Hello everyone,
I have three short questions (below I explain myself in a little bit more
detail):
1) Am I correct that it is not possible to use a covariance matrix, but only
a correlation matrix as input in SPSS?
2) After reading in a correlation matrix it is impossible to use the GUI to
perform any subsequent analyses?
3) After reading in a correlation matrix, how can I run a repeated measures
ANOVA?
ad 1) I tried to read in a covariance matrix as shown below. I did not
receive an error message but was not able to perform any subsequent analyes
("The ROWTYPE variable has a value that is invalid for matrix input").
MATRIX DATA VARIABLES=ROWTYPE_ x1 x2 x3 x4 x5 x6.
BEGIN DATA
MEAN 0 0 0 0 0 0
N 200 200 200 200 200 200
COV 8.32
COV -1.70 8.55
COV -1.71 -1.59 8.04
COV -1.54 -1.77 -1.68 8.21
COV -1.63 -1.74 -1.54 -1.56 8.12
COV -1.73 -1.74 -1.52 -1.66 -1.65 8.30
END DATA.
ad 2) Reading in a correlation matrix as shown below, I could run a
regression analysis (e.g., regression matrix=in(*) /dep x1/enter x2 x3.)
but was not able to run any other analysis using the "matrix=in(*)" trick,
e.g. in GLM
MATRIX DATA VARIABLES=ROWTYPE_ x1 x2 x3 x4 x5 x6.
BEGIN DATA
MEAN 0 0 0 0 0 0
STDDEV 1 1 1 1 1 1
N 200 200 200 200 200 200
COR 1
COR -0.70 1
COR -0.71 -0.59 1
COR -0.54 -0.77 -0.68 1
COR -0.63 -0.74 -0.54 -0.56 1
COR -0.73 -0.74 -0.52 -0.66 -0.65 1
END DATA.
THANK YOU SO MUCH FOR ANY HELP!
Sun
=====================
To manage your subscription to SPSSX-L, send a message to
[hidden email] (not to SPSSX-L), with no body text except the
command. To leave the list, send the command
SIGNOFF SPSSX-L
For a list of commands to manage subscriptions, send the command
INFO REFCARD
Locked
