1. Not finding help here? Sign up for a free 30min tutor trial with Chegg Tutors
    Dismiss Notice
Dismiss Notice
Join Physics Forums Today!
The friendliest, high quality science and math community on the planet! Everyone who loves science is here!

Physics Credentials for those who made it in Academia (physics)

  1. Aug 5, 2012 #1
    Is there any source to show you the credentials for those who got accepted into academic positions (research/professorship) in physics? If not, what credentials would one expect to be the minimum for having a posibility at academia?
     
  2. jcsd
  3. Aug 5, 2012 #2
    Community college teachers: Masters degree (although it can be in any topic)
    Online universities like University of Phoenix: Masters degree plus industry work experience (i.e. Stephen Hawking probably couldn't get a position at University of Phoenix)
    Research universities physics departments: Ph.D.'s with at least one post-doc

    These are the bare minimum (i.e. if you don't have those, your application gets *instantly* trashed without anyone spending any more time at it. There are other criteria that will cause your application to get trashed *quickly*.)

    The other thing is that I do know of physics Ph.D.'s that have gotten adjunct positions in big name business schools, and one that is a professor in a law school (http://www.utexas.edu/law/faculty/jg33285/). In order to get those positions, what you need is about a decade of work experience on Wall Street.

    Something that I've read is that law schools hate to hire professional lawyers to be teachers, and would prefer to hire a Ph.D. in another field that has passed the bar. (http://www.nytimes.com/2011/11/20/business/after-law-school-associates-learn-to-be-lawyers.html)

    Also, my plan is to make a ton of money, and then hang out at some major university, making minimum wage tutoring first year physics students or moonlighting as a janitor. As long as they give me a library card and don't call campus security if I show up at a seminar, I don't need funding or a title.

    The major reason I haven't considered law school is because I have nightmares about zombie lawyers. Think of it this way. Imagine a horror slasher film say, Nightmare on Elm Street, Halloween, or Friday the 13th. OK. horror over, they catch Freddy Krueger. Who shows up the next day? Armies of lawyers.
     
    Last edited: Aug 5, 2012
  4. Aug 5, 2012 #3
    Okay, I should have been more careful with my wording. I didn't really mean the absolute minimum.

    I often see people complain about the lack of job prospects in academia, particularly for research positions. What I should have asked is what credentials is needed to have a good chance at finding a job in academia.

    And to be even more specific, what credential is needed to have a good chance in theoretical research or professor positions? At this moment I'm a bit more inclined towards math than experiment (perhaps due to this internship lol), and generally there is more math in theory if I'm not mistaken.

    I've read that statement a couple of times and I sincerely hope you achieve that. It touches me because it shows me what real curiosity is like. How many years until you are able to achieve that goal? I hope you are able to do this soon!

    :rofl:

    I am all for law, but I hate lawyers -- it stops being about justice and starts being about just winning the case. The blame is on the way law is run, the rules are so stringent that it acts as a double-edged sword. It ends up being about who is the best at playing by the rules, like a big chess game. And life isn't something to play chess with.
     
  5. Aug 5, 2012 #4
    A Ph.D. in economics, business, or finance. People with those Ph.D.'s pretty much automatically get a tenure-track faculty positions. If you get a Ph.D. in finance from a tier N school, you have a job waiting for you at tier N+1.

    The catch is that they filter very heavily at the admissions level.

    Now as far as physics goes, there is absolutely no credential that will give you a "good" (N>50%) chance of getting an research professorship. None. Nada. Zip.

    A birth certificate which has you graduating in the 1960's or the late-1980's

    Barring that.... There ain't no such thing. In some ways that good news, once you figure out that there is absolutely no credential you can get that will give you a good chance of getting a faculty position, you give up on the credentials game. Cool!!!!

    I can access my pension accounts at age 59 1/2. That's the upper bound.

    It's actually not. In a lot of legal situations (i.e. dealing with the aftermath of a horror slasher film), there is no realistic way of "winning" (i.e. Freddy Kruger has already chopped up people). Figuring out what happens next is something that lawyers do. We know Freddy Kruger is going to get locked up. The question is how long and under what circumstances, and the defense and prosecution are usually a lot less confrontational than you see in the movies.

    The other thing is that most legal work doesn't involve "winning a case." For example, while Freddy Kruger was slashing people in the house, he probably chopped up some furniture which means an insurance claim which creates lawyer work. There are also going to be real estate disclosures in the contract of sale, which means more lawyer work. There's also a lot of negotiation work. There are going to be a series of meetings to determine what happens to Freddy. If you were actually living on Elm Street, you probably would rather not attend the meetings yourself, so you have a lawyer do it.

    I dislike lawyers for the same reason I dislike dentists. It's not that dentists are bad people, but it's not that great if you are in a situation were you are forced to see one (and if you are so bad off that you need to see teams of dentists, it's really unpleasant and expensive).

    No. I've seen the legal system close enough to think that it works rather well.

    Also one reason that physics people get involves in this sort of thing is that the universe is this big giant rule based chess game. And if you don't like games in which people win by figuring out complicated rules, then I really don't see why you want a job in academia. You get ahead in academia with grants and publications, and figuring out how to write a grant proposal and what to publish means figuring out a ton of unwritten, unspoken rules.

    For example, there is a *reason* why "what credentials do I need to get a research position is a self-defeating question." Suppose I tell you that to have a good chance of getting a faculty position, you just have to dye your head pink and wear zebra striped jeans. Well, *everyone* is going to do it at which point it becomes useless. The only credentials that could be useful are those that you don't have and can't get (i.e. finance Ph.D.'s have good job prospects because they make it so that most people can't get finance Ph.D.'s.)

    You might be able to get somewhere if you have early access to information so you know to dye your hair pink before anyone does it. But if you are asking here, then it's already too late.

    One reason I like physics is that the universe *is* a giant chess game, and you can get ahead by figuring out the rules of the game (i.e. energy is conserved, do not go faster than light, F=ma). A lot of "winning" in industry is to figure out how to use the rules of the universe to your advantage more quickly than the other guy.
     
    Last edited: Aug 6, 2012
  6. Aug 6, 2012 #5
    Why should I give up on the credentials game? To make sure we are on the same page, credentials to me is the research you have done, the grades you got, the adviser you had and the university you graduated from. As far as I'm concerned, I should get the best grades that I can, do the best research that I can, etc. and hope for the best. And I'll make sure I have a safety net in case it doesn't work out.

    Okay so how many years from now though? I really want to see the day that you talk about your goal in the past tense versus the future tense as I've seen it quite a few times -- it looks to be a very important goal of yours.

    Well I was more or less talking about the cases where ridiculous things happen. Such as when some are let off the hook so to speak due to clever use of the rules set in place, one who were obviously guilty. Another that comes to mind is the student in Rutgers who took a video of a homosexual and he reacted by committing suicide. http://www.reuters.com/article/2012/05/31/us-crime-student-rutgers-idUSBRE84U1HR20120531

    He was an idiot, but he did not deserve a sentence of 10 years in jail. The student who committed suicide had many other problems that he faced in life, sure the idiot caused the breaking point -- but it does not mean he was the cause of this.

    Most of all though, lawyers are paid to defend the case that is proposed to them. They are there to keep the dirty secrets shoved under the carpet. If someone murdered another human being, he simply goes up to the lawyer and tells him I did x,y,z and the lawyer will do his best to get him out of the situation or into a better sentence.

    Well it is, but it isn't always fun to play.

    Well I'm not concerned with what credentials I need in order to attain so and so position. What I am concerned with is the credentials that people who made it have.

    Surprisingly, a lot of people can get through by just dying their hair even pinker. It seems that the only problem is when the rules change and you have to dye your hair pink but with highlights. In this age for theory, I think that might translate to getting good experience in not only math and physics, but also in programming.

    But hey your the expert here not me, so feel free to correct me.
     
  7. Aug 6, 2012 #6
    So here are two people who came from a theory group at a school I'm familiar with.

    Student A took 5 years to get his phd, published 4 papers, one in science. He did a postdoc at MIT, a second postdoc at Madison Wisconsin, no faculty interviews so he did a third postdoc in Europe before leaving physics for a large bank.

    Student B came from the same advisor, took almost ten years to get his phd, publishing only one paper (in collaboration with student A, while student A was in his second postdoc). After that he did one postdoc at a UC school, and landed a tenure track position at a European university.

    Basically- its a craps shoot. Everyone who is qualified is basically entered into a job lottery. One key feature of the job process is that most of the factors are totally beyond your control, and are totally unpredictable on a gradschool timescale- did people pay attention to my research? are schools looking for someone in my sub-specialty? whats the multi-year funding outlook in my subspecialty?, etc. The only way you can be sure to have good prospects is to be entering into the postdoc part of your career in an area with growing funding, so unless you can predict the grant situation accurately 7+ years down the line from the start of grad school....

    To get some anecdata for particle physics theory, you could check out the particle physics rumor mill and grab the job winners CVs.

    My back of the envelope calculation is that 85+% or so of theorists program heavily for their research. Programming experience will not help you stand out.
     
    Last edited: Aug 6, 2012
  8. Aug 6, 2012 #7
    Because it's a silly game that you can't consistently win at.

    First of all, it's not a safety net. The way people should think of it is that getting some sort of job in industry is the *standard* route. Getting a job in academia is if you win the lottery.

    Also the factors you mentioned, quality of research, grades, adviser, and university are only related to "good research" in a tauntological sense. One problem when you get into research is that there aren't that many non-tauntological standards to determine "good."

    Upper bound is 15. Lower bound (realistically) is 5.

    Those are pretty rare. They seem much more common that they are because it doesn't make good drama. If you have decent, hardworking lawyers, and when the system works (which is most of the time), it's not going to make it into the news, and it makes for a pretty boring movie or TV show.

    Actually they aren't.

    No.

    If you are a defendant in a criminal case, and you start to tell the lawyer about what happened, the lawyer will most likely tell you *not* to say anything more, and then tell you the legal consequences of telling him what happened. If a lawyer knows that you committed the crime, they cannot argue in court that you didn't do it. So if you want your lawyer to argue that you are innocent, then you can't confess guilt to them. Most defense lawyers would prefer that you don't tell them what happened, and have them figure it out from the police reports and the investigation.

    Also in practice, if you are accused of a serious crime whether you did it or not, your brain is probably so scrambled, that you aren't going to be thinking clearly.

    That's information tends to be useless because 1) you aren't going to be graduating in 2010, 1990, 1980, or 1970 and 2) it's only part of the picture. For example, I have it on good authority that most people that get tenured faculty positions have ten fingers and two eyes. So if you have ten fingers and two eyes, then you should get in..... Um.... No....

    You might point out that's silly logic, but people do the same thing with universities.
     
  9. Aug 6, 2012 #8
    What happened to me is that once I figured out that everything was about money, I started to get very fascinated by this money thing, and figured out that it would be a good idea if I learned more about it.

    There's something called the "Keynesian beauty contest." So not only to have to to predict the grant situation, you have to predict people's reaction to the grant situation. From a game theory approach, my solution is "do something random and don't think about it." I have a strong suspicion that this is the optimal solution to problem.

    It's actually related to the lottery ticket problem. If you choose numbers at random, you are likely to lose. If you think about it, and if your thought processes are similar to other people that are also thinking about it, then your expected return *decreases*.

    You can then define the situations under which it's a bad idea to think.
     
  10. Aug 8, 2012 #9
    Very interesting post.

    My approach is almost always to analyze. This clearly shows that sometimes it is a bad choice.

    I'll have to add this to my ladder of analysis :)
     
Know someone interested in this topic? Share this thread via Reddit, Google+, Twitter, or Facebook




Similar Discussions: Credentials for those who made it in Academia (physics)
  1. To those who hire (Replies: 7)

Loading...