Nabble - SPSSX Discussion

GLM for factors coded as [-1;0;1]

2009-11-26T03:35:51Z

Dear Listers,
I'd like to perform a ANOVA on this dataset: I have 4 factors coded as [-1;0;1]

runs    A’    B’    C’    D’    outcome
1         -1    -1    -1    -1    1075
2         -1    0    0    0    633
3         -1    1    1    1    406
4         0    -1    0    1    860
5         0    0    1    -1    561
6         0    1    -1    0    868
7         1    -1    1    0    669
8         1    0    -1    1    1138
9         1    1    0    -1    749
10      -1    -1    -1    -1    1052
11     -1    -1    -1    -1    1037

Is it correct to use the factors A;B,C,D as covariates in the Univariate GLM procedure, or I'm wrong?

Thanks in advance for your support.

Cristiano.

Automatic Reply: SPSSX-L Digest - 24 Nov 2009 to 25 Nov 2009 (#2009-329)

2009-11-25T21:08:23Z

The University of Georgia is closed for a mandatory furlough day on Nov. 25 and the Thanksgiving holiday Nov. 26-27.

Please leave a message, and I'll get back to you as soon as I can.

Re: Macro for reordering variables

2009-11-25T17:39:48Z

Writing the macro or Python code may be a useful exercise, but if you just want to alphabetize your variables and you have version 17? or 18, go to Data Editor Variable View, right click on the header of the Name column, and choose Sort from the pop-up menu.

Regards,
Jon Peck
-----------------
Sent from my BlackBerry Handheld.

From: Ruben van den Berg [ruben_van_den_berg@...]
Sent: 11/25/2009 09:25 AM GMT
To: SPSSX-L@...
Subject: [SPSSX-L] Macro for reordering variables

Dear all,

I'm trying to write a macro that should reorder the variables in my file. Right now, I've Label.1 to Label.15 and then Leave.1 to Leave.15. I'd like the order to be Label.1 Leave.1 Label.2 Leave.2 (...) Label.15 Leave.15. Admittedly, I could write this by hand since it's only 30 variables but next time it could be 100 or 200 variables and I'd really like to learn some very basic macro language as well.

Thanks a lot in advance!

define !reorder ()
!do !t2=1 !to 15
!let !varlist=!concat('label.',!t2) !concat('leave.',!t2)
!doend
matc fil fil *
/kee variable_name !varlist
exe.
!end.

set mpr on.
!reorder.

Express yourself instantly with MSN Messenger! MSN Messenger

Re: How to read v14 output files in later versions/at all?

2009-11-25T11:39:46Z

Caroline,

After having this problem myself and also working with people who do not
have access to SPSS (I'm the only one in the office that has SPSS) I
export or copy/paste my output to Excel. The export syntax is the better
method as you can preserve your output.spo file.

Hope this helps,

Shannon Deike-Sims
Research Analyst III
(907) 269.0669
(907) 269.8166
-----Original Message-----
From: SPSSX(r) Discussion [mailto:SPSSX-L@...] On Behalf Of
CarolineUK
Sent: Tuesday, November 24, 2009 9:43 PM
To: SPSSX-L@...
Subject: How to read v14 output files in later versions/at all?

I've searched discussions and found that you've discussed before that
SPSS 16
can't read SPSS 14 output files. Has anyone any advice about what you
do
about it?
I'm mid way through a lot of analysis and my university is upgrading its
licence from 16 to 14, so reluctantly, I'm going to have to use 16 a
week
from now.
I've installed the 'legacy viewer' in the vain hope that it would work
backwards as well as forwards but it doesn't seem to make any difference
to
this problem.
Can I take out a licence for 14 myself? Should I waste a tree and print
everything out while I still can?
I'd really welcome some advice.
--
View this message in context:
http://old.nabble.com/How-to-read-v14-output-files-in-later-versions-at-
all--tp26508213p26508213.html
Sent from the SPSSX Discussion mailing list archive at Nabble.com.

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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

Re: Using means as Dependant Variable

2009-11-25T11:34:45Z

James,

The original design would have had one between factor and two within factors
(speed and throw). Somebody averaged across throws. So now you have a one
between and one within design. I think the results are statistically valid
given the data.

Gene Maguin

>>A friend of mine has sent me the results of some throwing accuracy tests
that he has done on cricketers and baseballers throwing at 2 different
speeds. The data I have been given is the mean accuracy for each
cricketer/ baseballer over ten throws. I cannot get acces to the raw data.

What statistical short comings would a Univariate Analysis of variance
using SPSS provide me with if I were to use the mean as the dependant
variable and the Between subject factors; Sport (baseball or cricket) and
Speed (80% or 100%. I know that I would not be able to provide any insight
as to whether a thrower's accuracy is affected by fatigue or repetition
but I would just like to know whether the results that I get from the
between subject factors would still be statistically valid despite using
the means as my dependant variable.

Sorry for sucha vague question but any response would be greatly
appreciated.

James Chiswell
Sportsdata - Statistician

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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

Re: dichotomous moderator with continuous IV and DV

2009-11-25T11:28:28Z

Protzko wrote:

So I have a problem with mediator variables. I have a dataset that has a continuous IV, a continuous DV but the mediator is dichotomous. Also, the data is non-normal and refuses to get so when it comes to transforming. Therefore, I am trying to find a robust method to run a regression on the data (bootstrapping, sobel, OLS, etc.) that can not only handle the non-normality, but can also deal with my dichotomous mediator. Anyone have any ideas?

many thanks

~John

John, you say "moderator" in the subject line, and "mediator" in the body of the message. Which is it? Regarding normality, bear in mind that it is the residuals from the model that should be approximately normal, but even that is not the most important requirement. It is more important that the residuals be mutually independent, and not correlated with the explanatory variables (or with the fitted value of Y).

--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/
"When all else fails, RTFM."

NOTE: My Hotmail account is not monitored regularly.
To send me an e-mail, please use the address shown above.

dichotomous moderator with continuous IV and DV

2009-11-25T11:16:08Z

So I have a problem with mediator variables. I have a dataset that has a continuous IV, a continuous DV but the mediator is dichotomous. Also, the data is non-normal and refuses to get so when it comes to transforming. Therefore, I am trying to find a robust method to run a regression on the data (bootstrapping, sobel, OLS, etc.) that can not only handle the non-normality, but can also deal with my dichotomous mediator. Anyone have any ideas?

many thanks

~John

Re: Macro for reordering variables

2009-11-25T11:14:41Z

Dear Bruce, Albert-Jan and Bob,

Thanks very much for the wonderful solutions! The two classic macro solutions work fine, the Python solution -unfortunately- made SPSS crash (it should work with V17.01 and Python 2.5, right?) but I'll try to run it again tomorrow at work because I'll have to start learning Python at some point anyway.

Kind regards,

Ruben

> Date: Wed, 25 Nov 2009 06:12:16 -0800
> From: bruce.weaver@...
> Subject: Re: Macro for reordering variables
> To: SPSSX-L@...
>
> Ruben van den Berg wrote:
> >
> >
> > Dear all,
> >
> >
> >
> > I'm trying to write a macro that should reorder the variables in my file.
> > Right now, I've Label.1 to Label.15 and then Leave.1 to Leave.15. I'd like
> > the order to be Label.1 Leave.1 Label.2 Leave.2 (...) Label.15 Leave.15.
> > Admittedly, I could write this by hand since it's only 30 variables but
> > next time it could be 100 or 200 variables and I'd really like to learn
> > some very basic macro language as well.
> >
> >
> >
> > Thanks a lot in advance!
> >
> >
> > define !reorder ()
> >
> > !do !t2=1 !to 15
> >
> > !let !varlist=!concat('label.',!t2) !concat('leave.',!t2)
> >
> > !doend
> >
> > matc fil fil *
> >
> > /kee variable_name !varlist
> >
> > exe.
> >
> > !end.
> >
> >
> >
> > set mpr on.
> >
> > !reorder.
> >
> >
>
> You need to append !varlist to itself each time through the loop, and you
> need to leave spaces between the variable names in the list. Try this:
>
> define !reorder ()
> !let !varlist=''
> !do !t2=1 !to 15
> !let !varlist=!concat(!varlist,' label.',!t2,' leave.',!t2)
> !doend
> match files file = *
> /keep variable_name !varlist
> exe.
> !end.
>
>
>
> -----
> --
> Bruce Weaver
> bweaver@...
> http://sites.google.com/a/lakeheadu.ca/bweaver/
> "When all else fails, RTFM."
>
> NOTE: My Hotmail account is not monitored regularly.
> To send me an e-mail, please use the address shown above.
> --
> View this message in context: http://old.nabble.com/Macro-for-reordering-variables-tp26509696p26513307.html
> Sent from the SPSSX Discussion mailing list archive at Nabble.com.
>
> =====================
> To manage your subscription to SPSSX-L, send a message to
> LISTSERV@... (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

Express yourself instantly with MSN Messenger! MSN Messenger

Re: Macro for reordering variables

2009-11-25T09:54:42Z

Hi Ruben,

Alternatively, using an example with five variables...

PRESERVE.

SET MEXPAND ON /MPRINT ON.

DATA LIST LIST /label.1 TO label.5 leave.1 TO leave.5.

BEGIN DATA

0 1 0 1 1 0 1 0 1 1

1 1 1 1 1 1 1 0 1 1

0 1 0 1 1 0 1 0 1 1

END DATA.

DEFINE RUBEN()

MATCH FILES FILE=* /KEEP !DO !Z = 1 !TO 5 !CONCAT("label.",!Z," leave.",!Z) !DOEND

!ENDDEFINE.

RUBEN.

RESTORE.

Here the looping is simply embedded within a /KEEP subcommand (which is OK)…

Regards,

Bob Walker

Surveys & Forecasts, LLC

www.safllc.com

From: SPSSX(r) Discussion [mailto:SPSSX-L@...] On Behalf Of Ruben van den Berg
Sent: Wednesday, November 25, 2009 4:26 AM
To: SPSSX-L@...
Subject: Macro for reordering variables

Express yourself instantly with MSN Messenger! MSN Messenger

Re: How to read v14 output files in later versions/at all?

2009-11-25T06:32:08Z

Brian - have tried smartviewer again. Thought it was a patch - didn't realise it was a standalone programme that simply allows you to 'view' older version files. Note to self - don't try to solve problems at 5am in the morning!
thanks... the smartviewer seems to do the trick.

Caroline

Re: Macro for reordering variables

2009-11-25T06:15:38Z

Hi!

Here's a Python approach. Don't know how to do it in macro. Perhaps with N of cases = 1, then FLIP, SORT WRITE, etc.

* vector label. (50) / leave. (50). /* test variables.

begin program.

import spssaux

labels = spssaux.VariableDict(pattern = r"label")

leaves = spssaux.VariableDict(pattern = r"leave")

vlist = " ".join([label + " " + leave for label, leave in zip (labels, leaves)])

spss.Submit("add files / file = * / keep = %s all." % vlist)

end program.

Cheers!!
Albert-Jan

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
In the face of ambiguity, refuse the temptation to guess.
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

--- On Wed, 11/25/09, Ruben van den Berg <ruben_van_den_berg@...> wrote:

From: Ruben van den Berg <ruben_van_den_berg@...>
Subject: [SPSSX-L] Macro for reordering variables
To: SPSSX-L@...
Date: Wednesday, November 25, 2009, 10:25 AM

Dear all,

I'm trying to write a macro that should reorder the variables in my file. Right now, I've Label.1 to Label.15 and then Leave.1 to Leave.15. I'd like the order to be Label.1 Leave.1 Label.2 Leave.2 (...) Label.15 Leave.15. Admittedly, I could write this by hand since it's only 30 variables but next time it could be 100 or 200 variables and I'd really like to learn some very basic macro language as well.

Thanks a lot in advance!

define !reorder ()
!do !t2=1 !to 15
!let !varlist=!concat('label.',!t2) !concat('leave.',!t2)
!doend
matc fil fil *
/kee variable_name !varlist
exe.
!end.

set mpr on.
!reorder.

Express yourself instantly with MSN Messenger! MSN Messenger

Re: Macro for reordering variables

2009-11-25T06:12:16Z

Ruben van den Berg wrote:

Dear all,

I'm trying to write a macro that should reorder the variables in my file. Right now, I've Label.1 to Label.15 and then Leave.1 to Leave.15. I'd like the order to be Label.1 Leave.1 Label.2 Leave.2 (...) Label.15 Leave.15. Admittedly, I could write this by hand since it's only 30 variables but next time it could be 100 or 200 variables and I'd really like to learn some very basic macro language as well.

Thanks a lot in advance!

define !reorder ()

!do !t2=1 !to 15

!let !varlist=!concat('label.',!t2) !concat('leave.',!t2)

!doend

matc fil fil *

/kee variable_name !varlist

exe.

!end.

set mpr on.

!reorder.

You need to append !varlist to itself each time through the loop, and you need to leave spaces between the variable names in the list. Try this:

define !reorder ()
!let !varlist=''
!do !t2=1 !to 15
!let !varlist=!concat(!varlist,' label.',!t2,' leave.',!t2)
!doend
match files file = *
/keep variable_name !varlist
exe.
!end.

Re: Sample power survival calculation

2009-11-25T06:10:47Z

Carol,

I think you really do need 66 persons. So, two scenarios given a starting
sample size of 60. 1) The day after enrollment, six people drop out. 2) The
day before the study ends, six people drop out. Given that the actual data
generates the expected sample size, I think a power calculation will show
scenario 1 to have power less than 80 and scenario 2 to have power
fractionally less than if the six people had not quit on the day before the
last day.

Gene Maguin

>>I'm running Sample Power 2.0 for a survival calculation. My question is
about attrition.
Suppose you include an attrition of 10% and come up with a 60 patient study
that is powered at 80%. Do you really need 66 to get the power of .80? Or
will you have adequate power with 54 patients (a 10% attrition)?

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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

Re: How to read v14 output files in later versions/at all?

2009-11-25T05:59:39Z

CarolineUK wrote:

I've searched discussions and found that you've discussed before that SPSS 16 can't read SPSS 14 output files. Has anyone any advice about what you do about it?
I'm mid way through a lot of analysis and my university is upgrading its licence from 16 to 14, so reluctantly, I'm going to have to use 16 a week from now.
I've installed the 'legacy viewer' in the vain hope that it would work backwards as well as forwards but it doesn't seem to make any difference to this problem.
Can I take out a licence for 14 myself? Should I waste a tree and print everything out while I still can?
I'd really welcome some advice.

I am surprised to hear that the legacy viewer (SmartViewer) does not work. You are talking about .spo files, are you?

I'm also not sure what you mean when you talk about the legacy viewer working forwards. Why would it need to work forwards?

Going forward, I think the general lesson here is to (as much as possible) save syntax that will generate the output in newer versions of the software.

literature of mediation

2009-11-25T05:58:22Z

Someone might know of articles on mediation effect from zero-order and parttial correlation to SEM approaches, please post here. I will have to study..

Thank you.

Eins

Yahoo! Mail Now Faster and Cleaner. Experience it today!

search argument and return the nth name

2009-11-25T05:45:26Z

Hi SPSS user,

Can someone tell me if SPSS has a function (similar to SAS %scan) that can search an argument (or string of variable names separated by ' ') and return the nth name. Below is an example of what I am trying to do using SAS code.

%do i = 1 %to 5;
%let varname = %scan(&varlist,&i, ' ');
%end.

Thanks in advance for your help.

AUTO: Daniel Edelstein is out of the office. (returning Mon 11/30/2009)

2009-11-25T05:06:43Z

I am out of the office from Wed 11/25/2009 until Mon 11/30/2009.

I will respond to your message when I return.

Note: This is an automated response to your message "SPSSX-L Digest - 23 Nov 2009 to 24 Nov 2009 (#2009-328)" sent on 11/25/09 0:00:16.

This is the only notification you will receive while this person is away.

Macro for reordering variables

2009-11-25T01:25:38Z

Express yourself instantly with MSN Messenger! MSN Messenger

How to read v14 output files in later versions/at all?

2009-11-24T22:43:22Z

I've searched discussions and found that you've discussed before that SPSS 16 can't read SPSS 14 output files. Has anyone any advice about what you do about it?
I'm mid way through a lot of analysis and my university is upgrading its licence from 16 to 14, so reluctantly, I'm going to have to use 16 a week from now.
I've installed the 'legacy viewer' in the vain hope that it would work backwards as well as forwards but it doesn't seem to make any difference to this problem.
Can I take out a licence for 14 myself? Should I waste a tree and print everything out while I still can?
I'd really welcome some advice.

Using means as Dependant Variable

2009-11-24T22:10:13Z

Hi,

A friend of mine has sent me the results of some throwing accuracy tests
that he has done on cricketers and baseballers throwing at 2 different
speeds. The data I have been given is the mean accuracy for each
cricketer/ baseballer over ten throws. I cannot get acces to the raw data.

What statistical short comings would a Univariate Analysis of variance
using SPSS provide me with if I were to use the mean as the dependant
variable and the Between subject factors; Sport (baseball or cricket) and
Speed (80% or 100%. I know that I would not be able to provide any insight
as to whether a thrower's accuracy is affected by fatigue or repetition
but I would just like to know whether the results that I get from the
between subject factors would still be statistically valid despite using
the means as my dependant variable.

Sorry for sucha vague question but any response would be greatly
appreciated.

James Chiswell
Sportsdata - Statistician

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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

Re: Vector Definition Issues

2009-11-24T19:28:19Z

At 04:42 PM 11/24/2009, Marks, Jim wrote:

>I thought a vector needed to refer to contiguous variables in a dataset?

Which, indeed, it does.

>If some of the variables are in the data set and some are added at
>the end of the variable list, will this still work?

Well, the only place you can add variables to a dataset is at the
end. So, the only way you can have pre-existing and new variables in
the same vector, is to have the pre-existing ones at the end of the dataset.

But, was that your problem? You wrote as if you're creating a
contiguous set of variables ("when the vector format creates new
variables we have no problem the first time we run the syntax"), and
were looking for syntax that would create these variables if
necessary, and define them as a vector whether newly created or not.
That was the problem I was addressing.

Tell us more about what you're trying to do: create a contiguous set
of variables, once? Create a contiguous set of variables, and then
extend the set on successive runs? Or, what else?

-Best wishes,
Richard Ristow

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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

Sample power survival calculation

2009-11-24T18:36:25Z

I'm running Sample Power 2.0 for a survival calculation. My question is about attrition.

Suppose you include an attrition of 10% and come up with a 60 patient study that is powered at 80%. Do you really need 66 to get the power of .80? Or will you have adequate power with 54 patients (a 10% attrition)?

thanks
carol

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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

Re: Windows 7 32 bit and PAWS 17 and 18 RAM

2009-11-24T18:17:27Z

SPSS techsupport is a lot helpful in Australia, they are not involved with sales thankfully.

Brian

> RalphG2703@... wrote:
>
> Thanks. This is very informative! Everyone I talked to by phone at SPSS
>
> said that SPSS v.17 and AMOS v.17 would not run under Windows 7. Of
> course,
> they wanted to sell me upgrades to versions 18, coming in just around
> $1000.
> Very uninformed and embarrassing, especially since they answer the
> phone
> now as "SPSS, an IBM Company." Ralph

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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

Re: Windows 7 32 bit and PAWS 17 and 18 RAM

2009-11-24T16:28:17Z

Thanks. This is very informative! Everyone I talked to by phone at SPSS said that SPSS v.17 and AMOS v.17 would not run under Windows 7. Of course, they wanted to sell me upgrades to versions 18, coming in just around $1000. Very uninformed and embarrassing, especially since they answer the phone now as "SPSS, an IBM Company." Ralph

Re: Windows 7 32 bit and PAWS 17 and 18 RAM

2009-11-24T14:05:10Z

Right click on spss 17 icon and select compatibility mode (XP service pack 2) allow the changes, and it should. If you are unfamiliar with Windows 7, ask you local PC dealer or help desk to show you how to. It should work for AMOS and other legacy products.

From: SPSSX(r) Discussion [mailto:SPSSX-L@...] On Behalf Of Ralph Grubb
Sent: Wednesday, 25 November 2009 1:51 AM
To: SPSSX-L@...
Subject: Re: Windows 7 32 bit and PAWS 17 and 18 RAM

How do you set SPSS v.17 to run in compatibility mode? is this also true for other SPSS products such as AMOS v.17? Thanks. Ralph

Re: Cartesian product in SPSS

2009-11-24T14:02:10Z

Ah, come on Richard, Don't hold back on your discovery.

Gene Maguin

>>Interestingly, I think that even that spread can be avoided, as can the
necessity of hard-coding the number of cases in the logic. I have discovered
a truly wonderful proof of this, which this message is insufficient to hold.

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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

Re: Vector Definition Issues

2009-11-24T13:42:50Z

Richard:

I thought a vector needed to refer to contiguous variables in a dataset? If some of the variables are in the data set and some are added at the end of the variable list, will this still work?

Jim Marks

Director, Market Research

x1616

From: SPSSX(r) Discussion [mailto:SPSSX-L@...] On Behalf Of Richard Ristow
Sent: Tuesday, November 24, 2009 3:10 PM
To: SPSSX-L@...
Subject: Re: Vector Definition Issues

At 02:26 PM 11/24/2009, Mark Vande Kamp wrote:

We are creating code that uses vectors and running into problems arising when the vector definition code is inconsistent with the presence or absence of the variables in the data file. For example, when the vector format creates new variables we have no problem the first time we run the syntax. However, if we happen to save the file and then run the syntax again, the vector commands do not work and the subsequent loops don't work either.

===================== To manage your subscription to SPSSX-L, send a message to LISTSERV@... (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

Re: [Unsubscribe]

2009-11-24T13:25:50Z

At 02:30 PM 11/24/2009, Christopher R. Ferland wrote:

>unsubscribe

As noted, nobody who reads the list can unsubscribe you. We have a
SIG that's supposed to go out with all messages, giving directions:

>To manage your subscription to SPSSX-L, send a message to
>LISTSERV@... (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

Does this do it for you?

Your posting is one of a number that hit the list without this SIG.
In hopes of tracking this down, could you let me know (off-list is fine),

- Do any of the postings you receive have that above SIG?

- How did you subscribe in the first place? That is, by sending a
SUBSCRIBE E-mail message, the Web site, or what?

- Are you receiving the messages straight from
SPSSX-L@..., or from one of the sites that 'shadow' the list?

- What ISP are you using to read messages? to post them? Is any
server other than usg.edu involved, that you're aware of?

- Or, do you read messages on some Web site?

-Thanks, a little mystified,
Richard Ristow

Anything

=====================
To manage your subscription to SPSSX-L, send a message to
LISTSERV@... (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

Re: Vector Definition Issues

2009-11-24T13:10:04Z

At 02:26 PM 11/24/2009, Mark Vande Kamp wrote:

We are creating code that uses vectors and running into problems arising when the vector definition code is inconsistent with the presence or absence of the variables in the data file. For example, when the vector format creates new variables we have no problem the first time we run the syntax. However, if we happen to save the file and then run the syntax again, the vector commands do not work and the subsequent loops don't work either.

Here's a possible work-around, untested. In this example, the vector elements are Datum01 TO Datum17. It should work, because assigning to a numeric variable works whether or not the variable already exists. Then, it should always work with 'already-exists' vector syntax.

DO IF $CASENUM EQ 1. . DO REPEAT Element = Datum01 TO Datum17. . COMPUTE Element = 0. . END REPEAT. END IF. VECTOR DATA=Datuum01 TO Datum17. ===================== To manage your subscription to SPSSX-L, send a message to LISTSERV@... (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

Re: Cartesian product in SPSS 2009-11-24T13:01:03Z At 09:26 AM 11/24/2009, Gene Maguin wrote: Lakshmikanth sent me a message late last week naming this message. Warning # 525 An attempt was made to store a value into an element of a vector the subscript of which was missing or otherwise invalid. The subscript must be a positive integer and must not be greater than the length of the vector. No store can occur. Command line: 12 Current case: 4 Current splitfile group: 1 I think the cause of the warning may have been that Lakshmikanth did not change the vector dimension to however many cases he had, I didn't confirm that with him. Yep. You learn the most amazing things actually reading the messages, don't you? As I wrote, the number of cases in file 1 have to be hard-coded in the syntax. This sure looks like what's happen if that isn't done. As I read Dave Marso's syntax it seemed to me that it was a pre varstocases-casestovars structure. Yes, though I think that would change surprisingly little. The following syntax: *spread file 1*. vector ID1.(4,A1). compute ID1.(case)=ID1. compute merge=1. FORMATS merge (F2). DATASET DECLARE file1. aggregate outfile file1 / break merge / ID1.1 TO ID1.4 =MAX(ID1.1 TO ID1.4). could easily be recast, and made simpler, as CASESTOVARS logic. But that isn't the crucial part of the code. It also concerned me that if the 40000 case file were to be converted to a wide format there would be trouble because, I think I remember you saying, there is a 32000 variable limit to spss. Look at David Marso's syntax: he never converts either file into 'wide' form. (That would be practically impossible without CASESTOVARS, anyway.) He is using an ingenious form of what I call 'spine' logic: the 'wide' portion of the file is file 2 with the indices of the records in file 1 spread over all records - not the contents of those records. This is his 'spread' file: ID2 var2 a2 merge ID1.1 ID1.2 ID1.3 ID1.4 I One 987 1 A B C D II Two 654 1 A B C D III Three 321 1 A B C D IV Four 9 1 A B C D Interestingly, I think that even that spread can be avoided, as can the necessity of hard-coding the number of cases in the logic. I have discovered a truly wonderful proof of this, which this message is insufficient to hold. ===================== To manage your subscription to SPSSX-L, send a message to LISTSERV@... (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 Re: (no subject) 2009-11-24T11:36:14Z No can do Christopher. -----Original Message----- From: Christopher R. Ferland <Chris.Ferland@...> To: SPSSX-L@... Sent: Tue, Nov 24, 2009 2:30 pm Subject: (no subject) unsubscribe Christopher R. Ferland Board of Regents of the University System of Georgia 270 Washington Street Atlanta GA 30334 Phone: 404-656-2220 or 706-583-2113 Fax: 404-657-4130 email: chris.ferland@... (no subject) 2009-11-24T11:30:53Z unsubscribe Christopher R. Ferland Board of Regents of the University System of Georgia 270 Washington Street Atlanta GA 30334 Phone: 404-656-2220 or 706-583-2113 Fax: 404-657-4130 email: chris.ferland@... Vector Definition Issues 2009-11-24T11:26:01Z We are creating code that uses vectors and running into problems arising when the vector definition code is inconsistent with the presence or absence of the variables in the data file. For example, when the vector format creates new variables we have no problem the first time we run the syntax. However, if we happen to save the file and then run the syntax again, the vector commands do not work and the subsequent loops don't work either. We tried the kludgy method of defining each vector using both the "new variable" and "existing variable" formats, but that creates a bunch of error messages in the log that make it difficult to find legitimate errors. I'm looking for syntax that will work whether or not the variables happen to be present in the data file. Any help will be appreciated. Mark ===================== To manage your subscription to SPSSX-L, send a message to LISTSERV@... (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 Re: Functional syntax editor weirdness 2009-11-24T10:17:10Z Hi Gene, We've filed three bugs: 1) The weirdness with text strings when pasting at the end of a line. 2) Overtype/overwrite mode pulling up the following line rather than just adding to the end of the current one when you reach the end. 3) Alt-R needing to be repeated in order to repeat the search and replace function. Dave > From: > > Gene Maguin <emaguin@...> > > To: > > SPSSX-L@... > > Date: > > 11/18/2009 09:40 AM > > Subject: > > Re: Functional syntax editor weirdness > > Sent by: > > "SPSSX(r) Discussion" <SPSSX-L@...> > > Hi Dave, > > (For your records, I've also copied this to the opened case.) > >>I don't replicate exactly what you're describing when pasting at the end > of a line, but I do see something similar on both 17 and 18. It behaves > better when you have the new features turned on (the pasted text appears in > full after a slight delay, where with the new features turned off it doesn't > fully appear until you hit return or type something else there. We'll get a > bug filed on this. > > I had an off-list reply from Anita yesterday and following her advice, I > turned on syntax color coding and set all colors to black. While I would > agree that the editor behaves oddly and could benefit from a fix, I'm happy > now on this point. So, closed. > > >>Ctrl-Del in 18 works as it does in 15, deleting from the current location > to the end of the line. > > Thank you. > > >>I'm not sure what you mean by nonstandard operation of overtype mode at > the end of line. Can you be more specific? > > This is easier to show, harder to describe. But here goes. Suppose you have > > VECTOR RATEI=ROWI TO PPWI. > DO IF (SEX EQ 1). /* FEMALE RESPONDENTS. > > Now suppose I want to change 'ROWI TO PPWI.' to 'RespSelfInt TO > PartRespInt.' Now, I agree that there's different ways of doing this. But > one way is to go into overtype mode and just type. So you do that and when > you are at this point > > VECTOR RATEI=RespSelfInt T > DO IF (SEX EQ 1). /* FEMALE RESPONDENTS. > > Typing the next character pulls the following line up as in > > VECTOR RATEI=RespSelfInt TODO IF (SEX EQ 1). /* FEMALE RESPONDENTS. > > I hope to god that's clear. My claim is that 15 and prior did not do this, > and that Word, outlook, and multiedit do not do this. > > > >>I'm also not sure what you're referring to with Alt-R. In the syntax > editor, Alt-R is the shortcut to the Run menu in 15, as it is in more recent > Java releases. > > Sorry, I wasn't detailed enough here. Suppose you have > > DO IF ($CASENUM EQ 1 OR ID NE LAG(ID)). > * RESPONDENT LEVEL. > + COMPUTE STEM=RV.NORMAL(0.0,1.0). > + COMPUTE B0RC=0.5*STEM+RV.NORMAL(0.0,0.866). > + COMPUTE B0RF=0.5*STEM+RV.NORMAL(0.0,0.866). > + COMPUTE B0RI=0.5*STEM+RV.NORMAL(0.0,0.866). > > And you want to change 'BOR' to 'BAR'. So, Ctrl-H for find and replace, > find: BOR, replace: BAR, 'find next'. Then: my claim is that previously > (<=15) Alt-R would execute the change-find next sequence repeatedly until > the Alt-R keys were released. Currently, Alt-R executes the sequence one > time. So maybe, find-and-replace wasn't supposed to work that was previously > but that's how it works in word and other editors. > > Gene Maguin > > ===================== > To manage your subscription to SPSSX-L, send a message to > LISTSERV@... (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 Re: Problem with Equal Variance Not Assumed 2009-11-24T10:12:27Z I believe v17 is being fussy about producing invalid tests. From the help: "When the variances are unequal, use Tamhane's T2 (conservative pairwise comparisons test based on a t test), Dunnett's T3 (pairwise comparison test based on the Studentized maximum modulus), Games-Howell pairwise comparison test (sometimes liberal), or Dunnett's C (pairwise comparison test based on the Studentized range). Note that these tests are not valid and will not be produced if there are multiple factors in the model." Alex From: "Smith, Barry" <BSmith@...> To: SPSSX-L@... Date: 11/24/2009 11:46 AM Subject: Problem with Equal Variance Not Assumed Sent by: "SPSSX(r) Discussion" <SPSSX-L@...> I am using Release Version 17.0.2 I am attempting to do a GLM Univariate analysis. The data set has two Fixed factors both of which are ordinal. The dependent variable is scalar. I have set a custom model with the two fixed factors as main effects and the combination of the two fixed factor as interaction. When I go to Post Hoc tests and move one or both of the fixed factors to the “Post Hoc Test for” box, the “Equal Variances Not Assumed” group of check boxes remain grayed out. I do not have the same problem when attempting this with Version 15. Is this a known bug in 17 or is 17 being fussy about something that Version 15 was not fussy about? By the way, I have a student using a grad student version 17 who is seeing the same problem. I am also seeing the same problem with other simple data files. Does anyone know was is going on? Thanks for your help. Barry

From:	"Smith, Barry" <BSmith@...>
To:	SPSSX-L@...
Date:	11/24/2009 11:46 AM
Subject:	Problem with Equal Variance Not Assumed
Sent by:	"SPSSX(r) Discussion" <SPSSX-L@...>