On Thu, 27 Jan 2000 20:29:29 +1000, "boonlert"
<[EMAIL PROTECTED]> wrote:
> Dear all
> The result of paired sample t-test analysis from SPSS shows both a
> significance(sig.) of paired sample correlation and a significance of paired
> sample test but they are opposite interpretation (one is <.05, one >.05).
> Could any one tell me why they are different and how should I
> interpret this result?
Two different hypotheses, two different tests.
The correlation is extra information concerning the pairing:
How correlated *were* they? The means could be identical, where the
correlation is nearly 1 (powerful testing), or -1 (not a good choice
for paired-testing), or near 0.
If the correlation is not positive, you might ask yourself, Was there
really a reason for pairing? If the correlation is not what you can
explain, you might ask if the data are messed up in the computer
program, or in the data file. I check those correlations; just like
the actual means have to make sense for what the variable is, the
correlation is another clue that the computer has analyzed what I
intended for it to analyze.
Anyway, the paired t-test is the test performed on the difference in
means.
--
Rich Ulrich, [EMAIL PROTECTED]
http://www.pitt.edu/~wpilib/index.html
===========================================================================
This list is open to everyone. Occasionally, people lacking respect
for other members of the list send messages that are inappropriate
or unrelated to the list's discussion topics. Please just delete the
offensive email.
For information concerning the list, please see the following web page:
http://jse.stat.ncsu.edu/
===========================================================================
