Multivariate analysis on a Power Mac
Jon Naude
Could anyone tell me why the "OK" button stays grey when trying to do a
multivariate analysis running SPSS 6.1.1 on my Power Mac 8100/110 with
32 Megs of RAM? It refuses to process more than two or three factors
simultaneously with univariate analysis, the message is that there is
not enough memory available. What sort of memory requirements does
multivariate analysis require?
san
If this message comes with additional messages such as the advise to
enlarge your working memory, than do so with the help of submenu
"preferences". Also, what happens if you run the analysis in the editor
? What kind of messages do you get ?
David Nichols
The OK button is controlled at the GUI front end level. It's not going
to know anything about memory. If it's greyed out, it means you haven't
specified what is needed to produce proper syntax for a multivariate
model. The minimum that is required is that there be at least two
dependent variables specified. If any factors are specified, the ranges
must be defined.
If you're running into memory problems with a small number of factors
with 32MB, I'd suspect that one or more of these factors has a huge
number of levels and may really need to be treated as continuous, or
perhaps is a nested factor that should be renumbered to run from 1 to
N(i) where N(i) is the number of levels of that factor nested under
the ith level of the other factor(s). MANOVA isn't exactly optimized
for memory usage, but it will definitely handle more than two or three
factors.
--
-----------------------------------------------------------------------------
David Nichols Senior Support Statistician SPSS, Inc.
Phone: (312) 329-3684 Internet: nic...@spss.com Fax: (312) 329-3668
-----------------------------------------------------------------------------
Jon Naude
David Nichols <nic...@spss.com> wrote:
> In article <322DF0...@med.ruu.nl>, san <s.l.sc...@med.ruu.nl> wrote:
> >Jon Naude wrote:
> >>
> >> Could anyone tell me why the "OK" button stays grey when trying to do a
> >> multivariate analysis running SPSS 6.1.1 on my Power Mac 8100/110 with
> >> 32 Megs of RAM? It refuses to process more than two or three factors
> >> simultaneously with univariate analysis, the message is that there is
> >> not enough memory available. What sort of memory requirements does
> >> multivariate analysis require?
> >
> >If this message comes with additional messages such as the advise to
> >enlarge your working memory, than do so with the help of submenu
> >"preferences". Also, what happens if you run the analysis in the editor
> >? What kind of messages do you get ?
>
> The OK button is controlled at the GUI front end level. It's not going
> to know anything about memory. If it's greyed out, it means you haven't
> specified what is needed to produce proper syntax for a multivariate
> model. The minimum that is required is that there be at least two
> dependent variables specified. If any factors are specified, the ranges
> must be defined.
I did this. The OK button remains grey. Thats why I was wondering about
its memory needs.. BTW, my syntax window (in the menue) is also greyed
out. Could this be due to incomplete installation?
>
> If you're running into memory problems with a small number of factors
> with 32MB, I'd suspect that one or more of these factors has a huge
> number of levels and may really need to be treated as continuous, or
> perhaps is a nested factor that should be renumbered to run from 1 to
> N(i) where N(i) is the number of levels of that factor nested under
> the ith level of the other factor(s). MANOVA isn't exactly optimized
> for memory usage, but it will definitely handle more than two or three
> factors.
Most of the factors had 3 to 4 levels. Only one (age) had numerous
levels. Some nesting of factors...
David Nichols
In article <1996090621...@plato-slip-150.univie.ac.at>,
If age is measured in say years, and you have people over a large range,
it should be treated as a covariate rather than as a factor.
On the front end question, I don't know what's going on. As I said before,
I'd be at a loss to explain physically how memory considerations could
affect that.