Discussion: View Thread

MS vs PhD?

  • 1.  MS vs PhD?

    Posted 09-27-2012 14:55
    This message has been cross posted to the following eGroups: Statistical Consulting Section and Young Professionals Group .
    -------------------------------------------

    Hi all,

    I realize this topic has somewhat been tackled in the comments of previous posts, but I'm hoping to organize the answers to this question: what are the real benefits of getting a PhD in statistics, for someone who already has an MS and doesn't necessarily intend to head into academia? I'm in Montreal, so the answers I've heard to this question may vary across cities/countries. Am looking forward to hearing what people think!

    Follow-up: Can going for a PhD close off certain non-academic jobs from you for being overqualified? I rarely see job postings in my area requiring a PhD.




    -------------------------------------------
    Jonathan Moscovici
    -------------------------------------------


  • 2.  RE:MS vs PhD?

    Posted 09-27-2012 15:04

    Hi Jonathan,

    I have a Master's degree in statistics and have worked in research hospitals for the past 15 years. In my experience in that field (and in the US), a PhD usually comes with a lot more administrative management, grant writing and funding expectations, and supervising Master's level statisticians.

    The main advantage to me of not getting a PhD is that I prefer to be hands-on with the data, doing much of the data analysis myself, rather than supervising someone else doing it. I am not expected to find my own funding, but most PhD level statisticians are expected to find funding via grant applications and teaching. 

    The main drawback is that I would love to get into teaching, but without a PhD I cannot teach my own classes. I can TA or lead labs, but it's not the same.

    Hope this helps.

    Lori Lyn

     



    -------------------------------------------
    Lori Lyn Price
    Tufts Medical Center
    -------------------------------------------








  • 3.  RE:MS vs PhD?

    Posted 09-27-2012 15:13
    Hi Lory,

    That's insightful, thank you! I also enjoy performing hands-on data analysis, but am looked at funny by my department when I mention that. Is salary/upward mobility a concern? In terms of funding, have you worked on staff, or on soft grant money that might not get renewed (i.e. constantly seeking new employment)? For example, I worked in medical research at a hospital during my master's, but the money has run dry/project ended.

    I have been told by some faculty members that MS statisticians are less likely to be happy with their employment in the long run. I'm looking to debunk this, since these faculty members are more theory-oriented and rarely perform data analysis.


    -------------------------------------------
    Jonathan Moscovici
    -------------------------------------------








  • 4.  RE:MS vs PhD?

    Posted 09-27-2012 15:32

    In my experience, what has been put forward is incorrect.  Many medical projects and grants have a PI who is an MD or epidemiologist, or researcher with some particular specialty.  They write the grants (with assistance from others) and have a statistician in the group to do the statistical methodology section and analyze the data.  That person normally has a PhD.  Also, the preference is to have PhD statisticians serve as co-authors, again, to write the statistical sections of the journal articles.  The first major grant funded project that hired me, did so because I was ABD at the time.  Additionally, for the best pharmaceutical jobs, one needs a PhD. 

    -------------------------------------------
    Nora Galambos, PhD
    -------------------------------------------








  • 5.  RE:MS vs PhD?

    Posted 09-27-2012 16:28

    In the NIH/AHRQ grants that I have been funded on, there is a PhD level statistican for a small percent effort and a MS statistican (me)  funded at a larger percent effort. I do the bulk of the data management and analytic work, with guidance from the PhD level statistician as necessary. For the primary paper for the grant, both the PhD level statistician and the MS level statistician appear as co-authors. For most other papers related to the grant, I work independently and am the sole statistical co-author on the paper.

    Again--this is just my experience.
    -------------------------------------------
    Lori Price
    Tufts Medical Center
    -------------------------------------------








  • 6.  RE:MS vs PhD?

    Posted 09-27-2012 22:30
    I agree with Nora.  Although I have done some methods development, most of my career has been spent analyzing health data for the federal government.  The MS statisticians in the group could handle standard analyses but I got the difficult (and thus more interesting) datasets. Finding a suitable method required reading the stat literature, sometimes extending the published methods.  When I started grad school in the 70s, I intended to only get the MS degree because all I wanted to do was analyze data.  However, my advisor convinced me to stay for the PhD and I am really glad he did. I love solving puzzles and hate repetitive tasks. The PhD allowed me more autonomy in choosing projects, more creativity in the analyses and writing. I have taught as an adjunct but prefer to mentor young statisticians one on one, something else the PhD has allowed me to do.

    -------------------------------------------
    Linda Pickle, Ph.D.
    StatNet Consulting, LLC
    -------------------------------------------








  • 7.  RE:MS vs PhD?

    Posted 09-28-2012 11:31

    First of all, thank you for all the good advice so far. I am dealing with the same question myself; Should I go for a PhD? I am a master's level Biostatistician with an MPH degree in Biostatistics. I've been thinking very hard for the past year whether to go for a PhD or not. I do not want to work in the academia...I work for a university hospital setting in a biostatistics consulting service center and sometime I feel like I am treated like a second class citizen because I don't have the piece of paper saying "PhD" although I do most of the work for a given project without being supervised.

    When I saw this discussion, I felt like, finally, someone felt my pain. So, thanks so much Jonathan for raising this issue.

    I will follow this discussion very closely to find the answers of the following questions:

    1. Is the salary increase I will get with a PhD worth spending 4-7 years in school, paying tuition, and taking time away from family?
    2. Looking at the future, how much I will really lose without a PhD?
    3. How much I will really gain with a PhD?
    4. How many door closes with a PhD vs. How many doors open with a PhD?

    I recently took GRE and am actually in the process of a PhD application and this discussion may change that. Looking forward to some great advice.

    Thanks
    Shah


    -------------------------------------------
    Shahidul Islam, MPH, CPH
    Biostatistician
    Winthrop University Hospital
    Mineola, NY 11590
    -------------------------------------------








  • 8.  RE:MS vs PhD?

    Posted 09-28-2012 12:54
    I worked in a large pharma company for over 25 years.  Having a Ph.D. in that environment was definitely worth the time and effort.  Its not entirely fair but Ph.D. statisticians in pharma have more career advancement opportunities, more freedom in their work responsibilities,  and are compensated at a much higher level.   

    As others have mentioned,  I don't think that the decision should be financially based.  Going on for a Ph.D. in statistics should be because you love learning about all aspects of statistics and because you honestly want to be a statistician (academic or non-academic) for most of your career. 


    -------------------------------------------
    Roy Tamura
    Associate Professor
    University of South Florida
    -------------------------------------------








  • 9.  RE:MS vs PhD?

    Posted 09-28-2012 13:12
    In my experience, statistics is a field in which one benefits from having a PhD.  There are many jobs advertised as being PhD-level. Do some searches to convince yourself. They include pharma, which has just been mentioned, FDA, NIH, Census Bureau, Social Security Admin, and many other US government agencies, and of course, SAS and SPSS.  I never regret for a minute having completed my degree and I believe my career has benefited.  As someone who has gone through the process, I can say that the difference between a BS and MS is small, but between an MS and PhD is large.  It took a lot of work to complete my dissertation, but what I learned about the area I studied as well as the skills I developed to do the research are immeasurable.  (My dissertation was an applied statistics dissertation, as opposed to a math stat one.)  

    -------------------------------------------
    Nora Galambos, PhD
    Stony Brook University
    -------------------------------------------








  • 10.  RE:MS vs PhD?

    Posted 09-28-2012 14:23
    Very interesting discussion and one I find very relevant.  I obtained an MS in applied stats two years ago supported by my employer, but I do not work as a statistician despite having gotten the degree, due unfortunately to a major change in business priorities at my particular site.  I spent much of this year looking for stat jobs elsewhere and have discovered that most positions targeting MS candidates are being filled with PhD's instead. I've since given up my search for personal reasons, but it was interesting (and discouraging) to find myself in competition with so many PhD's for MS (and even BS) level jobs.

    -------------------------------------------
    Shannon Wheat
    Adjuvant Development
    Glaxo Smith Kline
    -------------------------------------------








  • 11.  RE:MS vs PhD?

    Posted 09-28-2012 16:58

    It is not uncommon for job postings to have requirements lower than what is really wanted/needed.  Depending on the agency, they may be prevented from hiring someone if they appear to fall short of the advertised credentials. Also, times are rough, and companies may be getting PhD's at MS prices....
    -------------------------------------------
    Nora Galambos, PhD
    -------------------------------------------








  • 12.  RE:MS vs PhD?

    Posted 10-01-2012 13:24

    I've noticed that this discussion seems to distill into two directions:

        1) what are the benefits for the individual in getting a PhD, and
        2) what are the career benefits for getting a PhD.

    Regarding 1), one should certainly be better developed as a statistician with the additional time and effort spent in understanding concepts more deeply, and in demonstrating the ability to do independent original research.  If that matters in your specific career choice (i.e., do you want to be an academician?), then that should be important in your decision to whether or not to get a PhD.  And you will understand new methodology quicker with the deep theoretical foundation gleaned from getting a PhD.  But not having that deep theoretical foundation does not mean that you won't grasp new methodology. - you'll just have to work hard(er) to do so.

    Regarding 2), having a PhD certainly demonstrates your qualifications to a higher degree than having a MS does.  This makes a significant difference in some industries, particularly pharma.  And as others have noted, PhD statisticians often get to work on more interesting problems than MS statisticians do. 

    But often, it is assumed that having a PhD also better qualifies you for non-statistical responsibilities as well.  I thoroughly enjoy being an internal consulting statistician!  I like the client interactions, refining the project request and developing the analyses, writing the programs and conducting the analyses, communicating the results and their implications to the clients, and even running project teams.  But I greatly dislike management and all of the accompanying non-value-added "detail".  I have had to make specific career choices to ensure that I am doing what I like to do, because it was assumed that I wanted to develop as a manager simply because I had a doctorate in Statistics.

    So it absolutely makes a difference if you have a doctorate versus a Masters.  But does the difference make a difference?  I think that depends on the individual situation.  I started my graduate career on a fellowship from Bell Labs, and someone there (it may have been Ram Ganandesikan) told me that there were two reasons for getting a doctorate in Statistics:

        1) to be able to do what you want to do, and 
        2) to ensure that you have more options (when 1 is not possible).

    All this being said, I think the most important thing is to understand yourself and what you want your career to be like, and make your decision accordingly.  I would also remind you that our career desires evolve and may change over time, while the practical window of opportunity to get a PhD is often very narrow.  To quote an old saw, "It is better to have (the PhD) and not need it, than to need (the PhD) and not have it."

    -------------------------------------------
    James Gear
    Senior Statistician
    -------------------------------------------




  • 13.  RE:MS vs PhD?

    Posted 09-27-2012 15:37
    Hi Jonathan,

    I have a master degree in biostatistics and work in industry funded consulting. I actively manage and mentor PhD level analyst. There is a lot to be said for practical experience. Most of the analyst we have hired with PhDs are really lack luster in there practical analytic abilities. They are great for theoretical concerns, but not for the applied work we engage in as consultants. I would hire an experienced analyst with a MS over an inexperience PhD any day. 

    Also if you are interested in working in academia as an educator, there are plenty of opportunities for someone with a masters degree. I teach adjunct at a couple state universities in the evenings and regularly lecture at s very prestigious university. I love it, I don't have to deal the the political BS that a full time faculty encounter.

    If you want to work as an academic in a research institution, get your PhD. Otherwise you will always be treated like a second class citizen...at least that was my experience.

    To answer your last question. My favorite math professor in my undergraduate program always said the biggest mistake he ever made was getting a PhD. His case was somewhat unique. He was absolutely brilliant and his dissertation was very specialized. Because of this he had trouble getting an industry job. His work was not easy to understand and he was seen as over qualified. He did not want to work as an academic, but didn't have much choice. Finally, he did find an industry job after 10+ years as an academic. 

    I'm ABD and a few years into the doctoral program I began a cost benefit analysis of forgoing professional work while spending time in the program vs working in industry with an MS and I found that the opportunity cost of staying and completing my PhD would come to an equilibrium in approximately 20 years. 

    A PhD does lend credibility and is certainly necessary if you want to pursue a tenure track position. But it's not a necessity in industry. 

    These are solely my personal opinions and experiences. 

    Best wishes,
    Chris

    -------------------------------------------
    Christopher LaVallee
    Senior Analytic Consultant
    Precision Health Economics
    -------------------------------------------








  • 14.  RE:MS vs PhD?

    Posted 09-27-2012 17:00
    I went through college quickly and got the Bachelor's, Master's, and Doctorate all in a row without thinking too much about it. I've worked inside and outside academia since then, and I am very glad I got my Ph.D. The pay is a bit better, but the big thing is that the work is far more interesting for a Ph.D. statistician. You get a lot more variety and are expected to work more independently.

    I've never believed that you should turn down an applicant because he/she is overqualified. What's the downside for you? I know people say that an overqualified candidate will get bored and will move on more quickly, but I think they will contribute far more while they are there. That's enough to tip the balance in favor of the overqualified candidate, if you ask me. But I suspect I'm in a minority here. Yes, the Ph.D. might close a few doors, but it will also open other doors and give you a competitive edge on those jobs looking for either degree.

    A possible downside to the Ph.D. is that the jobs are fewer and you might have to make more geographical compromises to find the right job. If staying in your current location is important, you might stick with the Master's degree.

    The only other downside to the Ph.D. is that it takes a lot of endurance to get through all the requirements. You'd better love Statistics before you start the Ph.D. or you'll be really sick of it by the time you're done.

    -------------------------------------------
    Stephen Simon
    Independent Statistical Consultant
    P. Mean Consulting
    -------------------------------------------








  • 15.  RE:MS vs PhD?

    Posted 09-27-2012 17:10
    Hi Stephen,

    I've read your blog and recent magazine article regarding Web 2.0 tactics! They're great stuff.

    I think the main issue behind the being "overqualified" argument has been budget constraints. Naturally, PhDs expect higher starting salaries and perhaps expect an environment where they can climb the ladder quicker and/or higher. This might be possible for MS applicants as well, but require less of an investment from the employer's part right away, especially if the task at hand doesn't necessarily call for it.

    Simply put, I'm seeking an argument to keep me in a PhD program if a great job offer turns up tomorrow. So far, my happiest times have been consulting for research teams and collaborating in an applied environment. My background is in pure math and did an MS in statistics to learn to solve problems in a practical context.
    However, I would like to earn a nice salary and live a happy (preferably quiet) life. Is an MS likely to limit this?

    Thanks very much!

    -------------------------------------------
    Jonathan Moscovici
    -------------------------------------------








  • 16.  RE:MS vs PhD?

    Posted 09-27-2012 17:58
    I have my MS, earn a nice salary with great work/life balance, and have not met up with a ceiling (yet) in my 7-year career. I passed up an opportunity to pursue a PhD for financial reasons, and it was undoubtedly the right decision for me.

    I do not doubt that in many situations, Stephen's comment that those with a PhD "will contribute far more" is warranted. There are times when I think, "MAN I wish I had my PhD so I could have thought of that reply" to a Statistical Consulting Section thread. While many industry jobs will require a PhD (for good reason), you can be assured that those not requiring a PhD, along with opportunities for advancement, are plentiful. In those positions, how much you contribute is dictated by other factors apart from whether you have a MS or PhD. If those are your priorities, my personal belief is that an MS is sufficient.

    Another of Stephen's comments that I agree with is that the work probably is more interesting for a PhD statistician. I have worked on some cool projects that stretched my analytical skills, but I had to leave my first employer to find them. (In my first job out of grad school, I rarely used statistical methods outside of what is taught in Stats 101.) I think it would be cool to be an independent consultant one day, but even when I have years more applied experience, I wonder if I would be qualified without a PhD. If you have a passion for statistics, prefer more challenging assignments (more often) and aren't limited by financial/other constraints, then I think a PhD is worth considering. I can't imagine regretting it, anyway. Employment should be abundant and financially rewarding either way, it just depends on what kind of employment will keep you satisfied.

    Good luck!
    Tricia

    -------------------------------------------
    Tricia Larson
    Principal Statistician
    Deluxe Corporation
    -------------------------------------------








  • 17.  RE:MS vs PhD?

    Posted 09-27-2012 18:01
    Making the assumption you want to work outside a university. No a MS will not limit your life. An MS actually offers a greater deal of flexibility. You're as good as you are whether you have a PhD or not. A PhD is not a guarantee to success. Nor is it a guarantee to higher pay. I make significantly more than the PhD analyst that work under me. 

    When I left the university, it was because of money. My supervisor (20+ years at the university) wanted to counter the offer, but couldn't because my new salary was more than hers. 

    Someone, made a pharma comment...in pharma experience makes up for PhD in terms of pay scale and promotion. It also depends on how good a communicator and negotiator you are. If you want to working in pharma switch to biostatistics. Pharma, FDA, NIH, any health related field prefer biostatisticians. 

    If you want to work at a university, get your PhD. You will not make a lot of money for a very long time in at a university and grants are highly competitive. Where I was working if there were no grants, there was no job. 

    You're success and happiness in life should not dictated be three letters. 

    I'm not trying to discourage you to obtain your PhD, but there are pros and cons that should be weighed carefully. There is opportunity cost associated with the time spent in a doctoral program. You could be passing up a $100+K position to spend 4+ more years in the program plus 2+ years as a postdoc to get into a $70-80K job as at a university.

    If you would like to talk on the phone, send me a private message and we can coordinate a time to talk. 

    -------------------------------------------
    Christopher LaVallee
    Senior Analytic Consultant
    Precision Health Economics
    -------------------------------------------








  • 18.  RE:MS vs PhD?

    Posted 09-27-2012 18:06
    My only suggestion is to go deep within and feel what will make you happy. Then, do that!

    Best wishes,

    Nayak



    -------------------------------------------
    Nayak Polissar
    Principal Statistician
    The Mountain-Whisper-Light Statistics
    -------------------------------------------








  • 19.  RE:MS vs PhD?

    Posted 09-28-2012 09:35
    None of the replies so far have addressed what I think is the most important issue.  What will you learn in the particular PhD program in question that you would not learn in the masters program?  Will you take courses covering more advanced techniques that you may well need?  Will you learn more about how to do the sorts of things that you hope to do in your first job(s)?  What sorts of things are they?
    -------------------------------------------
    Emil M Friedman, PhD
    emil.friedman@alum.mit.edu (forwards to day job)
    emilfrie@alumni.princeton.edu (home)
    http://www.statisticalconsulting.org


  • 20.  RE:MS vs PhD?

    Posted 09-30-2012 05:17
    My understanding of the primary difference between the PhD and MS level is a solid 1-year sequence in both probability and statistics at the funcational analysis level (e.g. the Hogg & Craig text or similar), plus a thorough understanding of asymptotics, i.e. Gauss-Markov and Central Limit Theorem.  Not just understandig them but being "easily familiar" with the mathematics behind them.

    More importantly, I have always viewed the PhD as more of a crucible, wholly aside from the knowledge issue.  During my PhD program my advisor was very busy, and we met maybe once a month.  Sometimes I woke up early Monday morning with list of things to accomplish on my dissertation for that day numbered 1-5, only to find that 1(a) took two weeks.  Many bright people excelled in the coursework but then spun their wheels when it came to writing their dissertation.

    I would also add that unless you are in academics, and in partuclar if you undertake mainly applied statistical work, after 5-10 years the distinction between the PhD and MS becomes much less important.  Can you finish this on time for the client?  Great, thanks, whether you have a PhD or an MS.

    Matthew G. Mercurio, PhD

    Applied Statistical Consultant
     

    http://www.MGMAppliedConsulting.com










  • 21.  RE:MS vs PhD?

    Posted 10-01-2012 09:59
    I have to disagree slightly with the comment that "after 5-10 years the distinction between the PhD and MS becomes much less important." The big advantage that a PhD has and will always have over an MS is the requirement to produce independent scholarly research. It is very valuable in that it forces you to understand Statistics at a very fundamental level. This is nice by itself, but more importantly it makes you a quick learner. Once you understand the foundations of Statistics, you can pick up new methods faster because you can follow the logical underpinnings of those new methods.

    In many other ways, of course, experience is very critical, and if you are like most statisticians, you learn more on the first few years of your first job than you ever did in your graduate program. So Mr. Mercurio's comments still have validity in that sense.

    By the way, you could make an argument that TWO degrees at the MS level are better than ONE degree at the Doctoral level. As hard as I try, my scientific and medical knowledge will always be superficial to the people that I consult with. Someone who knows the science well AND knows Statistics well might be more valuable than someone who only knows Statistics well.

    -------------------------------------------
    Stephen Simon
    Independent Statistical Consultant
    P. Mean Consulting
    -------------------------------------------








  • 22.  RE:MS vs PhD?

    Posted 10-01-2012 11:09

    I have been following this long, interesting and controversial thread from time to time and have been motivated to chime in but did not do so for lack of time to do this justice.

    First of all I agree with Stephen Simon that a PhD is different from a masters in that the dissertation prepares you for future scholarly research and to be able to produce good original research is a talent that can be developed when working on the dissertation with an advisor.  But not everyone is cut out for it or has the talent fo5 it.  these skills are essential inacademia where research publications are highly important and it is also important outside academia at research institutes suchas the one I currently work at.


    I think when you decide on a career it is a personal choice and each individual should try for a satisfying career that suits them.  It can be a bachelor's degree, a master's degree and a PhD.  In some industries such as pharmaceuticals a personal who does not have a PhD is at a disadvantage.  I know this from experience in pharmaceuticals with about 10 years in that regulated industry.  To a lesser extent that was also true in medical devices where I work for 7 years.  The PhD gives you a easier chance to enter the industry and it is easier to advance to positions with visability and authority in the company if you have the PhD degree.

    I am not saying that it is impossible to accomplish some of these goals without the PhD.  But it is much more difficult.  In some industries a person with a PhD and the same experience as a person without the PhD will have a higher salary and better advancement opportunities.  In other areas it may not matter and in some case the person with the doctorate may be disadvantaged as he is viewed to be overqualified for a job he is capable of doing and finds interesting.

    It upset me to hear some people downgrade the PhD.  It takes at least 2 extra years of study taking the more advanced theoretical courses plus the struggle through a dissertation to obtain it.  It certainly has values and in many fields is valued.

    As technology advances the need for higher levels of education is increasing.  Statistics continues to gain importance and the demand for statisticians exceeds the supply, particualrly at the PhD level..  A PhD is not a guarantee for a high paying job.  It also does not lead to a path of fame and riches.  We know of the college dropouts and others with just bachelor's degrees whose entrepreneurial skills brought them fame and fortune without any advanced college degree.

    Success is based on skill and determination.  Sometimes an advanced degree can help with that. if you are brilliant and confident you will probably go far on whatever path you take.

    My father was a mathematician who was very brilliant and liked number theory.  But World War II slowed down his academic progress along with marriage and children.  He worked at Aberdeen Proving Ground during World War II working on important applied mathematics in ballistics.  After the war we moved to long Island where he became one of the pioneers in the emerging field of nuclear engineering.  During those years he work as an applied mathematician learning  and helping develop reactor theory.  He worked his way up to the Director of Reactor Theory and supervised several scientists, mathematicians and physicists who had PhDs.  He had the ability and interst to get a PhD and through urging from my mother and grandmother worked on a PhD in applied math at NYU.  He was working on a PhD in nonlinear differential equations and had made great progress using methods he developed for solving real problems that came up in his work at Brookhaven. When he presented his thesis to the dissertation committee at NYU the committee asked for uniqueness theorems related to the equations.  My father was stubborn and principled and felt that his hard work was important enough and sufficient for the degree. Whether or not he could have worked out the proofs was not the issue for him (I think he could have done it).  He felt he had more important things to do than to take a lot of time out to do the work that would satisfy the committee.  So he never got the degree.  This was a disappointment to my mother and especially my grandmother.  Nevertheless he went on to get a patent on a nuclear reactor and publish over 100 research papers (the first two or three were number theory results in the math journals back in the 1930s and 40s).  He also was a coholder of a patent on the design of the Brookhaven High Flux Beam Reactor.

    So my father is an example of someone who could be a success in research without getting a PhD.  But the situation of the early 1950s was different.  Nuclear physics was a new field with very few people trained in it.  There was an opportunity to be a part of a research area that was just getting off the ground.  It would not be as easy to do that in the nuclear field today.

    Statistics was a relatively young field with major developments occurring from 1900-1960.  Now it is very much established although computing advances led to the computer-intensive methods being developed primarily in the 1980s and 1990s.  I was an undergraduate in mathematics from 1965-1969.  At that time the number of schools with math or statistics departments offering degrees was much smaller than today.  There were no undergraduate degrees in statistics and nothing like the current high school courses in AP statistics.  so many of us got our degrees in mathematics.  I had only one course in probability and statistics as an undergraduate and statistics was not view with great importance inspite of its great success with sequential analysis and other applications during World War II. i was among a large number of statisticians that went back to school to get a degree in statistics because I discovered its importance in my work in applied math and operations research.

    Today we have children introduced to statistics in their elementary school math classes.  Ap statistics is a formal math course in statistics that can be taken with or substituted for calculus.  Students can find many colleges and universities where statistics degrees are offered at all levels.  The number of schools with statistics departments independent of mathematics departments is growing.  Some statistics departments can be found in business schools and biostatistics departments in medical schools. Statistics has close connections with others fields such as operations research, computer science, artificial intelligence, data mining and machine learning.  There are many options.  Great opportunities are out there for PhD statisticians to do theoretical and applied research in many disciplines and there are jobs in statistics for statisticians at all levels.  No wonder some people call it the sexy profession for the 21st century.

    The bottom line I think is that the PhD can be a great thing but there is nothing wrong with just getting a masters or a bachelors degree.  Where the degree will take you will depend on your interest and determination and the field you choose to work in.


    -------------------------------------------
    Michael Chernick
    Director of Biostatistical Services
    Lankenau Institute for Medical Research
    -------------------------------------------








  • 23.  RE:MS vs PhD?

    Posted 10-01-2012 11:47
    There are different angles on this, and they vary depending on the field and subfield. But I think it would be fair to say that there always has been demand for MS level statisticians, and there will always be.

    From an employer's angle, what matters is what you can bring to the organization. An opinion that I heard a couple of times that is really hard to argue with is, "There is little in Ph.D.'s tollbox that isn't in the MS toolbox". Unless you go to the company that has a fleet of 20-30 Ph.D.s, and is known to work on the projects that eventually end up in a series of academic publications, including a couple in JASA/Biometrika, you are being hired to do rather specific tasks (these series of cross-tabs, this type of power analysis for a proposal, this time series of customer satisfaction or student evaluations, etc.), and very few of them would require the development of new methodology (which is what Ph.D. students spend most of their time polishing after the first couple of years, and grow to think that statistics is all about this new methodology and publishing JASA papers).

    Second, there are formal job requirements, and you have to look at the ones in the field that you are interested in to see the breakdown of MS vs. Ph.D. There are many positions (certainly in academia, but also in pharma) where Ph.D. is a requirement, so you simply can't apply if you don't have that piece of paper. In the line of work I am in, I have taught myself 90% of what I know in survey statistics, so my Ph.D. is fairly nominal, and measure theoretic probability does not cross paths with me, no matter how much I loved the subject, the course and the instructor I had in grad school. (Survey methodology was something I was learning outside of the statistics program, and that part of education turned out to be more relevant, even though an order of magnitude less technical than the graduate courses in theoretical statistics.) I wouldn't be surprised to find people with MS degrees and titles similar to mine, as there simply aren't enough programs producing Ph.D.s in survey methodology (unlike the dozens of biostat programs stamping a few hundred Ph.D.s a year, and creating an intellectual overhang on that submarket). On the other hand, whether Ph.D.s are overqualified for MS positions is a judgement call for every employer (see the point above about the job requirements and the skills that will or will not be usable; there have been different opinions voiced already in this exchange), and hiring and keeping Ph.D.s on MS level positions is only sustainable in the long run if the overproduction of Ph.D.s continues into the future.

    Third, there are certain levels of task complexity, especially on the interfaces with other fields, where just the MS is likely to be unrealistically little (unless it is from a top university). If you want to do Bayesian computing for a job, or some cool *omics stuff, or "big data" with a very heavy computational component that requires a professional grade Java/Python/Ruby programming, then you would probably have to get a Ph.D. just to have enough time in school to get immersed in these multiple disciplines (assuming say one CS class per language, which is likely to be insufficient, anyway, and/or a bunch of biology classes). You would also need to have more of the Ph.D.-type frame of mind to learn to learn the new technological stuff that will be appearing every three to five years. A Carnegie Mellon MS in computational statistics, on top of CalTech BS in computer science, will probably do fine in a Silicon Valley startup, but a less stellar constellation of degrees will raise doubts of whether you simply has had a chance to learn enough to work in these multidisciplinary fields.

    Fourth, there's general deflation of degree credentials. Some thirty years ago (when I was in grade school... not in grad school), it probably would've been fine to hire an MS statistician to lead a university consulting center, as there weren't that many statisticians around, and MS was a solid degree. When I looked at Mardia, Kent and Bibby's multivariate book that I used to teach a doctoral multivariate class from (in combination with Anderson and Hastie-Tibshirani-Friedman though), the book, written in the 1970s, placed itself as an upper undergraduate. It might have been then, but it is certainly a graduate level text now; and most doctoral students in social sciences would be greatly challenged by it. Due to the combination of gradually declining education standards and larger fraction of the young adults getting their degrees, college and above, you may have to have a higher level degree just to stand above the crowd -- to send a signal about your intellectual ability, as a labor economist would say (and may continue to say that this is a very costly signal: to you personally in terms of the number of years you have to spend in school just to prove that you are smart, to the companies hiring on the market who have to sift through the larger number of applicants, and eventually to the society as a whole). If everybody has a BS now, you have to get at least an MS to show off that you have better skills than an "average" fresh-from-school statistician; if everybody has an MS, you have to get a Ph.D. -- but then we are back to the square one re: the actual skills, as discussed in the first point.

    Finally, some folks go get a nominal Stat MS degree just to learn enough SAS to get themselves a SAS programmer/data management/report prep job, with little if any statistical content. Again, there are jobs that require just that, and since calling SAS a programming language, let alone a modern programming language, is a bit of a stretch, this won't be something you would learn in a CS program. Instead, you would take a (bio)stat class titled "Statistical computing" or something like that, and would have to build up your SAS skills through both doing practical work in a statistical consulting center as a graduate RA, and collecting the various certificates through SAS training courses. Similar considerations apply to Stata and R, although building up R-no-statistics resume probably won't get you a job as easily as building up SAS-no-statistics resume would, and the certification/training is not nearly as structured. Stata users can just read J Scott Long's "Workflow" book that contains enough wisdom to count as a solid foundation for further practical work :).

    These are my personal opinions based on what I've seen on the market, and what I've heard from other people who have been on both the hiring and the being hired sides of the market.

    -------------------------------------------
    Stanislav Kolenikov
    Senior Survey Statistician
    Abt SRBI
    -------------------------------------------



  • 24.  RE:MS vs PhD?

    Posted 10-01-2012 12:24
    Stanislav's comments are very thoughtful and thought provoking.  I agree with him especially regarding the fact that a lot of job work can be and may need to be learned on the job.  Although I may never use measure theory or advanced probability in my applied work, I can't say that I didn't benefit from learning it.  Advanced mathematcis is always helpful in tuning your logic skills and thinking ability.  This can play an indirect role in many things we do in life whether career related or not.

    My dissertation topic has mainly been helpful for me to get a few papers published and contribute a little to the advances in extreme value theory.  But I don't use it much and dare say I haven't used any of it for the last 15 years (except maybe for answering questions on the StackExchange website).

    Keep in mind that hiring managers often don't know much about statistics and if they have no statisticians working for them they will go by your degrees, how you interview and maybe a reference or two (although surprisingly references are not checked very often these days).  So what may make sense to you as an argument for why you should get the job may not to them.  Written and oral presentation skills are usually a key to success.

    I have a PhD and am an ASA fellow and I have published some well-reviewed books and over 40 journal publications.  This means nothing to most hiring managers I run into when I apply for a job.  Also since I do private consulting it is not important to those that may need a statistical consultant on a part-time basis.  But I think accreditation through the ASA PSTAT designation might help with the consulting.  So I did get that.

    I am not an expert in survey sampling although I have dabbled in it a little.  I think that there are aspects of complex surveys that require a higher level of sophistication than Stanislav lets on.

    -------------------------------------------
    Michael Chernick
    Director of Biostatistical Services
    Lankenau Institute for Medical Research
    -------------------------------------------








  • 25.  RE:MS vs PhD?

    Posted 10-01-2012 09:41
    I am a master's-level statistician with a > 30 year career providing statistical support to the pharmaceutical, biotechnology and medical device industries. Since 1995 I have built a fairly successful consulting business for myself. My work has allowed me to grow from the most junior level to one where I'm now often involved in dealing with client company executives when planning clinical programs and individual trials and overseeing others' work (including PhDs). I've frequently gone to FDA as my client's statistician and have contributed at advisory panel meetings. I've recently taught math and statistics to undergraduate university students and have provided statistical guidance to several PhD candidates in areas other than statistics.

    Does this mean that a masters is sufficient and a PhD unneccesary? My answer is biased. I got my degree about 32 years ago. Times and technology were different. A masters was considered to be perfectly acceptable for applied work while a PhD was more for those with an academic or theoretical bent. My first manager was director of the data sciences group at that company and he had an MS. After several years and a couple of companies later I rose to a director-level position at a multinational company, supervising others and in charge of the departmental budget. And now, for 17 years, I've run my own consultancy and had good success with my clients. 

    In that time I've accumulated much experience working under varying circumstances and I'm convinced that has given me an advantage over many PhDs. However, if I were in school today and looking to make a career as a statistician, I would definitely try for that doctorate. Recently minted PhDs are today seem to be equivalent to what I was "back in the day" as a new masters-level statistician. 

    -------------------------------------------
    Nestor Rohowsky
    Integrated Data Consultation Svcs, Inc.
    -------------------------------------------








  • 26.  RE:MS vs PhD?

    Posted 10-01-2012 11:10
    HI Nestor:

    You explained, so very well, what has been going through my mind as I follow this thread - times have changed.  I have been an applied statistican for over 35 years.  For the last 28 years running a statistical consulting company that now has 10 employees (CRO in healthcare, pharam, presenting to FDA etc.).  Keeping current with new developments has always been "part of the job", and networking with the academics who develop the methods, or write the books, has been really helpful to ensure you are on the right track when new developments are adopted.

    Like you, the choice for an applied staistican "back then" was MSc - PhD if you wanted to be an academic.  That has changed, although in our company we still prefer the Masters graduates as they seem to have more of a love of "the data and the code", rather than getting lost in methodology and driving costs for the client.  We attend webinars, seminars etc. to keep up and read in the area have internal debates, so I feel the clients are well served.

    So there are still an excellent opportunities for a Masters graduate (as a number have indicated), but it is likely easier to keep up with more theoretical training early on, and also networking with some of the academics to help support you throughout your career - and the right PhD programme and supervisor can certainly open those doors.  Our statistical world is evolving much more rapidly so the extra dose of new theory - via a course, vs. self study, can help to smooth the path of continuous learning that statisticans embark on when they choose this as a career. 

    -------------------------------------------
    Janet McDougall
    President
    McDougall Scientific Ltd
    -------------------------------------------








  • 27.  RE:MS vs PhD?

    Posted 10-05-2012 02:42

    I've been catching up with this lively discussion.

    Its worth remembering that one of the most talented statisticians in our field, Nate Mantel, to the best of my knowledge, did not have a Ph.D. though over the years surely was awarded several honorary Ph.D's.   Certainly there are other statisticians who made as many contributions, without a Ph.D.

    http://en.wikipedia.org/wiki/Nathan_Mantel


    -------------------------------------------
    Chris Barker, Ph.D.
    Consultant
    www,barkerstats.com

    ---
    "In composition you have all the time you want to decide what to say in 15 seconds, in improvisation you have 15 seconds."
    -Steve Lacy
    -------------------------------------------








  • 28.  RE:MS vs PhD?

    Posted 10-05-2012 10:48
    If you want to go along the lines of Chris' post here is part of Bill Cochran's bio.

    William Cochran took the University of Glasgow Bursary Competition in 1927 and he was placed first. The award of a bursary allowed him to take his first degree at Glasgow and he was awarded an MA in mathematics and physics in 1931. He won the Logan Medal for the best student in the Faculty of Arts and won a scholarship to study mathematics at Cambridge.

    When PhDs in statistics were rare many prominent statisticians only had masters degrees yet became emeritis professors.

    -------------------------------------------
    Michael Chernick
    Director of Biostatistical Services
    Lankenau Institute for Medical Research
    -------------------------------------------








  • 29.  RE:MS vs PhD?

    Posted 10-05-2012 12:00
    "When PhDs in statistics were rare". 

    However, these days, PhDs seem to be more numerous in many fields.

    A person starting out will be working in a world with a different distribution of people with degrees in many fields.  The fields where my clients come from certainly have more doctorates today.

    In those days an elementary school teacher may not have even had a bachelors.  Today many (most?) school systems require a master's.  In order to have more people with master's, it was necessary to have more people with doctorates to teach them.  More people working on doctorates in education, nursing, social work, business, etc. means more people who have to use statistics in their coursework and dissertations.



    -------------------------------------------
    Arthur Kendall
    Social Research Consultants
    -------------------------------------------








  • 30.  RE:MS vs PhD?

    Posted 10-05-2012 12:38
    .... and now you need significant post-doc experience to get most faculty positions. What's wrong with our education system that it requires 12 years of post secondary education to get a starting university position? Its absurd.

    -------------------------------------------
    Neil Paton
    Lead Statistician
    Cargill
    -------------------------------------------








  • 31.  RE:MS vs PhD?

    Posted 10-05-2012 12:50
    I am straying a little off topic here, but I see the same thing happening in medicine, at least in non-rural areas of the US.

      Years ago if I was faced with a complicated hand and tendon laceration that I as a family physician did not know how to repair, I'd call my general surgeon colleague and he'd do it.  Later it somehow "became necessary" to call the orthopedic surgeon.  Nowadays even that is not good enough--need a fellowship-trained hand specialist orthopedic surgeon.

    Complicated pregnancies I'd refer to my OB colleague. Nowadays, the OBs don't even keep those patients--they refer them to the fellowship-trained perinatologist.

    Sheesh!

    -------------------------------------------
    Christopher Ryan
    Clinical Associate Professor of Family Medicine
    SUNY Upstate Clinical Campus At Binghamton
    -------------------------------------------








  • 32.  RE:MS vs PhD?

    Posted 10-05-2012 17:26
    One of the National Institutes of Health Giant statisticians, Jerry Cornfield did not have a PhD!.  John Bartko

    -------------------------------------------
    John Bartko
    Consulting Biostatistician
    -------------------------------------------








  • 33.  RE:MS vs PhD?

    Posted 10-05-2012 20:44
    It is not surprising that we all have personal interpretations of the value of a graduate degree in statistics. 

    In my view, this value is not a static concept - it is something that evolves over time based on experiences, opportunities, successes, etc., and something that constantly informs our standards for performing in the professional and even personal world. 

    Rather than thinking of the value of a degree in a binary fashion (e.g., MSc vs. PhD degree), we can think about how we can best add ongoing value to our degree.  How can we make it reflect our own strengths and passions and interests throughout our professional careers?  How can we use it to make a difference? 

    The most important thing for each of us is to not only believe in the value of our own degree but realize that that degree is just a platform for bigger and better things, many of which are directly within our control. 

    -------------------------------------------
    Isabella Ghement
    Ghement Statistical Consulting Co.
    -------------------------------------------








  • 34.  RE:MS vs PhD?

    Posted 10-05-2012 22:14
    I like Isabella's post very much. What seems important to me is what you do with what you have and not what you have.

    Best wishes,

    Nayak


    -------------------------------------------
    Nayak Polissar
    Principal Statistician
    The Mountain-Whisper-Light Statistics
    -------------------------------------------








  • 35.  RE:MS vs PhD?

    Posted 10-07-2012 12:04
    Following Nayak's and Isabella's comments, I found this recent piece in the NY Times: http://www.nytimes.com/2012/09/30/jobs/follow-a-career-passion-let-it-follow-you.html?ref=jobs

    I have a MS that I earned in 1979, and I work primarily as a SAS programming consultant. For the past several years, I've felt a bit limited by the degree. This may be a change in the culture such that MS degrees are more routine and PhDs more a basic job requirement, or it may be that I've gotten older and am more interested in greater responsibilities. However, the MS degree has allowed me to work with many different clients in many different areas. advise on development of SAS code for several projects, and mentor beginning grad students and programmers. And I've also had the opportunity to write 7 books for SAS Press. 

    -------------------------------------------
    Michele Burlew
    -------------------------------------------








  • 36.  RE:MS vs PhD?

    Posted 10-08-2012 12:22
    This is not a reply to Michele Burlew, but to the entire string; I just don't know how to reply to the whole enchilada.

    Usually I have an opinion on an issue but sooner or later someone else voices my stand, often better than I could. On the MS/PhD issue, no one has voiced my position, so here goes.

    All the replies have treated the degrees as if they were completed "things". They are processes. A good degree teaches the student how to learn. Yes, you learn facts, too, but that is secondary, since if you know how to learn you can learn the facts you haven't met. In broad generality, a good MS teaches you how to use and learn to use statistical methods; a good PhD teaches that, a little deeper, plus how to develop new methods to treat problems for which no methods have yet been developed. Neither is a finished product that ends the learning. Both represent the first steps on a lifelong path of learning statistics. I had my first statistics course in the 1940s. I am still learning, just as fast as ever.

    Now my personal take on the practicality of choosing the MS or PhD. Exceptions can be quoted, and have been on this string, but are rare. Most of the time, the following can be counted on. The MS can have lots of fun solving real problems with real data and can make a decent living. However, the MS is often subsidiary to a PhD and is assigned his work. The PhD has more selection in choosing tasks and professional duties. The PhD can do anything an MS can if he chooses to but can also better avoid what he chooses not to do, and he does it with a better salary. I still sweep dirty data off the floor, draw histograms, and perform simple t tests, as well as fit spooky data into three-way repeated measures ANCOVAs or find a way to detect a change in a time series where variances and data distributions change with time. In summary, an MS will have a good life; a PhD will have a better life.

    (Female statisticians are welcome to change the "he's" to "she's" above. I look forward to the development of a sexless pronoun. The process has already started with the use of categories like actor and aviator for both sexes; you never hear actress or aviatrix any more. One day we will have word for he/she that does that Ms did for Miss/Mrs.)

    -------------------------------------------
    Bob Riffenburgh
    Naval Medical Center
    -------------------------------------------








  • 37.  RE:MS vs PhD?

    Posted 10-08-2012 14:18
    This discussion has been turning more and more philosophical. I do agree that a PhD person will have, ON THE AVERAGE, more material and monetary success than an MS person. However, i do not agree that the PhD person will have a better life than the MS person. To me the "better" adjective deals with a welter of non-statistical issues. 

    I believe that the many responses (more than on any topic I can remember in the last year or two) have been very valuable and will help a person decide on what course is best for them. But, sometimes, it seems like we are asking, which is better, pistachio ice cream or banana ice cream? It depends on what you like! (I would prefer banana.) I do not feel that it is a better/worse decision. It is a what-do-you-like decision.

    Warmest regards and gratitude to all the many fine and thoughtful participants and readers.

    Nayak



    -------------------------------------------
    Nayak Polissar
    Principal Statistician
    The Mountain-Whisper-Light Statistics
    -------------------------------------------








  • 38.  RE:MS vs PhD?

    Posted 10-08-2012 14:40

    I agree with Nayak. I once had a colleague who said he "only had a Master's degree". His attitude should have been how special he is for having a Master's degree in Statistics. With such a self-deprecating attitude regarding his academic success,  I assume that he had the same attitude about life in general.
    David

    -------------------------------------------
    David Bristol
    Statistical Consulting Services
    -------------------------------------------








  • 39.  RE:MS vs PhD?

    Posted 10-09-2012 15:55
    I think that having a PhD in Statistics has much less career impact in many industries outside of Biomedical and Pharmceutical.  I have spent most of my career in food industry R&D as an internal consulting statistician and managing a global internal statistical consulting department.  My department peers have had both Masters and PhD degrees but there has been no distinction between them in terms of project work assignments and career opportunities.  Experience has been more important than degree level in our ability to effectively support projects.  And our internal clients have not cared about degree level, to them we are all statisticians.  I have statistician colleagues in other industries with similar experiences so I do not think that my career has been unique in this way.  But I do acknowledge that Biomedical and Pharmceutical industries represent a large proportion of jobs in Statistics outside of academia and government.

    I am happy that my Masters degree in Statistics has provided me with an enjoyable career to date.

    Thanks,
    Frank

    -------------------------------------------
    Francis Rossi
    Associate Director, Statistics
    Kraft Foods Group
    -------------------------------------------








  • 40.  RE:MS vs PhD?

    Posted 10-10-2012 08:35
    I would agree that that is the case in many industries;
    I don't expect that to be the case in the future -
    competition is stiff now, and will be increasing sharply
    for the foreseeable future.

    -------------------------------------------
    Barry DeCicco
    Statistician
    -------------------------------------------