Loading ...
Sorry, an error occurred while loading the content.
 

R article at the new york times

Expand Messages
  • gerrobrein
    http://www.nytimes.com/2009/01/07/technology/business- computing/07program.html?_r=2
    Message 1 of 7 , Jan 8, 2009
    • Will Hopkins
      But R has also quickly found a following because statisticians, engineers and scientists without computer programming skills find it easy to use. This is an
      Message 2 of 7 , Jan 8, 2009

        “But R has also quickly found a following because statisticians, engineers and scientists without computer programming skills find it easy to use.”

        This is an example of why I don’t read newspapers on or off line.

         

        Will

         

        From: sportsci_rtutorial@yahoogroups.com [mailto:sportsci_rtutorial@yahoogroups.com] On Behalf Of gerrobrein
        Sent: Friday, 9 January 2009 1:29 a.m.
        To: sportsci_rtutorial@yahoogroups.com
        Subject: [sportsci_rtutorial] R article at the new york times

         

      • Ian Shrier
        Will Within R, there is a feature called R Commander. Once loaded, it is menu-driven. Further, as it runs, it writes the code so you learn how to program as
        Message 3 of 7 , Jan 8, 2009
          Will

          Within R, there is a feature called R Commander. Once loaded, it is menu-driven. Further, as it runs, it writes the code so you learn how to program as you go along.

          Unfortunately, I only learned about it after I went through the steepest part of the learning curve. And although I probably should use it sometimes, I tend to rely on the help documentation or my statistician friend who is an R grand master...very helpful if you have one.


          Ian Shrier MD, PhD, Dip Sport Med, FACSM
          Associate Professor, Dep't of Fam Med, McGill University
          Past-President, Canadian Academy of Sport Medicine
          Check out: www.casm-acms.org
          SKYPE name: ian.shrier

          Centre for Clinical Epidemiology and Community Studies
          SMBD-Jewish General Hospital
          3755 Cote Ste-Catherine Rd
          Montreal, Qc  H3T 1E2
          Tel: 514-340-7563
          Fax: 514-340-7564





          On 8-Jan-09, at 4:05 PM, Will Hopkins wrote:

          “But R has also quickly found a following because statisticians, engineers and scientists without computer programming skills find it easy to use.”
          This is an example of why I don’t read newspapers on or off line.
           
          Will
           
          From: sportsci_rtutorial@yahoogroups.com [mailto:sportsci_rtutorial@yahoogroups.com] On Behalf Of gerrobrein
          Sent: Friday, 9 January 2009 1:29 a.m.
          To: sportsci_rtutorial@yahoogroups.com
          Subject: [sportsci_rtutorial] R article at the new york times
           

        • Will Hopkins
          I presume folks won t mind if this interaction continues via the list. Others might make some useful comments. I hadn t realized a menu-driven R was up and
          Message 4 of 7 , Jan 9, 2009

            I presume folks won’t mind if this interaction continues via the list.  Others might make some useful comments.

             

            I hadn’t realized a menu-driven R was up and running.  If that’s what the NYT article was referring to, it would imply that menu-driven R is easier to use than all the other menu-driven stats packages. Is R’s menu interface any different from any other?  Does the interface somehow capture R’s strange concepts/jargon for specifying data and models (which I still haven’t attempted to understand, I must admit)?

             

            Will

             

            From: sportsci_rtutorial@yahoogroups.com [mailto:sportsci_rtutorial@yahoogroups.com] On Behalf Of Ian Shrier
            Sent: Friday, 9 January 2009 11:17 a.m.
            To: sportsci_rtutorial@yahoogroups.com
            Subject: Re: [sportsci_rtutorial] R article at the new york times

             

            Will

             

            Within R, there is a feature called R Commander. Once loaded, it is menu-driven. Further, as it runs, it writes the code so you learn how to program as you go along.

             

            Unfortunately, I only learned about it after I went through the steepest part of the learning curve. And although I probably should use it sometimes, I tend to rely on the help documentation or my statistician friend who is an R grand master...very helpful if you have one.

             

            Ian Shrier MD, PhD, Dip Sport Med, FACSM
            Associate Professor, Dep't of Fam Med, McGill University

            Past-President, Canadian Academy of Sport Medicine

            Check out: www.casm-acms.org
            SKYPE name: ian.shrier

            Centre for Clinical Epidemiology and Community Studies
            SMBD-Jewish General Hospital
            3755 Cote Ste-Catherine Rd
            Montreal, Qc  H3T 1E2
            Tel: 514-340-7563
            Fax: 514-340-7564

             

             

             

             

             

            On 8-Jan-09, at 4:05 PM, Will Hopkins wrote:



            “But R has also quickly found a following because statisticians, engineers and scientists without computer programming skills find it easy to use.”

            This is an example of why I don’t read newspapers on or off line.

             

            Will

             

            From: sportsci_rtutorial@yahoogroups.com [mailto:sportsci_rtutorial@yahoogroups.com] On Behalf Of gerrobrein
            Sent: Friday, 9 January 2009 1:29 a.m.
            To: sportsci_rtutorial@yahoogroups.com
            Subject: [sportsci_rtutorial] R article at the new york times

             

             

          • gerrobrein
            Hi Will, Great to see some action happening on this list. Personally, I don t think that providing a menu-driven interface is of any help. We discussed it a
            Message 5 of 7 , Jan 9, 2009
              Hi Will,

              Great to see some action happening on this list.
              Personally, I don't think that providing a menu-driven interface is
              of any help. We discussed it a while ago when you came down to
              Dunedin. I really am convinced that tables and mice don't do the
              trick. Also, I'm 100% percent sure that an undergrad not being
              polluted by Excel, SPSS or the like (try SPSS's scripting language if
              you want to see strange concepts by the way) will learn R in no time.
              Getting used to the commands may seem complicated at first but once
              you deal with real data, which tends to be really messy, it will be
              difficult to find the right menu anyway, because typically it doesn't
              exist yet.
              Further, as the amount of data in our area is also constantly on the
              rise: From my own experience, for my master and my PhD I analyzed
              whole body joint kinematics:
              Master: ~216000 Data pts
              PhD: ~157500000 Data pts
              Today, there is just no escape from getting profficient in at least
              one programming language. Might as well be R.
              Anyway, I'll try the R commander over the weekend and give some
              feedback.

              Cheers
              Rob

              --- In sportsci_rtutorial@yahoogroups.com, "Will Hopkins" <will@...>
              wrote:
              >
              > I presume folks won't mind if this interaction continues via the
              list.
              > Others might make some useful comments.
              >
              >
              >
              > I hadn't realized a menu-driven R was up and running. If that's
              what the
              > NYT article was referring to, it would imply that menu-driven R is
              easier to
              > use than all the other menu-driven stats packages. Is R's menu
              interface any
              > different from any other? Does the interface somehow capture R's
              strange
              > concepts/jargon for specifying data and models (which I still
              haven't
              > attempted to understand, I must admit)?
              >
              >
              >
              > Will
              >
              >
              >
              > From: sportsci_rtutorial@yahoogroups.com
              > [mailto:sportsci_rtutorial@yahoogroups.com] On Behalf Of Ian Shrier
              > Sent: Friday, 9 January 2009 11:17 a.m.
              > To: sportsci_rtutorial@yahoogroups.com
              > Subject: Re: [sportsci_rtutorial] R article at the new york times
              >
              >
              >
              > Will
              >
              >
              >
              > Within R, there is a feature called R Commander. Once loaded, it is
              > menu-driven. Further, as it runs, it writes the code so you learn
              how to
              > program as you go along.
              >
              >
              >
              > Unfortunately, I only learned about it after I went through the
              steepest
              > part of the learning curve. And although I probably should use it
              sometimes,
              > I tend to rely on the help documentation or my statistician friend
              who is an
              > R grand master...very helpful if you have one.
              >
              >
              >
              > Ian Shrier MD, PhD, Dip Sport Med, FACSM
              > Associate Professor, Dep't of Fam Med, McGill University
              >
              > Past-President, Canadian Academy of Sport Medicine
              >
              > Check out: www.casm-acms.org
              > SKYPE name: ian.shrier
              >
              > Centre for Clinical Epidemiology and Community Studies
              > SMBD-Jewish General Hospital
              > 3755 Cote Ste-Catherine Rd
              > Montreal, Qc H3T 1E2
              > Tel: 514-340-7563
              > Fax: 514-340-7564
              >
              >
              >
              >
              >
              >
              >
              >
              >
              >
              >
              > On 8-Jan-09, at 4:05 PM, Will Hopkins wrote:
              >
              >
              >
              >
              >
              > "But R has also quickly found a following because statisticians,
              engineers
              > and scientists without computer programming skills find it easy to
              use."
              >
              > This is an example of why I don't read newspapers on or off line.
              >
              >
              >
              > Will
              >
              >
              >
              > From: sportsci_rtutorial@yahoogroups.com
              > [mailto:sportsci_rtutorial@yahoogroups.com] On Behalf Of gerrobrein
              > Sent: Friday, 9 January 2009 1:29 a.m.
              > To: sportsci_rtutorial@yahoogroups.com
              > Subject: [sportsci_rtutorial] R article at the new york times
              >
              >
              >
              > http://www.nytimes.com/2009/01/07/technology/business-
              > computing/07program.html?_r=2
              >
            • Ian Shrier
              I agree with Rob that the menu-driven approach is limited to simple analyses. But this is going to be adequate for many studies. I also have become more aware
              Message 6 of 7 , Jan 9, 2009
                I agree with Rob that the menu-driven approach is limited to simple analyses. But this is going to be adequate for many studies.

                I also have become more aware of all the grave statistical mistakes I have made in the past. If you don't have the time to learn how to program (in at least one of the software packages) for the more complicated analyses (and maybe even for the simple ones), then you probably don't have the time to learn all the nuances and things to avoid doing that everyone is else is doing but doing them wrong. I have had some wonderful collaborations with statisticians and if you can find one that is interested in your research question, use him/her - it is also a great way to get free private tutoring.

                Here is a little saying that maybe is a bit harsh and over the top, but it does make a point that we have to be very careful:
                _______
                “A mistake in the operating room can threaten the life of one patient; a mistake in statistical analysis or interpretation can lead to hundreds of early deaths. So it is perhaps odd that, while we allow a doctor to conduct surgery only after years of training, we give SPSS to almost anyone” 

                Vickers A. Interpreting data from randomized trials: the Scandinavian prostatectomy study illustrates two common errors. Nat Clin Pract Urol 2005;2(9):404-5.
                ________

                Ian Shrier MD, PhD, Dip Sport Med, FACSM
                Associate Professor, Dep't of Fam Med, McGill University
                Past-President, Canadian Academy of Sport Medicine
                Check out: www.casm-acms.org
                SKYPE name: ian.shrier

                Centre for Clinical Epidemiology and Community Studies
                SMBD-Jewish General Hospital
                3755 Cote Ste-Catherine Rd
                Montreal, Qc  H3T 1E2
                Tel: 514-340-7563
                Fax: 514-340-7564





                On 9-Jan-09, at 5:53 PM, gerrobrein wrote:

                Hi Will,

                Great to see some action happening on this list.
                Personally, I don't think that providing a menu-driven interface is
                of any help. We discussed it a while ago when you came down to
                Dunedin. I really am convinced that tables and mice don't do the
                trick. Also, I'm 100% percent sure that an undergrad not being
                polluted by Excel, SPSS or the like (try SPSS's scripting language if
                you want to see strange concepts by the way) will learn R in no time.
                Getting used to the commands may seem complicated at first but once
                you deal with real data, which tends to be really messy, it will be
                difficult to find the right menu anyway, because typically it doesn't
                exist yet.
                Further, as the amount of data in our area is also constantly on the
                rise: From my own experience, for my master and my PhD I analyzed
                whole body joint kinematics:
                Master: ~216000 Data pts
                PhD:    ~157500000 Data pts
                Today, there is just no escape from getting profficient in at least
                one programming language. Might as well be R.
                Anyway, I'll try the R commander over the weekend and give some
                feedback.

                Cheers
                Rob

                --- In sportsci_rtutorial@yahoogroups.com, "Will Hopkins" <will@...>
                wrote:

                I presume folks won't mind if this interaction continues via the
                list.
                Others might make some useful comments.



                I hadn't realized a menu-driven R was up and running.  If that's
                what the
                NYT article was referring to, it would imply that menu-driven R is
                easier to
                use than all the other menu-driven stats packages. Is R's menu
                interface any
                different from any other?  Does the interface somehow capture R's
                strange
                concepts/jargon for specifying data and models (which I still
                haven't
                attempted to understand, I must admit)?



                Will



                From: sportsci_rtutorial@yahoogroups.com
                [mailto:sportsci_rtutorial@yahoogroups.com] On Behalf Of Ian Shrier
                Sent: Friday, 9 January 2009 11:17 a.m.
                To: sportsci_rtutorial@yahoogroups.com
                Subject: Re: [sportsci_rtutorial] R article at the new york times



                Will



                Within R, there is a feature called R Commander. Once loaded, it is
                menu-driven. Further, as it runs, it writes the code so you learn
                how to
                program as you go along.



                Unfortunately, I only learned about it after I went through the
                steepest
                part of the learning curve. And although I probably should use it
                sometimes,
                I tend to rely on the help documentation or my statistician friend
                who is an
                R grand master...very helpful if you have one.



                Ian Shrier MD, PhD, Dip Sport Med, FACSM
                Associate Professor, Dep't of Fam Med, McGill University

                Past-President, Canadian Academy of Sport Medicine

                Check out: www.casm-acms.org
                SKYPE name: ian.shrier

                Centre for Clinical Epidemiology and Community Studies
                SMBD-Jewish General Hospital
                3755 Cote Ste-Catherine Rd
                Montreal, Qc  H3T 1E2
                Tel: 514-340-7563
                Fax: 514-340-7564











                On 8-Jan-09, at 4:05 PM, Will Hopkins wrote:





                "But R has also quickly found a following because statisticians,
                engineers
                and scientists without computer programming skills find it easy to
                use."

                This is an example of why I don't read newspapers on or off line.



                Will



                From: sportsci_rtutorial@yahoogroups.com
                [mailto:sportsci_rtutorial@yahoogroups.com] On Behalf Of gerrobrein
                Sent: Friday, 9 January 2009 1:29 a.m.
                To: sportsci_rtutorial@yahoogroups.com
                Subject: [sportsci_rtutorial] R article at the new york times



                http://www.nytimes.com/2009/01/07/technology/business-
                computing/07program.html?_r=2




                ------------------------------------

                Yahoo! Groups Links

                <*> To visit your group on the web, go to:
                   http://groups.yahoo.com/group/sportsci_rtutorial/

                <*> Your email settings:
                   Individual Email | Traditional

                <*> To change settings online go to:
                   http://groups.yahoo.com/group/sportsci_rtutorial/join
                   (Yahoo! ID required)

                <*> To change settings via email:
                   mailto:sportsci_rtutorial-digest@yahoogroups.com
                   mailto:sportsci_rtutorial-fullfeatured@yahoogroups.com

                <*> To unsubscribe from this group, send an email to:
                   sportsci_rtutorial-unsubscribe@yahoogroups.com

                <*> Your use of Yahoo! Groups is subject to:
                   http://docs.yahoo.com/info/terms/


              • gerrobrein
                Hello Guys, I ve checked out, although only brief and superficial, R commander. The interface provides most of the typical tests used in undergraduate
                Message 7 of 7 , Jan 11, 2009
                  Hello Guys,

                  I've checked out, although only brief and superficial, R commander.
                  The interface provides most of the typical tests used in
                  undergraduate training. Thus, it could be used to ease novices into
                  working with R.
                  Check out:
                  http://socserv.mcmaster.ca/jfox/Misc/Rcmdr/

                  It is also possible to extend the basic functionality and write add-
                  on menus for new routines. Could be usefull if you tend to use the
                  same routines again and again in your team and you have somebody
                  knowing how to programm in R and Tcl/Tk. This way, simply add a new
                  menu entry, and everybody can make happy mice clicking (at least in
                  principle). Thus, providing similiar out-of-the-box functionality
                  just like Will does through his Excel sheets.
                  Therefore check it out.

                  Cheers
                  Rob

                  >
                  > I agree with Rob that the menu-driven approach is limited to
                  simple
                  > analyses. But this is going to be adequate for many studies.
                  >
                  > I also have become more aware of all the grave statistical mistakes
                  I
                  > have made in the past. If you don't have the time to learn how to
                  > program (in at least one of the software packages) for the more
                  > complicated analyses (and maybe even for the simple ones), then
                  you
                  > probably don't have the time to learn all the nuances and things
                  to
                  > avoid doing that everyone is else is doing but doing them wrong. I
                  > have had some wonderful collaborations with statisticians and if
                  you
                  > can find one that is interested in your research question, use him/
                  her
                  > - it is also a great way to get free private tutoring.
                  >
                  > Here is a little saying that maybe is a bit harsh and over the
                  top,
                  > but it does make a point that we have to be very careful:
                  > _______
                  > "A mistake in the operating room can threaten the life of one
                  patient;
                  > a mistake in statistical analysis or interpretation can lead to
                  > hundreds of early deaths. So it is perhaps odd that, while we allow
                  a
                  > doctor to conduct surgery only after years of training, we give
                  SPSS
                  > to almost anyone"
                  >
                  > Vickers A. Interpreting data from randomized trials: the
                  Scandinavian
                  > prostatectomy study illustrates two common errors. Nat Clin Pract
                  Urol
                  > 2005;2(9):404-5.
                  > ________
                  >
                  > Ian Shrier MD, PhD, Dip Sport Med, FACSM
                  > Associate Professor, Dep't of Fam Med, McGill University
                  > Past-President, Canadian Academy of Sport Medicine
                  > Check out: www.casm-acms.org
                  > SKYPE name: ian.shrier
                  >
                  > Centre for Clinical Epidemiology and Community Studies
                  > SMBD-Jewish General Hospital
                  > 3755 Cote Ste-Catherine Rd
                  > Montreal, Qc H3T 1E2
                  > Tel: 514-340-7563
                  > Fax: 514-340-7564
                  >
                  >
                  >
                  >
                  >
                  > On 9-Jan-09, at 5:53 PM, gerrobrein wrote:
                  >
                  > > Hi Will,
                  > >
                  > > Great to see some action happening on this list.
                  > > Personally, I don't think that providing a menu-driven interface
                  is
                  > > of any help. We discussed it a while ago when you came down to
                  > > Dunedin. I really am convinced that tables and mice don't do the
                  > > trick. Also, I'm 100% percent sure that an undergrad not being
                  > > polluted by Excel, SPSS or the like (try SPSS's scripting
                  language if
                  > > you want to see strange concepts by the way) will learn R in no
                  time.
                  > > Getting used to the commands may seem complicated at first but
                  once
                  > > you deal with real data, which tends to be really messy, it will
                  be
                  > > difficult to find the right menu anyway, because typically it
                  doesn't
                  > > exist yet.
                  > > Further, as the amount of data in our area is also constantly on
                  the
                  > > rise: From my own experience, for my master and my PhD I analyzed
                  > > whole body joint kinematics:
                  > > Master: ~216000 Data pts
                  > > PhD: ~157500000 Data pts
                  > > Today, there is just no escape from getting profficient in at
                  least
                  > > one programming language. Might as well be R.
                  > > Anyway, I'll try the R commander over the weekend and give some
                  > > feedback.
                  > >
                  > > Cheers
                  > > Rob
                  > >
                  > > --- In sportsci_rtutorial@yahoogroups.com, "Will Hopkins" <will@>
                  > > wrote:
                  > >>
                  > >> I presume folks won't mind if this interaction continues via the
                  > > list.
                  > >> Others might make some useful comments.
                  > >>
                  > >>
                  > >>
                  > >> I hadn't realized a menu-driven R was up and running. If that's
                  > > what the
                  > >> NYT article was referring to, it would imply that menu-driven R
                  is
                  > > easier to
                  > >> use than all the other menu-driven stats packages. Is R's menu
                  > > interface any
                  > >> different from any other? Does the interface somehow capture R's
                  > > strange
                  > >> concepts/jargon for specifying data and models (which I still
                  > > haven't
                  > >> attempted to understand, I must admit)?
                  > >>
                  > >>
                  > >>
                  > >> Will
                  > >>
                  > >>
                  > >>
                  > >> From: sportsci_rtutorial@yahoogroups.com
                  > >> [mailto:sportsci_rtutorial@yahoogroups.com] On Behalf Of Ian
                  Shrier
                  > >> Sent: Friday, 9 January 2009 11:17 a.m.
                  > >> To: sportsci_rtutorial@yahoogroups.com
                  > >> Subject: Re: [sportsci_rtutorial] R article at the new york times
                  > >>
                  > >>
                  > >>
                  > >> Will
                  > >>
                  > >>
                  > >>
                  > >> Within R, there is a feature called R Commander. Once loaded, it
                  is
                  > >> menu-driven. Further, as it runs, it writes the code so you learn
                  > > how to
                  > >> program as you go along.
                  > >>
                  > >>
                  > >>
                  > >> Unfortunately, I only learned about it after I went through the
                  > > steepest
                  > >> part of the learning curve. And although I probably should use it
                  > > sometimes,
                  > >> I tend to rely on the help documentation or my statistician
                  friend
                  > > who is an
                  > >> R grand master...very helpful if you have one.
                  > >>
                  > >>
                  > >>
                  > >> Ian Shrier MD, PhD, Dip Sport Med, FACSM
                  > >> Associate Professor, Dep't of Fam Med, McGill University
                  > >>
                  > >> Past-President, Canadian Academy of Sport Medicine
                  > >>
                  > >> Check out: www.casm-acms.org
                  > >> SKYPE name: ian.shrier
                  > >>
                  > >> Centre for Clinical Epidemiology and Community Studies
                  > >> SMBD-Jewish General Hospital
                  > >> 3755 Cote Ste-Catherine Rd
                  > >> Montreal, Qc H3T 1E2
                  > >> Tel: 514-340-7563
                  > >> Fax: 514-340-7564
                  > >>
                  > >>
                  > >>
                  > >>
                  > >>
                  > >>
                  > >>
                  > >>
                  > >>
                  > >>
                  > >>
                  > >> On 8-Jan-09, at 4:05 PM, Will Hopkins wrote:
                  > >>
                  > >>
                  > >>
                  > >>
                  > >>
                  > >> "But R has also quickly found a following because statisticians,
                  > > engineers
                  > >> and scientists without computer programming skills find it easy
                  to
                  > > use."
                  > >>
                  > >> This is an example of why I don't read newspapers on or off line.
                  > >>
                  > >>
                  > >>
                  > >> Will
                  > >>
                  > >>
                  > >>
                  > >> From: sportsci_rtutorial@yahoogroups.com
                  > >> [mailto:sportsci_rtutorial@yahoogroups.com] On Behalf Of
                  gerrobrein
                  > >> Sent: Friday, 9 January 2009 1:29 a.m.
                  > >> To: sportsci_rtutorial@yahoogroups.com
                  > >> Subject: [sportsci_rtutorial] R article at the new york times
                  > >>
                  > >>
                  > >>
                  > >> http://www.nytimes.com/2009/01/07/technology/business-
                  > >> computing/07program.html?_r=2
                  > >>
                  > >
                  > >
                  > >
                  > > ------------------------------------
                  > >
                  > > Yahoo! Groups Links
                  > >
                  > >
                  > >
                  >
                Your message has been successfully submitted and would be delivered to recipients shortly.