Discussion:
Any chance of matrix input being implemented?
(too old to reply)
Jaap Verhage
2017-04-10 08:29:23 UTC
Permalink
Hello reader,
I hope this is the right place to pose my question; I couldn't find any other. I'm doing a literature research on reading problems in children, and I like to do my own re-analysis of the data that the researchers report. This is possible if the authors of an article report N of cases, means, standard deviations and a complete correlation matrix of the variables measured. SPSS then allows me to use this as input with procedures like FACTOR and REGRESSION. But that means that matrix input must be accepted by these procedures. As I'm not sure how long I will be able to use SPSS in the future, I'm looking for alternatives. Naturally, my first choice is PSPP, but unfortunately, PSPP doesn't support matrix input of the kind I'm talking about. Is there a chance of it being implemented? I am quite ready to do something in return. I'm not a programmer, but I could do translation work from English into Dutch.

Regards, Jaap.
Dr. Oliver Walter
2017-04-10 12:26:42 UTC
Permalink
Dear Mr Verhage,

you are right. The matrix input command has not yet been implemented in
PSPP, as far as I know. It would be a nice feature to have. In the
meantime you could work with R. It seems to be possible to do a factor
analysis based on the covariance or correlation matrix in it.

Kind regards

Oliver Walter
Post by Jaap Verhage
Hello reader,
I hope this is the right place to pose my question; I couldn't find
any other. I'm doing a literature research on reading problems in
children, and I like to do my own re-analysis of the data that the
researchers report. This is possible if the authors of an article
report N of cases, means, standard deviations and a complete
correlation matrix of the variables measured. SPSS then allows me to
use this as input with procedures like FACTOR and REGRESSION. But that
means that matrix input must be accepted by these procedures. As I'm
not sure how long I will be able to use SPSS in the future, I'm
looking for alternatives. Naturally, my first choice is PSPP, but
unfortunately, PSPP doesn't support matrix input of the kind I'm
talking about. Is there a chance of it being implemented? I am quite
ready to do something in return. I'm not a programmer, but I could do
translation work from English into Dutch.
Regards, Jaap.
_______________________________________________
Pspp-users mailing list
https://lists.gnu.org/mailman/listinfo/pspp-users
John Darrington
2017-05-05 11:09:52 UTC
Permalink
On Mon, Apr 10, 2017 at 10:29:23AM +0200, Jaap Verhage wrote:
I hope this is the right place to pose my question; I couldn't find any other. I'm doing a literature research on reading problems in children, and I like to do my own re-analysis of the data that the researchers report. This is possible if the authors of an article report N of cases, means, standard deviations and a complete correlation matrix of the variables measured. SPSS then allows me to use this as input with procedures like FACTOR and REGRESSION. But that means that matrix input must be accepted by these procedures. As I'm not sure how long I will be able to use SPSS in the future, I'm looking for alternatives. Naturally, my first choice is PSPP, but unfortunately, PSPP doesn't support matrix input of the kind I'm talking about. Is there a chance of it being implemented? I am quite ready to do something in return. I'm not a programmer, but I could do translation work from English into Dutch.



I've pushed a few changes to git implementing the MATRIX DATA command, and a
MATRIX IN subcommand to FACTOR. REGRESSION hasn't been done yet - but you might
like to give some feedback in the meantime.

PSPP's standard lack of warranty applies.

J'
--
Avoid eavesdropping. Send strong encrypted email.
PGP Public key ID: 1024D/2DE827B3
fingerprint = 8797 A26D 0854 2EAB 0285 A290 8A67 719C 2DE8 27B3
See http://sks-keyservers.net or any PGP keyserver for public key.
Continue reading on narkive:
Loading...