[R] R reports

[Frank Harrell]

On Sat, 21 Aug 2010, Donald Winston wrote:

> The point is SAS has had simple reporting for 30 years. R apparently doesn't have any. Why is it so hard to accept that a report function analogous to a plot function would be a good thing?

R has had more advanced reporting features that SAS since not long 
after R's first production release in the 1990s.  Users who are 
unwilling to study R documentation and examples will not see that of 
course.

Let's see if you want to put your money where your mouth is.  How much 
are you willing to pay every year for SAS for its rigid approach to 
reporting?  What is the current licensing fee to your organization?

You are wrong on yet another piece of this discussion.  SAS legally 
removed the phrase "Statistical Analysis System" from the company more 
than 15 years ago.  SAS doesn't stand for anything.

Frank

  >
> On Aug 21, 2010, at 8:38 AM, Frank Harrell wrote:
>
>>
>> Your notes are not well thought out.
>>
>> You'll find that r-help is a friendly place for new users that do not come in with an attitude.
>>
>> I once used SAS (for 23 years) and know it very well.  I wrote the first SAS procedures for a graphics device, percentiles, logistic regression, and Cox regression.  Its reporting for the first 30 years of SAS' existence was quite primitive, and since then it is not what I'd call advanced and certainly not esthetically pleasing.  Considering that SAS has had tens of billions of $ at its disposal for research and development the result is quite unimpressive.  Look at the fake PROC EXPORT if you want to get a real laugh - it can't even put out valid csv files if there are any unmatched quotes inside of character variable values.  It is not even a procedure, just a front end to a trivial macro.
>>
>> The "report" function you outlined is in many ways more primitive than many functions already available in R.   See the summary.formula function for example in the Hmisc package.  Among other things, some of the functions give you flexibility in specifying the criteria by which a variable is considered continuous vs. discrete numeric.  They also allow you to override statistical tests to include in the table with your own functions.  Now one of the functions even automatically creates micrographics inside of table cells.
>>
>> You are welcome to write any R functions you'd like.  The language for doing so is richer than the SAS language by a significant margin.
>>
>> Frank E Harrell Jr   Professor and Chairman        School of Medicine
>>                     Department of Biostatistics   Vanderbilt University
>>
>> On Sat, 21 Aug 2010, Donald Paul Winston wrote:
>>
>>>
>>> People have been generating reports with a computer for many years. R is
>>> supposed to be an analytical engine. Report writing is fundamental to any
>>> kind of analysis tool. SAS has had several report procedures/functions since
>>> the very beginning(1960's?). SAS stands for Statistical Analysis System. Do
>>> you really expect users to have to piece together a half dozen or so bits of
>>> R code to create a report?
>>>
>>> It's not like it's difficult to do! I see this new company called Revolution
>>> Analytics who thinks R is the next big thing. Good grief. Maybe they can
>>> rescue it from the ghettoized academic world.
>>> --
>>> View this message in context: http://r.789695.n4.nabble.com/R-reports-tp2330733p2333267.html
>>> Sent from the R help mailing list archive at Nabble.com.
>>>
>>> ______________________________________________
>>> R-help at r-project.org mailing list
>>> https://stat.ethz.ch/mailman/listinfo/r-help
>>> PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
>>> and provide commented, minimal, self-contained, reproducible code.
>>>
>