R

[Thomas MacFarland]

Dale and Everyone:

Here are two Elluminate-based videos that may be helpful for those who wish to try the R Commander GUI as a first step to using R:

To enable guests to view this recording Winter 2013 Biostats Chat 10b - How to Install the GUI (Graphical User Interface) R Commander, use the following link:
https://mako.nova.edu/webapps/bb-collaborate-bb_bb60/guest.recording.launch.event?uid=5739113e-1c1d-49b3-a3ee-90085c38a682

To enable guests to view this recording Winter 2013 Biostats Chat 11 - How to Import Data Into R, Using the GUI (Graphical User Interface) R Commander, use the following link:
https://mako.nova.edu/webapps/bb-collaborate-bb_bb60/guest.recording.launch.event?uid=f99a2044-d650-407d-aa27-6792dbd2021b

Of course, full functionality in R comes with using syntax and not the GUI but training wheels are always a good beginning.

Please contact me ([hidden email]) off list if you have questions.  I do not want to stray too far from the purpose of this list.

Best wishes.

Tom
----------
Thomas W. MacFarland, Ed.D.
Senior Research Associate; Institutional Effectiveness and Associate Professor
Nova Southeastern University
Voice 954-262-5395 [hidden email]


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Pietrzak, Dale
Sent: Friday, May 10, 2013 7:46 AM
To: [hidden email]
Subject: Re: R

Albert-Jan,

Even your description of how simple it is, does not appear all that simple.

Dale


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Salbod, Mr. Stephen
Sent: Thursday, May 09, 2013 2:07 PM
To: [hidden email]
Subject: Re: R

Hi Albert-Jan,

I'm just an R newbie. I suffer stat-fright until I get comfortable with a package.

Cheers, Steve


-----Original Message-----
From: Albert-Jan Roskam [mailto:[hidden email]]
Sent: Thursday, May 09, 2013 5:21 AM
To: Salbod, Mr. Stephen; [hidden email]
Subject: Re: [SPSSX-L] R

Hi Stephen,

Why do you find Rcmdr tricky to install?
options(repos=structure(c(CRAN="http://cran.xl-mirror.nl" )))  # if you don want to select a mirror install.packages(c("Rcmdr"), dependencies=TRUE) If you download all the uninstalled packages (.zip) you need to a network location, you can use the "repo" argument to download it from there.
If I understood it correctly, you can set the environment R_LIBS_USER to a netwrok location that does not get wiped.
Then, when you got a new image, you just need to set that variable (Win7: setx R_LIBS_USER=\\myserver\myshare\myblah,
next time R starts up, it will look there for the installed packages.


At first instance, R cmdr seems promising, but in the end I have rarely used it. IIRC, writing additional functionality (plugins) also requires TCL knowledge.


Regards,
Albert-Jan


~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
All right, but apart from the sanitation, the medicine, education, wine, public order, irrigation, roads, a fresh water system, and public health, what have the Romans ever done for us?
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~


----- Original Message -----

> From: "Salbod, Mr. Stephen" <[hidden email]>
> To: [hidden email]
> Cc:
> Sent: Thursday, May 9, 2013 1:32 AM
> Subject: Re: [SPSSX-L] R
>
> Hi Bob, you might want to check out Rcmdr. Google to get installation notes.
> It's got an interface like what you would expect from a statistical
> package.  But it is tricky to install and use.  --Steve
>
>
> PS At this moment that is exactly what I'm trying to do. Every time,
> at this University, my computer gets reimaged I have to reinstall all
> my additional programs.
>
> -----Original Message-----
> From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf
> Of Bob Green
> Sent: Wednesday, May 08, 2013 5:02 PM
> To: [hidden email]
> Subject: Re: R
>
> I changed to R simply because I couldn't afford SPSS. Without any
> background in programming the learning curve was extremely steep. I
> still struggle with things. If you want to run a frequency count of
> all your variables and obtain the respective percentages SPSS is way
> easier to use.  Having said that R forces you to look more carefully at what variables to include in analyses.
>
> They are totally different beasts in many ways.
>
> The user forums are also very different.  I think this is a major
> issue, asking an extremely naive question (and as a naive user), this
> can met with stony silence or rebuke, though some people on the R list
> have been enormously helpful.  First impressions can make a big difference.
>
>
> Bob
>
>
>
>
>
>
> nAt 02:50 AM 9/05/2013, John F Hall wrote:
>> I too downloaded R.  Forget it: too much learning to be able to write
>> sensible tutorials or for students to learn enough in 13 hours to be
>> able to do anything useful.
>>
>> I taught survey processing and analysis using SPSS for 23 years from
>> 1972 until I (early) retired in 1992, and I still use it in my
>> website tutorials.  My students got jobs: would they with R?  Tell
>> the business managers that (although in the UK the Deans in the
>> Polytechnics which became Universities in 1992 were also part of corporate management).
>> One of my ex-researchers was once promoted over my head by Stalinist
>> senior management to be Director of Research (aka Business Manager,
>> Research): when I repeated to her the late Angus Campbell's advice to
>> me, "Never employ anyone whose job you couldn't take over if you
> had
>> to." she stormed off in a huff.
>>
>> I've said it before and I'll say it again: the SPSS rot started when
>> they went after the business market and let marketing people act as
>> their agents.  I appreciate that the success of SPSSI was threatening
>> the charitable status of the University of Chicago, but once IBM took
>> over, academic users were doomed.  IBM are only ever, and always have
>> been, interested in
>> profits: didn't anyone tell you?  That's why so many academic
>> institutions are turning to Stata, but will their students still get
>> jobs?  Tell the Deans/Business Managers that as well.
>>
>> John F Hall (Mr)
>> [Retired academic survey researcher]
>>
>> Email:   <mailto:[hidden email]>[hidden email]
>> Website:
>> <http://www.surveyresearch.weebly.com>www.surveyresearch.weebly.com
>> Start
>> page:
>> <http://www.surveyresearch.weebly.com/spss-without-tears.html>www.sur
>> ve yresearch.weebly.com/spss-without-tears.html
>>
>>
>>
>>
>>
>>
>>
>> From: SPSSX(r) Discussion
>> [mailto:[hidden email]] On Behalf Of Dr. Thomas W.
>> MacFarland
>> Sent: 08 May 2013 15:04
>> To: [hidden email]
>> Subject: Re: R
>>
>> Stan and everyone else:
>>
>> SPSS is great software and I've been using it since the mid-1980s
>> when it was housed at the University on a VAX machine, using vi as
>> the editor (No - I don't think we need too many more posts related to
>> the traffic thread.).  Separate from the statistical analyses
>> supported by SPSS, the ctables module alone is a pearl without price.
>>
>> Saying that, I've also moved new projects over to R whenever possible.
>>
>> Why?
>>
>> Diversification is good, if nothing else for self-protection.
>>
>> I am concerned that SPSS will soon price itself out of reach for many
>> of us in higher education.
>>
>> Do you remember when Deans were the major decision-makers in higher
>> education?  Increasingly (at least here is the U.S.), higher
>> education has become more corporate (even at the state schools and
>> not-for-profits) and we now have business managers who have little to
>> no classroom experience influencing decision-making, and some are now
>> starting to question the high cost of SPSS license fees.
>>
>> So, both for administrative jobs as well as classroom teaching, I'm
>> moving more activities over to R.
>>
>> Is R good software since those who develop the base package and
>> external packages do not receive a paycheck for their efforts?  We
>> all have separate ideas, but I certainly respect those who work with
>> missionary zeal as opposed to those who are only concerned about last
>> quarter's profit statement, ship dates (think dBase IV), and today's
>> stock price.
>>
>> Ultimately, however, the market will determine the winners and
>> losers, whether the product is expensive or free.
>>
>> If you need a brief introduction to R please take a look at the
>> videos listed below, even if they are slightly dated:
>>
>> Are You Ready for R - Open Source Software for Institutional Research
>> <https://tegr.it/y/ty2b>https://tegr.it/y/ty2b
>> Are You Ready for R - Background Information About R
>> <https://tegr.it/y/ty15>https://tegr.it/y/ty15
>> Are You Ready for R - Orientation of This Training Session
>> <https://tegr.it/y/ty0v>https://tegr.it/y/ty0v
>> Are You Ready for R - Download R
>> <https://tegr.it/y/ty0n>https://tegr.it/y/ty0n
>> Are You Ready for R - Exposure to the R Interface
>> <https://tegr.it/y/ty0b>https://tegr.it/y/ty0b
>> Are You Ready for R - Use R in Interactive Mode
>> <https://tegr.it/y/txzj>https://tegr.it/y/txzj
>> Are You Ready for R - Use R Syntax in a .R Script File
>> <https://tegr.it/y/txz1>https://tegr.it/y/txz1
>> Are You Ready for R - Import a .csv Spreadsheet-Type Data File into R
>> <https://tegr.it/y/txyv>https://tegr.it/y/txyv
>> Are You Ready for R - Organize the Data
>> <https://tegr.it/y/txyn>https://tegr.it/y/txyn
>> Are You Ready for R - Conduct a Visual Data Check
>> <https://tegr.it/y/txwf>https://tegr.it/y/txwf
>> Are You Ready for R - Supplemental R Packages
>> <https://tegr.it/y/txw6>https://tegr.it/y/txw6
>> Are You Ready for R - Summary Descriptive Statistics
>> <https://tegr.it/y/txvx>https://tegr.it/y/txvx
>> Are You Ready for R - Breakout Descriptive Statistics
>> <https://tegr.it/y/txvt>https://tegr.it/y/txvt
>> Are You Ready for R - Use R for Student's t-Test
>> <https://tegr.it/y/txvb>https://tegr.it/y/txvb
>> Are You Ready for R - Use R for Oneway Analysis of Variance (ANOVA)
>> <https://tegr.it/y/txux>https://tegr.it/y/txux
>> Are You Ready for R - Use R for Multivariate Analysis of Variance
>> (MANOVA) <https://tegr.it/y/txul>https://tegr.it/y/txul
>> Are You Ready for R - Use R for Correlation and Measures of
>> Association <https://tegr.it/y/txtj>https://tegr.it/y/txtj
>> Are You Ready for R - Prepare to Exit, Save, and Later Retrieve This
>> R Session <https://tegr.it/y/txrv>https://tegr.it/y/txrv
>> Are You Ready for R - Demonstrate Batch Processing with a Script File
>> <https://tegr.it/y/txrj>https://tegr.it/y/txrj
>>
>> Comment:  The R tables package
>> (<http://cran.r-project.org/web/packages/tables/vignettes/tables.pdf>
>> ht tp://cran.r-project.org/web/packages/tables/vignettes/tables.pdf),
>> specifically the tables::tabular() function, does not yet equal the
>> SPSS ctables module, but it is getting close.  When that happens SPSS
>> will have an end-product rival that may be hard to hold back.
>>
>> Full Disclosure (and not a shameless plug):  I have a R-based text
>> (<http://www.springer.com/statistics/social+sciences+%26+law/book/978
>> -1
>> -4614-2133-7>http://www.springer.com/statistics/social+sciences+%26+l
>> -4614-2133-7>aw
>> /book/978-1-4614-2133-7) that is still available and a new text (R
>> for
>> biostatistics) will be available before too long.
>>
>> Best wishes.
>>
>> Tom
>>
>>
>> ----------
>> Thomas W. MacFarland, Ed.D.
>> Senior Research Associate; Institutional Effectiveness and Associate
>> Professor Nova Southeastern University Voice 954-262-5395
>> <mailto:[hidden email]>[hidden email]
>>
>>
>> -----Original Message-----
>> From: SPSSX(r) Discussion
>> [<mailto:[hidden email]>mailto:[hidden email]]
>> On Behalf Of David Marso
>> Sent: Wednesday, May 08, 2013 3:57 AM
>> To: <mailto:[hidden email]>[hidden email]
>> Subject: Re: R
>>
>> AGAIN INTERSPERSED:
>>
>> Stan Gorodenski wrote
>>  > On 5/7/2013 7:07 PM, Bruce Weaver wrote:
>>  >> I bet you'd get some very interesting responses if you posted
> your
>>  >> question
>>  >> here:
>>  >>
>>  >>
> <http://r.789695.n4.nabble.com/>http://r.789695.n4.nabble.com/
>>  >>
>>  >
>>  > I am sure you are absolutely correct. :-)  >  > But, this raises
>> another issue I have with R. The link you provide  > is a discussion
>> group. I have come to dislike having to rely solely  > on discussion
>> groups to get answers to questions.
>>  >
>>  > DMM:  Well, there is always package documentation and various
>> online  > training resources.
>>  >
>>  > In astronomy there is a
>>  > new protocol called ASCOM. One has to rely completely on a  >
>> discussion group to get answers to questions. As a result one has to
>> > deal with personalities and their willingness to help.
>>  >
>>  > DMM:  I'll bet ASCOM has some sort of documentation.  I don't  >
>> personally know anything about it and am not interested enough in  >
>> astronomy to bother researching the issue.
>>  > Personalities are like A$$#@!%$ everyone has one.  I for one am  >
>> sometimes mistaken for an A$$#@!% due to my personality.
>>  > Willingness to help is a crapshoot.  If you have a question that
>> > maybe one person in the DG knows the answer, you might have a problem.
>>  > OTOH:  There is this mysterious element in the universe called  >
>> reciprocity.  Put in take out.
>>  > People who frequently take but don't reciprocate get to know the
>> > A$$#@!% side of my personality.
>>  >
>>  > I get so tired of having to rephrase something.
>>  >
>>  > DMM:  I get so tired of people who think they can rely on others
>> > InterNeTelepathy and ESPss skills to read their minds when they  >
>> don't provide sufficient information to understand the issue much
> or
>>  > less solve the problem.  Phrase it explicitly with examples in the
>> > first place and there is not so much of a need to rephrase.  Lately
>> > people in this group seem to think of it as a place to twitter
>> their  > plights rather than carefully explicate the nature of the
>> issue,  > what was tried, what failed, what came close, sample of
>> data in/data
> out.
>>  > Almost as though it is expected that the solvers here spend more
>> > time thinking about the issue than the OP (I am almost tempted to
>> > name some names).  For these peeps I post one liners or  subject
>> them to the Socratic method.
>>  >
>>  > BTW Stan:  You are not guilty of such.  Your posted questions here
>> > are typically well thought out and you take the effort to be clear
>> > and do the heavy lifting WRT RTFM and background research prior to
> posting.
>>  >
>>  > It is amazing how difficult communication can be at times. SPSS
>> and  > Stata, in contrast, have manuals that come from the companies.
>> I  > believe there are manuals for R, but they were not produced by a
>> > company and so may be limited in their scope.
>>  >
>>  > DMM:  You seem to relish the comfort of 'companies'.  I for
> one am
>>  > exploring open source options at every opportunity!  For example,
>> I  > would rather use GIMP or ImageJ to do photo editing or
>> generating my  > video art then shell out several hundred dollars for PhotoShop.
>>  > Why would I spend a ton on upgrading M$ office when I can acquire
>> > OpenOffice or NeoOffice for FREE.
>>  > Why would I dump $20+K on SPSS Modeller when I can DL Rapid Miner?
>>  > List goes on and on!
>>  > Yeah Stan, trust the 'companies'... They have your best
> interest and
>>  > comfort at heart!
>>  >
>>  > I may be talking out
>>  > of ignorance regarding R, but that is my impression. There are two
>> > extremes, ASCOM which totally relies on a discussion group, and SAS.
>>  > It is incredible the number of manuals SAS has. The last time I  >
>> checked about 12 years ago it would have taken about 2 or 3 feet of
>> > manuals, probably even more, if one wanted to be complete (that is,
>> > if one wants hard copies).
>>  >
>>  > DMM: AFAICK: you can likely find pretty in depth info by querying
>> > something as simple as R Manuals  >
>>
> <http://www.search.ask.com/web?q=R%20Manuals&o=15554&l=dis&qsrc=2871>h
>>
> ttp://www.search.ask.com/web?q=R%20Manuals&o=15554&l=dis&qsrc=2871
>>  >
>>  >
>>  > Stan
>>  >
>>  >
>>  >> ;-)
>>  >>
>>  >>
>>  >>
>>  >> Stan Gorodenski wrote
>>  >>
>>  >>> A number of individuals highly recommend R. I do not have much
>> >>> inclination to switch to R (I had downloaded it about four
> years
>>  >>> ago and didn't know where to go from there) for two
> reasons. As
>>  >>> someone said, there is a steep learning curve. The other
> reason is
>>  >>> that because R is freeware, I believe all the statistical  >>>
>> procedures were written by users. Is this correct? If so, I
> feel
>>  >>> more comfortable with a package like SPSS or Stata because
> there
>>  >>> is a profit motive, and because there is a profit motive these
>> >>> companies have a definite interest in testing the product
> before
>>  >>> release and correcting any errors after release. In other
> words,
>>  >>> one can be more cerain the statistical results from the
> various
>>  >>> procedures are correct. I do not think this can be said of R.
> Is this correct?
>>  >>> Stan
>>  >>>
>>  >>> =====================
>>  >>> To manage your subscription to SPSSX-L, send a message to  >>>
>> >>  >>> <mailto:LISTSERV@.UGA>LISTSERV@.UGA
>>  >>>
>>  >>
>>  >>>   (not to SPSSX-L), with no body text except the command. To
> leave
>>  >>> the list, send the command SIGNOFF SPSSX-L For a list of
> commands
>>  >>> to manage subscriptions, send the command INFO REFCARD  >>>  >>
>> >>  >>  >>  >> -----  >> --  >> Bruce Weaver  >>
>>
>>  > bweaver@
>>
>>  >>
>>
> <http://sites.google.com/a/lakeheadu.ca/bweaver/>http://sites.google.c
>>  om/a/lakeheadu.ca/bweaver/
>>  >>
>>  >> "When all else fails, RTFM."
>>  >>
>>  >> NOTE: My Hotmail account is not monitored regularly.
>>  >> To send me an e-mail, please use the address shown above.
>>  >>
>>  >> --
>>  >> View this message in context:
>>  >>
>>
>> <http://spssx-discussion.1045642.n5.nabble.com/R-tp5720058p5720060.ht
>> m
>> >http://spssx-discussion.1045642.n5.nabble.com/R-tp5720058p5720060.ht
>> m  >> l Sent from the SPSSX Discussion mailing list archive at
> Nabble.com.
>>  >>
>>  >> =====================
>>  >> To manage your subscription to SPSSX-L, send a message to  >>
>>
>>  > <mailto:LISTSERV@.UGA>LISTSERV@.UGA
>>
>>  >  (not to SPSSX-L), with no body text except the  >> command. To
>> leave the list, send the command SIGNOFF SPSSX-L For a  >> list of
>> commands to manage subscriptions, send the command INFO  >> REFCARD
>> >>  >>  >>  >>  >  > =====================  > To manage your
>> subscription to SPSSX-L, send a message to
>>
>>  > <mailto:LISTSERV@.UGA>LISTSERV@.UGA
>>
>>  >  (not to SPSSX-L), with no body text except the command. To leave
>> > the list, send the command SIGNOFF SPSSX-L For a list of commands
>> to  > manage subscriptions, send the command INFO REFCARD
>>
>>
>>
>>
>>
>> -----
>> Please reply to the list and not to my personal email.
>> Those desiring my consulting or training services please feel free to
>> email me.
>> ---
>> "Nolite dare sanctum canibus neque mittatis margaritas vestras ante
>> porcos ne forte conculcent eas pedibus suis."
>> Cum es damnatorum possederunt porcos iens ut salire off sanguinum
>> cliff in abyssum?"
>> --
>> View this message in context:
>> <http://spssx-discussion.1045642.n5.nabble.com/R-tp5720058p5720067.ht
>> ml
>>> http://spssx-discussion.1045642.n5.nabble.com/R-tp5720058p5720067.ht
>>> ml
>> Sent from the SPSSX Discussion mailing list archive at Nabble.com.
>>
>> =====================
>> To manage your subscription to SPSSX-L, send a message to
>> <mailto:[hidden email]>[hidden email]
>> (not to SPSSX-L), with no body text except the command. To leave the
>> list, send the command SIGNOFF SPSSX-L For a list of commands to
>> manage subscriptions, send the command INFO REFCARD
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except
> the command. To leave the list, send the command SIGNOFF SPSSX-L For a
> list of commands to manage subscriptions, send the command INFO
> REFCARD
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> [hidden email] (not to SPSSX-L), with no body text except
> the command. To leave the list, send the command SIGNOFF SPSSX-L For a
> list of commands to manage subscriptions, send the command INFO
> REFCARD
>

=====================
To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to [hidden email] (not to SPSSX-L), with no body text except the command. To leave the list, send the command SIGNOFF SPSSX-L For a list of commands to manage subscriptions, send the command INFO REFCARD

=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD