Recoding multiple dichotomous variables into new categorical variable

[jreyespo]

Hi all,

I would like to create a mental health services variable with 4 categories
(no treatment, formal, online, and combined) using the following six
dichotomous variables (No=0; Yes = 1):

Mental Health Professional (MH_2)
Primary Care Doctor (MH_3)
Psychiatrist (MH_4)
Crisis Hotline (MH_5)
Online Program for Mental Health (MH_6)
Mental Health App (MH_7)

Participants were allowed to select multiple choices. No treatment would be
if a participant selected No for all 6 variables. Formal would be if a
participant selected Yes for any variable from MH_2 to MH_5, but No for MH_6
and MH_7. Online would be if participant selected Yes to MH_6 or MH_7, but
No for all MH_2 to MH_5. Combined would be if participant selected Yes to
MH_6 or MH_7 and Yes to any MH_2 to MH_5.

I have some experience recoding and combining categorical variables, but not
one with this many permutations. I know similar topics have been discussed
on this forum in the past, but I have been unable to find a response that
matches my problem. Any suggestions appreciated!

Thanks,
Jazmin




--
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

[jreyespo]

Hi all,

I would like to create a mental health services variable with 4 categories
(no treatment, formal, online, and combined) using the following six
dichotomous variables (No=0; Yes = 1):

Mental Health Professional (MH_2)
Primary Care Doctor (MH_3)
Psychiatrist (MH_4)
Crisis Hotline (MH_5)
Online Program for Mental Health (MH_6)
Mental Health App (MH_7)

Participants were allowed to select multiple choices. No treatment would be
if a participant selected No for all 6 variables. Formal would be if a
participant selected Yes for any variable from MH_2 to MH_5, but No for MH_6
and MH_7. Online would be if participant selected Yes to MH_6 and/or MH_7,
but No for all MH_2 to MH_5. Combined would be if participant selected Yes
to MH_6 and/or MH_7 and Yes to any MH_2 to MH_5.

I have some experience recoding and combining categorical variables, but not
one with this many permutations. I know similar topics have been discussed
on this forum in the past, but I have been unable to find a response that
matches my problem. Any suggestions appreciated!

Thanks,
Jazmin





--
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

[Jeff A]

It's fairly straightforward following a template similar to:

Compute NewVar1 = 1.
If (OldVar1 = 0 and OldVar2 = 1) NewVar1 = 0.

Repeat for the 4 different categories with alterations as necessary.

...although there are many ways to do this and you might run into a problem
with the simple code above if you have missing cases.

Best,

Jeff


-----Original Message-----
From: SPSSX(r) Discussion <[hidden email]> On Behalf Of jreyespo
Sent: Wednesday, November 25, 2020 2:56 PM
To: [hidden email]
Subject: Recoding multiple dichotomous variables into new categorical
variable

Hi all,

I would like to create a mental health services variable with 4 categories
(no treatment, formal, online, and combined) using the following six
dichotomous variables (No=0; Yes = 1):

Mental Health Professional (MH_2)
Primary Care Doctor (MH_3)
Psychiatrist (MH_4)
Crisis Hotline (MH_5)
Online Program for Mental Health (MH_6)
Mental Health App (MH_7)

Participants were allowed to select multiple choices. No treatment would be
if a participant selected No for all 6 variables. Formal would be if a
participant selected Yes for any variable from MH_2 to MH_5, but No for MH_6
and MH_7. Online would be if participant selected Yes to MH_6 or MH_7, but
No for all MH_2 to MH_5. Combined would be if participant selected Yes to
MH_6 or MH_7 and Yes to any MH_2 to MH_5.

I have some experience recoding and combining categorical variables, but not
one with this many permutations. I know similar topics have been discussed
on this forum in the past, but I have been unable to find a response that
matches my problem. Any suggestions appreciated!

Thanks,
Jazmin




--
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


--
This email has been checked for viruses by Avast antivirus software.
https://www.avast.com/antivirus

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

Hi,

try this:

DATA LIST /MH_2 1 MH_3 2 MH_4 3 MH_5 4 MH_6 5 MH_7 6.

BEGIN DATA

000000

000001

000010

000011

000100

000101

000110

000111

001000

001001

001010

001011

001100

001101

001110

001111

010000

010001

010010

010011

010100

010101

010110

010111

011000

011001

011010

011011

011100

011101

011110

011111

100000

100001

100010

100011

100100

100101

100110

100111

101000

101001

101010

101011

101100

101101

101110

101111

110000

110001

110010

110011

110100

110101

110110

110111

111000

111001

111010

111011

111100

111101

111110

111111

END DATA.

DO IF MAX(MH_2,MH_3,MH_4,MH_5,MH_6,MH_7) = 0.

- COMPUTE treatment = 1.

ELSE IF MAX(MH_2,MH_3,MH_4,MH_5) = 1 AND MAX(MH_6,MH_7) = 0.

- COMPUTE treatment = 2.

ELSE IF MAX(MH_2,MH_3,MH_4,MH_5) = 0 AND MAX(MH_6,MH_7) = 1.

- COMPUTE treatment = 3.

ELSE IF MAX(MH_2,MH_3,MH_4,MH_5) = 1 AND MAX(MH_6,MH_7) = 1.

- COMPUTE treatment = 4.

END IF.

VAL LAB treatment

1 'None'

2 'Formal'

3 'Online'

4 'Combined'.

FORMATS treatment (F1).

FREQUENCIES treatment.

Mario Giesel

Munich, Germany

Am Mittwoch, 25. November 2020, 05:56:15 MEZ hat jreyespo <[hidden email]> Folgendes geschrieben:

Hi all,

I would like to create a mental health services variable with 4 categories

(no treatment, formal, online, and combined) using the following six

dichotomous variables (No=0; Yes = 1):

Mental Health Professional (MH_2)

Primary Care Doctor (MH_3)

Psychiatrist (MH_4)

Crisis Hotline (MH_5)

Online Program for Mental Health (MH_6)

Mental Health App (MH_7)

Participants were allowed to select multiple choices. No treatment would be

if a participant selected No for all 6 variables. Formal would be if a

participant selected Yes for any variable from MH_2 to MH_5, but No for MH_6

and MH_7. Online would be if participant selected Yes to MH_6 or MH_7, but

No for all MH_2 to MH_5. Combined would be if participant selected Yes to

MH_6 or MH_7 and Yes to any MH_2 to MH_5.

I have some experience recoding and combining categorical variables, but not

one with this many permutations. I know similar topics have been discussed

on this forum in the past, but I have been unable to find a response that

matches my problem. Any suggestions appreciated!

Thanks,

Jazmin

--

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

[Maguin, Eugene]

I think there's several ways to do this but one way that I like is to make a new string variable that combines responses from the component variables and then recode that into to my new category variables. I like doing this because I can easily see all the combinations of responses. The alternative is crosstabs, which I think would be unmanageable beyond two 'by' keywords. I would do this (and I'm going to assume that each variable is coded only as 0=No or 1=Yes and whatever odd things are in the data values have been fixed already.

String pattern(a6).
Do repeat x=mh_2  to mh_7.
Compute pattern=concat(rtrim(pattern),string(x,f1.0)).
End repeat.

Gene Maguin



-----Original Message-----
From: SPSSX(r) Discussion <[hidden email]> On Behalf Of jreyespo
Sent: Tuesday, November 24, 2020 11:51 PM
To: [hidden email]
Subject: Recoding multiple dichotomous variables into new categorical variable

Hi all,

I would like to create a mental health services variable with 4 categories (no treatment, formal, online, and combined) using the following six dichotomous variables (No=0; Yes = 1):

Mental Health Professional (MH_2)
Primary Care Doctor (MH_3)
Psychiatrist (MH_4)
Crisis Hotline (MH_5)
Online Program for Mental Health (MH_6)
Mental Health App (MH_7)

Participants were allowed to select multiple choices. No treatment would be if a participant selected No for all 6 variables. Formal would be if a participant selected Yes for any variable from MH_2 to MH_5, but No for MH_6 and MH_7. Online would be if participant selected Yes to MH_6 and/or MH_7, but No for all MH_2 to MH_5. Combined would be if participant selected Yes to MH_6 and/or MH_7 and Yes to any MH_2 to MH_5.

I have some experience recoding and combining categorical variables, but not one with this many permutations. I know similar topics have been discussed on this forum in the past, but I have been unable to find a response that matches my problem. Any suggestions appreciated!

Thanks,
Jazmin





--
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

[Bruce Weaver]

Hi Gene.  Jazmin must have posted the same question twice, as there are two
separate threads in the Nabble archive.  Mario G posted a solution in the
other thread.  I'll merge the threads after I finish this post!  

Cheers,
Bruce


Maguin, Eugene wrote

> I think there's several ways to do this but one way that I like is to make
> a new string variable that combines responses from the component variables
> and then recode that into to my new category variables. I like doing this
> because I can easily see all the combinations of responses. The
> alternative is crosstabs, which I think would be unmanageable beyond two
> 'by' keywords. I would do this (and I'm going to assume that each variable
> is coded only as 0=No or 1=Yes and whatever odd things are in the data
> values have been fixed already.
>
> String pattern(a6).
> Do repeat x=mh_2  to mh_7.
> Compute pattern=concat(rtrim(pattern),string(x,f1.0)).
> End repeat.
>
> Gene Maguin

-----
--
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

[Jon Peck]

There are lots of ways to construct such a combination using standard syntax: Gene has shown one here.

But this reminds me of a function accessible through the SPSSINC TRANS extension command that packs up to 53 dummy variables

into a single variable.  It has a companion function that extracts selected dummies.  I wrote this initially to provide a convenient

way of storing a large number of dummies where you might need to exact one or a few at a time.

Here is an example.

data list list/x1 to x10.

begin data.

1 1 1 1 1 1 1 1 1 1

0 0 0 0 0 0 0 0 0 0

1 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 1

1 0 1 0 1 0 1 0 1 0

end data.

dataset name binary.

* pack 10 variables.  The <> in the formula stands for the list in /variables.

spssinc trans result=packed /variables x1 to x10

/formula "extendedTransforms.packDummies([<>])".

* extract two specific bits.  Bits are numbered from 0.

spssinc trans result=bit3 bit4

/formula "extendedTransforms.extractDummies(packed, indexes='2,3')".

It's up to the user to remember which bit is which, but a custom attribute on the packed variable could be helpful here.

This creates an attribute named dict with the displayed text.

variable attribute variables = packed attribute=dict("x1 to x10").

The extendedTransforms module is installed with Statistics; SPSSINC TRANS can be installed from the Extensions > Extension Hub menu.

On Fri, Nov 27, 2020 at 8:26 AM Maguin, Eugene <[hidden email]> wrote:

I think there's several ways to do this but one way that I like is to make a new string variable that combines responses from the component variables and then recode that into to my new category variables. I like doing this because I can easily see all the combinations of responses. The alternative is crosstabs, which I think would be unmanageable beyond two 'by' keywords. I would do this (and I'm going to assume that each variable is coded only as 0=No or 1=Yes and whatever odd things are in the data values have been fixed already.

String pattern(a6).
Do repeat x=mh_2  to mh_7.
Compute pattern=concat(rtrim(pattern),string(x,f1.0)).
End repeat.

Gene Maguin



-----Original Message-----
From: SPSSX(r) Discussion <[hidden email]> On Behalf Of jreyespo
Sent: Tuesday, November 24, 2020 11:51 PM
To: [hidden email]
Subject: Recoding multiple dichotomous variables into new categorical variable

Hi all,

I would like to create a mental health services variable with 4 categories (no treatment, formal, online, and combined) using the following six dichotomous variables (No=0; Yes = 1):

Mental Health Professional (MH_2)
Primary Care Doctor (MH_3)
Psychiatrist (MH_4)
Crisis Hotline (MH_5)
Online Program for Mental Health (MH_6)
Mental Health App (MH_7)

Participants were allowed to select multiple choices. No treatment would be if a participant selected No for all 6 variables. Formal would be if a participant selected Yes for any variable from MH_2 to MH_5, but No for MH_6 and MH_7. Online would be if participant selected Yes to MH_6 and/or MH_7, but No for all MH_2 to MH_5. Combined would be if participant selected Yes to MH_6 and/or MH_7 and Yes to any MH_2 to MH_5.

I have some experience recoding and combining categorical variables, but not one with this many permutations. I know similar topics have been discussed on this forum in the past, but I have been unable to find a response that matches my problem. Any suggestions appreciated!

Thanks,
Jazmin





--
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

[Rich Ulrich]

It looks to me like logic reduces the problem enormously, from 7

dichotomies to two.  Combining two dichtomies is simple. 

I use ANY( )  as a handy tool, though the example with MAX( ) works

the same if there are no stray (error) values in the dataset.

I write this code as several lines, for the ease of reading -

COMMENT  Create four categories, scored from 1 to 4.

Compute  any25 = ANY(1, MH_2 to MH_5).

Compute  any67= ANY(1, MH_6, MH_7).

Compute newvar= 1 + any25 + 2*any67.

--

Rich Ulrich

---

From: SPSSX(r) Discussion <[hidden email]> on behalf of jreyespo <[hidden email]>
Sent: Tuesday, November 24, 2020 11:51 PM
To: [hidden email] <[hidden email]>
Subject: Recoding multiple dichotomous variables into new categorical variable

 

Hi all,

I would like to create a mental health services variable with 4 categories
(no treatment, formal, online, and combined) using the following six
dichotomous variables (No=0; Yes = 1):

Mental Health Professional (MH_2)
Primary Care Doctor (MH_3)
Psychiatrist (MH_4)
Crisis Hotline (MH_5)
Online Program for Mental Health (MH_6)
Mental Health App (MH_7)

Participants were allowed to select multiple choices. No treatment would be
if a participant selected No for all 6 variables. Formal would be if a
participant selected Yes for any variable from MH_2 to MH_5, but No for MH_6
and MH_7. Online would be if participant selected Yes to MH_6 and/or MH_7,
but No for all MH_2 to MH_5. Combined would be if participant selected Yes
to MH_6 and/or MH_7 and Yes to any MH_2 to MH_5.

I have some experience recoding and combining categorical variables, but not
one with this many permutations. I know similar topics have been discussed
on this forum in the past, but I have been unable to find a response that
matches my problem. Any suggestions appreciated!

Thanks,
Jazmin





--
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

[Bruce Weaver]

Very nice, Rich.  I ran your code on the dataset Mario posted earlier, and it
yields the same results as his DO-IF structure.  I hope a text version of
the cross-tab displays okay here:

MG * RU Crosstabulation
Count
                RU Total
                1 2 3 4
MG 1 1 0 0 0 1
        2 0 15 0 0 15
        3 0 0 3 0 3
        4 0 0 0 45 45
Tot 1 15 3 45 64




Rich Ulrich wrote

> It looks to me like logic reduces the problem enormously, from 7
> dichotomies to two.  Combining two dichtomies is simple.
> I use ANY( )  as a handy tool, though the example with MAX( ) works
> the same if there are no stray (error) values in the dataset.
> I write this code as several lines, for the ease of reading -
>
> COMMENT  Create four categories, scored from 1 to 4.
> Compute  any25 = ANY(1, MH_2 to MH_5).
> Compute  any67= ANY(1, MH_6, MH_7).
> Compute newvar= 1 + any25 + 2*any67.
>
> --
> Rich Ulrich

-----
--
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

[MLIves]

This sounded like something I could really use, but when I tried it, there was a warning for both spssinc trans commands:

Warnings

module 'string' has no attribute 'replace'

 

Did I miss something?  Is there documentation for packed and bits?

 

Melissa

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Jon Peck
Sent: Friday, November 27, 2020 12:21 PM
To: [hidden email]
Subject: Re: [SPSSX-L] Recoding multiple dichotomous variables into new categorical variable

 

EXTERNAL EMAIL: This email originated from outside of the organization. Do not click any links or open any attachments unless you trust the sender and know the content is safe.

There are lots of ways to construct such a combination using standard syntax: Gene has shown one here.

 

But this reminds me of a function accessible through the SPSSINC TRANS extension command that packs up to 53 dummy variables

into a single variable.  It has a companion function that extracts selected dummies.  I wrote this initially to provide a convenient

way of storing a large number of dummies where you might need to exact one or a few at a time.

 

Here is an example.

data list list/x1 to x10.

begin data.

1 1 1 1 1 1 1 1 1 1

0 0 0 0 0 0 0 0 0 0

1 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 1

1 0 1 0 1 0 1 0 1 0

end data.

dataset name binary.

 

* pack 10 variables.  The <> in the formula stands for the list in /variables.

spssinc trans result=packed /variables x1 to x10

/formula "extendedTransforms.packDummies([<>])".

 

* extract two specific bits.  Bits are numbered from 0.

spssinc trans result=bit3 bit4

/formula "extendedTransforms.extractDummies(packed, indexes='2,3')".

 

It's up to the user to remember which bit is which, but a custom attribute on the packed variable could be helpful here.

This creates an attribute named dict with the displayed text.

 

variable attribute variables = packed attribute=dict("x1 to x10").

 

The extendedTransforms module is installed with Statistics; SPSSINC TRANS can be installed from the Extensions > Extension Hub menu.

 

 

 

 

On Fri, Nov 27, 2020 at 8:26 AM Maguin, Eugene <[hidden email]> wrote:

I think there's several ways to do this but one way that I like is to make a new string variable that combines responses from the component variables and then recode that into to my new category variables. I like doing this because I can easily see all the combinations of responses. The alternative is crosstabs, which I think would be unmanageable beyond two 'by' keywords. I would do this (and I'm going to assume that each variable is coded only as 0=No or 1=Yes and whatever odd things are in the data values have been fixed already.

String pattern(a6).
Do repeat x=mh_2  to mh_7.
Compute pattern=concat(rtrim(pattern),string(x,f1.0)).
End repeat.

Gene Maguin



-----Original Message-----
From: SPSSX(r) Discussion <[hidden email]> On Behalf Of jreyespo
Sent: Tuesday, November 24, 2020 11:51 PM
To: [hidden email]
Subject: Recoding multiple dichotomous variables into new categorical variable

Hi all,

I would like to create a mental health services variable with 4 categories (no treatment, formal, online, and combined) using the following six dichotomous variables (No=0; Yes = 1):

Mental Health Professional (MH_2)
Primary Care Doctor (MH_3)
Psychiatrist (MH_4)
Crisis Hotline (MH_5)
Online Program for Mental Health (MH_6)
Mental Health App (MH_7)

Participants were allowed to select multiple choices. No treatment would be if a participant selected No for all 6 variables. Formal would be if a participant selected Yes for any variable from MH_2 to MH_5, but No for MH_6 and MH_7. Online would be if participant selected Yes to MH_6 and/or MH_7, but No for all MH_2 to MH_5. Combined would be if participant selected Yes to MH_6 and/or MH_7 and Yes to any MH_2 to MH_5.

I have some experience recoding and combining categorical variables, but not one with this many permutations. I know similar topics have been discussed on this forum in the past, but I have been unable to find a response that matches my problem. Any suggestions appreciated!

Thanks,
Jazmin





--
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

 

--

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

---

This correspondence contains proprietary information some or all of which may be legally privileged; it is for the intended recipient only. If you are not the intended recipient you must not use, disclose, distribute, copy, print, or rely on this correspondence and completely dispose of the correspondence immediately. Please notify the sender if you have received this email in error. NOTE: Messages to or from the State of Connecticut domain may be subject to the Freedom of Information statutes and regulations.

===================== 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]

Please send me ([hidden email])  or post your syntax, and I'll take a look.  

On Mon, Dec 7, 2020 at 9:37 AM Ives, Melissa L <[hidden email]> wrote:

This sounded like something I could really use, but when I tried it, there was a warning for both spssinc trans commands:

Warnings

module 'string' has no attribute 'replace'

 

Did I miss something?  Is there documentation for packed and bits?

 

Melissa

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Jon Peck
Sent: Friday, November 27, 2020 12:21 PM
To: [hidden email]
Subject: Re: [SPSSX-L] Recoding multiple dichotomous variables into new categorical variable

 

EXTERNAL EMAIL: This email originated from outside of the organization. Do not click any links or open any attachments unless you trust the sender and know the content is safe.

There are lots of ways to construct such a combination using standard syntax: Gene has shown one here.

 

But this reminds me of a function accessible through the SPSSINC TRANS extension command that packs up to 53 dummy variables

into a single variable.  It has a companion function that extracts selected dummies.  I wrote this initially to provide a convenient

way of storing a large number of dummies where you might need to exact one or a few at a time.

 

Here is an example.

data list list/x1 to x10.

begin data.

1 1 1 1 1 1 1 1 1 1

0 0 0 0 0 0 0 0 0 0

1 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 1

1 0 1 0 1 0 1 0 1 0

end data.

dataset name binary.

 

* pack 10 variables.  The <> in the formula stands for the list in /variables.

spssinc trans result=packed /variables x1 to x10

/formula "extendedTransforms.packDummies([<>])".

 

* extract two specific bits.  Bits are numbered from 0.

spssinc trans result=bit3 bit4

/formula "extendedTransforms.extractDummies(packed, indexes='2,3')".

 

It's up to the user to remember which bit is which, but a custom attribute on the packed variable could be helpful here.

This creates an attribute named dict with the displayed text.

 

variable attribute variables = packed attribute=dict("x1 to x10").

 

The extendedTransforms module is installed with Statistics; SPSSINC TRANS can be installed from the Extensions > Extension Hub menu.

 

 

 

 

On Fri, Nov 27, 2020 at 8:26 AM Maguin, Eugene <[hidden email]> wrote:

I think there's several ways to do this but one way that I like is to make a new string variable that combines responses from the component variables and then recode that into to my new category variables. I like doing this because I can easily see all the combinations of responses. The alternative is crosstabs, which I think would be unmanageable beyond two 'by' keywords. I would do this (and I'm going to assume that each variable is coded only as 0=No or 1=Yes and whatever odd things are in the data values have been fixed already.

String pattern(a6).
Do repeat x=mh_2  to mh_7.
Compute pattern=concat(rtrim(pattern),string(x,f1.0)).
End repeat.

Gene Maguin



-----Original Message-----
From: SPSSX(r) Discussion <[hidden email]> On Behalf Of jreyespo
Sent: Tuesday, November 24, 2020 11:51 PM
To: [hidden email]
Subject: Recoding multiple dichotomous variables into new categorical variable

Hi all,

I would like to create a mental health services variable with 4 categories (no treatment, formal, online, and combined) using the following six dichotomous variables (No=0; Yes = 1):

Mental Health Professional (MH_2)
Primary Care Doctor (MH_3)
Psychiatrist (MH_4)
Crisis Hotline (MH_5)
Online Program for Mental Health (MH_6)
Mental Health App (MH_7)

Participants were allowed to select multiple choices. No treatment would be if a participant selected No for all 6 variables. Formal would be if a participant selected Yes for any variable from MH_2 to MH_5, but No for MH_6 and MH_7. Online would be if participant selected Yes to MH_6 and/or MH_7, but No for all MH_2 to MH_5. Combined would be if participant selected Yes to MH_6 and/or MH_7 and Yes to any MH_2 to MH_5.

I have some experience recoding and combining categorical variables, but not one with this many permutations. I know similar topics have been discussed on this forum in the past, but I have been unable to find a response that matches my problem. Any suggestions appreciated!

Thanks,
Jazmin





--
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

 

--

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

---

This correspondence contains proprietary information some or all of which may be legally privileged; it is for the intended recipient only. If you are not the intended recipient you must not use, disclose, distribute, copy, print, or rely on this correspondence and completely dispose of the correspondence immediately. Please notify the sender if you have received this email in error. NOTE: Messages to or from the State of Connecticut domain may be subject to the Freedom of Information statutes and regulations.

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

[MLIves]

Hi Jon,

I copied it from the prior message and commented out the 2 lines before the var attribute line.

(I’m using v27.0.1)

 

I did try replacing the <> and the [<>] in various combinations, including dropping the square brackets.

 

M

 

data list list/x1 to x10.

begin data.

1 1 1 1 1 1 1 1 1 1

0 0 0 0 0 0 0 0 0 0

1 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 1

1 0 1 0 1 0 1 0 1 0

end data.

dataset name binary.

 

* pack 10 variables.  The <> in the formula stands for the list in /variables.

spssinc trans result=packed /variables x1 to x10

/formula "extendedTransforms.packDummies([<>])".

 

* extract two specific bits.  Bits are numbered from 0.

spssinc trans result=bit3 bit4

/formula "extendedTransforms.extractDummies(packed, indexes='2,3')".

 

*It's up to the user to remember which bit is which, but a custom attribute on the packed variable could be helpful here.

*This creates an attribute named dict with the displayed text.

 

variable attribute variables = packed attribute=dict("x1 to x10").

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Jon Peck
Sent: Monday, December 7, 2020 12:10 PM
To: [hidden email]
Subject: Re: [SPSSX-L] Recoding multiple dichotomous variables into new categorical variable

 

EXTERNAL EMAIL: This email originated from outside of the organization. Do not click any links or open any attachments unless you trust the sender and know the content is safe.

Please send me ([hidden email])  or post your syntax, and I'll take a look.  

 

On Mon, Dec 7, 2020 at 9:37 AM Ives, Melissa L <[hidden email]> wrote:

This sounded like something I could really use, but when I tried it, there was a warning for both spssinc trans commands:

Warnings

module 'string' has no attribute 'replace'

 

Did I miss something?  Is there documentation for packed and bits?

 

Melissa

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Jon Peck
Sent: Friday, November 27, 2020 12:21 PM
To: [hidden email]
Subject: Re: [SPSSX-L] Recoding multiple dichotomous variables into new categorical variable

 

EXTERNAL EMAIL: This email originated from outside of the organization. Do not click any links or open any attachments unless you trust the sender and know the content is safe.

There are lots of ways to construct such a combination using standard syntax: Gene has shown one here.

 

But this reminds me of a function accessible through the SPSSINC TRANS extension command that packs up to 53 dummy variables

into a single variable.  It has a companion function that extracts selected dummies.  I wrote this initially to provide a convenient

way of storing a large number of dummies where you might need to exact one or a few at a time.

 

Here is an example.

data list list/x1 to x10.

begin data.

1 1 1 1 1 1 1 1 1 1

0 0 0 0 0 0 0 0 0 0

1 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 1

1 0 1 0 1 0 1 0 1 0

end data.

dataset name binary.

 

* pack 10 variables.  The <> in the formula stands for the list in /variables.

spssinc trans result=packed /variables x1 to x10

/formula "extendedTransforms.packDummies([<>])".

 

* extract two specific bits.  Bits are numbered from 0.

spssinc trans result=bit3 bit4

/formula "extendedTransforms.extractDummies(packed, indexes='2,3')".

 

It's up to the user to remember which bit is which, but a custom attribute on the packed variable could be helpful here.

This creates an attribute named dict with the displayed text.

 

variable attribute variables = packed attribute=dict("x1 to x10").

 

The extendedTransforms module is installed with Statistics; SPSSINC TRANS can be installed from the Extensions > Extension Hub menu.

 

 

 

 

On Fri, Nov 27, 2020 at 8:26 AM Maguin, Eugene <[hidden email]> wrote:

I think there's several ways to do this but one way that I like is to make a new string variable that combines responses from the component variables and then recode that into to my new category variables. I like doing this because I can easily see all the combinations of responses. The alternative is crosstabs, which I think would be unmanageable beyond two 'by' keywords. I would do this (and I'm going to assume that each variable is coded only as 0=No or 1=Yes and whatever odd things are in the data values have been fixed already.

String pattern(a6).
Do repeat x=mh_2  to mh_7.
Compute pattern=concat(rtrim(pattern),string(x,f1.0)).
End repeat.

Gene Maguin



-----Original Message-----
From: SPSSX(r) Discussion <[hidden email]> On Behalf Of jreyespo
Sent: Tuesday, November 24, 2020 11:51 PM
To: [hidden email]
Subject: Recoding multiple dichotomous variables into new categorical variable

Hi all,

I would like to create a mental health services variable with 4 categories (no treatment, formal, online, and combined) using the following six dichotomous variables (No=0; Yes = 1):

Mental Health Professional (MH_2)
Primary Care Doctor (MH_3)
Psychiatrist (MH_4)
Crisis Hotline (MH_5)
Online Program for Mental Health (MH_6)
Mental Health App (MH_7)

Participants were allowed to select multiple choices. No treatment would be if a participant selected No for all 6 variables. Formal would be if a participant selected Yes for any variable from MH_2 to MH_5, but No for MH_6 and MH_7. Online would be if participant selected Yes to MH_6 and/or MH_7, but No for all MH_2 to MH_5. Combined would be if participant selected Yes to MH_6 and/or MH_7 and Yes to any MH_2 to MH_5.

I have some experience recoding and combining categorical variables, but not one with this many permutations. I know similar topics have been discussed on this forum in the past, but I have been unable to find a response that matches my problem. Any suggestions appreciated!

Thanks,
Jazmin





--
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

 

--

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

 

---

This correspondence contains proprietary information some or all of which may be legally privileged; it is for the intended recipient only. If you are not the intended recipient you must not use, disclose, distribute, copy, print, or rely on this correspondence and completely dispose of the correspondence immediately. Please notify the sender if you have received this email in error. NOTE: Messages to or from the State of Connecticut domain may be subject to the Freedom of Information statutes and regulations.

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

---

This correspondence contains proprietary information some or all of which may be legally privileged; it is for the intended recipient only. If you are not the intended recipient you must not use, disclose, distribute, copy, print, or rely on this correspondence and completely dispose of the correspondence immediately. Please notify the sender if you have received this email in error. NOTE: Messages to or from the State of Connecticut domain may be subject to the Freedom of Information statutes and regulations.

===================== 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]

Unfortunately, the conversion of this module to Python 3 with V27 was not done correctly.  I have reported this to Development, but I have fixed these problems.

(The packing/unpacking code did not change, but changes were required in other places in order to get the module to load.)

Anyone can download the fixed version from this link.

https://1drv.ms/u/s!AoWcE61g_FAdisFvwuW16bqhQDdNVA?e=mamUvk

Replace the one you have with the attached.  To find out where it lives, run this code.

begin program python3.

import extendedTransforms

print(extendedTransforms)

end program.

On Mon, Dec 7, 2020 at 10:21 AM Ives, Melissa L <[hidden email]> wrote:

Hi Jon,

I copied it from the prior message and commented out the 2 lines before the var attribute line.

(I’m using v27.0.1)

 

I did try replacing the <> and the [<>] in various combinations, including dropping the square brackets.

 

M

 

data list list/x1 to x10.

begin data.

1 1 1 1 1 1 1 1 1 1

0 0 0 0 0 0 0 0 0 0

1 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 1

1 0 1 0 1 0 1 0 1 0

end data.

dataset name binary.

 

* pack 10 variables.  The <> in the formula stands for the list in /variables.

spssinc trans result=packed /variables x1 to x10

/formula "extendedTransforms.packDummies([<>])".

 

* extract two specific bits.  Bits are numbered from 0.

spssinc trans result=bit3 bit4

/formula "extendedTransforms.extractDummies(packed, indexes='2,3')".

 

*It's up to the user to remember which bit is which, but a custom attribute on the packed variable could be helpful here.

*This creates an attribute named dict with the displayed text.

 

variable attribute variables = packed attribute=dict("x1 to x10").

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Jon Peck
Sent: Monday, December 7, 2020 12:10 PM
To: [hidden email]
Subject: Re: [SPSSX-L] Recoding multiple dichotomous variables into new categorical variable

 

EXTERNAL EMAIL: This email originated from outside of the organization. Do not click any links or open any attachments unless you trust the sender and know the content is safe.

Please send me ([hidden email])  or post your syntax, and I'll take a look.  

 

On Mon, Dec 7, 2020 at 9:37 AM Ives, Melissa L <[hidden email]> wrote:

This sounded like something I could really use, but when I tried it, there was a warning for both spssinc trans commands:

Warnings

module 'string' has no attribute 'replace'

 

Did I miss something?  Is there documentation for packed and bits?

 

Melissa

 

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Jon Peck
Sent: Friday, November 27, 2020 12:21 PM
To: [hidden email]
Subject: Re: [SPSSX-L] Recoding multiple dichotomous variables into new categorical variable

 

EXTERNAL EMAIL: This email originated from outside of the organization. Do not click any links or open any attachments unless you trust the sender and know the content is safe.

There are lots of ways to construct such a combination using standard syntax: Gene has shown one here.

 

But this reminds me of a function accessible through the SPSSINC TRANS extension command that packs up to 53 dummy variables

into a single variable.  It has a companion function that extracts selected dummies.  I wrote this initially to provide a convenient

way of storing a large number of dummies where you might need to exact one or a few at a time.

 

Here is an example.

data list list/x1 to x10.

begin data.

1 1 1 1 1 1 1 1 1 1

0 0 0 0 0 0 0 0 0 0

1 0 0 0 0 0 0 0 0 0

0 0 0 0 0 0 0 0 0 1

1 0 1 0 1 0 1 0 1 0

end data.

dataset name binary.

 

* pack 10 variables.  The <> in the formula stands for the list in /variables.

spssinc trans result=packed /variables x1 to x10

/formula "extendedTransforms.packDummies([<>])".

 

* extract two specific bits.  Bits are numbered from 0.

spssinc trans result=bit3 bit4

/formula "extendedTransforms.extractDummies(packed, indexes='2,3')".

 

It's up to the user to remember which bit is which, but a custom attribute on the packed variable could be helpful here.

This creates an attribute named dict with the displayed text.

 

variable attribute variables = packed attribute=dict("x1 to x10").

 

The extendedTransforms module is installed with Statistics; SPSSINC TRANS can be installed from the Extensions > Extension Hub menu.

 

 

 

 

On Fri, Nov 27, 2020 at 8:26 AM Maguin, Eugene <[hidden email]> wrote:

I think there's several ways to do this but one way that I like is to make a new string variable that combines responses from the component variables and then recode that into to my new category variables. I like doing this because I can easily see all the combinations of responses. The alternative is crosstabs, which I think would be unmanageable beyond two 'by' keywords. I would do this (and I'm going to assume that each variable is coded only as 0=No or 1=Yes and whatever odd things are in the data values have been fixed already.

String pattern(a6).
Do repeat x=mh_2  to mh_7.
Compute pattern=concat(rtrim(pattern),string(x,f1.0)).
End repeat.

Gene Maguin



-----Original Message-----
From: SPSSX(r) Discussion <[hidden email]> On Behalf Of jreyespo
Sent: Tuesday, November 24, 2020 11:51 PM
To: [hidden email]
Subject: Recoding multiple dichotomous variables into new categorical variable

Hi all,

I would like to create a mental health services variable with 4 categories (no treatment, formal, online, and combined) using the following six dichotomous variables (No=0; Yes = 1):

Mental Health Professional (MH_2)
Primary Care Doctor (MH_3)
Psychiatrist (MH_4)
Crisis Hotline (MH_5)
Online Program for Mental Health (MH_6)
Mental Health App (MH_7)

Participants were allowed to select multiple choices. No treatment would be if a participant selected No for all 6 variables. Formal would be if a participant selected Yes for any variable from MH_2 to MH_5, but No for MH_6 and MH_7. Online would be if participant selected Yes to MH_6 and/or MH_7, but No for all MH_2 to MH_5. Combined would be if participant selected Yes to MH_6 and/or MH_7 and Yes to any MH_2 to MH_5.

I have some experience recoding and combining categorical variables, but not one with this many permutations. I know similar topics have been discussed on this forum in the past, but I have been unable to find a response that matches my problem. Any suggestions appreciated!

Thanks,
Jazmin





--
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

 

--

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

 

---

This correspondence contains proprietary information some or all of which may be legally privileged; it is for the intended recipient only. If you are not the intended recipient you must not use, disclose, distribute, copy, print, or rely on this correspondence and completely dispose of the correspondence immediately. Please notify the sender if you have received this email in error. NOTE: Messages to or from the State of Connecticut domain may be subject to the Freedom of Information statutes and regulations.

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

---

This correspondence contains proprietary information some or all of which may be legally privileged; it is for the intended recipient only. If you are not the intended recipient you must not use, disclose, distribute, copy, print, or rely on this correspondence and completely dispose of the correspondence immediately. Please notify the sender if you have received this email in error. NOTE: Messages to or from the State of Connecticut domain may be subject to the Freedom of Information statutes and regulations.

--
===================== 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