converting string to numerical

[Muir Houston-2]

Hi Listers,
I am running V18 (but not Python)

I have data from multiple responses which are entered in SPSS as a
string variable - respondents were asked to tick all that apply and
there were 7 options. The question asked young people where they hung
out in the local area - so 1 = own home; 2 = friends home; 3 = outside;
4 = online; etc.

Data is in the form:
1,2,4,6
2,3,7
1,4,6,7
1,2,3,4,5,6,7
4,6
Etc.

They are always in order, but do not necessarily start at 1.

I had a look in Raynald's programming book and tried to adapt the syntax
he uses in an example about splitting telephone numbers into component
parts - but ended up getting stuck. Also had a look at archives but not
the easiest to search

What I want is to create seven new dummy variables (v1 to v7) where the
variables correspond to the seven possible options

I will then use the multiple response option to analyze (thanks to John
Hall for assistance there)

Thanks in advance



Muir Houston, HNC, BA (Hons), M.Phil., PhD, FHEA
Research Fellow
School of Education
University of Glasgow
0044+141-330-4699

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

Muir Houston-2 wrote

Hi Listers,
I am running V18 (but not Python)

I have data from multiple responses which are entered in SPSS as a
string variable - respondents were asked to tick all that apply and
there were 7 options. The question asked young people where they hung
out in the local area - so 1 = own home; 2 = friends home; 3 = outside;
4 = online; etc.

Data is in the form:
1,2,4,6
2,3,7
1,4,6,7
1,2,3,4,5,6,7
4,6
Etc.

They are always in order, but do not necessarily start at 1.

I had a look in Raynald's programming book and tried to adapt the syntax
he uses in an example about splitting telephone numbers into component
parts - but ended up getting stuck. Also had a look at archives but not
the easiest to search

What I want is to create seven new dummy variables (v1 to v7) where the
variables correspond to the seven possible options

I will then use the multiple response option to analyze (thanks to John
Hall for assistance there)

Thanks in advance

data list list / stringvar (a15).
begin data
"1,2,4,6"
"2,3,7"
"1,4,6,7"
"1,2,3,4,5,6,7"
"4,6"
end data.

numeric r1 to r7 (f1.0).
do repeat r = r1 to r7 / #i = 1 to 7.
- compute r = char.index(stringvar,string(#i,f1)) GT 0.
end repeat.
list.

[David Marso]

On Thu, 7 Oct 2010 07:42:33 -0700, Bruce Weaver <[hidden email]>
wrote:

>Muir Houston-2 wrote:
>>
>> Hi Listers,
>> I am running V18 (but not Python)
>>
>> I have data from multiple responses which are entered in SPSS as a
>> string variable - respondents were asked to tick all that apply and
>> there were 7 options. The question asked young people where they hung
>> out in the local area - so 1 = own home; 2 = friends home; 3 = outside;
>> 4 = online; etc.
>>
>> Data is in the form:
>> 1,2,4,6
>> 2,3,7
>> 1,4,6,7
>> 1,2,3,4,5,6,7
>> 4,6
>> Etc.
>>
>> They are always in order, but do not necessarily start at 1.
>>
>> I had a look in Raynald's programming book and tried to adapt the syntax
>> he uses in an example about splitting telephone numbers into component
>> parts - but ended up getting stuck. Also had a look at archives but not
>> the easiest to search
>>
>> What I want is to create seven new dummy variables (v1 to v7) where the
>> variables correspond to the seven possible options
>>
>> I will then use the multiple response option to analyze (thanks to John
>> Hall for assistance there)
>>
>> Thanks in advance
>>
>
>data list list / stringvar (a15).
>begin data
>"1,2,4,6"
>"2,3,7"
>"1,4,6,7"
>"1,2,3,4,5,6,7"
>"4,6"
>end data.
>
>numeric r1 to r7 (f1.0).
>do repeat r = r1 to r7 / #i = 1 to 7.
>- compute r = char.index(stringvar,string(#i,f1)) GT 0.
>end repeat.
>list.
>
>
>
>-----
>--
>Bruce Weaver
>[hidden email]
>http://sites.google.com/a/lakeheadu.ca/bweaver/
>
>"When all else fails, RTFM."
>
>

Very nice Bruce,
 Here's a slightly different version using a VECTOR.
---
VECTOR V(7).
LOOP #=1 to 7.
+  COMPUTE V(#)=CHAR.INDEX(strvar,string(#,f1)) GT 0.
END LOOP.

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