drop out, for a job offer or a change in direction/discipline, or when students are not qualified. If it's a change in intention, it's not in the best interest of the programs to admit those candidates in the first place, and had they known they probably wouldn't have made the offer, but is there really a way to know? Most often times the candidates did not know this themselves; things just come up and they have to make a choice. Companies do try to minimize this from happening, but it does still happen, so whatever method they're using is not 100% effective, just like in academia.

The only case which I think which might call for better screening is when students don't qualify, but then again I don't know how common it is as the reason for non-completion. If it's a very low percentage, perhaps it's just the same risk inherent in any other endeavor and there is nothing more to be done. Adcoms are not omniscient after all, no matter how much they seem to be so to us applicants.

[Eigen]

If you want to speak to the cost to the institution, a number of top programs, at least in STEM fields seem to think it's cheaper to take on double the number of people you expect to stay, and weed them out in the first x years. You can see this with TT hires at ivy leagues, and you can also see this with top 10 STEM programs. 

 

I can think of several that take on ~200 new PhD students, and expect around half to leave in the first 12-18 months. Getting into the program is the first step, and then you compete for spots in labs, etc. 

 

Since it's done this way on purpose, I'd imagine it's not more expensive to the institution. 

 

Also, you seem to be making specious arguments with stats. The fact that 1/3rd of students don't finish a degree doesn't imply that the same stats can be applied to the whole applicant pool- ie, you can't say for sure that they should have taken someone else, and then they'd have 100% completion rate. 

[Loric]

That's just it - i think it's quite the opposite. There is not a lack of qualified applicants.. it's that many of those "qualified" applicants will jump ship at the first opportunity they see.

 

It's a basic characteristic of the "climber" mentality. They see something better, they'll go for it. In organizations that have well developed internal culture and promote loyalty you commonly wont find the climbers - they dont get hired. The companies will choose "less attractive" candidates than the climbers - who by their own volition have the better score and better pedigree - because those people will choose loyalty over the status, pay raise, etc..

 

You can spot a climber. For whatever reason, graduate programs aim for them instead of avoiding them.

[Eigen]

I don't think that's true, at least not in the STEM fields. 

 

It's quite common to see "the best" applicant turned down if they haven't demonstrated that they fit well at the school, or if the school thinks they're likely to take a better offer or try to "climb" out of the school. 

 

The same is definitely true for faculty searches in STEM fields, at least, where showing you want to stay at the school and like it is as important as your qualifications. 

[Seeking]

I think the comparison between a company and a university is not appropriate.

 

The primary aim of the Grad School is to train candidates in research and teaching skills, which these candidates can use in their future career. Whatever teaching and research employment they get in Grad School is by way of giving them the opportunity to hone their research and teaching skills.

 

On the other hand, companies hire candidates who already have skills and who are willing to use their skills for the benefit of the company. In the process they also hone their skills further or acquire new skills, but that's the not the main aim of the companies while hiring employees.

 

Precisely because the Grad School is into training the candidates in research and teaching, they don't necessarily regard it as a loss if some of their best candidates find a job half way through and leave. If they find a lucrative job, it is a matter of prestige for the Grad Program that even a few semesters' training they gave to these candidates has turned out to be fruitful. They may even use it as an index of their top-grade training quality to attract more top-grade candidates.

 

Grad School has well-conceptualized degree-programs of 2-5/7 years for candidates who would like to use the full range of training for the entire 2-5/7 years. But if some candidates can have a successful career with less than the entire term of a degree Program, it's not much problem for the Grad School. It means that they have been able to train these candidates so much that they could achieve their professional aim with less than the full-term training. In this sense, the Grad School has achieved its aim.

 

As for the lab work getting affected, there are always candidates available to carry further the work that was done by the previous candidate.

 

This is one of the reasons that often more candidates are taken than they can accommodate, with a view that some of these will leave. And this is the reason the most qualified candidates are selected - they are most likely to get the desired training in the shortest period of time.

 

In fact, the candidates who stay for a long period in Grad School are going to cost more because more resources are required to be invested on their training.

 

We need to understand that Grad School is not a regular employment in the sense a regular job in a company is an employment. The dynamics of Grad School are different from the job dynamics in the companies. 

 

PS - Deliberately hiring the less qualified candidates who will stay loyal by avoiding the top-quality candidates is what the Third World countries do, because loyalty at the cost of quality is more important to them. And it shows in the cumulative results the countries show. 

Edited January 5, 2014 by Seeking

[Loric]

You realize the cost of tuition far outweighs the cost educating students, right?

[Seeking]

Yes and you know that most candidates get tuition waiver in Grad School - which means the Grad School is paying for the tuition of most candidates. The longer they stay, the more the university pays for them - and also spends on their research and teaching training and is also giving them RA/TA salaries.

 

The Grad School is not just a business about tuition. As I said, its primary aim is to train people in research and teaching. It has to show how many people it trained and with what results. Candidates who leave for a lucrative job offer in fact help positively in this index - and they cost less than those who stay longer. 

Edited January 5, 2014 by Seeking

[DerpTastic]

I'm not sure I would consider it a huge lose for tuition costs. In general, they have a set amount of students they will pay for. The school knows they will lose some students. They still have these students performing research, working as a TA, (hopefully) publishing or helping professors that publish. Either way, the money isn't simply a waste for the school.

[TakeruK]

If you assume that a student who drops out has contributed nothing at all, then yes, losing a student is a waste of money. Also, if you assume that a graduated PhD who does not stay in academia is also zero contribution, then yes, people who don't go into academia also "wasted" the school's money.

 

However, in STEM fields, this is simply not the case. I would say the majority of the money/costs that go into having a graduate student is returned in the value of the student's research labour/output and any teaching duties. Most PhD programs realise that the majority of their graduates will not be working in academia. The good programs don't consider this a failure and instead want to train PhD graduates to have broadly employable skills (e.g. data analysis). It's not a failure if a PhD in Physics ends up working in a career that uses no physics at all but does rely on the computing and critical thinking skills developed by the program.

 

And if you want to say that there is a certain value in a PhD graduate remaining in academia and their future good work brings honour/recognition to their alma mater, then I think the increased value of 2/3 PhD graduates outweighs the loss of the remaining non-graduate. This is why some grad schools will accept graduate students who are in a committed relationship even if one partner isn't as qualified as the first. If one partner is very desirable to the grad school, they might prefer to take 1 great candidate and 1 mediocre candidate instead of 2 other good candidates. This happens for tenure-track position hires as well.