Discussion: View Thread

First time consulting?

  • 1.  First time consulting?

    Posted 09-19-2011 14:23
    This message has been cross posted to the following eGroups: New York City Metro Area Chapter and Statistical Consulting Section .
    -------------------------------------------
    Hello everyone,

    I am thinking of doing some independent Biostatistics consulting..How do I start? I've been a Biostatistician for a year and half with 7 years of experience in Clinical IT (software development).

    Any thought?

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


  • 2.  RE:First time consulting?

    Posted 09-19-2011 16:28

    To be frank I don't think you are ready to do consulting if you only have a little over 1 year experience as a statistician.   I don't think the experience in clinical IT helps you either.  Most successful consultants get a lot of experience first, learn how to interact with client and make contacts through work and professional meetings.  It is a slow process.  I think some of us have said a number of times on this eGroup site that good consulting requires much more than technical expertise.  You need to listen to the client, try to become acquainted with his discipline and develop skill for find the right and if possible simple approach that answers the right question.+
    -------------------------------------------
    Michael Chernick
    Director of Biostatistical Services
    Lankenau Institute for Medical Research
    -------------------------------------------








  • 3.  RE:First time consulting?

    Posted 09-19-2011 16:37
    Given Michael's comments which I totally agree with, perhaps your best bet is to focus on Clinical IT Software consulting where you interface with biostats groups. Good luck.

    -------------------------------------------
    Patrick Spagon
    -------------------------------------------








  • 4.  RE:First time consulting?

    Posted 09-19-2011 17:13

    1 1/2 years experience is insufficient to begin statistical consulting. I worked
    in corporate America and academia for over 20 years before I began consulting.
    I cannot imagine even attempting that route with only 1 1/2 years!

    Shirlene

    -------------------------------------------
    Dr. N. Shirlene Pearson
    Statistical Consultant & Research Support Specialist
    Southern Methodist University
    Dallas, Texas USA
    -------------------------------------------








  • 5.  RE:First time consulting?

    Posted 09-19-2011 17:34

    There is statistical consulting and then there is statistical consulting. He (or she) did not say that the wanted to start a career as a full-time independent consultant. To me it sounds more like looking for some independent work on the side. There is nothing wrong with that.

    My sense is that being in a university hospital, there may be many physicians doing independent work themselves. Often they need help that requires nothing more than basic descriptive statistics or cleaning data from a spreadsheet. It is a way to get started.

    As pointed out, making use of his software skills can be a good source for statistical programming work. That also counts as consulting.

    Best regards, 
    -------------------------------------------
    Carlos Alzola
    President
    Data Insights Inc.
    -------------------------------------------








  • 6.  RE:First time consulting?

    Posted 09-19-2011 18:27

    Sorry Carlos, I respectful disagree! I do consulting for physicians all the time.  Usually they don't know what they want and it takes a good consultant to provide them the proper service that they really need.  Sure an inexperienced consultant can turn the crank and provide the requested services.  I was clear to say that you cannot be a "good" consultant with 1.5 years of experience.  I think to do data analysis to the request of a client you don't even need a degree in statistics.  But what you describe is not consulting the way I think of it and I doubt that I have ever done exact what a client asked me to do without discussion that leads to something a little different.  This is especially true for doctors and research nurses.
    -------------------------------------------
    Michael Chernick
    Director of Biostatistical Services
    Lankenau Institute for Medical Research
    -------------------------------------------








  • 7.  RE:First time consulting?

    Posted 09-19-2011 19:00
    I would like to suggest that a good way would be to ally yourself with a senior consultant (this would be like an apprenticeship for you--you pay the consultant) and learn. When you are ready, you can go independent, perhaps gradually. You can also just hire (and bill to the client) such an ally, and just keep asking them what to do. When I first started consulting, there was one extremely experienced statistician whom i would often go to for advice. That was very valuable. 

    Michael is right--so many of the physicians and researchers do not really know what they want to do, they just have a feeling about what area they want to work in. The statistician needs to help them design specific questions and studies and formalize vague notions into a bonafide study. The role of experience is very important here. In a parallel area, would you rather have your dear one's brain surgery done by a 20+ year surgeon or a resident? 

    Best wishes,

    Nayak





    -------------------------------------------
    Nayak Polissar
    Consultant
    The Mountain Whisper Light
    -------------------------------------------








  • 8.  RE:First time consulting?

    Posted 09-19-2011 20:01

    Of course I agree with Michael's remarks but eventhough I worked for big pharma for 30 years before becoming an independent statistical consultant, I recalled doing some stat work on the side especially in my early years.  I did some work at the local university hospital.  Their local stat group was very expensive with a lot of additional overhead and I got a few stat jobs where I just did simple stat support.  I even got on a few papers.  I didn't make a lot of money doing this but I always wondered what would have happened if I advertised a little - probably put up a few flyers or send out a letter to each of the physicians at the hospital.  I would suggest charging a modest fee at first and build up some clients.  Another idea is to provide some SAS support for local existing statistical consultants - there are times where we all could use some support.  I have two SAS experts I use on occassions when my work gets a little heavy.

    -------------------------------------------
    Rocco Brunelle
    Senior Statistician
    Bowsher Brunelle Smith LLC
    -------------------------------------------








  • 9.  RE:First time consulting?

    Posted 09-19-2011 20:33
    While 20 years as an applied statistician is a good way to gain experience to become a good statistical consultant, if someone wants to get experience consulting it is not necessary to wait 20 years.  One or two years serving as an apprentice to a senior statistical consultant would be the best way to get experience quickly.  I have over the past three years had a young college student work with me as a volunteer.  I have given her direction and explained why we do things certain ways. I think the most valuable thing i can do as a mentor is to have the apprentice attend the consulting sessions and allow her to ask questions to the client as well.  This year my apprentice who was an undergraduate stat major when she volunteered for me now has her bachelor's degree and is going on for a masters while doing volunteer work for me again.  She enjoys the experience and is learning and performing well.  When she graduates from Temple I am sure she will be prepared for a career in statistics and a stab at independent consulting.

    So I agree that working with an experienced consultant is the way to go but finding one may be the problem.  I think that there are small consulting groups that take on supporting statisticians. You may find them right here in this section.

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








  • 10.  RE:First time consulting?

    Posted 09-19-2011 20:20
    I am certainly hoping that the requirement for being an independent consultant does not require many years of working for a company doing statistics.  What if you worked in other disciplines before getting formal statistical training?  I am doing my training now with the intent of working from home while my kids are still young.  I am switching fields from teaching and am in my 2nd year of getting a PhD in Statistics.  If this doesn't work, then I am making a colossal mistake.

    Let me give my own case to explain,
    I already have some experience in other disciplines:
    1) A Master's in Biochemistry (8 years of laboratory experience)
    2) Middle/High school teaching license in Physics, Math, and Biology in two states with 4+ years of teaching experience in public and private school.
    3) 2 years University teaching experience
    4) 5 years experience in town organizations and political action groups.
    5) 10+ years web design experience, including knowledge of coding in html, Joomla, Perl, C++, Java
    6) Trained HVAC technician with 7 years on the job experience.
    7) Multilingual: English, French, Haitian Creole

    I would echo an earlier comment on the board that the original poster, Mr. Islam, should try to work in areas that he has special knowledge of, and therefore knows the lingo and any special conditions of the field.  I plan to focus my practice in Educational consulting (providing assistance to schools and teachers in developing data driven assessments/instruction) or perhaps basic science research support, since I have a lot of contacts and can speak the language in those fields better than a typical consultant.  I have seen a number of statisticians make basic errors simply because they don't truly understand the context the client is working in, yet 99.9% of the people who need statistics don't know the first thing about basic statistics like p values or population assumptions.  You have to know the context because its critical to communicate across the knowledge gap for your clients and get it right.  The people who are going to pay out significant dollars need super consultants.

    In addition, as someone with a background in small business, I think you need to make yourself stand out to be successful when you have little independent experience.  Working with a senior consultant can be a good idea, but I wouldn't expect to earn much money during the learning phase.  However, I wouldn't guarantee that striking out on your own is advisable to anyone who doesn't have a backup plan or a "day job" to support the early days while building your client-base and learning the landscape, however.  I would work on making a business plan and think about answering questions like the following:

    1.) Who are my ideal clients? Where do they work? What do they need? What software are they used to?
    2.) How much is my service worth to clients?  How much will they pay?
    3.) How much will it cost to acquire the software needed to do the consulting?
    4.) Who is my competition?  Why would a client pick me over them?





    -------------------------------------------
    Michael Bilica
    Graduate Student
    University of Connecticut
    -------------------------------------------








  • 11.  RE:First time consulting?

    Posted 09-19-2011 21:37

    Welcome to the discussion and to the Section. 

    The short answer, is- yes-, you can start a consulting business, though I think more experience and possibly more certification/degrees would be helpful.

    I think its fair to say you are just beginning.

    A few questions (and you are welcome to  send answers  by private email)


    Some things that would be useful to know are your degree (BA, MS, PHD), and the kind of -statistical- and -IT- projects you work on.

    As to the IT, are you programming in SAS or other language(s)? In advance, are you SAS certified/

    And, have you looked into the ASA certification program?


    -------------------------------------------
    Chris Barker, Ph.D.
    President - San Francisco Bay Area Chapter of the American Statistical Association
    www,barkerstats.com
    -------------------------------------------