R help - Feb 2010 - How to do: Correlation with "blocks" (or - "repeated measures" ?!) ?

Hello dear R help group,

I have the following setup to analyse:
We have about 150 subjects, and for each subject we performed a pair of
tests (under different conditions) 18 times.
The 18 different conditions of the test are complementary, in such a way so
that if we where to average over the tests (for each subject), we would get
no correlation between the tests (between subjects).
What we wish to know is the correlation (and P value) between the tests, in
within subjects, but over all the subjects.

The way I did this by now was to perform the correlation for each subject,
and then look at the distribution of the correlations received so to see if
it's mean is different then 0.
But I suspect there might be a better way for answering the same question
(someone said to me something about "geographical correlation", but a
shallow search didn't help).

p.s: I understand there might be a place here to do some sort of mixed
model, but I would prefer to present a "correlation", and am not sure
how to
extract such an output from a mixed model.

Also, here is a short dummy code to give an idea of what I am talking about:

attach(longley)
N <- length(Unemployed)
block <- c(
rep( "a", N),
rep( "b", N),
 rep( "c", N)
)
 Unemployed.3 <- c(Unemployed + rnorm(1),
Unemployed + rnorm(1),
Unemployed + rnorm(1))

GNP.deflator.3 <- c(GNP.deflator + rnorm(1),
GNP.deflator + rnorm(1),
GNP.deflator + rnorm(1))

cor(Unemployed, GNP.deflator)
cor(Unemployed.3, GNP.deflator.3)
cor(Unemployed.3[block == "a"], GNP.deflator.3[block ==
"a"])
cor(Unemployed.3[block == "b"], GNP.deflator.3[block ==
"b"])
cor(Unemployed.3[block == "c"], GNP.deflator.3[block ==
"c"])

(I would like to somehow combine the last three correlations...)



Any ideas will be welcomed.

Best,
Tal






----------------Contact
Details:-------------------------------------------------------
Contact me: Tal.Galili@gmail.com |  972-52-7275845
Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) |
www.r-statistics.com (English)
----------------------------------------------------------------------------------------------

	[[alternative HTML version deleted]]

I didn't follow your question completely. But do a search for
intraclass correlation with nlme or lmer and see if those results
relate to the question you are asking. If so, I would
suggest following up on the mixed model list. I know you
wanted to avoid mixed models, but if I have understood
your question, that is the way to use all of your data
to estimate the parameters you seek.

On Thu, Feb 25, 2010 at 10:57 AM, Tal Galili <tal.galili at gmail.com>
wrote:
> Hello dear R help group,
>
> I have the following setup to analyse:
> We have about 150 subjects, and for each subject we performed a pair of
> tests (under different conditions) 18 times.
> The 18 different conditions of the test are complementary, in such a way so
> that if we where to average over the tests (for each subject), we would get
> no correlation between the tests (between subjects).
> What we wish to know is the correlation (and P value) between the tests, in
> within subjects, but over all the subjects.
>
> The way I did this by now was to perform the correlation for each subject,
> and then look at the distribution of the correlations received so to see if
> it's mean is different then 0.
> But I suspect there might be a better way for answering the same question
> (someone said to me something about "geographical correlation",
but a
> shallow search didn't help).
>
> p.s: I understand there might be a place here to do some sort of mixed
> model, but I would prefer to present a "correlation", and am not
sure how to
> extract such an output from a mixed model.
>
> Also, here is a short dummy code to give an idea of what I am talking
about:
>
> attach(longley)
> N <- length(Unemployed)
> block <- c(
> rep( "a", N),
> rep( "b", N),
> ?rep( "c", N)
> )
> ?Unemployed.3 <- c(Unemployed + rnorm(1),
> Unemployed + rnorm(1),
> Unemployed + rnorm(1))
>
> GNP.deflator.3 <- c(GNP.deflator + rnorm(1),
> GNP.deflator + rnorm(1),
> GNP.deflator + rnorm(1))
>
> cor(Unemployed, GNP.deflator)
> cor(Unemployed.3, GNP.deflator.3)
> cor(Unemployed.3[block == "a"], GNP.deflator.3[block ==
"a"])
> cor(Unemployed.3[block == "b"], GNP.deflator.3[block ==
"b"])
> cor(Unemployed.3[block == "c"], GNP.deflator.3[block ==
"c"])
>
> (I would like to somehow combine the last three correlations...)
>
>
>
> Any ideas will be welcomed.
>
> Best,
> Tal
>
>
>
>
>
>
> ----------------Contact
> Details:-------------------------------------------------------
> Contact me: Tal.Galili at gmail.com | ?972-52-7275845
> Read me: www.talgalili.com (Hebrew) | www.biostatistics.co.il (Hebrew) |
> www.r-statistics.com (English)
>
----------------------------------------------------------------------------------------------
>
> ? ? ? ?[[alternative HTML version deleted]]
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> PLEASE do read the posting guide
http://www.R-project.org/posting-guide.html
> and provide commented, minimal, self-contained, reproducible code.
>