FW: Group Based Trajectory Analysis

[Maguin, Eugene]

This is the reply that Katherine sent off-list about her dataset and may be
useful to others thinking about offering replies to her.

-----Original Message-----
From: Courtney, Katherine, CYFD [mailto:[hidden email]]
Sent: Friday, September 30, 2011 6:07 PM
To: [hidden email]
Subject: Re: Group Based Trajectory Analysis

Yes, we have arrest dates. I have arrest for individuals referred to the
juvenile justice system in the state for a number of years (would have to
check for the exact number). The range in number of arrests per year could
be anything from one to well over ten.


On Fri, 30 Sep 2011 17:04:34 -0400, Gene Maguin <[hidden email]> wrote:

>Ooooh. You say you have arrests measured on a continuous basis. So what
does

>that mean? Does that mean you know the arrest dates? And, does everybody
>have at least one arrest? I see you are in state government. To the extent
>that you can, would you describe your dataset in terms of who's in it, the
>range in the time spans for which you have data for dataset members and the
>range in the number of arrests per time unit, like maybe, a year.
>
>Gene Maguin
>
>
>
>
>
>-----Original Message-----
>From: Courtney, Katherine, CYFD [mailto:[hidden email]]
>Sent: Friday, September 30, 2011 4:50 PM
>To: [hidden email]; [hidden email]
>Subject: Re: Group Based Trajectory Analysis
>
>I do not have predifined groups. Your second guess is right, I have a big
>bunch of people and want to find groups who have similar trajectories. I
>have measures of something on a continuous basis- specifically arrests.
>That data could be grouped if necessary into number of arrests per year for
>several years for each person. Relating to your question about variance,
>I'm not entirely sure because this is a new method I am trying to learn,
>but I believe that there would be variance within the groups. The goal is
>to determine if there are underlying trajectories or patterns of offending.
>
>On Fri, 30 Sep 2011 16:01:06 -0400, Gene Maguin <[hidden email]>

wrote:

>
>>Kat,
>>
>>Tell us what you mean by 'Group' and as Bruce indicated what you mean by
>>'trajectory'. More specifically, do you have predefined groups, like males
>>and females; 8th, 9th and 10th grade students? OR do you have a big bunch
>of
>>people and you want to find groups of people who have 'similar'
>>trajectories? On to trajectory. Do you have measures of something at
>>multiple time points (how many??) for each person? Or something else? And.
>>How are you thinking of trajectory? Is it something that the group has,
>>meaning that the variance of the trajectory coefficients for the group is
>>0.0 or is it something that has  a distribution among members of a group,
>>meaning that the trajectory coefficents for the group have non-zero
>>variance.
>>
>>Gene Maguin
>>
>>
>>
>>-----Original Message-----
>>From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
>>Bruce Weaver
>>Sent: Friday, September 30, 2011 3:23 PM
>>To: [hidden email]
>>Subject: Re: Group Based Trajectory Analysis
>>
>>Sometimes, the same statistical animal goes by different names in

different
>>disciplines.  In case that's happening here, what procedure would you use
>in
>>SAS for this type of analysis?  And are you aware of any other names for
>>"group based trajectory analysis"?  ("Trajectory" is a word that crops up
>in
>>analysis of longitudinal data with multilevel models, with occasions
>>clustered within subjects.  So I wonder if this is what you're getting
at.)
>>
>>
>>
>>Mariajose Romero, PhD wrote:
>>>
>>> As far as I know, the answer is no. You must use SAS. If anyone knows
how

>>> to
>>> do this type of analysis in SPSS, I would love to hear about it.
>>>
>>> Mariajos� Romero, PhD
>>> Associate Professor & Coordinator Education Programs
>>> LaGuardia Community College, CUNY
>>>
>>> -----Original Message-----
>>> From: SPSSX(r) Discussion [mailto:SPSSX-L@.UGA] On Behalf Of Kat
>>> Sent: Friday, September 30, 2011 11:42 AM
>>> To: SPSSX-L@.UGA
>>> Subject: Group Based Trajectory Analysis
>>>
>>> Does anybody know if it is possible to do group based trajectory

analysis

>>> in SPSS? I would like to find latent groups of trajectories for juvenile
>>> arrests.
>>> Thank you in advance!
>>>
>>> =====================
>>> 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
>>>
>>> =====================
>>> 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
>>>
>>
>>
>>-----
>>--
>>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.
>>
>>--
>>View this message in context:
>>http://spssx-discussion.1045642.n5.nabble.com/Group-Based-Trajectory-
>Analysi
>>s-tp4857707p4858134.html
>>Sent from the SPSSX Discussion mailing list archive at 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

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