COW wanted

[Dr. O]

I typically examine hundreds of correlations using the form of:

CORRELATIONS VARIABLES X Y Z WITH A B C.

USING THE DRAFT VIEWER (I REFUSE TO USE THOSE PIVOT TABLES) I WOULD LIKE
TO USE A "CORRELATION OUTPUT WIZARD" IN ORDER TO PRODUCE A SIMPLE
CORRELATION MATRIX WITHOUT N or SIG IN THE CELLS -- THIS TYPE OF OUTPUT
WAS LAST PROVIDED IN SPSS 6.1

MANY THANKS
DR. O

[Peck, Jon]

There are several ways to do this.  The easiest one, however, does require you to use the regular pivot tables.  If you go that route, you can double click on a table to activate it, and pivot the statistics dimension into the layers.  Then your table will just show the correlations.

That process can be automated with an autoscript applied to the Correlations table automatically.

You could also create an autoscript that hid the rows you don't want to see, but pivoting may be a better strategy, because if you do want to check the other statistics sometimes, you can more easily view them by choosing that layer.

If you don't want to use the regular Viewer, you can get a Data Editor view of the table using syntax such as the following.

dataset declare fred.
oms /if subtypes='Correlations' /destination outfile=fred format=sav.
corr salary salbegin with educ jobtime.
omsend.
dataset activate fred.
select if (var2 eq 'Pearson Correlation').
exec.
delete variables command_ subtype_ label_ var2.

This assumes SPSS 14 or later in order to use multiple datasets, but the output could be sent to a sav file with earlier versions.  Using OMS, you could alternatively execute the pivot to hide the other statistics and write the output to an html or plain text file.

Of course, this syntax could be wrapped up as a macro

HTH,
Jon Peck

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Dr. O
Sent: Saturday, March 17, 2007 8:47 AM
To: [hidden email]
Subject: [SPSSX-L] COW wanted

I typically examine hundreds of correlations using the form of:

CORRELATIONS VARIABLES X Y Z WITH A B C.

USING THE DRAFT VIEWER (I REFUSE TO USE THOSE PIVOT TABLES) I WOULD LIKE
TO USE A "CORRELATION OUTPUT WIZARD" IN ORDER TO PRODUCE A SIMPLE
CORRELATION MATRIX WITHOUT N or SIG IN THE CELLS -- THIS TYPE OF OUTPUT
WAS LAST PROVIDED IN SPSS 6.1

MANY THANKS
DR. O

[Maguin, Eugene]

Seeing this message reminded me of something that irritates me over and over
when I come up against it. In the better text output verion (i.e., pre 7),
correlation matrices were printed with asterisks if, I think I remember this
correctly, the nosig option was specified. And those asterisks remained when
output was copied to, in those days, wordperfect. Now with the improved
output, nosig does not suppress the significance line in the output but does
add the asterisks. However, when copied to word, asterisks disappear.

Several questions come to mind on this topic. The most important is why do
the asterisks disappear (or, conversely, why can't they be made to stick?

But also, why has the operation of nosig been changed.

And, underlying all this is just why are these scripts a better thing than
procedure keyword options. I regard them as undocumented. I think I know
where they are (edit --> options --> output) and that they can be selected
by marking or unmarking them, as the case may be. But, beyond that--nothing.

Gene Maguin

[Maguin, Eugene]

I see I have some crow to eat. Could ya pass the salt, pepper and butter
please? My complaint about the asterisks not coming through when tables are
copied to word is incorrect. The asterisks come through in 14. I also
checked a version of 11.5 I have and, although I could not copy tables, the
asterisks came through on export. Although I could swear that they didn't
used to, maybe I misremember that also. So. Thanks to spss development for
fixing that.

Gene Maguin

[Peck, Jon]

Perhaps  you were thinking of Excel.  The asterisks are removed there, since they would ruin the numbers for computations.

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Gene Maguin
Sent: Monday, March 19, 2007 4:27 PM
To: [hidden email]
Subject: Re: [SPSSX-L] COW wanted

I see I have some crow to eat. Could ya pass the salt, pepper and butter
please? My complaint about the asterisks not coming through when tables are
copied to word is incorrect. The asterisks come through in 14. I also
checked a version of 11.5 I have and, although I could not copy tables, the
asterisks came through on export. Although I could swear that they didn't
used to, maybe I misremember that also. So. Thanks to spss development for
fixing that.

Gene Maguin

[Maguin, Eugene]

All,

I have some text data contains a numeric field and several string fields.
The string fields are enclosed in double quotes (e.g., "qwerty" and
separated by a ampersand as a delimiter. Using data list I can read it in;
however, the double quotes come with it. I don't think this is possible but
can spss be gotten to strip off the double quotes enclosing string fields? I
looked at both data list and get data and it doesn't seem that anything is
possible in either of those two commands. Let me add that while I can strip
off the double quotes after the data has been read in, I am focused on the
reading process.

Thanks, Gene Maguin

[Peck, Jon]

GET DATA works fine for this.
Suppose the file looks like this:
"abc"&"def" 123,"xyz"

so blank, comma, and & may all show up as separators.  Then the following syntax reads this in and strips the quotes.
GET DATA  /TYPE = TXT
 /FILE = 'C:\temp\abcdef.txt'
 /DELCASE = LINE
 /DELIMITERS = "& ,"
 /QUALIFIER = '"'
 /ARRANGEMENT = DELIMITED
 /FIRSTCASE = 1
 /IMPORTCASE = ALL
 /VARIABLES = V1 A3 V2 A3 V3 F2.0 V4 A3.

HTH,
Jon Peck
SPSS

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Gene Maguin
Sent: Thursday, March 22, 2007 8:36 AM
To: [hidden email]
Subject: [SPSSX-L] Data reading question

All,

I have some text data contains a numeric field and several string fields.
The string fields are enclosed in double quotes (e.g., "qwerty" and
separated by a ampersand as a delimiter. Using data list I can read it in;
however, the double quotes come with it. I don't think this is possible but
can spss be gotten to strip off the double quotes enclosing string fields? I
looked at both data list and get data and it doesn't seem that anything is
possible in either of those two commands. Let me add that while I can strip
off the double quotes after the data has been read in, I am focused on the
reading process.

Thanks, Gene Maguin

[Maguin, Eugene]

Jon,

Thanks. While I did look at get data, I looked at my version 10 syntax book.
I see that since then greater flexibilty has been written into the code.
Very nice.

Gene Maguin