creating a variable based on a sequence

[Paul Mcgeoghan]

Hi,

 

I want to work out the number of cases in each group and then store that as a variable.

So I have variable Group

A has 20 cases

B has 30 cases

 

I want data to look like:

A  1

A  2

A  3

.

.

.

A 20

B 1

B 2

.

.

B 30

 

What is easiest way of achieving this in SPSS?

Thanks,

Paul

 

 

Paul McGeoghan,
Application Support Specialist (Statistics and Databases),
University Infrastructure Group (UIG),
Information Services
Cardiff University
Tel. [phone]

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

Paul,

Actually you want to number cases within group. There are several ways to do
this. I'd do it this way. Let rec be the new case number variable.

Compute rec=1.
If (group eq lag(group)) rec=lag(rec)+1.

The 'trick' is that with every case having the value of rec preset to 1, the
first case in group B has rec=1 when the comparison is made, which fails.

Gene Maguin


>>I want to work out the number of cases in each group and then store that
as a variable.
So I have variable Group
A has 20 cases
B has 30 cases

I want data to look like:
A  1
A  2
A  3
.
.
.
A 20
B 1
B 2
.
.
B 30

What is easiest way of achieving this in SPSS?
Thanks,
Paul

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

[Paul Mcgeoghan]

Gene,

Thanks, yes that is what I was looking for.

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

[Paul Mcgeoghan]

Thanks,
The following syntax will produce
Time
1
2
3
.
.
82
1
2
3
.
.
82

However, I now want to produce the following instead which represents 1-82
periods

Time
07:01
07:06
07:11
.
.
13:46
07:01
07:06
07:11
.
.
13:46
I could use Value Labels to define each of the 82 periods. Any other way?

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

[David Marso]

                                
Since SPSS stores time as number of seconds since midnight (ie 12:00 Am = 0) it follows that 7:00 AM = 7 * 60 * 60 = 25200.
And 5 minutes = 5 * 60 = 300 .
-------------------------------------
Hence your equation = NewTime = 25200 + Time * 300 .
So in SPSS:
COMPUTE NewTime = 25200 + Time * 300 .
FORMAT NewTime (TIME5).
------------------------------------
HTH, David
-------------

Thanks,
The following syntax will produce
Time
1
2
3
.
.
82
1
2
3
.
.
82

However, I now want to produce the following instead which represents 1-82
periods

Time
07:01
07:06
07:11
.
.
13:46
07:01
07:06
07:11
.
.
13:46

[Richard Ristow]

At 09:07 AM 3/26/2009, Paul McGeoghan wrote:

>However, I now want to produce the following which represents 1-82 periods
>
>Time
>07:01
>07:06

David Marso's given you a solution. However, I recommend using SPSS
time functions instead of calculating with seconds. The meaning of
the code is much clearer, and it's much easier to write accurately.

This is Gene Maguin's syntax, modified to increment by 5 minutes
instead of 1 unit:

NUMERIC tod(TIME5).

Compute tod=TIME.HMS(7,1).
If (group eq lag(group)) tod=lag(tod)+TIME.HMS(0,5).

>07:11
>.
>.
>13:46
>07:01
>07:06
>07:11
>.
>.
>13:46
>I could use Value Labels to define each of the 82 periods. Any other way?
>
>=====================
>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
>--=======AVGMAIL-49CBACCD0000=======
>Content-Type: multipart/alternative;
>         boundary="=======AVGMAIL-49CBACCD0000======="
>
>--=======AVGMAIL-49CBACCD0000=======
>Content-Type: text/plain; x-avg=cert; charset=us-ascii
>Content-Transfer-Encoding: quoted-printable
>Content-Disposition: inline
>Content-Description: "AVG certification"
>
>
>No virus found in this incoming message.
>Checked by AVG - www.avg.com
>Version: 8.0.238 / Virus Database: 270.11.29/2024 - Release Date: 03/26/09 0=
>7:12:00
>
>--=======AVGMAIL-49CBACCD0000=======--
>--=======AVGMAIL-49CBACCD0000=======--

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