[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Stata vs SPSS
2-points (just an FYI):
1. PL/I was created by IBM to combine the best of the numerical
processing of Fortran, the record (actually, early database
operations) of Cobol and the string processing capabilities of Algol
and create a single language, which they did somewhat successfully. I
won't get into what is meant by 'somewhat', but anyone who has
actually used it for any commercial system would swear by it.
Academics hated it.
2. C on the other hand, was designed to do one thing: write UNIX.
It's designed to write an operating system. It's the classical
"attempt to use a screwdriver to paint a wall" language. It became
popular because it was the "language-of-the-moment", given that a
large percentage of computing since 1970 is a popularity contest, and
not based on whether it's a better tool to do a job or not.
I'm not the most statistical savvy, and I have used Stata, SAS and
SPSS and if I need a tool to do the job, then Stata is the best I've
used. Maybe it has shortcomings like any software package, but it
does what I want it to do. So what if it can't do everything? Nothing
else I own does everything either. Either does my screwdriver.
Dept. of Biomedical Sciences
U. Liverpool, John Moores University
On Oct 15, 2006, at 3:22 PM, Clive Nicholas wrote:
Nick Cox replied:
Difficult to dissent from any of this (especially as I've never
Well, yes and no in my view. One general reason behind
many of the more positive comments in this thread is
that StataCorp (and its predecessors) have maintained
a very steady market focus on researchers and their
needs and desires.
Conversely, many a statistical program has lost its
way to a greater or lesser extent -- as far as researchers
are concerned -- insofar as it has gone for a wider
business market. I really am not clear that you
can go for both without major loss.
Similarly, various people have commented that for many
purposes they need to switch to other software (e.g.
MS Excel) for other things. This really doesn't sound
a source of surprise! I suppose it is flattering that
some Stata users seem to want to do everything in a
project in Stata.
I am often reminded in these discussions of Dennis
Ritchie's semi-apocryphal comments when he was repeatedly
asked why C -- famously lean and mean, like Stata in its
youth -- didn't include this feature or that feature:
"If you want PL/1, you know where to find it." PL/1
was, by comparison, a relative behemoth, but where
is it now?
PL/1 before!). Let me come back and add some more comments.
On balance, I think you're right that StataCorp shouldn't be in the
business of prostituting itself at the altar of the private sector. If
commerce and industry can't see why Stata would be of benefit to them,
then frankly that's their lookout (but it may have consequences for
statistically-minded individuals who are looking for jobs in the
sector but may have to settle working with packages they know to be
inferior for one reason or another). I suspect Stata's lack of 'market
penetration', for the want of a better experession, isn't just down to
StataCorp's retience about getting involved in a fight it senses it
ultimately 'win', but also because
(a) businesses aren't fully aware of the alternatives to SPSS (and
and that some of them might be better; and
(b) businesses can't be bothered to look for any other
more to complacency than anything else.
On this, I know whereof I speak. I remember having a casual webroom
conversation with a lady who told me that she worked as a market
in the City of London (and probably still does). It went something
Me: "Oh really? So you try and predict market movements, do you?"
Me: "Wow! So tell me; do you use time-series methods for prediction?"
Her: "Er, nooooo."
Me: "Um, OK. So what do you use, then?"
Her: "Nothing other than our professional intuition. We stick a
our mouth, take it out, hold it up in the air, and judge
which way the
Me: "Right. Don't you think you should be using something more, um,
Her: "No. Why should we when this works just as well? We've had
young, 'wet-behind-the-ears' kids come to us trying to teach
this crap before, and we soon sent them packing."
The discourse went downhill from there. With attitudes like this, no
package that aims to be more comprehensive than MS Excel can hope
a slice of the commerical cake, as it were, either here in the UK or
Of course, this isn't the only problem that Stata faces, because it
faces what one might call 'barriers to entry' those parts of the
sector that are already accustomed to using statistical packages as
of their work, notably market research. In this sector, SPSS have been
rolling out a specially-tailored package called SPSS-MR for a
years now, and these companies - who normally do little more than
crosstabulations, pie charts and maybe a bit of cluster analysis -
snapped them up by the shedload. As Richard Williams rightly says,
you're going to use a stat-package for is counting tallies, run bar
and maybe the odd bit of OLS, then why should StataCorp engage in a
for business that it may lose (in wasted dollars?) in the long run?
Some may see this view as a defeatist, but one has to look at it from
their point of view.
CLIVE NICHOLAS |t: 0(044)7903 397793
Politics |e: email@example.com
Newcastle University |http://www.ncl.ac.uk/geps
Whereever you go and whatever you do, just remember this. No matter
many like you, admire you, love you or adore you, the number of people
turning up to your funeral will be largely determined by local weather
* For searches and help try:
* For searches and help try: