Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: RE: What is Popularity? (with respect to R, SAS, SPSS, Stata...)

From   "Nick Cox" <>
To   <>
Subject   st: RE: What is Popularity? (with respect to R, SAS, SPSS, Stata...)
Date   Tue, 29 Jun 2010 12:08:19 +0100

I agree. In fact, I'd say that there were several latent variables here.

If popularity is the criterion, then Excel probably trumps all other
programs, even focusing on statistical uses alone. 


Michael N. Mitchell

   I think this is a very interesting, unique, and challenging endeavor.
I think this is a 
classic issue of having a latent construct that we want to measure
(popularity) and some 
easy measures lying around. So far, I have not seen a discussion of how
we would define 
this latent variable of "popularity" of a statistical package. In terms
of defining 
"popularity" some random thoughts are...

   1. Number of "users" (but of course there are different kinds of
   2. Number of copies "in use" (but of course there are different kinds
of use, 
occasional vs. vs. daily).
   3. Amount of use (people hours, publications completed).
   4. Preferred utilization (I must use X, but I would prefer to use Y).
   5. Quality ratings.

   I think that it would be useful to solicit discussion of how people
would define 
"popularity" with respect to statistical packages. Then, each of the
available measures 
could be assessed to the degree it reflects the underlying unmeasured
idea of "popularity".

On 2010-06-28 6.52 AM, Muenchen, Robert A (Bob) wrote:

> At I have added plots, data, and/or
> discussion of:
> 1. Scholarly impact of each package across the years
> 2. The number of subscribers to some of the listservs
> 3. How popular each package is among Google searches across the years
> 4. Survey results from a Rexer Analytics poll
> 5. Survey results from a KDnuggests poll
> 6. A rudimentary analysis of the software skills that employers are
> seeking
> Thanks very much to all the folks who helped on this project
> John Fox, Marc Schwartz, Duncan Murdoch, Martin Weiss, John
> HU, Andre Wielki, Kjetil Halvorsen, Dario Solari, Joris Meys, Keo
> Ormsby, Karl Rexer, and Gregory Piatetsky-Shapiro.
> If anyone can think of other angles, please let me know.

*   For searches and help try:

© Copyright 1996–2017 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index