SPSSX Discussion

Compute New Variable - Based on Responses to Other Variables

Classic

List

Threaded

11 messages Options

Chao yawo-2

Compute New Variable - Based on Responses to Other Variables

Hello Listservers,

I have variables:

X (use of anal-condom)

Y (use of oral condom)

Z (use of vaginal condom)

Each as is coded 1=Yes, 2=No, 0,8,9=Missing.

I want to create a new variable that is set to 1 (CONDUSE = 1), if the respondent uses condom for any type of sex, 2, if not

I tried the following command, however but it is giving me an error regarding the ELSE statement.

If i delete the ELSE statement, the command is not giving me the values for case 2 (those who do not use condom for at all.

COMPUTE CONDUSE=0.

IF ((ANALCOND=1) OR (ORALCOND=1) OR (VAGCOND=1)) CONDUSE=1.

IF ((ANALCOND=0) AND (ORALCOND=0) AND (VAGCOND=0)) CONDUSE=0.

IF ((ANALCOND=8) AND (ORALCOND=8) AND (VAGCOND=8)) CONDUSE=8.

IF ((ANALCOND=9) AND (ORALCOND=9) AND (VAGCOND=9)) CONDUSE=9.

IF ELSE CONDUSE=2.

EXECUTE.

I will appreciate some help from you all.

thanks - Cy

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

peter link

Re: Compute New Variable - Based on Responses to Other Variables

IF ELSE is not supported syntax. See DO IF command.

peter

From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Chao yawo
Sent: Tuesday, March 20, 2018 9:49 AM
To: [hidden email]
Subject: [SPSSX-L] Compute New Variable - Based on Responses to Other Variables

Hello Listservers,

I have variables:

X (use of anal-condom)

Y (use of oral condom)

Z (use of vaginal condom)

Each as is coded 1=Yes, 2=No, 0,8,9=Missing.

I want to create a new variable that is set to 1 (CONDUSE = 1), if the respondent uses condom for any type of sex, 2, if not

I tried the following command, however but it is giving me an error regarding the ELSE statement.

If i delete the ELSE statement, the command is not giving me the values for case 2 (those who do not use condom for at all.

COMPUTE CONDUSE=0.

IF ((ANALCOND=1) OR (ORALCOND=1) OR (VAGCOND=1)) CONDUSE=1.

IF ((ANALCOND=0) AND (ORALCOND=0) AND (VAGCOND=0)) CONDUSE=0.

IF ((ANALCOND=8) AND (ORALCOND=8) AND (VAGCOND=8)) CONDUSE=8.

IF ((ANALCOND=9) AND (ORALCOND=9) AND (VAGCOND=9)) CONDUSE=9.

IF ELSE CONDUSE=2.

EXECUTE.

I will appreciate some help from you all.

thanks - Cy

John F Hall

Re: Compute New Variable - Based on Responses to Other Variables

Try something like (untested)

COUNT concount = ANALCOND ORALCOND VAGCOND (1).

IF concount gt 0 conduse = ANALCOND * 100 + ORALCOND*10 + VAGCOND.

FREQ conduse.

Values should range from 001 (vagcon only) to 111 (all three).

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 Peter Link
Sent: 20 March 2018 18:05
To: [hidden email]
Subject: Re: Compute New Variable - Based on Responses to Other Variables

IF ELSE is not supported syntax. See DO IF command.

peter

From: SPSSX(r) Discussion [[hidden email]] On Behalf Of Chao yawo
Sent: Tuesday, March 20, 2018 9:49 AM
To: [hidden email]
Subject: [SPSSX-L] Compute New Variable - Based on Responses to Other Variables

Hello Listservers,

I have variables:

X (use of anal-condom)

Y (use of oral condom)

Z (use of vaginal condom)

Each as is coded 1=Yes, 2=No, 0,8,9=Missing.

I want to create a new variable that is set to 1 (CONDUSE = 1), if the respondent uses condom for any type of sex, 2, if not

I tried the following command, however but it is giving me an error regarding the ELSE statement.

If i delete the ELSE statement, the command is not giving me the values for case 2 (those who do not use condom for at all.

COMPUTE CONDUSE=0.

IF ((ANALCOND=1) OR (ORALCOND=1) OR (VAGCOND=1)) CONDUSE=1.

IF ((ANALCOND=0) AND (ORALCOND=0) AND (VAGCOND=0)) CONDUSE=0.

IF ((ANALCOND=8) AND (ORALCOND=8) AND (VAGCOND=8)) CONDUSE=8.

IF ((ANALCOND=9) AND (ORALCOND=9) AND (VAGCOND=9)) CONDUSE=9.

IF ELSE CONDUSE=2.

EXECUTE.

I will appreciate some help from you all.

thanks - Cy

Rich Ulrich

Re: Compute New Variable - Based on Responses to Other Variables

In reply to this post by Chao yawo-2

I think you have not clearly defined what to do for various combinations of MISSings.

But John Hall's IF imples the useful step,

COMPUTE conduse = ANALCOND * 100 + ORALCOND*10 + VAGCOND.

That result shows you all your values, and you might RECODE to your heart's content.

To rely on IF statements to good effective for a hierarchy of preferred answers,

start by setting your lowest priority answer, and end with the one that over-rules all the others.

IF ANY(0, < three vars > ) conduse= 0.

IF ANY(9, < three vars > ) conduse= 9.

IF ANY(8, < three vars > ) conduse= 8.

IF ANY(2, < three vars > ) conduse= 2.

IF ANY(1, < three vars > ) conduse= 1.

Those could be placed in a DO REPEAT, with testval= 0, 9, 8, 2, 1.

DO REPEAT testval= 0,9,8,2,1.

IF ANY( testval, ANALCOND, ORALCOND, VAGCOND ) conduse= testval.

END REPEAT.

Rich Ulrich

From: SPSSX(r) Discussion <[hidden email]> on behalf of Chao yawo <[hidden email]>
Sent: Tuesday, March 20, 2018 12:49:25 PM
To: [hidden email]
Subject: Compute New Variable - Based on Responses to Other Variables

Hello Listservers,

I have variables:

X (use of anal-condom)

Y (use of oral condom)

Z (use of vaginal condom)

Each as is coded 1=Yes, 2=No, 0,8,9=Missing.

I want to create a new variable that is set to 1 (CONDUSE = 1), if the respondent uses condom for any type of sex, 2, if not

I tried the following command, however but it is giving me an error regarding the ELSE statement.

If i delete the ELSE statement, the command is not giving me the values for case 2 (those who do not use condom for at all.

COMPUTE CONDUSE=0.

IF ((ANALCOND=1) OR (ORALCOND=1) OR (VAGCOND=1)) CONDUSE=1.

IF ((ANALCOND=0) AND (ORALCOND=0) AND (VAGCOND=0)) CONDUSE=0.

IF ((ANALCOND=8) AND (ORALCOND=8) AND (VAGCOND=8)) CONDUSE=8.

IF ((ANALCOND=9) AND (ORALCOND=9) AND (VAGCOND=9)) CONDUSE=9.

IF ELSE CONDUSE=2.

EXECUTE.

I will appreciate some help from you all.

thanks - Cy

David Marso

Re: Compute New Variable - Based on Responses to Other Variables

Administrator

In reply to this post by Chao yawo-2

DO IF (condition).
blah blah blah.
ELSE IF (condition).
........
....
ELSE
....
END IF.

But that does NOT address the utterly piss poor mapping of the 3D space to
your proposed values.

Does {0,8,9} REALLY map to a 2?
Look at a 5x5 cube. That is your 3D space. What goes into what?
Back to the drawing board.

Chao yawo-2 wrote

> Hello Listservers,
>
> I have variables:
>
> X (use of anal-condom)
> Y (use of oral condom)
> Z (use of vaginal condom)
>
> Each as is coded 1=Yes, 2=No, 0,8,9=Missing.
>
> I want to create a new variable that is set to 1 (CONDUSE = 1), if the
> respondent uses condom for any type of sex, 2, if not
>
> I tried the following command, however but it is giving me an error
> regarding the ELSE statement.
> If i delete the ELSE statement, the command is not giving me the values
> for
> case 2 (those who do not use condom for at all.
>
> COMPUTE CONDUSE=0.
> IF ((ANALCOND=1) OR (ORALCOND=1) OR (VAGCOND=1)) CONDUSE=1.
> IF ((ANALCOND=0) AND (ORALCOND=0) AND (VAGCOND=0)) CONDUSE=0.
> IF ((ANALCOND=8) AND (ORALCOND=8) AND (VAGCOND=8)) CONDUSE=8.
> IF ((ANALCOND=9) AND (ORALCOND=9) AND (VAGCOND=9)) CONDUSE=9.
> IF ELSE CONDUSE=2.
> EXECUTE.
>
> I will appreciate some help from you all.
>
> thanks - Cy
>
> =====================
> 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

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?"

Javier Figueroa

Re: Compute New Variable - Based on Responses to Other Variables

To use ELSE you have to use the DO IF command, not just with an IF,

I hope this serves you,

/* procedimiento recodificacion en nuevas variables con condiciones.

data list list

/ANALCOND ORALCOND VAGCOND.

begin data.

1 1 2

0 1 5

2 0 1

1 1 5

2 0 1

0 8 2

8 1 1

5 2 1

1 0 2

2 1 5

1 8 0

2 1 2

end data.

execute.

numeric id(f8.0).

compute id=$casenum.

execute.

variable level all (scale).

IF any(1, ANALCOND, ORALCOND, VAGCOND) CONDUSE = 1.

EXE.

RECODE CONDUSE (SYSMIS=2).

EXE.

Sincerely,

2018-03-20 15:51 GMT-06:00 David Marso <[hidden email]>:

DO IF (condition).
blah blah blah.
ELSE IF (condition).
........
....
ELSE
....
END IF.

But that does NOT address the utterly piss poor mapping of the 3D space to
your proposed values.

Does {0,8,9} REALLY map to a 2?
Look at a 5x5 cube. That is your 3D space. What goes into what?
Back to the drawing board.

Chao yawo-2 wrote
> Hello Listservers,
>
> I have variables:
>
> X (use of anal-condom)
> Y (use of oral condom)
> Z (use of vaginal condom)
>
> Each as is coded 1=Yes, 2=No, 0,8,9=Missing.
>
> I want to create a new variable that is set to 1 (CONDUSE = 1), if the
> respondent uses condom for any type of sex, 2, if not
>
> I tried the following command, however but it is giving me an error
> regarding the ELSE statement.
> If i delete the ELSE statement, the command is not giving me the values
> for
> case 2 (those who do not use condom for at all.
>
> COMPUTE CONDUSE=0.
> IF ((ANALCOND=1) OR (ORALCOND=1) OR (VAGCOND=1)) CONDUSE=1.
> IF ((ANALCOND=0) AND (ORALCOND=0) AND (VAGCOND=0)) CONDUSE=0.
> IF ((ANALCOND=8) AND (ORALCOND=8) AND (VAGCOND=8)) CONDUSE=8.
> IF ((ANALCOND=9) AND (ORALCOND=9) AND (VAGCOND=9)) CONDUSE=9.
> IF ELSE CONDUSE=2.
> EXECUTE.
>
> I will appreciate some help from you all.
>
> thanks - Cy
>
> =====================
> 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

Javier Figueroa
Procesamiento y Análisis de bases de datos

Cel: 5927-4748 / 4970-1940

Casa: 2289-0184

Jon Peck

Re: Compute New Variable - Based on Responses to Other Variables

Please don't use those EXECUTE commands. They are forcing four data passes when one would do.

On Tue, Mar 20, 2018 at 4:15 PM Javier Figueroa <[hidden email]> wrote:

To use ELSE you have to use the DO IF command, not just with an IF,

I hope this serves you,

/* procedimiento recodificacion en nuevas variables con condiciones.

data list list
/ANALCOND ORALCOND VAGCOND.
begin data.
1 1 2
0 1 5
2 0 1
1 1 5
2 0 1
0 8 2
8 1 1
5 2 1
1 0 2
2 1 5
1 8 0
2 1 2
end data.
execute.

numeric id(f8.0).
compute id=$casenum.
execute.
variable level all (scale).

IF any(1, ANALCOND, ORALCOND, VAGCOND) CONDUSE = 1.
EXE.
RECODE CONDUSE (SYSMIS=2).
EXE.

Sincerely,

2018-03-20 15:51 GMT-06:00 David Marso <[hidden email]>:
DO IF (condition).
blah blah blah.
ELSE IF (condition).
........
....
ELSE
....
END IF.

But that does NOT address the utterly piss poor mapping of the 3D space to
your proposed values.

Does {0,8,9} REALLY map to a 2?
Look at a 5x5 cube. That is your 3D space. What goes into what?
Back to the drawing board.

Chao yawo-2 wrote
> Hello Listservers,
>
> I have variables:
>
> X (use of anal-condom)
> Y (use of oral condom)
> Z (use of vaginal condom)
>
> Each as is coded 1=Yes, 2=No, 0,8,9=Missing.
>
> I want to create a new variable that is set to 1 (CONDUSE = 1), if the
> respondent uses condom for any type of sex, 2, if not
>
> I tried the following command, however but it is giving me an error
> regarding the ELSE statement.
> If i delete the ELSE statement, the command is not giving me the values
> for
> case 2 (those who do not use condom for at all.
>
> COMPUTE CONDUSE=0.
> IF ((ANALCOND=1) OR (ORALCOND=1) OR (VAGCOND=1)) CONDUSE=1.
> IF ((ANALCOND=0) AND (ORALCOND=0) AND (VAGCOND=0)) CONDUSE=0.
> IF ((ANALCOND=8) AND (ORALCOND=8) AND (VAGCOND=8)) CONDUSE=8.
> IF ((ANALCOND=9) AND (ORALCOND=9) AND (VAGCOND=9)) CONDUSE=9.
> IF ELSE CONDUSE=2.
> EXECUTE.
>
> I will appreciate some help from you all.
>
> thanks - Cy
>
> =====================
> 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

--
Javier Figueroa
Procesamiento y Análisis de bases de datos
Cel: 5927-4748 / 4970-1940
Casa: 2289-0184

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

Javier Figueroa

Re: Compute New Variable - Based on Responses to Other Variables

Ok, thanks for the recommendation. It is custom, but I will try not to use EXECUTE a lot, I will try to use them properly. I had already heard something about that issue of using many EXECUTE commands.

Sincerely,

2018-03-20 16:34 GMT-06:00 Jon Peck <[hidden email]>:

Please don't use those EXECUTE commands. They are forcing four data passes when one would do.

On Tue, Mar 20, 2018 at 4:15 PM Javier Figueroa <[hidden email]> wrote:
To use ELSE you have to use the DO IF command, not just with an IF,

I hope this serves you,

/* procedimiento recodificacion en nuevas variables con condiciones.

data list list
/ANALCOND ORALCOND VAGCOND.
begin data.
1 1 2
0 1 5
2 0 1
1 1 5
2 0 1
0 8 2
8 1 1
5 2 1
1 0 2
2 1 5
1 8 0
2 1 2
end data.
execute.

numeric id(f8.0).
compute id=$casenum.
execute.
variable level all (scale).

IF any(1, ANALCOND, ORALCOND, VAGCOND) CONDUSE = 1.
EXE.
RECODE CONDUSE (SYSMIS=2).
EXE.

Sincerely,

2018-03-20 15:51 GMT-06:00 David Marso <[hidden email]>:
DO IF (condition).
blah blah blah.
ELSE IF (condition).
........
....
ELSE
....
END IF.

But that does NOT address the utterly piss poor mapping of the 3D space to
your proposed values.

Does {0,8,9} REALLY map to a 2?
Look at a 5x5 cube. That is your 3D space. What goes into what?
Back to the drawing board.

Chao yawo-2 wrote
> Hello Listservers,
>
> I have variables:
>
> X (use of anal-condom)
> Y (use of oral condom)
> Z (use of vaginal condom)
>
> Each as is coded 1=Yes, 2=No, 0,8,9=Missing.
>
> I want to create a new variable that is set to 1 (CONDUSE = 1), if the
> respondent uses condom for any type of sex, 2, if not
>
> I tried the following command, however but it is giving me an error
> regarding the ELSE statement.
> If i delete the ELSE statement, the command is not giving me the values
> for
> case 2 (those who do not use condom for at all.
>
> COMPUTE CONDUSE=0.
> IF ((ANALCOND=1) OR (ORALCOND=1) OR (VAGCOND=1)) CONDUSE=1.
> IF ((ANALCOND=0) AND (ORALCOND=0) AND (VAGCOND=0)) CONDUSE=0.
> IF ((ANALCOND=8) AND (ORALCOND=8) AND (VAGCOND=8)) CONDUSE=8.
> IF ((ANALCOND=9) AND (ORALCOND=9) AND (VAGCOND=9)) CONDUSE=9.
> IF ELSE CONDUSE=2.
> EXECUTE.
>
> I will appreciate some help from you all.
>
> thanks - Cy
>
> =====================
> 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

--
Javier Figueroa
Procesamiento y Análisis de bases de datos
Cel: 5927-4748 / 4970-1940
Casa: 2289-0184

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

Javier Figueroa
Procesamiento y Análisis de bases de datos

Cel: 5927-4748 / 4970-1940

Casa: 2289-0184

John F Hall

Re: Compute New Variable - Based on Responses to Other Variables

In haste and tested using Javier's dummy data:

data list

/analcond 1 oralcond 3 vagcond 5.

begin data

1 1 2

0 1 5

2 0 1

1 1 5

2 0 1

0 8 2

8 1 1

5 2 1

1 0 2

2 1 5

1 8 0

2 1 2

end data.

count concount = analcond oralcond vagcond (1).

recode concount (2,1=1)(0=0) into conduse.

var lev concount (scale) conuse (Nominal).

var lab concount 'Number of methods used'

/conduse 'Use any methods'.

val lab conduse 1 'Yes' 0 'No'.

freq concount conduse.

Number of methods used
		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	0	1	8.3	8.3	8.3
	1	8	66.7	66.7	75.0
	2	3	25.0	25.0	100.0
	Total	12	100.0	100.0

Use any methods
		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	No	1	8.3	8.3	8.3
	Yes	11	91.7	91.7	100.0
	Total	12	100.0	100.0

mult resp groups = method 'Method(s) used'

(analcond oralcond vagcond (1))

/var concount (1,3) conduse (0,1)

/freq method

/tab concount conduse by method.

method Frequencies
		Responses		Percent of Cases
		N	Percent	Percent of Cases
Method(s) used^a	ANALCOND	4	28.6%	36.4%
	ORALCOND	6	42.9%	54.5%
	VAGCOND	4	28.6%	36.4%
Total		14	100.0%	127.3%
a. Dichotomy group tabulated at value 1.

*concountmethod Crosstabulation**
			Method(s) used^a			Total
			ANALCOND	ORALCOND	VAGCOND	Total
Number of methods used	1	Count	2	3	3	8
Number of methods used	2	Count	2	3	1	3
Total		Count	4	6	4	11
Percentages and totals are based on respondents.
a. Dichotomy group tabulated at value 1.

*condusemethod Crosstabulation**
			Method(s) used^a			Total
			ANALCOND	ORALCOND	VAGCOND
Use any methods	Yes	Count	4	6	4	11
Total		Count	4	6	4	11
Percentages and totals are based on respondents.
a. Dichotomy group tabulated at value 1.

Hope this is close to what you want.

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 Javier Figueroa
Sent: 20 March 2018 23:48
To: [hidden email]
Subject: Re: Compute New Variable - Based on Responses to Other Variables

Sincerely,

2018-03-20 16:34 GMT-06:00 Jon Peck <[hidden email]>:

Please don't use those EXECUTE commands. They are forcing four data passes when one would do.

On Tue, Mar 20, 2018 at 4:15 PM Javier Figueroa <[hidden email]> wrote:
To use ELSE you have to use the DO IF command, not just with an IF,

I hope this serves you,

/* procedimiento recodificacion en nuevas variables con condiciones.

data list list
/ANALCOND ORALCOND VAGCOND.
begin data.
1 1 2
0 1 5
2 0 1
1 1 5
2 0 1
0 8 2
8 1 1
5 2 1
1 0 2
2 1 5
1 8 0
2 1 2
end data.
execute.

numeric id(f8.0).
compute id=$casenum.
execute.
variable level all (scale).

IF any(1, ANALCOND, ORALCOND, VAGCOND) CONDUSE = 1.
EXE.
RECODE CONDUSE (SYSMIS=2).
EXE.

Sincerely,

2018-03-20 15:51 GMT-06:00 David Marso <[hidden email]>:
DO IF (condition).
blah blah blah.
ELSE IF (condition).
........
....
ELSE
....
END IF.

But that does NOT address the utterly piss poor mapping of the 3D space to
your proposed values.

Does {0,8,9} REALLY map to a 2?
Look at a 5x5 cube. That is your 3D space. What goes into what?
Back to the drawing board.

Chao yawo-2 wrote

> Hello Listservers,
>
> I have variables:
>
> X (use of anal-condom)
> Y (use of oral condom)
> Z (use of vaginal condom)
>
> Each as is coded 1=Yes, 2=No, 0,8,9=Missing.
>
> I want to create a new variable that is set to 1 (CONDUSE = 1), if the
> respondent uses condom for any type of sex, 2, if not
>
> I tried the following command, however but it is giving me an error
> regarding the ELSE statement.
> If i delete the ELSE statement, the command is not giving me the values
> for
> case 2 (those who do not use condom for at all.
>
> COMPUTE CONDUSE=0.
> IF ((ANALCOND=1) OR (ORALCOND=1) OR (VAGCOND=1)) CONDUSE=1.
> IF ((ANALCOND=0) AND (ORALCOND=0) AND (VAGCOND=0)) CONDUSE=0.
> IF ((ANALCOND=8) AND (ORALCOND=8) AND (VAGCOND=8)) CONDUSE=8.
> IF ((ANALCOND=9) AND (ORALCOND=9) AND (VAGCOND=9)) CONDUSE=9.
> IF ELSE CONDUSE=2.
> EXECUTE.
>
> I will appreciate some help from you all.
>
> thanks - Cy
>
> =====================
> 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

--
Javier Figueroa
Procesamiento y Análisis de bases de datos
Cel: 5927-4748 / 4970-1940
Casa: 2289-0184
===================== 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]

Javier Figueroa
Procesamiento y Análisis de bases de datos

Cel: 5927-4748 / 4970-1940

Casa: 2289-0184

John F Hall

Re: Compute New Variable - Based on Responses to Other Variables

In reply to this post by Javier Figueroa

Left this one out:

mult resp groups = method 'Method(s) used'
(ANALCOND ORALCOND VAGCOND (1))
/tab method by method.

*methodmethod Crosstabulation**
			Method(s) used^a			Total
			ANALCOND	ORALCOND	VAGCOND	Total
Method(s) used^a	ANALCOND	Count	4	2	0	4
	ORALCOND	Count	2	6	1	6
	VAGCOND	Count	0	1	4	4
Total		Count	4	6	4	11
Percentages and totals are based on respondents.
a. Dichotomy group tabulated at value 1.

You can also do it three ways! Does DJT?

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: John F Hall <[hidden email]>
Sent: 21 March 2018 10:18
To: '[hidden email]' <[hidden email]>
Subject: RE: Compute New Variable - Based on Responses to Other Variables

In haste and tested using Javier's dummy data:

data list

/analcond 1 oralcond 3 vagcond 5.

begin data

1 1 2

0 1 5

2 0 1

1 1 5

2 0 1

0 8 2

8 1 1

5 2 1

1 0 2

2 1 5

1 8 0

2 1 2

end data.

count concount = analcond oralcond vagcond (1).

recode concount (2,1=1)(0=0) into conduse.

var lev concount (scale) conuse (Nominal).

var lab concount 'Number of methods used'

/conduse 'Use any methods'.

val lab conduse 1 'Yes' 0 'No'.

freq concount conduse.

Number of methods used
		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	0	1	8.3	8.3	8.3
	1	8	66.7	66.7	75.0
	2	3	25.0	25.0	100.0
	Total	12	100.0	100.0

Use any methods
		Frequency	Percent	Valid Percent	Cumulative Percent
Valid	No	1	8.3	8.3	8.3
	Yes	11	91.7	91.7	100.0
	Total	12	100.0	100.0

mult resp groups = method 'Method(s) used'

(analcond oralcond vagcond (1))

/var concount (1,3) conduse (0,1)

/freq method

/tab concount conduse by method.

method Frequencies
		Responses		Percent of Cases
		N	Percent	Percent of Cases
Method(s) used^a	ANALCOND	4	28.6%	36.4%
	ORALCOND	6	42.9%	54.5%
	VAGCOND	4	28.6%	36.4%
Total		14	100.0%	127.3%
a. Dichotomy group tabulated at value 1.

*concountmethod Crosstabulation**
			Method(s) used^a			Total
			ANALCOND	ORALCOND	VAGCOND	Total
Number of methods used	1	Count	2	3	3	8
Number of methods used	2	Count	2	3	1	3
Total		Count	4	6	4	11
Percentages and totals are based on respondents.
a. Dichotomy group tabulated at value 1.

*condusemethod Crosstabulation**
			Method(s) used^a			Total
			ANALCOND	ORALCOND	VAGCOND
Use any methods	Yes	Count	4	6	4	11
Total		Count	4	6	4	11
Percentages and totals are based on respondents.
a. Dichotomy group tabulated at value 1.

Hope this is close to what you want.

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)

Sincerely,

2018-03-20 16:34 GMT-06:00 Jon Peck <[hidden email]>:

Please don't use those EXECUTE commands. They are forcing four data passes when one would do.

On Tue, Mar 20, 2018 at 4:15 PM Javier Figueroa <[hidden email]> wrote:
To use ELSE you have to use the DO IF command, not just with an IF,

I hope this serves you,

/* procedimiento recodificacion en nuevas variables con condiciones.

data list list
/ANALCOND ORALCOND VAGCOND.
begin data.
1 1 2
0 1 5
2 0 1
1 1 5
2 0 1
0 8 2
8 1 1
5 2 1
1 0 2
2 1 5
1 8 0
2 1 2
end data.
execute.

numeric id(f8.0).
compute id=$casenum.
execute.
variable level all (scale).

IF any(1, ANALCOND, ORALCOND, VAGCOND) CONDUSE = 1.
EXE.
RECODE CONDUSE (SYSMIS=2).
EXE.

Sincerely,

2018-03-20 15:51 GMT-06:00 David Marso <[hidden email]>:
DO IF (condition).
blah blah blah.
ELSE IF (condition).
........
....
ELSE
....
END IF.

But that does NOT address the utterly piss poor mapping of the 3D space to
your proposed values.

Does {0,8,9} REALLY map to a 2?
Look at a 5x5 cube. That is your 3D space. What goes into what?
Back to the drawing board.

Chao yawo-2 wrote

> Hello Listservers,
>
> I have variables:
>
> X (use of anal-condom)
> Y (use of oral condom)
> Z (use of vaginal condom)
>
> Each as is coded 1=Yes, 2=No, 0,8,9=Missing.
>
> I want to create a new variable that is set to 1 (CONDUSE = 1), if the
> respondent uses condom for any type of sex, 2, if not
>
> I tried the following command, however but it is giving me an error
> regarding the ELSE statement.
> If i delete the ELSE statement, the command is not giving me the values
> for
> case 2 (those who do not use condom for at all.
>
> COMPUTE CONDUSE=0.
> IF ((ANALCOND=1) OR (ORALCOND=1) OR (VAGCOND=1)) CONDUSE=1.
> IF ((ANALCOND=0) AND (ORALCOND=0) AND (VAGCOND=0)) CONDUSE=0.
> IF ((ANALCOND=8) AND (ORALCOND=8) AND (VAGCOND=8)) CONDUSE=8.
> IF ((ANALCOND=9) AND (ORALCOND=9) AND (VAGCOND=9)) CONDUSE=9.
> IF ELSE CONDUSE=2.
> EXECUTE.
>
> I will appreciate some help from you all.
>
> thanks - Cy
>
> =====================
> 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

--
Javier Figueroa
Procesamiento y Análisis de bases de datos
Cel: 5927-4748 / 4970-1940
Casa: 2289-0184
===================== 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]

Javier Figueroa
Procesamiento y Análisis de bases de datos

Cel: 5927-4748 / 4970-1940

Casa: 2289-0184

Art Kendall

Re: Compute New Variable - Based on Responses to Other Variables

In reply to this post by Chao yawo-2

Rich gave a good suggestion so that you find out what patterns of responses
actually occur in your data.
in <help> look at AUTORECODE. It is sometimes helpful in further use of the
new variable.

A few suggestions for your future work to increase readability and thereby
help yourself, QA reviewers, and possible helpers:

-- After creating a new variable be sure to add a variable label and value
labels.
-- use <autoindent> to clarify the structure of your syntax.
-- reserve "=" as an assignment operator and use "EQ" as the logical
operator.
IF ((ANALCOND EQ 1) OR (ORALCOND=1) OR (VAGCOND=1)) CONDUSE=1.

at the end of your DO logic structure end with something like this (untested
PRINT command).
This will document to yourself and to others that you have provided for all
alternative combinations
ELSE.
PRINT / 'oops should not be here'.
END IF.

In this instance, I would go with Rich's suggestion about concatenating
variables into a new variable.

but there will be times when ANY will be useful.

-----
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