excel file issue

[Dave Taber]

has anyone run across this issue?

i open an excel 2007 file with syntax in spss and then i save the file as an spss data file and the run additional syntax creating new files  (with the original excel file not being used at the end of the syntax)

but when i try and then edit the original excel file in excel  i get a message saying that file is being used so it can't be saved 

when i close down spss the problem goes away so it appears as if spss is somehow still holding on to my excel file behind the scenes

i'm using v17 and excel 2007 with new copy of Vista on a fairly high-end pc

i didn't have this issues with my old setup which was xp / excel 2003 and v16

any suggestions on how to avoid this would be greatly appreciated

thanks

dave

[Bud Bliss]

Since I usually export from SPSS to Excel, I haven't seen this exact problem, but I do have similar problems when reading from Excel 2007 files in other apps on XP. Not a major problem in what I do, but I would be interested in hearing a solution!

Meredith (Bud) Bliss
Crime Analyst - Beaverton Police Dept.
4755 SW Griffith Drive
Beaverton, OR 97005
503-526-2294 - Fax: [phone]
________________________________
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Dave Taber
Sent: Thursday, July 16, 2009 8:54 AM
To: [hidden email]
Subject: excel file issue

has anyone run across this issue?

i open an excel 2007 file with syntax in spss and then i save the file as an spss data file and the run additional syntax creating new files  (with the original excel file not being used at the end of the syntax)

but when i try and then edit the original excel file in excel  i get a message saying that file is being used so it can't be saved

when i close down spss the problem goes away so it appears as if spss is somehow still holding on to my excel file behind the scenes

i'm using v17 and excel 2007 with new copy of Vista on a fairly high-end pc

i didn't have this issues with my old setup which was xp / excel 2003 and v16

any suggestions on how to avoid this would be greatly appreciated

thanks
dave






======================================================================
PUBLIC RECORDS LAW DISCLOSURE

This e-mail is a public record of the City of Beaverton and is subject to public disclosure unless exempt from disclosure under Oregon Public Records Law.  This email is subject to the State Retention Schedule.

=====================
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

[Richard Ristow]

At 11:54 AM 7/16/2009, Dave Taber wrote:

>I open an Excel 2007 file with syntax in SPSS and then I save the
>file as an SPSS data file and the run additional syntax creating new
>files  (with the original Excel file not being used at the end of
>the syntax). But when I try and then edit the original excel file in
>Excel I get a message saying that file is being used so it can't be
>saved. It appears as if SPSS is somehow still holding on to my Excel
>file behind the scenes.

That may be happening as designed. Since, I think, release 12.5, SPSS
has used a 'virtual active fit' -- that is, the file you build and
run isn't saved, but the syntax to build it is, and when you use the
file again, that syntax is re-run to rebuild the file as you need it.

That save disk space, since you don't need a scratch disk file. But
it can well be slower; and it means SPSS has to hang onto its input files.

Try using 'CACHE' after you read the first file. That forces SPSS to
write a scratch file, after all, and should get it to 'let go' of the
Excel file.

On the other hand, it's common enough for programs to hang onto files
and other resources they aren't going to use anymore. I'm not aware
of any such problem documented for SPSS, but I wouldn't dogmatically
rule it out.

=====================
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

[Dave Taber]

thank you

i will try the CACHE  command

dave

On Jul 19, 2009, at 2:50 AM, Richard Ristow wrote:

At 11:54 AM 7/16/2009, Dave Taber wrote:

I open an Excel 2007 file with syntax in SPSS and then I save the file as an SPSS data file and the run additional syntax creating new files  (with the original Excel file not being used at the end of the syntax). But when I try and then edit the original excel file in Excel I get a message saying that file is being used so it can't be saved. It appears as if SPSS is somehow still holding on to my Excel file behind the scenes.

That may be happening as designed. Since, I think, release 12.5, SPSS has used a 'virtual active fit' -- that is, the file you build and run isn't saved, but the syntax to build it is, and when you use the file again, that syntax is re-run to rebuild the file as you need it.

That save disk space, since you don't need a scratch disk file. But it can well be slower; and it means SPSS has to hang onto its input files.

Try using 'CACHE' after you read the first file. That forces SPSS to write a scratch file, after all, and should get it to 'let go' of the Excel file.

On the other hand, it's common enough for programs to hang onto files and other resources they aren't going to use anymore. I'm not aware of any such problem documented for SPSS, but I wouldn't dogmatically rule it out.

[mpirritano]

I am trying to calculate risk from my logistic regression results. The data is whether computer access to medical records at the time of an emergency dept visit is related to an ed visit within three months or not (DV). I’m also including the amount of time that they have been enrolled in the safety net health program. Medical records access is coded as 0 for access occurred, and 1 no access (this is actually the reverse of how I coded it, but how the logistic regression procedure is coding it).

 

The strange thing is that I’m getting the same risk for those who had their records accessed and those who did not. I’m using an equation I’ve gotten from Scott Menard’s Applied Logistic Regression.

 

My beta’s are: for time in the program, -.019.

For the corresponding X I’m just using the avg time spent in the program, 11.32 months

 

For no access,-1.571.

 

The intercept is 3.577.

 

The equation for the probability of ED visit for those with no access is:

 

e^ -.019*(11.32)+ 3.577

----------------------------------

1+ e^ -.019*(11.32) + 3.577

 

Equals 0.367879441

 

 

For those with their information accessed (including the beta for access):

 

e^ -.019*(11.32) -1.517 + 3.577

--------------------------------------------

1 + e^ -.019*(11.32) – 1.571 + 3.577

 

Equals 0.367879441

 

 

How could this be possible? The probability is exactly the same for both?

 

Am I missing something here?

 

Thanks

Matt

 

Matthew Pirritano, Ph.D.

Research Analyst IV

Medical Services Initiative (MSI)

Orange County Health Care Agency

[phone]

 

[Gary Rosin]

I (or MS Excel(r)) got probabilities of 0.966 (no access) vs. 0.857 (access),
using

                y = 3.577 + (-0.019*duration) + (-1.571*access), and
     duration = 11.32

access = 0                access = 1

        y = 3.362                   y = 1.791
exp(y) = 28.845        exp(y) = 5.995
  prob = 0.966            prob  = 0.857

That's consistent with access making the odds 

    exp(-1.571) = 0.208 times smaller

I'll send you the Excel worksheet directly.

Gary

.


At 12:31 PM 7/20/2009, Pirritano, Matthew wrote:

I am trying to calculate risk from my logistic regression results. The data is whether computer access to medical records at the time of an emergency dept visit is related to an ed visit within three months or not (DV). I’m also including the amount of time that they have been enrolled in the safety net health program. Medical records access is coded as 0 for access occurred, and 1 no access (this is actually the reverse of how I coded it, but how the logistic regression procedure is coding it).
 
The strange thing is that I’m getting the same risk for those who had their records accessed and those who did not. I’m using an equation I’ve gotten from Scott Menard’s Applied Logistic Regression.
 
My beta’s are: for time in the program, -.019.
For the corresponding X I’m just using the avg time spent in the program, 11.32 months
 
For no access,-1.571.
 
The intercept is 3.577.
 
The equation for the probability of ED visit for those with no access is:
 
e^ -.019*(11.32)+ 3.577
----------------------------------
1+ e^ -.019*(11.32) + 3.577
 
Equals 0.367879441
 
 
For those with their information accessed (including the beta for access):
 
e^ -.019*(11.32) -1.517 + 3.577
--------------------------------------------
1 + e^ -.019*(11.32) – 1.571 + 3.577
 
Equals 0.367879441
 
 
How could this be possible? The probability is exactly the same for both?
 
Am I missing something here?
 
Thanks
Matt
 
Matthew Pirritano, Ph.D.
Research Analyst IV
Medical Services Initiative (MSI)
Orange County Health Care Agency
[phone]
 

[Albert-Jan Roskam]

Somebody recommended using the CACHE command. In the past using CACHE, followed by an EXECUTE command worked for me, but that was (I believe) in an sql query using GET DATA.

Cheers!!
Albert-Jan


--- On Fri, 7/17/09, Bud Bliss <[hidden email]> wrote:

> From: Bud Bliss <[hidden email]>
> Subject: Re: excel file issue
> To: [hidden email]
> Date: Friday, July 17, 2009, 12:15 AM
> Since I usually export from SPSS to
> Excel, I haven't seen this exact problem, but I do have
> similar problems when reading from Excel 2007 files in other
> apps on XP. Not a major problem in what I do, but I would be
> interested in hearing a solution!
>
> Meredith (Bud) Bliss
> Crime Analyst - Beaverton Police Dept.
> 4755 SW Griffith Drive
> Beaverton, OR 97005
> 503-526-2294 - Fax: [phone]
> ________________________________
> From: SPSSX(r) Discussion [mailto:[hidden email]]
> On Behalf Of Dave Taber
> Sent: Thursday, July 16, 2009 8:54 AM
> To: [hidden email]
> Subject: excel file issue
>
> has anyone run across this issue?
>
> i open an excel 2007 file with syntax in spss and then i
> save the file as an spss data file and the run additional
> syntax creating new files  (with the original excel
> file not being used at the end of the syntax)
>
> but when i try and then edit the original excel file in
> excel  i get a message saying that file is being used
> so it can't be saved
>
> when i close down spss the problem goes away so it appears
> as if spss is somehow still holding on to my excel file
> behind the scenes
>
> i'm using v17 and excel 2007 with new copy of Vista on a
> fairly high-end pc
>
> i didn't have this issues with my old setup which was xp /
> excel 2003 and v16
>
> any suggestions on how to avoid this would be greatly
> appreciated
>
> thanks
> dave
>
>
>
>
>
>
> ======================================================================
> PUBLIC RECORDS LAW DISCLOSURE
>
> This e-mail is a public record of the City of Beaverton and
> is subject to public disclosure unless exempt from
> disclosure under Oregon Public Records Law.  This email
> is subject to the State Retention Schedule.
>
> =====================
> 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
>

=====================
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

[Dave Taber]

unfortunately, the CACHE solution didn't work for me, however,  I tried saving the excel file as an excel 2003 type and the problem went away -- so it seems as though the problem only happens with excel 2007 files

On Jul 19, 2009, at 2:50 AM, Richard Ristow wrote:

At 11:54 AM 7/16/2009, Dave Taber wrote:

I open an Excel 2007 file with syntax in SPSS and then I save the file as an SPSS data file and the run additional syntax creating new files  (with the original Excel file not being used at the end of the syntax). But when I try and then edit the original excel file in Excel I get a message saying that file is being used so it can't be saved. It appears as if SPSS is somehow still holding on to my Excel file behind the scenes.

That may be happening as designed. Since, I think, release 12.5, SPSS has used a 'virtual active fit' -- that is, the file you build and run isn't saved, but the syntax to build it is, and when you use the file again, that syntax is re-run to rebuild the file as you need it.

That save disk space, since you don't need a scratch disk file. But it can well be slower; and it means SPSS has to hang onto its input files.

Try using 'CACHE' after you read the first file. That forces SPSS to write a scratch file, after all, and should get it to 'let go' of the Excel file.

On the other hand, it's common enough for programs to hang onto files and other resources they aren't going to use anymore. I'm not aware of any such problem documented for SPSS, but I wouldn't dogmatically rule it out.

Dave Taber

novaQuant

listening - thinking - solving

[hidden email]

914.238.1310

www.novaQuant.com

[mpirritano]

Befuddled again,

Maybe I'm just overtired but I've looked at this every which way. My DV
has exactly 2 values. I can see that in the data, and frequencies,
crosstabs all list the two values as well. But when I try to run a
logistic regression I get the error message that there are less than two
nonmissing values.

I am trying to predict the number of emergency dept visits by whether
individuals have ever had their personal health information accessed
through a web based portal. I'd ideally like to be able to use a poison
regression, predicting total number of ED visits with whether their info
was accessed through the portal or not, but that doesn't seem to work
either. That also seems to include only one level of my dichotomous
variable. So I'm trying to predict the dichotomous variable, accessed
through portal or not (1,0 respectively) with number of ED visits.

Why is this seemingly illogical error message occurring?

Thanks
matt

Matthew Pirritano, Ph.D.
Research Analyst IV
Medical Services Initiative (MSI)
Orange County Health Care Agency
[phone]

=====================
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

[Maguin, Eugene]

Matthew,

Count missing values (both user and sysmis) across the variables used in the
logistic regression. One possibility is that when all the variables are in
the command statement, missing data on the IVs wipes out all the cases with
one value of the DV.

Gene Maguin

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Pirritano, Matthew
Sent: Tuesday, July 21, 2009 11:50 AM
To: [hidden email]
Subject: logistic regression error: less than two nonmissing values

Befuddled again,

Maybe I'm just overtired but I've looked at this every which way. My DV
has exactly 2 values. I can see that in the data, and frequencies,
crosstabs all list the two values as well. But when I try to run a
logistic regression I get the error message that there are less than two
nonmissing values.

I am trying to predict the number of emergency dept visits by whether
individuals have ever had their personal health information accessed
through a web based portal. I'd ideally like to be able to use a poison
regression, predicting total number of ED visits with whether their info
was accessed through the portal or not, but that doesn't seem to work
either. That also seems to include only one level of my dichotomous
variable. So I'm trying to predict the dichotomous variable, accessed
through portal or not (1,0 respectively) with number of ED visits.

Why is this seemingly illogical error message occurring?

Thanks
matt

Matthew Pirritano, Ph.D.
Research Analyst IV
Medical Services Initiative (MSI)
Orange County Health Care Agency
[phone]

=====================
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

=====================
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

[Hector Maletta]

Either one of the values of your DV has been declared as missing, or (which
is the most probable case) all valid cases FOR THE ANALYSIS have the same
value in the DV. This may arise because any case having a missing value in
ANY (dependent or independent) variable is discarded from the analysis. Some
cases may be discarded because they have a missing value in X, some because
they have a missing value in Z, and so on. It is just possible that each and
every case that finally made it into the analysis have the same value in the
dependent variable Y.
Hector

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Pirritano, Matthew
Sent: 21 July 2009 12:50
To: [hidden email]
Subject: logistic regression error: less than two nonmissing values

Befuddled again,

Maybe I'm just overtired but I've looked at this every which way. My DV
has exactly 2 values. I can see that in the data, and frequencies,
crosstabs all list the two values as well. But when I try to run a
logistic regression I get the error message that there are less than two
nonmissing values.

I am trying to predict the number of emergency dept visits by whether
individuals have ever had their personal health information accessed
through a web based portal. I'd ideally like to be able to use a poison
regression, predicting total number of ED visits with whether their info
was accessed through the portal or not, but that doesn't seem to work
either. That also seems to include only one level of my dichotomous
variable. So I'm trying to predict the dichotomous variable, accessed
through portal or not (1,0 respectively) with number of ED visits.

Why is this seemingly illogical error message occurring?

Thanks
matt

Matthew Pirritano, Ph.D.
Research Analyst IV
Medical Services Initiative (MSI)
Orange County Health Care Agency
[phone]

=====================
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
No virus found in this incoming message.
Checked by AVG - www.avg.com
Version: 8.5.392 / Virus Database: 270.13.21/2252 - Release Date: 07/21/09
05:58:00

=====================
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

[Jason Ferris-3]

To check what Hector (and Gene) is suggesting run a frequency or
cross-tabulation of your DV with the filtering that each of the IV's do
not equal missing (sysmiss or user missing).  This should result in only
one of your DV categories being displayed.

Jason

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Hector Maletta
Sent: Wednesday, 22 July 2009 6:17 AM
To: [hidden email]
Subject: Re: logistic regression error: less than two nonmissing values

Either one of the values of your DV has been declared as missing, or
(which
is the most probable case) all valid cases FOR THE ANALYSIS have the
same
value in the DV. This may arise because any case having a missing value
in
ANY (dependent or independent) variable is discarded from the analysis.
Some
cases may be discarded because they have a missing value in X, some
because
they have a missing value in Z, and so on. It is just possible that each
and
every case that finally made it into the analysis have the same value in
the
dependent variable Y.
Hector

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Pirritano, Matthew
Sent: 21 July 2009 12:50
To: [hidden email]
Subject: logistic regression error: less than two nonmissing values

Befuddled again,

Maybe I'm just overtired but I've looked at this every which way. My DV
has exactly 2 values. I can see that in the data, and frequencies,
crosstabs all list the two values as well. But when I try to run a
logistic regression I get the error message that there are less than two
nonmissing values.

I am trying to predict the number of emergency dept visits by whether
individuals have ever had their personal health information accessed
through a web based portal. I'd ideally like to be able to use a poison
regression, predicting total number of ED visits with whether their info
was accessed through the portal or not, but that doesn't seem to work
either. That also seems to include only one level of my dichotomous
variable. So I'm trying to predict the dichotomous variable, accessed
through portal or not (1,0 respectively) with number of ED visits.

Why is this seemingly illogical error message occurring?

Thanks
matt

Matthew Pirritano, Ph.D.
Research Analyst IV
Medical Services Initiative (MSI)
Orange County Health Care Agency
[phone]

=====================
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
No virus found in this incoming message.
Checked by AVG - www.avg.com
Version: 8.5.392 / Virus Database: 270.13.21/2252 - Release Date:
07/21/09
05:58:00

=====================
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

------------------------------------------
DISCLAIMER: This message (including any attachments) is intended solely for the addressee(s) named and may contain confidential or privileged information.
If you are not the intended recipient, please delete it and notify the sender.
Views expressed in this message are those of the individual sender,and are not necessarily the views of the Turning Point Alcohol and Drug Centre Inc (ABN: 17 302 055 629).

<a href="http://www.turningpoint.org.au">Turning Point Alcohol and Drug Centre Inc</a>

Although this message and any attachments have been scanned for viruses by 'Trend Micro InterScan' at the time of sending,
you are advised to rescan on receipt.

The whole or parts of this email may be subject to copyright of Turning Point Alcohol and Drug Centre Inc (ABN: 17 302 055 629), and/or third parties.
You can only re-transmit, distribute or use the material if you are authorised to do so.

Please consider the environment before printing this email or attachments.

=====================
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

[Matthew Pirritano]

Thanks folks,

I found the problem. There was some missing data where there shouldn't have been. Back to the drawing board!

thanks
matt

 

Matthew Pirritano, Ph.D.
Email: [hidden email]

---

From: Jason Ferris <[hidden email]>
To: [hidden email]
Sent: Tuesday, July 21, 2009 4:38:32 PM
Subject: Re: logistic regression error: less than two nonmissing values

To check what Hector (and Gene) is suggesting run a frequency or
cross-tabulation of your DV with the filtering that each of the IV's do
not equal missing (sysmiss or user missing).  This should result in only
one of your DV categories being displayed.

Jason

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Hector Maletta
Sent: Wednesday, 22 July 2009 6:17 AM
To: [hidden email]
Subject: Re: logistic regression error: less than two nonmissing values

Either one of the values of your DV has been declared as missing, or
(which
is the most probable case) all valid cases FOR THE ANALYSIS have the
same
value in the DV. This may arise because any case having a missing value
in
ANY (dependent or independent) variable is discarded from the analysis.
Some
cases may be discarded because they have a missing value in X, some
because
they have a missing value in Z, and so on. It is just possible that each
and
every case that finally made it into the analysis have the same value in
the
dependent variable Y.
Hector

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Pirritano, Matthew
Sent: 21 July 2009 12:50
To: [hidden email]
Subject: logistic regression error: less than two nonmissing values

Befuddled again,

Maybe I'm just overtired but I've looked at this every which way. My DV
has exactly 2 values. I can see that in the data, and frequencies,
crosstabs all list the two values as well. But when I try to run a
logistic regression I get the error message that there are less than two
nonmissing values.

I am trying to predict the number of emergency dept visits by whether
individuals have ever had their personal health information accessed
through a web based portal. I'd ideally like to be able to use a poison
regression, predicting total number of ED visits with whether their info
was accessed through the portal or not, but that doesn't seem to work
either. That also seems to include only one level of my dichotomous
variable. So I'm trying to predict the dichotomous variable, accessed
through portal or not (1,0 respectively) with number of ED visits.

Why is this seemingly illogical error message occurring?

Thanks
matt

Matthew Pirritano, Ph.D.
Research Analyst IV
Medical Services Initiative (MSI)
Orange County Health Care Agency
[phone]

=====================
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
No virus found in this incoming message.
Checked by AVG - www.avg.com
Version: 8.5.392 / Virus Database: 270.13.21/2252 - Release Date:
07/21/09
05:58:00

=====================
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

------------------------------------------
DISCLAIMER: This message (including any attachments) is intended solely for the addressee(s) named and may contain confidential or privileged information.
If you are not the intended recipient, please delete it and notify the sender.
Views expressed in this message are those of the individual sender,and are not necessarily the views of the Turning Point Alcohol and Drug Centre Inc (ABN: 17 302 055 629).

Turning'>http://www.turningpoint.org.au">Turning Point Alcohol and Drug Centre Inc

Although this message and any attachments have been scanned for viruses by 'Trend Micro InterScan' at the time of sending,
you are advised to rescan on receipt.

The whole or parts of this email may be subject to copyright of Turning Point Alcohol and Drug Centre Inc (ABN: 17 302 055 629), and/or third parties.
You can only re-transmit, distribute or use the material if you are authorised to do so.

Please consider the environment before printing this email or attachments.

=====================
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

[Richard Ristow]

At 09:53 AM 7/21/2009, Dave Taber wrote:

>unfortunately, the CACHE solution didn't work for me, however,  I
>tried saving the excel file as an excel 2003 type and the problem
>went away -- so it seems as though the problem only happens with
>excel 2007 files

Interesting. YOu're running 17 under Vista, which is about as current
a version as they come. If you can document different behavior for
Excel 2003 files and Excel 2007 files, I think it's worth reporting
to SPSS support. (And post, as well. Maybe there'll be somebody with
Excel 2003, Excel 2007, and a beta version of SPSS 18, who can try it.)

At 06:15 PM 7/16/2009, Bud Bliss wrote:
>I have similar problems when reading from Excel 2007 files in other
>apps on XP.

I'd never heard of a file that wouldn't let itself be closed. In
fact, I'd call it logically impossible. But mere impossibility rarely
stops a bug. :-P If you can document the behavior, perhaps it is some
quirk or Excel 2007, though I can't imagine what the quirk would be.

As I say, CACHE *should* work. Or, the effect of CACHE can be
duplicated pretty well with logic using XSAVE. But if you're
interested in investigating, as opposed to just finding a
work-around, see what happens if you open the excel 2007 file, save
it, and then read something else entirely, so there's no open dataset
depending on the Excel file. Then try opening the Excel file in Excel.

-Onward, with best of luck,
  Richard

=====================
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

[Eric Langston]

I am SO glad that someone else was having that issue!  I had gone back and
forth with SPSS Tech Support only to have it filed as "unreplicatable " and
dismissed.  I think that because SPSS itself wasn't giving me the error,
Windows / Excel was -- even though it was SPSS causing the error -- that it
was dismissed without resolution.  Here's what my experience was though ...

It seems to be a problem with SPSS communicating with Windows and Excel.
For example, I would open an Excel file in SPSS (SPSS would tell Windows it
was using the file), save the file as an SPSS file and then close the
dataset, but not close SPSS all together (I'd keep a blank dataset open for
instance).  Theoretically, SPSS should tell Windows and it's done with the
Excel file.  However, there have been many times where I would open an Excel
file and if I tried to save it, then Windows would give me a file contention
error.  If I saved it with a different file new -- no problem.  If I saved
it as an Excel 2003 file (changing the extension rather than the file name)
-- no problem.

So, my workaround has been to restart SPSS after I read in a bunch of Excel
data if I also need to go back to the original Excel files and edit the data
at the source again.


I originally found this error because after I imported an Excel file, I
noticed that one of my records had an error in a date field, which messed up
the entire variable in SPSS -- everything was saved as a string, rather than
as a date. When I went back to Excel to correct the error and re-save the
file, Windows / Excel gave me the file contention error.


I hope that helps.  I remember I was exceptionally frustrated as the problem
happened to me because it appears to be a new problem and hard to stumble
upon.

Cheers,
-Eric

=====================
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

[mpirritano]

I've seen this.

The spssengine.exe is still running.

Hit ctl-alt-del and go into the task manager and kill the spssengine.

matt

Matthew Pirritano, Ph.D.
Research Analyst IV
Medical Services Initiative (MSI)
Orange County Health Care Agency
[phone]
-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Eric Langston
Sent: Wednesday, July 22, 2009 10:52 AM
To: [hidden email]
Subject: Re: excel file issue

I am SO glad that someone else was having that issue!  I had gone back
and
forth with SPSS Tech Support only to have it filed as "unreplicatable "
and
dismissed.  I think that because SPSS itself wasn't giving me the error,
Windows / Excel was -- even though it was SPSS causing the error -- that
it
was dismissed without resolution.  Here's what my experience was though
...

It seems to be a problem with SPSS communicating with Windows and Excel.
For example, I would open an Excel file in SPSS (SPSS would tell Windows
it
was using the file), save the file as an SPSS file and then close the
dataset, but not close SPSS all together (I'd keep a blank dataset open
for
instance).  Theoretically, SPSS should tell Windows and it's done with
the
Excel file.  However, there have been many times where I would open an
Excel
file and if I tried to save it, then Windows would give me a file
contention
error.  If I saved it with a different file new -- no problem.  If I
saved
it as an Excel 2003 file (changing the extension rather than the file
name)
-- no problem.

So, my workaround has been to restart SPSS after I read in a bunch of
Excel
data if I also need to go back to the original Excel files and edit the
data
at the source again.


I originally found this error because after I imported an Excel file, I
noticed that one of my records had an error in a date field, which
messed up
the entire variable in SPSS -- everything was saved as a string, rather
than
as a date. When I went back to Excel to correct the error and re-save
the
file, Windows / Excel gave me the file contention error.


I hope that helps.  I remember I was exceptionally frustrated as the
problem
happened to me because it appears to be a new problem and hard to
stumble
upon.

Cheers,
-Eric

=====================
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

=====================
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

[Dave Taber]

i'm passing this on to spss support but for those who are interested i did the most basic simple test possible

i created 2 small excel files in excel 2007 - i saved one as am excel 2003 file and the other as excel 2007 file
i grabbed both files with spss syntax and then ran a few more commands

i then tried to make a change to each excel file -- i was able to save my change for the 2003 file but could not save my change to the 2007 file - i received an error that said i couldn't save due to sharing violations -- when i closed spss i was able to change and save the 2007 file

i did this exercise immediately after rebooting my system -- i'm using Vista and v17 on a high end PC

test syntax below.

DEFINE !path() 'file path here' !ENDDEFINE.

GET DATA /TYPE=XLSX
  /FILE= !path + 'spss_2007.xlsx'
  /SHEET=name 'spss_2007'
  /CELLRANGE=full
  /READNAMES=on
  /ASSUMEDSTRWIDTH=32767.
SAVE OUTFILE=!path + 'file_07.sav'
  /COMPRESSED.



GET DATA /TYPE=XLS
  /FILE=!path + 'spss_2003.xls'
  /SHEET=name 'spss_2003'
  /CELLRANGE=full
  /READNAMES=on
  /ASSUMEDSTRWIDTH=32767.
SAVE OUTFILE=!path + 'file_03.sav'
  /COMPRESSED.


add files
/FILE=!path + 'file_03.sav'
/FILE=!path + 'file_07.sav'.
SAVE OUTFILE=!path + 'BOTH_01.sav'
  /COMPRESSED.

FREQUENCIES VARIABLES=Team
  /ORDER=ANALYSIS.


 * excel files are 2 column 10 rows as follows:
ID    Team
1    Giants
2    Dodgers
3    Yankees
4    Mets
5    Braves
6    Red Sox
7    Cardinals
8    Cubs
9    Phillies
10    Tigers

On Wed, Jul 22, 2009 at 1:06 PM, Richard Ristow <[hidden email]> wrote:

At 09:53 AM 7/21/2009, Dave Taber wrote:

unfortunately, the CACHE solution didn't work for me, however,  I tried saving the excel file as an excel 2003 type and the problem went away -- so it seems as though the problem only happens with excel 2007 files

Interesting. YOu're running 17 under Vista, which is about as current a version as they come. If you can document different behavior for Excel 2003 files and Excel 2007 files, I think it's worth reporting to SPSS support. (And post, as well. Maybe there'll be somebody with Excel 2003, Excel 2007, and a beta version of SPSS 18, who can try it.)

At 06:15 PM 7/16/2009, Bud Bliss wrote:

I have similar problems when reading from Excel 2007 files in other apps on XP.

I'd never heard of a file that wouldn't let itself be closed. In fact, I'd call it logically impossible. But mere impossibility rarely stops a bug. :-P If you can document the behavior, perhaps it is some quirk or Excel 2007, though I can't imagine what the quirk would be.

As I say, CACHE *should* work. Or, the effect of CACHE can be duplicated pretty well with logic using XSAVE. But if you're interested in investigating, as opposed to just finding a work-around, see what happens if you open the excel 2007 file, save it, and then read something else entirely, so there's no open dataset depending on the Excel file. Then try opening the Excel file in Excel.

-Onward, with best of luck,
 Richard

[Bud Bliss]

This may not be an issue with SPSS/PASW ... the problem I typically see is with Excel files used in ESRI's ArcMAP: once I've opened a map that calls data from an Excel spreadsheet, I can't open that spreadsheet in Excel until I've actually closed ArcMAP, just opening a different map that doesn't use that spreadsheet is not sufficient, I need to completely close ArcMAP before it will release the file. Whether that's related to the SPSS issue, I don't know. This is with Office 2007 on XP.

Meredith (Bud) Bliss
Crime Analyst - Beaverton Police Dept.
4755 SW Griffith Drive
Beaverton, OR 97005
503-526-2294 - Fax: [phone]
__________________________________________
From: Dave Taber [mailto:[hidden email]]
Sent: Thursday, July 23, 2009 9:43 PM
To: Richard Ristow
Cc: [hidden email]; Bud Bliss
Subject: Re: excel file issue

i'm passing this on to spss support but for those who are interested i did the most basic simple test possible

i created 2 small excel files in excel 2007 - i saved one as am excel 2003 file and the other as excel 2007 file
i grabbed both files with spss syntax and then ran a few more commands

i then tried to make a change to each excel file -- i was able to save my change for the 2003 file but could not save my change to the 2007 file - i received an error that said i couldn't save due to sharing violations -- when i closed spss i was able to change and save the 2007 file

i did this exercise immediately after rebooting my system -- i'm using Vista and v17 on a high end PC

test syntax below.

DEFINE !path() 'file path here' !ENDDEFINE.

GET DATA /TYPE=XLSX
  /FILE= !path + 'spss_2007.xlsx'
  /SHEET=name 'spss_2007'
  /CELLRANGE=full
  /READNAMES=on
  /ASSUMEDSTRWIDTH=32767.
SAVE OUTFILE=!path + 'file_07.sav'
  /COMPRESSED.



GET DATA /TYPE=XLS
  /FILE=!path + 'spss_2003.xls'
  /SHEET=name 'spss_2003'
  /CELLRANGE=full
  /READNAMES=on
  /ASSUMEDSTRWIDTH=32767.
SAVE OUTFILE=!path + 'file_03.sav'
  /COMPRESSED.


add files
/FILE=!path + 'file_03.sav'
/FILE=!path + 'file_07.sav'.
SAVE OUTFILE=!path + 'BOTH_01.sav'
  /COMPRESSED.

FREQUENCIES VARIABLES=Team
  /ORDER=ANALYSIS.


 * excel files are 2 column 10 rows as follows:
ID    Team
1    Giants
2    Dodgers
3    Yankees
4    Mets
5    Braves
6    Red Sox
7    Cardinals
8    Cubs
9    Phillies
10    Tigers



On Wed, Jul 22, 2009 at 1:06 PM, Richard Ristow <[hidden email]> wrote:
At 09:53 AM 7/21/2009, Dave Taber wrote:
unfortunately, the CACHE solution didn't work for me, however,  I tried saving the excel file as an excel 2003 type and the problem went away -- so it seems as though the problem only happens with excel 2007 files

Interesting. YOu're running 17 under Vista, which is about as current a version as they come. If you can document different behavior for Excel 2003 files and Excel 2007 files, I think it's worth reporting to SPSS support. (And post, as well. Maybe there'll be somebody with Excel 2003, Excel 2007, and a beta version of SPSS 18, who can try it.)


At 06:15 PM 7/16/2009, Bud Bliss wrote:
I have similar problems when reading from Excel 2007 files in other apps on XP.

I'd never heard of a file that wouldn't let itself be closed. In fact, I'd call it logically impossible. But mere impossibility rarely stops a bug. :-P If you can document the behavior, perhaps it is some quirk or Excel 2007, though I can't imagine what the quirk would be.

As I say, CACHE *should* work. Or, the effect of CACHE can be duplicated pretty well with logic using XSAVE. But if you're interested in investigating, as opposed to just finding a work-around, see what happens if you open the excel 2007 file, save it, and then read something else entirely, so there's no open dataset depending on the Excel file. Then try opening the Excel file in Excel.

-Onward, with best of luck,
 Richard

======================================================================
PUBLIC RECORDS LAW DISCLOSURE

This e-mail is a public record of the City of Beaverton and is subject to public disclosure unless exempt from disclosure under Oregon Public Records Law.  This email is subject to the State Retention Schedule.

=====================
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

[Dave Taber]

for those who are interested...

from spss support:

"This is a defect in PASW 17.0.2 specifically with just Excel 2007 as you have seen. The workaround is to use Excel 2003 or close PASW before editing your Excel 2007 file. I apologize for the inconvenience this presents. "

FYI - it has been my experience that you can still use excel 2007 if you save the files as a 2003 file

dave

On Jul 28, 2009, at 4:59 PM, Bud Bliss wrote:

This may not be an issue with SPSS/PASW ... the problem I typically see is with Excel files used in ESRI's ArcMAP: once I've opened a map that calls data from an Excel spreadsheet, I can't open that spreadsheet in Excel until I've actually closed ArcMAP, just opening a different map that doesn't use that spreadsheet is not sufficient, I need to completely close ArcMAP before it will release the file. Whether that's related to the SPSS issue, I don't know. This is with Office 2007 on XP.

Meredith (Bud) Bliss

Crime Analyst - Beaverton Police Dept.

4755 SW Griffith Drive

Beaverton, OR 97005

503-526-2294 - Fax: [phone] 

__________________________________________

From: Dave Taber [[hidden email]] 

Sent: Thursday, July 23, 2009 9:43 PM

To: Richard Ristow

Cc: [hidden email]; Bud Bliss

Subject: Re: excel file issue

i'm passing this on to spss support but for those who are interested i did the most basic simple test possible

i created 2 small excel files in excel 2007 - i saved one as am excel 2003 file and the other as excel 2007 file

i grabbed both files with spss syntax and then ran a few more commands

i then tried to make a change to each excel file -- i was able to save my change for the 2003 file but could not save my change to the 2007 file - i received an error that said i couldn't save due to sharing violations -- when i closed spss i was able to change and save the 2007 file

i did this exercise immediately after rebooting my system -- i'm using Vista and v17 on a high end PC

test syntax below.

DEFINE !path() 'file path here' !ENDDEFINE.

GET DATA /TYPE=XLSX 

  /FILE= !path + 'spss_2007.xlsx' 

  /SHEET=name 'spss_2007' 

  /CELLRANGE=full 

  /READNAMES=on 

  /ASSUMEDSTRWIDTH=32767.

SAVE OUTFILE=!path + 'file_07.sav' 

  /COMPRESSED.

GET DATA /TYPE=XLS 

  /FILE=!path + 'spss_2003.xls' 

  /SHEET=name 'spss_2003' 

  /CELLRANGE=full 

  /READNAMES=on 

  /ASSUMEDSTRWIDTH=32767.

SAVE OUTFILE=!path + 'file_03.sav' 

  /COMPRESSED.

add files

/FILE=!path + 'file_03.sav' 

/FILE=!path + 'file_07.sav'.

SAVE OUTFILE=!path + 'BOTH_01.sav' 

  /COMPRESSED.

FREQUENCIES VARIABLES=Team 

  /ORDER=ANALYSIS.

 * excel files are 2 column 10 rows as follows:

ID    Team

1    Giants

2    Dodgers

3    Yankees

4    Mets

5    Braves

6    Red Sox

7    Cardinals

8    Cubs

9    Phillies

10    Tigers

On Wed, Jul 22, 2009 at 1:06 PM, Richard Ristow <[hidden email]> wrote:

At 09:53 AM 7/21/2009, Dave Taber wrote:

unfortunately, the CACHE solution didn't work for me, however,  I tried saving the excel file as an excel 2003 type and the problem went away -- so it seems as though the problem only happens with excel 2007 files

Interesting. YOu're running 17 under Vista, which is about as current a version as they come. If you can document different behavior for Excel 2003 files and Excel 2007 files, I think it's worth reporting to SPSS support. (And post, as well. Maybe there'll be somebody with Excel 2003, Excel 2007, and a beta version of SPSS 18, who can try it.)

At 06:15 PM 7/16/2009, Bud Bliss wrote:

I have similar problems when reading from Excel 2007 files in other apps on XP.

I'd never heard of a file that wouldn't let itself be closed. In fact, I'd call it logically impossible. But mere impossibility rarely stops a bug. :-P If you can document the behavior, perhaps it is some quirk or Excel 2007, though I can't imagine what the quirk would be.

As I say, CACHE *should* work. Or, the effect of CACHE can be duplicated pretty well with logic using XSAVE. But if you're interested in investigating, as opposed to just finding a work-around, see what happens if you open the excel 2007 file, save it, and then read something else entirely, so there's no open dataset depending on the Excel file. Then try opening the Excel file in Excel.

-Onward, with best of luck,

 Richard

======================================================================

PUBLIC RECORDS LAW DISCLOSURE

This e-mail is a public record of the City of Beaverton and is subject to public disclosure unless exempt from disclosure under Oregon Public Records Law.  This email is subject to the State Retention Schedule.

Dave Taber

novaQuant

listening - thinking - solving

[hidden email]

914.238.1310

www.novaQuant.com