SPSSX Discussion

Paring/Trimming a Dataset

Classic

List

Threaded

19 messages Options

J.D. Haltigan

Paring/Trimming a Dataset

I have a very high dimensional dataset with many variables (e.g., X1-X200). I want to created a pared down file from this with only a subset of those variables (e.g., X1, X30-X45) which correspond to the ID variable and then a particular measure. I know I can do this using the /DROP and /SAVE commands but I'm having a bit of trouble getting the syntax precisely right. If not too much trouble would someone mind providing an example of the syntax I need to use? Many thanks in advance.

Sent from the SPSSX Discussion mailing list archive at Nabble.com.
===================== 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

Jon Peck

Re: Paring/Trimming a Dataset

It's easier to just use the DELETE VARIABLES command (and then save the file under a different name). It's functionally the same as using doing

ADD FILES /FILE=* /KEEP X1 X30 to X45.

followed by SAVE (again, under a different name).

On Thu, Nov 22, 2018 at 2:08 PM J.D. Haltigan <[hidden email]> wrote:

I have a very high dimensional dataset with many variables (e.g., X1-X200). I want to created a pared down file from this with only a subset of those variables (e.g., X1, X30-X45) which correspond to the ID variable and then a particular measure. I know I can do this using the /DROP and /SAVE commands but I'm having a bit of trouble getting the syntax precisely right. If not too much trouble would someone mind providing an example of the syntax I need to use? Many thanks in advance.
Sent from the SPSSX Discussion mailing list archive at Nabble.com.
===================== 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

Jon K Peck
[hidden email]

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

Bruce Weaver

Re: Paring/Trimming a Dataset

Administrator

In reply to this post by J.D. Haltigan

The FM includes some examples here:

https://www.ibm.com/support/knowledgecenter/en/SSLVMB_25.0.0/statistics_reference_project_ddita/spss/base/syn_get_drop_and_keep_s.html

J.D. Haltigan wrote

> I have a very high dimensional dataset with many variables (e.g.,
> X1-X200). I
> want to created a pared down file from this with only a subset of those
> variables (e.g., X1, X30-X45) which correspond to the ID variable and then
> a
> particular measure. I know I can do this using the /DROP and /SAVE
> commands
> but I'm having a bit of trouble getting the syntax precisely right.If not
> too much trouble would someone mind providing an example of the syntax I
> need to use?Many thanks in advance.
>
>
>
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

> LISTSERV@.UGA

> (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

-----
--
Bruce Weaver
[hidden email]
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.

--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

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

--
Bruce Weaver
bweaver@lakeheadu.ca
http://sites.google.com/a/lakeheadu.ca/bweaver/

"When all else fails, RTFM."

PLEASE NOTE THE FOLLOWING:
1. My Hotmail account is not monitored regularly. To send me an e-mail, please use the address shown above.
2. The SPSSX Discussion forum on Nabble is no longer linked to the SPSSX-L listserv administered by UGA (https://listserv.uga.edu/).

David Marso

Re: Paring/Trimming a Dataset

Administrator

In reply to this post by J.D. Haltigan

Three commands ( SAVE,ADD FILES & MATCH FILES support the *KEEP* subcommand)
read all about it in the FM ( Fine Manual).

"but I'm having a bit of trouble getting the syntax precisely right."

Difficult to imagine that since it is pretty damned near idiot proof.

J.D. Haltigan wrote

> LISTSERV@.UGA

> (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

-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

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

Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in abyssum?"

John F Hall

Re: Paring/Trimming a Dataset

> I have a very high dimensional dataset with many variables (e.g.,
> X1-X200). I want to created a pared down file from this with only a
> subset of those variables (e.g., X1, X30-X45) which correspond to the
> ID variable and then a particular measure. I know I can do this using
> the /DROP and /SAVE commands but I'm having a bit of trouble getting
> the syntax precisely right.If not too much trouble would someone mind
> providing an example of the syntax I need to use?Many thanks in
> advance.
>
>
>
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> To manage your subscription to SPSSX-L, send a message to

> LISTSERV@.UGA

> (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

-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email
me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos
ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in
abyssum?"
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

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

Christopher Stride

Re: Paring/Trimming a Dataset

...however Id strongly advise NOT to hack around at a dataset in the way that John describes... no audit trail for what you've done.

As with all data manipulation, using syntax, whether selfwritten or pasted, is strongly recommended, both for long-term efficiency and documentation reasons.

From: [hidden email]
Sent: ‎23/‎11/‎2018 08:42
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

John F Hall

Re: Paring/Trimming a Dataset

I agree, if you're new to data handling with SPSS, but if you save the file with a different name, you can always go back to the original.

John F Hall MA (Cantab) Dip Ed (Dunelm)

[Retired academic survey researcher]

Email: [hidden email]

Website: Journeys in Survey Research

Course: Survey Analysis Workshop (SPSS)

Research: Subjective Social Indicators (Quality of Life)

From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Chris Stride
Sent: 23 November 2018 10:14
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

From: [hidden email]
Sent: ‎23/‎11/‎2018 08:42
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

> [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

-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email
me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos
ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in
abyssum?"
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

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

Christopher Stride

Re: Paring/Trimming a Dataset

Hmmm, would have to disagree again there I'm afraid.

First - even if you are not new to SPSS, best practice is best practice for a reason. Just being good at SPSS doesn't mean that 6 months down the line you can remember why you did something without proper documentation to jog your memory. Lazily deleting vars using the data editor is never a good idea.

Second, creating multiple files is not a solution, just like trying to put out a fire with petrol isn't. Try to have one master version of your data file, and documented syntax that creates it from the raw data. Creating multiple versions is likely to lead to medium/long-term confusion over which results were obtained from which dataset / doubling of work if new calcs needed.

So J.D.'s initial leaning towards /keep or /drop is 100% correct - suspect someone already answered this but if not, the code is, for example,

get file = '<filepath>'

/keep IDNO X X30 to X45.

exe.

On 23/11/2018 09:36, John F Hall wrote:

I agree, if you're new to data handling with SPSS, but if you save the file with a different name, you can always go back to the original.

John F Hall MA (Cantab) Dip Ed (Dunelm)

[Retired academic survey researcher]

Email:          [hidden email]

Website:     Journeys in Survey Research

Course:       Survey Analysis Workshop (SPSS)

Research:   Subjective Social Indicators (Quality of Life)

From: SPSSX(r) Discussion [hidden email] On Behalf Of Chris Stride
Sent: 23 November 2018 10:14
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

...however Id strongly advise NOT to hack around at a dataset in the way that John describes... no audit trail for what you've done.

As with all data manipulation, using syntax, whether selfwritten or pasted, is strongly recommended, both for long-term efficiency and documentation reasons.

From: [hidden email]
Sent: ‎23/‎11/‎2018 08:42
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

You can also do it manually in the Data Editor: highlight the variable(s)
you wish to delete by dragging the cursor down the row number(s) then, with
the cursor on the highlighted variable(s)
Right click >> Clear
File >> Save As >> new file name.sav

John F Hall MA (Cantab) Dip Ed (Dunelm)
[Retired academic survey researcher]

Email:          [hidden email]
Website:     Journeys in Survey Research
Course:       Survey Analysis Workshop (SPSS)
Research:   Subjective Social Indicators (Quality of Life)

-----Original Message-----
From: SPSSX(r) Discussion <[hidden email]> On Behalf Of David
Marso
Sent: 22 November 2018 23:20
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

Three commands ( SAVE,ADD FILES & MATCH FILES support the *KEEP* subcommand)
read all about it in the FM ( Fine Manual).

"but I'm having a bit of trouble getting the syntax precisely right."

Difficult to imagine that since it is pretty damned near idiot proof.

J.D. Haltigan wrote
> I have a very high dimensional dataset with many variables (e.g.,
> X1-X200). I want to created a pared down file from this with only a
> subset of those variables (e.g., X1, X30-X45) which correspond to the
> ID variable and then a particular measure. I know I can do this using
> the /DROP and /SAVE commands but I'm having a bit of trouble getting
> the syntax precisely right.If not too much trouble would someone mind
> providing an example of the syntax I need to use?Many thanks in
> advance.
>
>
>
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> 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

-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email
me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos
ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in
abyssum?"
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

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

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

-- 

--

Dr Chris Stride, C. Stat, Statistician, Institute of Work Psychology,
University of Sheffield
Telephone: 0114 2223262
Fax: 0114 2727206

"Figure It Out"
Statistical Consultancy and Training Service for Social Scientists

Visit www.figureitout.org.uk for details of my consultancy services, and
forthcoming training courses, which are also available on an in-house basis:

 - Data management using SPSS syntax
 - Advanced SPSS syntax and SPSS macros
 - Testing for Mediation and Moderation using SPSS
 - Multi-level Modelling using SPSS
 - Introduction to Structural Equation Modelling using Mplus
 - Testing for Mediation and Moderation using Mplus
 - Multi-level Modelling using Mplus
 - Latent Growth Curve Modelling using Mplus

spss.giesel@yahoo.de

Re: Paring/Trimming a Dataset

Well, if I just wanna try sth. out with no intention to replicate later Johns approach looks like acceptable and the quickest to me.

Mario Giesel
Munich, Germany

Am Freitag, 23. November 2018, 11:06:53 MEZ hat Dr Chris Stride <[hidden email]> Folgendes geschrieben:

Hmmm, would have to disagree again there I'm afraid.

So J.D.'s initial leaning towards /keep or /drop is 100% correct - suspect someone already answered this but if not, the code is, for example,

get file = '<filepath>'

/keep IDNO X X30 to X45.

exe.

On 23/11/2018 09:36, John F Hall wrote:

I agree, if you're new to data handling with SPSS, but if you save the file with a different name, you can always go back to the original.

John F Hall MA (Cantab) Dip Ed (Dunelm)

[Retired academic survey researcher]

Email: [hidden email]

Website: Journeys in Survey Research

Course: Survey Analysis Workshop (SPSS)

Research: Subjective Social Indicators (Quality of Life)

From: SPSSX(r) Discussion [hidden email] On Behalf Of Chris Stride
Sent: 23 November 2018 10:14
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

From: [hidden email]
Sent: ‎23/‎11/‎2018 08:42
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

You can also do it manually in the Data Editor: highlight the variable(s)
you wish to delete by dragging the cursor down the row number(s) then, with
the cursor on the highlighted variable(s)
Right click >> Clear
File >> Save As >> new file name.sav

John F Hall MA (Cantab) Dip Ed (Dunelm)
[Retired academic survey researcher]

Email:          [hidden email]
Website:     Journeys in Survey Research
Course:       Survey Analysis Workshop (SPSS)
Research:   Subjective Social Indicators (Quality of Life)

-----Original Message-----
From: SPSSX(r) Discussion <[hidden email]> On Behalf Of David
Marso
Sent: 22 November 2018 23:20
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

Three commands ( SAVE,ADD FILES & MATCH FILES support the *KEEP* subcommand)
read all about it in the FM ( Fine Manual).

"but I'm having a bit of trouble getting the syntax precisely right."

Difficult to imagine that since it is pretty damned near idiot proof.

J.D. Haltigan wrote
> I have a very high dimensional dataset with many variables (e.g.,
> X1-X200). I want to created a pared down file from this with only a
> subset of those variables (e.g., X1, X30-X45) which correspond to the
> ID variable and then a particular measure. I know I can do this using
> the /DROP and /SAVE commands but I'm having a bit of trouble getting
> the syntax precisely right.If not too much trouble would someone mind
> providing an example of the syntax I need to use?Many thanks in
> advance.
>
>
>
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> 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

-----
Please reply to the list and not to my personal email.
Those desiring my consulting or training services please feel free to email
me.
---
"Nolite dare sanctum canibus neque mittatis margaritas vestras ante porcos
ne forte conculcent eas pedibus suis."
Cum es damnatorum possederunt porcos iens ut salire off sanguinum cliff in
abyssum?"
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

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

-- 

--

Dr Chris Stride, C. Stat, Statistician, Institute of Work Psychology,
University of Sheffield
Telephone: 0114 2223262
Fax: 0114 2727206

"Figure It Out"
Statistical Consultancy and Training Service for Social Scientists

Visit www.figureitout.org.uk for details of my consultancy services, and
forthcoming training courses, which are also available on an in-house basis:

 - Data management using SPSS syntax
 - Advanced SPSS syntax and SPSS macros
 - Testing for Mediation and Moderation using SPSS
 - Multi-level Modelling using SPSS
 - Introduction to Structural Equation Modelling using Mplus
 - Testing for Mediation and Moderation using Mplus
 - Multi-level Modelling using Mplus
 - Latent Growth Curve Modelling using Mplus

Christopher Stride

Re: Paring/Trimming a Dataset

From: [hidden email]
Sent: ‎23/‎11/‎2018 10:38
To: [hidden email]; [hidden email]
Subject: Re: Paring/Trimming a Dataset

Well, if I just wanna try sth. out with no intention to replicate later Johns approach looks like acceptable and the quickest to me.

Mario Giesel
Munich, Germany

Am Freitag, 23. November 2018, 11:06:53 MEZ hat Dr Chris Stride <[hidden email]> Folgendes geschrieben:

Hmmm, would have to disagree again there I'm afraid.

So J.D.'s initial leaning towards /keep or /drop is 100% correct - suspect someone already answered this but if not, the code is, for example,

get file = '<filepath>'

/keep IDNO X X30 to X45.

exe.

On 23/11/2018 09:36, John F Hall wrote:

I agree, if you're new to data handling with SPSS, but if you save the file with a different name, you can always go back to the original.

John F Hall MA (Cantab) Dip Ed (Dunelm)

[Retired academic survey researcher]

Email: [hidden email]

Website: Journeys in Survey Research

Course: Survey Analysis Workshop (SPSS)

Research: Subjective Social Indicators (Quality of Life)

From: SPSSX(r) Discussion [hidden email] On Behalf Of Chris Stride
Sent: 23 November 2018 10:14
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

From: [hidden email]
Sent: ‎23/‎11/‎2018 08:42
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

-- 

--

Dr Chris Stride, C. Stat, Statistician, Institute of Work Psychology,
University of Sheffield
Telephone: 0114 2223262
Fax: 0114 2727206

"Figure It Out"
Statistical Consultancy and Training Service for Social Scientists

Visit www.figureitout.org.uk for details of my consultancy services, and
forthcoming training courses, which are also available on an in-house basis:

 - Data management using SPSS syntax
 - Advanced SPSS syntax and SPSS macros
 - Testing for Mediation and Moderation using SPSS
 - Multi-level Modelling using SPSS
 - Introduction to Structural Equation Modelling using Mplus
 - Testing for Mediation and Moderation using Mplus
 - Multi-level Modelling using Mplus
 - Latent Growth Curve Modelling using Mplus

Christopher Stride

Re: Paring/Trimming a Dataset

In reply to this post by spss.giesel@yahoo.de

Can't see how going into the data editor and scrolling down multiple lines looking for rows to delete is quicker than typing /drop and then copying-pasting-cutting a set of names?

Also if you just want to try something out, why would you seek to drop a set of variables? And why would you save a copy of that dataset minus those vars?!

From: [hidden email]
Sent: ‎23/‎11/‎2018 10:38
To: [hidden email]; [hidden email]
Subject: Re: Paring/Trimming a Dataset

Well, if I just wanna try sth. out with no intention to replicate later Johns approach looks like acceptable and the quickest to me.

Mario Giesel
Munich, Germany

Am Freitag, 23. November 2018, 11:06:53 MEZ hat Dr Chris Stride <[hidden email]> Folgendes geschrieben:

Hmmm, would have to disagree again there I'm afraid.

So J.D.'s initial leaning towards /keep or /drop is 100% correct - suspect someone already answered this but if not, the code is, for example,

get file = '<filepath>'

/keep IDNO X X30 to X45.

exe.

On 23/11/2018 09:36, John F Hall wrote:

I agree, if you're new to data handling with SPSS, but if you save the file with a different name, you can always go back to the original.

John F Hall MA (Cantab) Dip Ed (Dunelm)

[Retired academic survey researcher]

Email: [hidden email]

Website: Journeys in Survey Research

Course: Survey Analysis Workshop (SPSS)

Research: Subjective Social Indicators (Quality of Life)

From: SPSSX(r) Discussion [hidden email] On Behalf Of Chris Stride
Sent: 23 November 2018 10:14
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

From: [hidden email]
Sent: ‎23/‎11/‎2018 08:42
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

-- 

--

Dr Chris Stride, C. Stat, Statistician, Institute of Work Psychology,
University of Sheffield
Telephone: 0114 2223262
Fax: 0114 2727206

"Figure It Out"
Statistical Consultancy and Training Service for Social Scientists

Visit www.figureitout.org.uk for details of my consultancy services, and
forthcoming training courses, which are also available on an in-house basis:

 - Data management using SPSS syntax
 - Advanced SPSS syntax and SPSS macros
 - Testing for Mediation and Moderation using SPSS
 - Multi-level Modelling using SPSS
 - Introduction to Structural Equation Modelling using Mplus
 - Testing for Mediation and Moderation using Mplus
 - Multi-level Modelling using Mplus
 - Latent Growth Curve Modelling using Mplus

spss.giesel@yahoo.de

Re: Paring/Trimming a Dataset

Well, like today I gave an excerpt of data to a colleague.
He wants to write a tool that can code open answers automatically and needs some example data.

I picked some data using Johns approach. I didn't need to type syntax and it was quickest to delete all numeric variables by hand. Really easy.

Mario Giesel
Munich, Germany

Am Freitag, 23. November 2018, 11:53:17 MEZ hat Chris Stride <[hidden email]> Folgendes geschrieben:

From: [hidden email]
Sent: ‎23/‎11/‎2018 10:38
To: [hidden email]; [hidden email]
Subject: Re: Paring/Trimming a Dataset

Well, if I just wanna try sth. out with no intention to replicate later Johns approach looks like acceptable and the quickest to me.

Mario Giesel
Munich, Germany

Am Freitag, 23. November 2018, 11:06:53 MEZ hat Dr Chris Stride <[hidden email]> Folgendes geschrieben:

Hmmm, would have to disagree again there I'm afraid.

So J.D.'s initial leaning towards /keep or /drop is 100% correct - suspect someone already answered this but if not, the code is, for example,

get file = '<filepath>'

/keep IDNO X X30 to X45.

exe.

On 23/11/2018 09:36, John F Hall wrote:

I agree, if you're new to data handling with SPSS, but if you save the file with a different name, you can always go back to the original.

John F Hall MA (Cantab) Dip Ed (Dunelm)

[Retired academic survey researcher]

Email: [hidden email]

Website: Journeys in Survey Research

Course: Survey Analysis Workshop (SPSS)

Research: Subjective Social Indicators (Quality of Life)

From: SPSSX(r) Discussion [hidden email] On Behalf Of Chris Stride
Sent: 23 November 2018 10:14
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

From: [hidden email]
Sent: ‎23/‎11/‎2018 08:42
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

-- 

--

Dr Chris Stride, C. Stat, Statistician, Institute of Work Psychology,
University of Sheffield
Telephone: 0114 2223262
Fax: 0114 2727206

"Figure It Out"
Statistical Consultancy and Training Service for Social Scientists

Visit www.figureitout.org.uk for details of my consultancy services, and
forthcoming training courses, which are also available on an in-house basis:

 - Data management using SPSS syntax
 - Advanced SPSS syntax and SPSS macros
 - Testing for Mediation and Moderation using SPSS
 - Multi-level Modelling using SPSS
 - Introduction to Structural Equation Modelling using Mplus
 - Testing for Mediation and Moderation using Mplus
 - Multi-level Modelling using Mplus
 - Latent Growth Curve Modelling using Mplus

J.D. Haltigan

Re: Paring/Trimming a Dataset

In reply to this post by Christopher Stride

To be sure: my original post was made with the intent to use ONLY the syntax
option. The dataset is too large to practically and efficiently drag and
delete which is awkward to begin with IMO.

The DELETE command worked perfectly and efficiently.

As for creating a roadmap with the saved syntax, while I strongly encourage
that as well, there was no need to do so in this particular instance since I
was using the pared dataset to compare with another as I had to deal with
some issues relevant to duplicate cases in which data for one measure was
collected in the primary dataset while data for *same* measure was collected
in a different database resulting in two records for a given subset of cases
of which I had to more or less manually create 'one record' based on a
correct calendar date of administration.

In short, the pared dataset needs no syntax to explain how it was created
since it is essentially a single measure from the larger (self-evident) with
no unique *variables* to either. Working with only the pared measure (rather
than the whole dataset) was much more feasible. I did have to drag and copy
about 10 cases' worth of data (in a sample of 2k+) nonetheless and paste
into the 'primary' dataset. I didn't see any other option in this instance
(i.e., with syntax). I do always document such in a working data doc
regardless.

Cheers & Thanks

--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

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

Art Kendall

Re: Paring/Trimming a Dataset

In reply to this post by Christopher Stride

I also strongly advise against the data view approach.
The audit trail is not only needed for quality assurance, it also provides a
way to go back and do the task that you come to intend.

Depending on why you want to pare down, it may not be worthwhile. 200
variables is a not very many unless you have several million cases.

I addition, using the x1 to x45 convention is something the software can
deal with.
However, I strongly emphasize the jelly-ware. (jelly is the soft tissue
between the ears.)

It is good practice to use much more mnemonic names. a good rule of thumb
is
The variables must be complete in variable names, variable, labels,
definition of missing values, and printing format before doing any further
processing.

-----
Art Kendall
Social Research Consultants
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

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

Art Kendall
Social Research Consultants

Art Kendall

Re: Paring/Trimming a Dataset

In reply to this post by Christopher Stride

Memory after 6 months? How about much more short term, such as after a phone
call or bathroom break.

Based on over 40 years experience with SPSS, even short-term memory needs th
support of the syntax approach.

Also, the variable view should be complete before sharing any of it. A
completed variable view help all involved -- the writer, QA reviewers, other
team members, those with whom the data is shared, etc.

-----
Art Kendall
Social Research Consultants
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

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

Art Kendall
Social Research Consultants

Timothy Hennigar-2

Re: Paring/Trimming a Dataset

In reply to this post by Art Kendall

We are the last of an era!

I am inundated all the time with new analytic software promising - remove the burden of writing syntax ... our program automates everything - point and click your way to bliss.

I HATE IT! And, I cant express how wrong I think this is.

I can go back 20 years and re-run (but hopefully not) anything I have ever done - and while my noting of the things may be sparse - the CODE does not lie - what I did is there in black and white
(or whatever color I have chosen currently to view things in). Good luck to anyone without a syntax record going back a month (or a week) to reproduce something at all complicated - and be hoping
you are never called to a courtroom to outline what you did.

If I were teaching a course in analytics - the first words out of my mouth every single day would be - make a record of everything - and I mean everything - you do to a data file from the moment
you touch it until the moment you are finished - AND NEVER ALTER THE ORIGINAL DATA FILE. If you must (and occasionally it happens) do something manually to a file - note at the point
done all the edits in the syntax. Or one day - it will happen - 'Geez - what happened to my data - and how long have I been reporting from this'.

We seem to be moving towards a world (if we are not already there) of no accountablility for anything. Lies, things that do not work and errors - its all OK.

Data without context. I just don't get it - some data is worse that no data at all!

My practice and my advice - never work in the data editor of SPSS - do everything in syntax.

Thanks!

Notice: This e-mail and any attachments may contain confidential and privileged information. If you are not the intended recipient, please notify the sender immediately by return e-mail, do not use the information, delete this e-mail and destroy any copies. Any dissemination or use of this information by a person other than the intended recipient is unauthorized and may be illegal. Email transmissions cannot be guaranteed to be secure or error free. The sender therefore does not accept any liability for errors or omissions in the contents of this message that arise as a result of email transmissions.

-----Original Message-----
From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Art Kendall
Sent: Friday, November 23, 2018 8:32 AM
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

I also strongly advise against the data view approach.
The audit trail is not only needed for quality assurance, it also provides a way to go back and do the task that you come to intend.

Depending on why you want to pare down, it may not be worthwhile. 200 variables is a not very many unless you have several million cases.

I addition, using the x1 to x45 convention is something the software can deal with.
However, I strongly emphasize the jelly-ware. (jelly is the soft tissue between the ears.)

It is good practice to use much more mnemonic names. a good rule of thumb is The variables must be complete in variable names, variable, labels, definition of missing values, and printing format before doing any further processing.

-----
Art Kendall
Social Research Consultants
--
Sent from: http://spssx-discussion.1045642.n5.nabble.com/

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

Christopher Stride

Re: Paring/Trimming a Dataset

I'd actually disagree... not with the sentiment, with which I agree
entirely, but with your prognosis.

I think that what has happened with computing is rather like what
happened with food. We have lived through the fast food/microwave meals
generation, but at least some people are now seeing that that's
unhealthy and proper meals cooked from scratch from fresh ingredients
are better for your health.

Likewise, in the UK at least, after 20-25 years of ever more push
towards using GUI/menus, it's been recognised that programming skills
will be an essential skill in the future, hence the Raspberry Pi
initiative in schools. Unfortunately it will still be a few years before
that filters through properly to Uni level, but over the past 2-3 years
I've noticed a slightly increased willingness to embrace as opposed to
just tolerate simple programming/documentation amongst my social science
postgraduates.

On 23/11/2018 14:04, Timothy Hennigar wrote:

> We are the last of an era!
>
> I am inundated all the time with new analytic software promising - remove the burden of writing syntax ... our program automates everything - point and click your way to bliss.
>
> I HATE IT! And, I cant express how wrong I think this is.
>
> I can go back 20 years and re-run (but hopefully not) anything I have ever done - and while my noting of the things may be sparse - the CODE does not lie - what I did is there in black and white
> (or whatever color I have chosen currently to view things in). Good luck to anyone without a syntax record going back a month (or a week) to reproduce something at all complicated - and be hoping
> you are never called to a courtroom to outline what you did.
>
> If I were teaching a course in analytics - the first words out of my mouth every single day would be - make a record of everything - and I mean everything - you do to a data file from the moment
> you touch it until the moment you are finished - AND NEVER ALTER THE ORIGINAL DATA FILE. If you must (and occasionally it happens) do something manually to a file - note at the point
> done all the edits in the syntax. Or one day - it will happen - 'Geez - what happened to my data - and how long have I been reporting from this'.
>
> We seem to be moving towards a world (if we are not already there) of no accountablility for anything. Lies, things that do not work and errors - its all OK.
>
> Data without context. I just don't get it - some data is worse that no data at all!
>
> My practice and my advice - never work in the data editor of SPSS - do everything in syntax.
>
>
>
>
>
>
> Thanks!
>
> Notice: This e-mail and any attachments may contain confidential and privileged information. If you are not the intended recipient, please notify the sender immediately by return e-mail, do not use the information, delete this e-mail and destroy any copies. Any dissemination or use of this information by a person other than the intended recipient is unauthorized and may be illegal. Email transmissions cannot be guaranteed to be secure or error free. The sender therefore does not accept any liability for errors or omissions in the contents of this message that arise as a result of email transmissions.
>
> -----Original Message-----
> From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Art Kendall
> Sent: Friday, November 23, 2018 8:32 AM
> To: [hidden email]
> Subject: Re: Paring/Trimming a Dataset
>
> I also strongly advise against the data view approach.
> The audit trail is not only needed for quality assurance, it also provides a way to go back and do the task that you come to intend.
>
> Depending on why you want to pare down, it may not be worthwhile. 200 variables is a not very many unless you have several million cases.
>
> I addition, using the x1 to x45 convention is something the software can deal with.
> However, I strongly emphasize the jelly-ware. (jelly is the soft tissue between the ears.)
>
> It is good practice to use much more mnemonic names. a good rule of thumb is The variables must be complete in variable names, variable, labels, definition of missing values, and printing format before doing any further processing.
>
>
>
> -----
> Art Kendall
> Social Research Consultants
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> 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

--

--

Dr Chris Stride, C. Stat, Statistician, Institute of Work Psychology,
University of Sheffield
Telephone: 0114 2223262
Fax: 0114 2727206

"Figure It Out"
Statistical Consultancy and Training Service for Social Scientists

Visit www.figureitout.org.uk for details of my consultancy services, and
forthcoming training courses, which are also available on an in-house basis:

- Data management using SPSS syntax
- Advanced SPSS syntax and SPSS macros
- Testing for Mediation and Moderation using SPSS
- Multi-level Modelling using SPSS
- Introduction to Structural Equation Modelling using Mplus
- Testing for Mediation and Moderation using Mplus
- Multi-level Modelling using Mplus
- Latent Growth Curve Modelling using Mplus

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

John F Hall

Re: Paring/Trimming a Dataset

In reply to this post by spss.giesel@yahoo.de

Absolutely: I'm with Mario on this one. I'm a sociologist/survey researcher, not a statistical (SPSS) Puritan. It's our time (and sometimes a deadline) that matters.

John F Hall MA (Cantab) Dip Ed (Dunelm)

[Retired academic survey researcher]

Email: [hidden email]

Website: Journeys in Survey Research

Course: Survey Analysis Workshop (SPSS)

Research: Subjective Social Indicators (Quality of Life)

From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Mario Giesel
Sent: 23 November 2018 12:14
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

Well, like today I gave an excerpt of data to a colleague.

He wants to write a tool that can code open answers automatically and needs some example data.

I picked some data using Johns approach. I didn't need to type syntax and it was quickest to delete all numeric variables by hand. Really easy.

Mario Giesel

Munich, Germany

Am Freitag, 23. November 2018, 11:53:17 MEZ hat Chris Stride <[hidden email]> Folgendes geschrieben:

From: [hidden email]
Sent: ‎23/‎11/‎2018 10:38
To: [hidden email]; [hidden email]
Subject: Re: Paring/Trimming a Dataset

Well, if I just wanna try sth. out with no intention to replicate later Johns approach looks like acceptable and the quickest to me.

Mario Giesel

Munich, Germany

Am Freitag, 23. November 2018, 11:06:53 MEZ hat Dr Chris Stride <[hidden email]> Folgendes geschrieben:

Hmmm, would have to disagree again there I'm afraid.

So J.D.'s initial leaning towards /keep or /drop is 100% correct - suspect someone already answered this but if not, the code is, for example,

get file = '<filepath>'

/keep IDNO X X30 to X45.

exe.

On 23/11/2018 09:36, John F Hall wrote:

I agree, if you're new to data handling with SPSS, but if you save the file with a different name, you can always go back to the original.

John F Hall MA (Cantab) Dip Ed (Dunelm)

[Retired academic survey researcher]

Email: [hidden email]

Website: Journeys in Survey Research

Course: Survey Analysis Workshop (SPSS)

Research: Subjective Social Indicators (Quality of Life)

From: SPSSX(r) Discussion [hidden email] On Behalf Of Chris Stride
Sent: 23 November 2018 10:14
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

From: [hidden email]
Sent: ‎23/‎11/‎2018 08:42
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

--

--

Dr Chris Stride, C. Stat, Statistician, Institute of Work Psychology,

University of Sheffield

Telephone: 0114 2223262

Fax: 0114 2727206

"Figure It Out"

Statistical Consultancy and Training Service for Social Scientists

Visit www.figureitout.org.uk for details of my consultancy services, and

forthcoming training courses, which are also available on an in-house basis:

 - Data management using SPSS syntax

 - Advanced SPSS syntax and SPSS macros

 - Testing for Mediation and Moderation using SPSS

 - Multi-level Modelling using SPSS

 - Introduction to Structural Equation Modelling using Mplus

 - Testing for Mediation and Moderation using Mplus

 - Multi-level Modelling using Mplus

 - Latent Growth Curve Modelling using Mplus

John F Hall

Re: Paring/Trimming a Dataset

In reply to this post by Christopher Stride

Just for the record, I still have all the original *.sps and *.sav files for my own and others' research going back as far as 1971. These include *.sps files with pre-GUI Fortran type commands such as INPUT FORMAT etc.

Being an SPSS veteran, I have even been able to restore binary and standard files from the 1960s (long before SPSS) for the UK Data Service (which they couldn't) so that others may use them.

I did it my way, and always will.

John F Hall MA (Cantab) Dip Ed (Dunelm)
[Retired academic survey researcher]

Email: [hidden email]
Website: Journeys in Survey Research
Course: Survey Analysis Workshop (SPSS)
Research: Subjective Social Indicators (Quality of Life)

-----Original Message-----
From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Dr Chris Stride
Sent: 23 November 2018 16:25
To: [hidden email]
Subject: Re: Paring/Trimming a Dataset

I'd actually disagree... not with the sentiment, with which I agree entirely, but with your prognosis.

I think that what has happened with computing is rather like what happened with food. We have lived through the fast food/microwave meals generation, but at least some people are now seeing that that's unhealthy and proper meals cooked from scratch from fresh ingredients are better for your health.

Likewise, in the UK at least, after 20-25 years of ever more push towards using GUI/menus, it's been recognised that programming skills will be an essential skill in the future, hence the Raspberry Pi initiative in schools. Unfortunately it will still be a few years before that filters through properly to Uni level, but over the past 2-3 years I've noticed a slightly increased willingness to embrace as opposed to just tolerate simple programming/documentation amongst my social science postgraduates.

On 23/11/2018 14:04, Timothy Hennigar wrote:

> We are the last of an era!
>
> I am inundated all the time with new analytic software promising - remove the burden of writing syntax ... our program automates everything - point and click your way to bliss.
>
> I HATE IT! And, I cant express how wrong I think this is.
>
> I can go back 20 years and re-run (but hopefully not) anything I have
> ever done - and while my noting of the things may be sparse - the CODE
> does not lie - what I did is there in black and white (or whatever color I have chosen currently to view things in). Good luck to anyone without a syntax record going back a month (or a week) to reproduce something at all complicated - and be hoping you are never called to a courtroom to outline what you did.
>
> If I were teaching a course in analytics - the first words out of my
> mouth every single day would be - make a record of everything - and I
> mean everything - you do to a data file from the moment you touch it until the moment you are finished - AND NEVER ALTER THE ORIGINAL DATA FILE. If you must (and occasionally it happens) do something manually to a file - note at the point done all the edits in the syntax. Or one day - it will happen - 'Geez - what happened to my data - and how long have I been reporting from this'.
>
> We seem to be moving towards a world (if we are not already there) of no accountablility for anything. Lies, things that do not work and errors - its all OK.
>
> Data without context. I just don't get it - some data is worse that no data at all!
>
> My practice and my advice - never work in the data editor of SPSS - do everything in syntax.
>
>
>
>
>
>
> Thanks!
>
> Notice: This e-mail and any attachments may contain confidential and privileged information. If you are not the intended recipient, please notify the sender immediately by return e-mail, do not use the information, delete this e-mail and destroy any copies. Any dissemination or use of this information by a person other than the intended recipient is unauthorized and may be illegal. Email transmissions cannot be guaranteed to be secure or error free. The sender therefore does not accept any liability for errors or omissions in the contents of this message that arise as a result of email transmissions.
>
> -----Original Message-----
> From: SPSSX(r) Discussion <[hidden email]> On Behalf Of Art
> Kendall
> Sent: Friday, November 23, 2018 8:32 AM
> To: [hidden email]
> Subject: Re: Paring/Trimming a Dataset
>
> I also strongly advise against the data view approach.
> The audit trail is not only needed for quality assurance, it also provides a way to go back and do the task that you come to intend.
>
> Depending on why you want to pare down, it may not be worthwhile. 200 variables is a not very many unless you have several million cases.
>
> I addition, using the x1 to x45 convention is something the software can deal with.
> However, I strongly emphasize the jelly-ware. (jelly is the soft
> tissue between the ears.)
>
> It is good practice to use much more mnemonic names. a good rule of thumb is The variables must be complete in variable names, variable, labels, definition of missing values, and printing format before doing any further processing.
>
>
>
> -----
> Art Kendall
> Social Research Consultants
> --
> Sent from: http://spssx-discussion.1045642.n5.nabble.com/
>
> =====================
> 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

--

--

Dr Chris Stride, C. Stat, Statistician, Institute of Work Psychology, University of Sheffield
Telephone: 0114 2223262
Fax: 0114 2727206

"Figure It Out"
Statistical Consultancy and Training Service for Social Scientists

Visit www.figureitout.org.uk for details of my consultancy services, and forthcoming training courses, which are also available on an in-house basis:

- Data management using SPSS syntax
- Advanced SPSS syntax and SPSS macros
- Testing for Mediation and Moderation using SPSS
- Multi-level Modelling using SPSS
- Introduction to Structural Equation Modelling using Mplus
- Testing for Mediation and Moderation using Mplus
- Multi-level Modelling using Mplus
- Latent Growth Curve Modelling using Mplus

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