R help - Nov 2005 - use of the 'by' command & converting SPSS ANOVA/GLM syntax into R syntax

I have two questions I would appreciate assistance with:

(1)  I believe "by" is the command used to split a file. In the
following
example, "mydat" is the dataframe , "group" the variable I
want to split
the analysis by & "WK1FREQ,WK2FREQ" the variables

attach(mydat)
by (GROUP)
cor.test (WK1FREQ,WK2FREQ)

I have also tried: (by (GROUP) cor.test (WK1FREQ,WK2FREQ))

(2)

I also wanted to run the following analyses, written in SPSS syntax. I have 
searched the R archives and gather I may need to use lm instead of glm, but 
got lost.

UNIANOVA
avfreq BY sex  WITH blik
/METHOD = SSTYPE(3)
/INTERCEPT = INCLUDE
/EMMEANS = TABLES(OVERALL)
/EMMEANS = TABLES(sex)
/PRINT = DESCRIPTIVE ETASQ OPOWER HOMOGENEITY
/CRITERIA = ALPHA(.05)
/DESIGN = blik sex blik*sex .

  GLM
    avmth3fq avfreq BY SEX
   /WSFACTOR = timex 2 Polynomial
   /METHOD = SSTYPE (3)
   /PRINT = DESCRIPTIVE ETASQ TEST(MMATRIX) HOMOGENEITY
   /CRITERIA = ALPHA (.05)
   /WSDESIGN = timex
   /DESIGN = sex .

Any assistance with either of these enquiries is most appreciated.

Bob

On 11/20/2005 3:44 PM, Bob Green wrote:
> I have two questions I would appreciate assistance with:
> 
> (1)  I believe "by" is the command used to split a file. In the
following
> example, "mydat" is the dataframe , "group" the
variable I want to split
> the analysis by & "WK1FREQ,WK2FREQ" the variables
> 
> attach(mydat)
> by (GROUP)
> cor.test (WK1FREQ,WK2FREQ)
> 
> I have also tried: (by (GROUP) cor.test (WK1FREQ,WK2FREQ))
I think you want

by(mydat, mydat$GROUP, function(subset) cor.test(subset$WK1FREQ, 
subset$WK2FREQ))

> 
> (2)
> 
> I also wanted to run the following analyses, written in SPSS syntax. I have
> searched the R archives and gather I may need to use lm instead of glm, but
> got lost.
> 
> UNIANOVA
> avfreq BY sex  WITH blik
> /METHOD = SSTYPE(3)
> /INTERCEPT = INCLUDE
> /EMMEANS = TABLES(OVERALL)
> /EMMEANS = TABLES(sex)
> /PRINT = DESCRIPTIVE ETASQ OPOWER HOMOGENEITY
> /CRITERIA = ALPHA(.05)
> /DESIGN = blik sex blik*sex .
> 
>   GLM
>     avmth3fq avfreq BY SEX
>    /WSFACTOR = timex 2 Polynomial
>    /METHOD = SSTYPE (3)
>    /PRINT = DESCRIPTIVE ETASQ TEST(MMATRIX) HOMOGENEITY
>    /CRITERIA = ALPHA (.05)
>    /WSDESIGN = timex
>    /DESIGN = sex .
Sorry, I don't know SPSS syntax, so I can't help you with this one.

Duncan Murdoch
> 
> Any assistance with either of these enquiries is most appreciated.
> 
> Bob
> 
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide!
http://www.R-project.org/posting-guide.html