| |
[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
Re: st: Stata vs SPSS
On 16 DF�mh 2006, at 19:17, Richard Williams wrote:
At 11:45 AM 10/16/2006, Georgeta Vidican wrote:
hi there,
try statistics/nonparametric/binomial
i hope it works.
georgeta
I don't think that gives you the confidence interval though - at
least it didn't as of SPSS 12. My comparisons of how SPSS and
Stata handle this sort of problem can be found in
What I meant was a simple descriptive statistic
. ci loneliness, bin wil
------
Wilson ------
Variable | Obs Mean Std. Err. [95% Conf.
Interval]
-------------
+---------------------------------------------------------------
loneliness | 1337 .3522812 .0130639 .
3271329 .378276
Stata gives the proportion of people who are lonely and the
confidence interval. I could not get SPSS to give me the confidence
interval and, a couple of years ago, I wrote to tech support to ask
how you did it.
They told me that SPSS didn't calculate the confidence interval for a
proportion but showed me how it was easy to obtain this, giving me
the textbook formula for the standard error of a proportion.
I found this incredible, since any fule kno that using the standard
error formula will produce confidence intervals that are too wide and
may well exceed the range 0-1.
Stata, on the other hand, embraced Wilson and Jeffreys confidence
intervals almost instantly the methods were shown to be the best
available - they were slightly beaten to it by Nick Cox. And Stata
has had a confidence interval command for a proportion since I
started using it (Stata 3).
If it's for social scientists, why does SPSS not have confidence
intervals for percentages? These aren't exactly alpha geek stuff...
=========
Ron�n Conroy
Royal College of Surgeons in Ireland
[email protected]
+353 (0) 1 402 2431
+353 (0) 87 799 97 95
www.flickr.com/photos/ronanconroy
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/