Combining Date and Service Fields so its sortable
Locked
 
|
Hi, all
Sorry if this email shows up twice. I am having problem with my email.
This may seem like an odd question and you guys may have a better way of doing this. I have files of clients that have had many different services across time. I just want to analyze the time variables but do so in a way that the time reflects the service. Clients can have 0 – 27 possible different time data points. One service does not build on another service. For example, a client does not necessarily go through training, before intervention before employment. I will be restructuring the data into long format. I want to be able to sort the dates and have the time reflect the order of the service for each client. If I just have the intervention name, I can’t tell which service came before another service. If I use the date, I can’t tell which service the dates represent. For example:
111 Training Intervention Intervention Training Training Employment Employment Intervention Employment
Other than pre-sorting prior to restructuring, which I don't want to do because I want to track the date files, is there a way to do this?
Thanks
George
|
Re: Combining Date and Service Fields so its sortable
|
|
Are you saying that the incoming data file has four variables (ClientID, Training, Intervention, Employment)? And you want to restructure the file to something like this? ClientID ServiceType Date 111 Training 3/4/2005 111 Intervention 6/1/2005 111 Intervention 7/1/2005 111 Training 7/5/2005 111 Training 9/7/2005 111 Employment 9/21/2005 111 Employment 10/1/2005 111 Employment 12/9/2005 111 Intervention 10/15/2006 If so, I think but I’m not sure that you will first need to do a Casestovars to get one record per id and then a Varstocases to get 9 (or however many there) records per id. And, I kind of think there may need to be some work in between the Casestovars and the Varstocases to get the Varstocases to fall out into three columns but I’d need to work with your example data to check that out. Gene Maguin From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of George Hamlin Hi, all Sorry if this email shows up twice. I am having problem with my email. This may seem like an odd question and you guys may have a better way of doing this. I have files of clients that have had many different services across time. I just want to analyze the time variables but do so in a way that the time reflects the service. Clients can have 0 – 27 possible different time data points. One service does not build on another service. For example, a client does not necessarily go through training, before intervention before employment. I will be restructuring the data into long format. I want to be able to sort the dates and have the time reflect the order of the service for each client. If I just have the intervention name, I can’t tell which service came before another service. If I use the date, I can’t tell which service the dates represent. For example: 111 Training Intervention Intervention Training Training Employment Employment Intervention Employment Other than pre-sorting prior to restructuring, which I don't want to do because I want to track the date files, is there a way to do this?
Thanks George |
Re: Combining Date and Service Fields so its sortable
Administrator
|
In reply to this post by GeorgeH
George,
It is unclear how the first data representation maps to the second wrt dates being associated with the different events. Perhaps posting or attaching an actual example of the data might be helpful? David
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?" |
Re: Combining Date and Service Fields so its sortable
|
|
In reply to this post by Maguin, Eugene
George, I wonder if it would be simple enough to save three files (a training file, an intervention file and an employment file) with just the id and the date, calculate
a type and add them all together. I think Date is a reserved word so I called the new variables SDate (service date). Something like this….(untested. Save outfile=’<pathname>/Training.sav’/keep=clientid Training. Save outfile=’<pathname>/Intervention.sav’/keep=clientid Intervention. Save outfile=’<pathname>/Employment.sav’/keep=clientid Employment. Get file=’<pathname>/Training.sav’. Dataset name training. Rename Training=SDate. Compute ServiceType=1. Get file=’<pathname>/Intervention.sav’. Dataset name Intervention. Rename Intervention=SDate. Compute ServiceType=2. Get file=’<pathname>/ Employment.sav’. Dataset name Employment. Rename Employment =SDate. Compute ServiceType=3. Add files file=Training/File=Intervention/File=Employment. Dataset name combo. Val labels ServiceType 1 ‘Training’ 2 ‘Intervention’ 3 ‘Employment’. Dataset activate combo. Sort cases by clientid SDate. Melissa From: SPSSX(r) Discussion [mailto:[hidden email]]
On Behalf Of Maguin, Eugene Are you saying that the incoming data file has four variables (ClientID, Training, Intervention, Employment)? And you want to restructure the file to something like this? ClientID ServiceType Date 111 Training
3/4/2005 111 Intervention
6/1/2005 111 Intervention
7/1/2005 111 Training
7/5/2005 111 Training
9/7/2005 111 Employment
9/21/2005 111 Employment
10/1/2005 111 Employment
12/9/2005 111 Intervention
10/15/2006 If so, I think but I’m not sure that you will first need to do a Casestovars to get one record per id and then a Varstocases to get 9 (or however many there)
records per id. And, I kind of think there may need to be some work in between the Casestovars and the Varstocases to get the Varstocases to fall out into three columns but I’d need to work with your example data to check that out. Gene Maguin From: SPSSX(r) Discussion [[hidden email]]
On Behalf Of George Hamlin Hi, all
Sorry if this email shows up twice. I am having problem with my email.
This may seem like an odd question and you guys may have a better way of doing this. I have files of clients that have had many different services across time. I just want to analyze
the time variables but do so in a way that the time reflects the service. Clients can have 0 – 27 possible different time data points. One service does not build on another service. For example, a client does not necessarily go through training, before intervention
before employment. I will be restructuring the data into long format. I want to be able to sort the dates and have the time reflect the order of the service for each client. If I just have the intervention name, I can’t tell which service came before another
service. If I use the date, I can’t tell which service the dates represent. For example:
111 Training Intervention Intervention Training Training Employment Employment Intervention Employment
Other than pre-sorting prior to restructuring, which I don't want to do because I want to track the date files, is there a way to do this?
Thanks
George PRIVILEGED AND CONFIDENTIAL INFORMATION This transmittal and any attachments may contain PRIVILEGED AND CONFIDENTIAL information and is intended only for the use of the addressee. If you are not the designated recipient, or an employee or agent authorized to deliver such transmittals to the designated recipient, you are hereby notified that any dissemination, copying or publication of this transmittal is strictly prohibited. If you have received this transmittal in error, please notify us immediately by replying to the sender and delete this copy from your system. You may also call us at (309) 827-6026 for assistance. |
|
|
In reply to this post by GeorgeH
Tamra Boyce Principal Planning Analyst Hennepin County Research, Planning and Development (238) 300 South 6th Street
Minneapolis, MN 55487 PH:(612) 348-7059 Blackberry 612-432-7701 The plural of anecdote is not data. ~ Roger Brinner From: "Maguin, Eugene" <[hidden email]>
To: [hidden email] Date: 05/07/2013 10:19 AM Subject: Re: Combining Date and Service Fields so its sortable
Sent by: "SPSSX(r) Discussion" <[hidden email]> Are you saying that the incoming data file has four variables (ClientID, Training, Intervention, Employment)? Gene I don't want to restructure it like that - at least not yet. I'm not at that point yet. To be honest, I'm still working through the logic of this whole thing.
I want to recode the data variable, for example 3/4/2005 to a 3/4/2005T to note that a training occurred on that date - thus the date also becomes the service type. But I want to keep the sort capability of the date. I want to combine the service type with the date and then sort that date/service to get a kind of service pathway.
So it would look like this for client 111: 3/4/2005T 6/1/2005I 7/1/2005I 7/5/2005T 9/7/2005T
9/21/2005E I really don't want to lose that date sort ability because the date order matters. Thinking more about this, I might just sort by ID then the date, concatenate the date and service type, and then restructure - which would give me my "service pathway".
George ____________________________________________________________________ From: "Maguin, Eugene" <[hidden email]>
To: [hidden email] Date: 05/07/2013 10:19 AM Subject: Re: Combining Date and Service Fields so its sortable
Sent by: "SPSSX(r) Discussion" <[hidden email]> Are you saying that the incoming data file has four variables (ClientID, Training, Intervention, Employment)? And you want to restructure the file to something like this? ClientID ServiceType Date 111 Training 3/4/2005 111 Intervention 6/1/2005 111 Intervention 7/1/2005 111 Training 7/5/2005
111 Training 9/7/2005 111 Employment 9/21/2005 111 Employment 10/1/2005 111 Employment 12/9/2005 111 Intervention 10/15/2006
If so, I think but I’m not sure that you will first need to do a Casestovars to get one record per id and then a Varstocases to get 9 (or however many there) records per id. And, I kind of think there may need to be some work in between the Casestovars and the Varstocases to get the Varstocases to fall out into three columns but I’d need to work with your example data to check that out.
Gene Maguin From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of George Hamlin Sent: Monday, May 06, 2013 1:48 PM
To: [hidden email] Subject: Combining Date and Service Fields so its sortable Hi, all Sorry if this email shows up twice. I am having problem with my email.
This may seem like an odd question and you guys may have a better way of doing this. I have files of clients that have had many different services across time. I just want to analyze the time variables but do so in a way that the time reflects the service. Clients can have 0 – 27 possible different time data points. One service does not build on another service. For example, a client does not necessarily go through training, before intervention before employment. I will be restructuring the data into long format. I want to be able to sort the dates and have the time reflect the order of the service for each client. If I just have the intervention name, I can’t tell which service came before another service. If I use the date, I can’t tell which service the dates represent. For example:
111 Training Intervention Intervention Training Training Employment Employment Intervention Employment Other than pre-sorting prior to restructuring, which I don't want to do because I want to track the date files, is there a way to do this?
ClientID Training Intervention Employment 111 3/4/2005 6/1/2005 9/21/2005
111 7/5/2005 7/1/2005 10/1/2005 111 9/7/2005 10/15/2006 12/9/2005
222 5/27/2009 7/23/2009 5/15/2009 222 8/14/2009 10/1/2009 1/23/2009
333 12/1/2009 1/15/2010 5/2/2008 333 2/10/2010 3/23/2010 4/15/2009
333 4/21/2010 6/1/2010 8/25/2010 333 7/15/2010 9/21/2010 2/1/2011
Thanks George |
Re: Combining Date and Service Fields so its sortable
|
Administrator
|
In reply to this post by Melissa Ives
Assuming (ESPSS here) that the table presented is the actual initial data structure rather than the result from some unknown structure (in latter case please redefine actual problem). The problem can be resolved very simply without intermediate files.
-- DATA LIST FREE/ ClientID (F3) Training (ADATE) Intervention (ADATE) Employment (ADATE) . BEGIN DATA 111 03/04/2005 06/01/2005 09/21/2005 111 07/05/2005 07/01/2005 10/01/2005 111 09/07/2005 10/15/2006 12/09/2005 222 05/27/2009 07/23/2009 05/15/2009 222 08/14/2009 10/01/2009 01/23/2009 333 12/01/2009 01/15/2010 05/02/2008 333 02/10/2010 03/23/2010 04/15/2009 333 04/21/2010 06/01/2010 08/25/2010 333 07/15/2010 09/21/2010 02/01/2011 END DATA. VARSTOCASES /MAKE Dates FROM Training Intervention Employment /INDEX=Service(Dates) /KEEP=ClientID . SORT CASES BY ClientID Dates. LIST. ClientID Service Dates 111 Training 03/04/2005 111 Intervention 06/01/2005 111 Intervention 07/01/2005 111 Training 07/05/2005 111 Training 09/07/2005 111 Employment 09/21/2005 111 Employment 10/01/2005 111 Employment 12/09/2005 111 Intervention 10/15/2006 222 Employment 01/23/2009 222 Employment 05/15/2009 222 Training 05/27/2009 222 Intervention 07/23/2009 222 Training 08/14/2009 222 Intervention 10/01/2009 333 Employment 05/02/2008 333 Employment 04/15/2009 333 Training 12/01/2009 333 Intervention 01/15/2010 333 Training 02/10/2010 333 Intervention 03/23/2010 333 Training 04/21/2010 333 Intervention 06/01/2010 333 Training 07/15/2010 333 Employment 08/25/2010 333 Intervention 09/21/2010 333 Employment 02/01/2011 Number of cases read: 27 Number of cases listed: 27
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?" |
Re: Combining Date and Service Fields so its sortable
|
Administrator
|
In reply to this post by GeorgeH
"To be honest, I'm still working through the logic of this whole thing."
When you sort out your inputs and desired outputs get back to us! NOTE the format you propose id NOT SORTABLE!! Rethink that. OTOH: There is an SDATE (sortable date) format yyyy-mm-dd I would keep the date and service as separate variables (see my solution below or above ). Doing otherwise simply opens you to a world of pain!
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?" |
Automatic reply: Combining Date and Service Fields so its sortable
|
|
I will be out of the office until May 13th with limited access to e-mail.
===================== 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 |
|
|
In reply to this post by GeorgeH
I appreciate all the syntax and help I received both on and off-line - and I have been given a number of things to think about. I'll work through the solutions offered to see what works best for me.
Thanks to all those that offered help. George |
| Free forum by Nabble | Edit this page |