Syntax for Changing Level of Measurement?

[Dennis Deck]

ViAnn

 

Since you’ve thrown your glove (and given a wink) I’ll bite:

 

Most data files consist of more than a couple variables and many records it can be tricky to inspect the resulting values without resorting to LIST.  In my case, the special conditions I need to check are usually buried in large files (many variables and many records).  I concur that manual inspection using the editor is useful for many purposes, but it seems impractical and inefficient for the selective checking of large files that I need to do.  

 

Used with TEMPORARY and SELECT IF  (or FILTER) to capture just those records of interest and a list of variables of interest, LIST serves as an efficient tool for inspecting both the underlying data and the results of transformations.  Admittedly it is a very limited command, but it serves this particular purpose very well. 

 

Thus, I would argue, LIST is an indispensible tool with real life data sets - especially large complex ones. 

But to be clear, I concur that the data editor is very useful and should be the first tool considered.

 

Dennis Deck

RMC Research

---

From: ViAnn Beadle [mailto:[hidden email]]
Sent: Monday, October 12, 2009 5:14 PM
Subject: Why use LIST (was Syntax for Changing Level of Measurement)

 

People use LIST in sample syntax  on this listserv because the listserv doesn’t like to forward anything but plain text and LIST creates a simple text object. In real life, why would I use LIST? When I want to check my logic on transformations, I just look at the results of my transformations in the data editor. When I want to look at my filter logic, I still look at the data editor because it indicates with a slash whether the case is in the filter.

 

If the LIST enthusiast tried to explain what it does to a spreadsheet enthusiast, the spreadsheet enthusiast would think this is a very strange thing to do.

 

I await vigorous defenses of the use of LIST ;-)

 

[Maguin, Eugene]

ViAnn,

I'm curious as to where this question comes from?

As for myself, I don't know of an alternative to list (excepting print,
which is in some ways is more useful but also more work). For instance,
yesterday I was working with somebody and we needed to find cases meeting a
certain criterion, dig out the questionnaires and then edit the data file.
Either we scramble through the data editor and write down the id numbers or
we print them, along with the variables in question. Perhaps a custom report
could be developed. But why? Admittedly I have used reports only a couple of
times. It's nice but for listing cases. Why?

Gene Maguin

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

[ViAnn Beadle]

OK, please explain what you mean by the SPSS tail is wagging the user dog again?

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of John F Hall
Sent: Tuesday, October 13, 2009 9:38 AM
To: [hidden email]
Subject: Re: Why use LIST (was Syntax for Changing Level of Measurement)

 

Some good replies here from experienced front-line users.  Sounds like the SPSS tail is wagging the user dog again.  Mustn't upset the LISTSERV protocols, must we?   If anyone wants to see my draft tutorial on LIST (and DISPLAY)  just mail me.

----- Original Message -----

From: [hidden email]

To: [hidden email]

Sent: Tuesday, October 13, 2009 3:50 PM

Subject: Re: Why use LIST (was Syntax for Changing Level of Measurement)

 

> Sounds like you need the Data Validation option ;-)

Unfortunately, that option is not included in the base-version :(

Francien

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

[ViAnn Beadle]

I ask this because I never use LIST and see it referred to all the time on
this list. The biggest problem I have with LIST is that it's only useful for
about 9 or so variables at a time and if I see something that is clearly a
typo, I find it much, much easier in the data editor data view and fix it
there anyway.

It seems to me from the replies so far that there are three reasons for
using LIST:

1. Dump dirty data to the output window to check for it.
2. Check transformation logic with a view of a small subset of variables
from a wider file.
3. Provide a narrower view of wide data.

If the data validation option were free, I wonder how many would actually
use it? It's a lot simpler to define rules via it, and find unusual cases.
The rules are automatically saved so that you can run them against new or
changed cases.

I'd suspect that people using a site license might already have access to it
and not know about it. It has got to be easier and much less error prone
that the manual method.

The data editor already provides panning and pinning for wide files, and
variable sets to hide variables not in the set.

If it was much easier to reorder variables, or hide intervening variables
would people use that? (To reorder variables, you can do it with one file
matches, dragging and dropping the variable view, or define a custom
attribute and sort on it in the variable view).

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Gene Maguin
Sent: Tuesday, October 13, 2009 12:21 PM
To: [hidden email]
Subject: Re: Why use LIST (was Syntax for Changing Level of Measurement)

ViAnn,

I'm curious as to where this question comes from?

As for myself, I don't know of an alternative to list (excepting print,
which is in some ways is more useful but also more work). For instance,
yesterday I was working with somebody and we needed to find cases meeting a
certain criterion, dig out the questionnaires and then edit the data file.
Either we scramble through the data editor and write down the id numbers or
we print them, along with the variables in question. Perhaps a custom report
could be developed. But why? Admittedly I have used reports only a couple of
times. It's nice but for listing cases. Why?

Gene Maguin

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

[Art Kendall]

I think that the reason why this particular list does not take attachments is that the listserv (U GA) host either has an old version of the listserv software or has it set to disallow attachments. SPSS does not host this listserv list.


Art

John F Hall wrote:

Some good replies here from experienced front-line users.  Sounds like the SPSS tail is wagging the user dog again.  Mustn't upset the LISTSERV protocols, must we?   If anyone wants to see my draft tutorial on LIST (and DISPLAY)  just mail me.

----- Original Message -----

From: [hidden email]

To: [hidden email]

Sent: Tuesday, October 13, 2009 3:50 PM

Subject: Re: Why use LIST (was Syntax for Changing Level of Measurement)

> Sounds like you need the Data Validation option ;-)

Unfortunately, that option is not included in the base-version :(

Francien

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

[Melissa Ives]

Try SUMMARIZE instead,  with /FORMAT=LIST to list all records or if you only want 10 records listed: /FORMAT=LIST LIMIT=10.

With summarize, you will see the value labels instead of the values, it also doesn't wrap as LIST does so is useful for more records.

With LIST you could try the subcommand /FORMAT=SINGLE to keep it to a single line, but onl variables that fit on 1 line will be displayed.

Melissa

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ViAnn Beadle
Sent: Tuesday, October 13, 2009 3:01 PM
To: [hidden email]
Subject: Re: [SPSSX-L] Why use LIST (was Syntax for Changing Level of Measurement)

I ask this because I never use LIST and see it referred to all the time on this list. The biggest problem I have with LIST is that it's only useful for about 9 or so variables at a time and if I see something that is clearly a typo, I find it much, much easier in the data editor data view and fix it there anyway.

It seems to me from the replies so far that there are three reasons for using LIST:

1. Dump dirty data to the output window to check for it.
2. Check transformation logic with a view of a small subset of variables from a wider file.
3. Provide a narrower view of wide data.

If the data validation option were free, I wonder how many would actually use it? It's a lot simpler to define rules via it, and find unusual cases.
The rules are automatically saved so that you can run them against new or changed cases.

I'd suspect that people using a site license might already have access to it and not know about it. It has got to be easier and much less error prone that the manual method.

The data editor already provides panning and pinning for wide files, and variable sets to hide variables not in the set.

If it was much easier to reorder variables, or hide intervening variables would people use that? (To reorder variables, you can do it with one file matches, dragging and dropping the variable view, or define a custom attribute and sort on it in the variable view).

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Gene Maguin
Sent: Tuesday, October 13, 2009 12:21 PM
To: [hidden email]
Subject: Re: Why use LIST (was Syntax for Changing Level of Measurement)

ViAnn,

I'm curious as to where this question comes from?

As for myself, I don't know of an alternative to list (excepting print, which is in some ways is more useful but also more work). For instance, yesterday I was working with somebody and we needed to find cases meeting a certain criterion, dig out the questionnaires and then edit the data file.
Either we scramble through the data editor and write down the id numbers or we print them, along with the variables in question. Perhaps a custom report could be developed. But why? Admittedly I have used reports only a couple of times. It's nice but for listing cases. Why?

Gene Maguin

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

PRIVILEGED AND CONFIDENTIAL INFORMATION
This transmittal and any attachments may contain PRIVILEGED AND
CONFIDENTIAL information and is intended only for the use of the
addressee. If you are not the designated recipient, or an employee
or agent authorized to deliver such transmittals to the designated
recipient, you are hereby notified that any dissemination,
copying or publication of this transmittal is strictly prohibited. If
you have received this transmittal in error, please notify us
immediately by replying to the sender and delete this copy from your
system. You may also call us at [phone] for assistance.

=====================
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 08:13 PM 10/12/2009, ViAnn Beadle wrote:

>People use LIST in sample syntax  on this listserv because the
>listserv doesn't like to forward anything but plain text and LIST
>creates a simple text object. In real life, why would I use LIST?
>When I want to check my logic on transformations, I just look at the
>results of my transformations in the data editor.

I think you're asking, why use any tool but the data editor, for viewing data?

Data listings are valuable in any context where it isn't easy to open
SPSS. Sample-data postings on the list are indeed an example. Mostly,
I think an in-line list of the data is much easier to use than would
be a .SAV file sent as an attachment: the in-line listing saves the
step of opening SPSS, and it makes it easy to juxtapose the code and
the listing.

You can use LIST to make a copy of the data, or of selected records,
to send to anyone who doesn't have SPSS, or who will be working
without easy access to a computer.

LIST output can be used for arching. You can paste or bind the
printed output into the project notebook, and it won't ever be
unavailable, or change.

And when inspecting data for anomalies, I can read a printed listing
much more quickly than I can the screen; and it can show many more
records at a time. Besides, one can write comments about cases in the
margin of the printed page.

So, that's what I do.

-Peace and good will,
  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