Is R overtaking SAS at universities and colleges?
in [SAS]
|
From: Andre Wielki on 4 Jul 2008 11:19 In this day where perhaps the majority of us (US a ) are not on line, i can react about the two trends Re: Is R overtaking SAS at universities and colleges? and Re: A powerpoint slide show comparing R with SAS and SPSS i have readed 1) i have started a little with R as there was here at beginning june a presentation of FactoMiner by their team and in fact in the demography area R is gaining places together with Stata at our school EDSD (european school for demography) without going in the twist (because i am not a statistician), i recognize completely that you have to "think" different in R 2) but nevertheless i have study and test a lot after reading on both list (sas and R) the difficulties encountered by people on the topic coming from sas with data tables i have to offer a pdf illustrated note IN FRENCH with the topic "Aller/Retour entre SAS/R" tested through multiple way to go to R from SAS and return and showing the structure of the results upon the class table enriched with dates formatted or not and missing data (character and numeric) Andre PS:send me a request if interested at wielki(a)ined.fr so that i will send it monday afternoon this note is in french but with a lot of code so may be other people can be interrested too RolandRB a �crit : > On Jun 30, 7:34 pm, blindgl...(a)GMAIL.COM ("A.J. Rossini") wrote: > >> On Mon, Jun 30, 2008 at 4:17 PM, RolandRB <rolandbe...(a)hotmail.com> wrote: >> >>> That Sweave looks very interesting. And I have found the "rreports" >>> and "hmisc" packages which might do something like in powerpoint show. >>> And R seems extendable so maybe I could write such "packages" that can >>> produce the reports I want. And I like Latex for the formatting I like >>> the idea of including the code that actually runs to produce what is >>> in the document.It seems light years ahead of the current systems for >>> producing clinical reports. >>> >> Look at the ODFweave package, which combined with Sun's ODF to MS >> Office converts Sweave-style documents to MS Word or MS Powerpoint. >> >> (or just use TeX/LaTeX, and the Sweave package combined with ones >> favoritelatex/tex styles). >> >> To bring this back to SAS, ODS should be able to do most of this, >> however, it probably requires a similar level of processing (RTF >> production rather than MS Word native format). Does anyone know? >> >> -- >> best, >> -tony >> >> blindgl...(a)gmail.com >> Muttenz, Switzerland. >> "Commit early,commit often, and commit in a repository from which we >> can easily roll-back your mistakes" (AJR, 4Jan05). >> >> Drink Coffee: Do stupid things faster with more energy! >> Andr� WIELKI INED (Institut National d'Etudes D�mographiques) Service Informatique 133 Boulevard Davout 75980 Paris Cedex 20 m�l : wielki(a)ined.fr t�l : 33 (0) 1 56 06 21 54
From: Angel on 8 Jul 2008 15:50 On Jun 30, 2:40 am, RolandRB <rolandbe...(a)hotmail.com> wrote: > Is R overtaking SAS at universities and colleges? I hear that it is > but only have to go on hearsay. I also heard that SPlus might be > discontinued. Should we be moving across to R in the clinical > reporting world? It would be good to read of accounts where this has > been done. At UCLA here's the deal (as I see it from a PH student standpoint) Stat department uses primarily R. Im unaware of what they do for data management. Biostat students are taught primarily SAS and STATA. R (graphics) is available as an elective. So any R you want to learn is on your own. Since quite a few biostat students end up going into pharma or medical consulting, I can't see SAS being overthrown that easily. SAS also seems to be used primarily by academic researchers (at least in public health), some use SPSS but i'm not sure where they teach it on campus. Personally, i do all data management and some analysis/modeling in SAS, but for better graphics/more advanced models and/or bayesian analysis I use R. As for reporting, I've used latex a little, and will use it a lot later (dissertation), but never really was fond of direct ODS reporting. I tend to restructure/rearrange/organize only the output i want to display in SAS and then ODS it and put the finishing touches on it personally using Excel.
From: AnnMaria DeMars on 8 Jul 2008 16:09 I certainly don't see that happening here. Our department offers course in SAS, SPSS and we will start offering a Stata course in the fall. We teach these because this is where there is demand. The recent addition of a Stata course is because we have had people requesting it. SPSS, SAS and Stata are all available free in our computer labs on campus, so it may be a self-fulfilling prophecy in that people ask for training on what is readily available. I have had one person, who was taking an Advanced SAS class, ask me how he could do something in SAS that he had done in R previously. He was switching to SAS because that was what the PI on the grant that now employed him preferred they use. AnnMaria DeMars Customer Support Center ademars(a)usc.edu Information Technology Services (213) 740-2840 University of Southern California On Jul 8, 2008, at 12:50 PM, Angel wrote: > On Jun 30, 2:40 am, RolandRB <rolandbe...(a)hotmail.com> wrote: >> Is R overtaking SAS at universities and colleges? I hear that it is >> but only have to go on hearsay. I also heard that SPlus might be >> discontinued. Should we be moving across to R in the clinical >> reporting world? It would be good to read of accounts where this has >> been done. > > > At UCLA here's the deal (as I see it from a PH student standpoint) > > Stat department uses primarily R. Im unaware of what they do for data > management. > Biostat students are taught primarily SAS and STATA. R (graphics) is > available as an elective. So any R you want to learn is on your own. > > Since quite a few biostat students end up going into pharma or medical > consulting, I can't see SAS being overthrown that easily. SAS also > seems to be used primarily by academic researchers (at least in public > health), some use SPSS but i'm not sure where they teach it on > campus. > > Personally, i do all data management and some analysis/modeling in > SAS, but for better graphics/more advanced models and/or bayesian > analysis I use R. > > As for reporting, I've used latex a little, and will use it a lot > later (dissertation), but never really was fond of direct ODS > reporting. I tend to restructure/rearrange/organize only the output i > want to display in SAS and then ODS it and put the finishing touches > on it personally using Excel.
From: Mary on 8 Jul 2008 16:10 I was just looking at the R web page (OK, my brick wall has really come = down in the past 5 days!!), and noted the supporting institutions; = interesting that there are three biostatistics schools and 5 statistics = programs. Being from Iowa, I was surprised to see both the Univ. of = Iowa (where I work) and Iowa State University on the list, but only for = the statistics program, where the Biostatistics program at the Univ. of = Iowa is not. =20 a.. Department of Biostatistics, Johns Hopkins University, Maryland, USA = a.. Department of Biostatistics, Vanderbilt University School of = Medicine, USA=20 a.. Division of Biostatistics, University of California, Berkeley, USA=20 a.. Department of Mathematics and Statistics, Utah State University, USA = a.. Department of Statistics, University of California at Los Angeles, = USA=20 a.. Department of Statistics, University of Wisconsin-Madison, = Wisconsin, USA=20 a.. Department of Statistics, Iowa State University, USA=20 a.. Department of Statistics & Actuarial Science, University of Iowa, = USA=20 Also, I couldn't help noticing a familiar name to SAS-L on the R donors = list: a.. Peter L. Flom (USA)=20 I wonder why the difference in more statistics programs going to using R = whereas the biostatistics programs are not; perhaps biostatistics = considers itself to be more applied than theoretical? However, in terms = of prestige, John Hopkins is certainly one of the best colleges in the = country, so that their biostatistics program is supporting R is of = interest.=20 -Mary ----- Original Message -----=20 From: Angel=20 To: SAS-L(a)LISTSERV.UGA.EDU=20 Sent: Tuesday, July 08, 2008 2:50 PM Subject: Re: Is R overtaking SAS at universities and colleges? On Jun 30, 2:40 am, RolandRB <rolandbe...(a)hotmail.com> wrote: > Is R overtaking SAS at universities and colleges? I hear that it is > but only have to go on hearsay. I also heard that SPlus might be > discontinued. Should we be moving across to R in the clinical > reporting world? It would be good to read of accounts where this has > been done. At UCLA here's the deal (as I see it from a PH student standpoint) Stat department uses primarily R. Im unaware of what they do for data management. Biostat students are taught primarily SAS and STATA. R (graphics) is available as an elective. So any R you want to learn is on your own. Since quite a few biostat students end up going into pharma or medical consulting, I can't see SAS being overthrown that easily. SAS also seems to be used primarily by academic researchers (at least in public health), some use SPSS but i'm not sure where they teach it on campus. Personally, i do all data management and some analysis/modeling in SAS, but for better graphics/more advanced models and/or bayesian analysis I use R. As for reporting, I've used latex a little, and will use it a lot later (dissertation), but never really was fond of direct ODS reporting. I tend to restructure/rearrange/organize only the output i want to display in SAS and then ODS it and put the finishing touches on it personally using Excel.
From: Ben on 9 Jul 2008 11:24
Why bother? On Tue, 8 Jul 2008 15:10:28 -0500, Mary <mlhoward(a)AVALON.NET> wrote: >I was just looking at the R web page (OK, my brick wall has really come down in the past 5 days!!), and noted the supporting institutions; interesting that there are three biostatistics schools and 5 statistics programs. Being from Iowa, I was surprised to see both the Univ. of Iowa (where I work) and Iowa State University on the list, but only for the statistics program, where the Biostatistics program at the Univ. of Iowa is not. > >a.. Department of Biostatistics, Johns Hopkins University, Maryland, USA >a.. Department of Biostatistics, Vanderbilt University School of Medicine, USA >a.. Division of Biostatistics, University of California, Berkeley, USA >a.. Department of Mathematics and Statistics, Utah State University, USA >a.. Department of Statistics, University of California at Los Angeles, USA >a.. Department of Statistics, University of Wisconsin-Madison, Wisconsin, USA >a.. Department of Statistics, Iowa State University, USA >a.. Department of Statistics & Actuarial Science, University of Iowa, USA > >Also, I couldn't help noticing a familiar name to SAS-L on the R donors list: >a.. Peter L. Flom (USA) > >I wonder why the difference in more statistics programs going to using R whereas the biostatistics programs are not; perhaps biostatistics considers itself to be more applied than theoretical? However, in terms of prestige, John Hopkins is certainly one of the best colleges in the country, so that their biostatistics program is supporting R is of interest. > >-Mary > > >----- Original Message ----- > From: Angel > To: SAS-L(a)LISTSERV.UGA.EDU > Sent: Tuesday, July 08, 2008 2:50 PM > Subject: Re: Is R overtaking SAS at universities and colleges? > > > On Jun 30, 2:40 am, RolandRB <rolandbe...(a)hotmail.com> wrote: > > Is R overtaking SAS at universities and colleges? I hear that it is > > but only have to go on hearsay. I also heard that SPlus might be > > discontinued. Should we be moving across to R in the clinical > > reporting world? It would be good to read of accounts where this has > > been done. > > > At UCLA here's the deal (as I see it from a PH student standpoint) > > Stat department uses primarily R. Im unaware of what they do for data > management. > Biostat students are taught primarily SAS and STATA. R (graphics) is > available as an elective. So any R you want to learn is on your own. > > Since quite a few biostat students end up going into pharma or medical > consulting, I can't see SAS being overthrown that easily. SAS also > seems to be used primarily by academic researchers (at least in public > health), some use SPSS but i'm not sure where they teach it on > campus. > > Personally, i do all data management and some analysis/modeling in > SAS, but for better graphics/more advanced models and/or bayesian > analysis I use R. > > As for reporting, I've used latex a little, and will use it a lot > later (dissertation), but never really was fond of direct ODS > reporting. I tend to restructure/rearrange/organize only the output i > want to display in SAS and then ODS it and put the finishing touches > on it personally using Excel. |