Logistic regression with nested data
Locked
Logistic regression with nested data
 
|
I'm getting stuck on some analyses & would appreciate anyone's ideas.
I want to do a logistic regression controlling for a clustering variable. I'm predicting whether or not somehone has a family doc from various demographic factors but controlling for people being nested within counties. Which SPSS procedure can handle this? Thanks! Graham ===================== 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 |
Re: Logistic regression with nested data
|
|
Graham,
You can use either the Logistic Regression procedure or the Multinomial Logistic Regression procedure available in the Regression add-on module. Pay particular attention to the coding of the county variable. Best of luck, Jim On Nov 2, 2007 8:45 AM, Graham Reid <[hidden email]> wrote: > I'm getting stuck on some analyses & would appreciate anyone's ideas. > > I want to do a logistic regression controlling for a clustering variable. > I'm predicting whether or not somehone has a family doc from various > demographic factors but controlling for people being nested within > counties. > > Which SPSS procedure can handle this? > > Thanks! > Graham > > ===================== > 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 |
Re: Logistic regression with nested data
|
|
James,
>>You can use either the Logistic Regression procedure or the Multinomial Logistic Regression procedure available in the Regression add-on module. What is the Regression add-on module? What version does this apply to? Will this module truly do a multilevel logistic model? When Graham said '... people being nested within > counties.' I had understood that to imply a multilevel model. Thanks, Gene Maguin ===================== 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 |
Re: Logistic regression with nested data
|
|
Gene,
I'm not familiar enough with the different versions of SPSS to answer your question about when this module became available, but I do know it is available in SPSS 16. You might take a look at this website for a more precise explanation of using SPSS Logistic Regression for nested models: http://www2.chass.ncsu.edu/garson/PA765/logistic.htm Hope this helps, Jim On Nov 2, 2007 1:30 PM, Gene Maguin <[hidden email]> wrote: > James, > > >>You can use either the Logistic Regression procedure or the > Multinomial Logistic Regression procedure available in the Regression > add-on module. > > What is the Regression add-on module? What version does this apply to? Will > this module truly do a multilevel logistic model? When Graham said '... > people being nested within > > counties.' I had understood that to imply a multilevel model. > > Thanks, Gene Maguin > > > ===================== > 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 |
Re: Logistic regression with nested data
|
|
You may want to consider hierarchical generalized linear models. I am not
sure whether SPSS can do this though. The GEE idea is also a good possibility. On 11/2/07, James C. Whanger <[hidden email]> wrote: > > Gene, > > I'm not familiar enough with the different versions of SPSS to answer > your question about when this module became available, but I do know > it is available in SPSS 16. > > You might take a look at this website for a more precise explanation > of using SPSS Logistic Regression for nested models: > > http://www2.chass.ncsu.edu/garson/PA765/logistic.htm > > Hope this helps, > > Jim > > On Nov 2, 2007 1:30 PM, Gene Maguin <[hidden email]> wrote: > > James, > > > > >>You can use either the Logistic Regression procedure or the > > Multinomial Logistic Regression procedure available in the Regression > > add-on module. > > > > What is the Regression add-on module? What version does this apply to? > Will > > this module truly do a multilevel logistic model? When Graham said '... > > people being nested within > > > counties.' I had understood that to imply a multilevel model. > > > > Thanks, Gene Maguin > > > > > > ===================== > > 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 > -- Pablo Mora, Ph.D. 30 College Avenue The Institute for Health Rutgers University New Brunswick, NJ 08901-1293 Office Phone: (732) 932-1941 FAX: (732) 932-1945 ===================== 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 |
Re: Logistic regression with nested data
|
|
as far as I know it is not possible to do a multilevel analysis with a
binary dependent variable in spss. you might consider using mlwin, hlm, stata (version 10, or gllamm), sas or r. please correct me if I'm wrong. perhaps there is a new possibility in spss 16? christian On Fri, 2 Nov 2007 19:57:15 -0400 "Pablo A. Mora" <[hidden email]> wrote: > You may want to consider hierarchical generalized linear models. I am not > sure whether SPSS can do this though. > The GEE idea is also a good possibility. > > > On 11/2/07, James C. Whanger <[hidden email]> wrote: >> >> Gene, >> >> I'm not familiar enough with the different versions of SPSS to answer >> your question about when this module became available, but I do know >> it is available in SPSS 16. >> >> You might take a look at this website for a more precise explanation >> of using SPSS Logistic Regression for nested models: >> >> http://www2.chass.ncsu.edu/garson/PA765/logistic.htm >> >> Hope this helps, >> >> Jim >> >> On Nov 2, 2007 1:30 PM, Gene Maguin <[hidden email]> wrote: >> > James, >> > >> > >>You can use either the Logistic Regression procedure or the >> > Multinomial Logistic Regression procedure available in the Regression >> > add-on module. >> > >> > What is the Regression add-on module? What version does this apply to? >> Will >> > this module truly do a multilevel logistic model? When Graham said '... >> > people being nested within >> > > counties.' I had understood that to imply a multilevel model. >> > >> > Thanks, Gene Maguin >> > >> > >> > ===================== >> > 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 >> > > > > -- > Pablo Mora, Ph.D. > 30 College Avenue > The Institute for Health > Rutgers University > New Brunswick, NJ 08901-1293 > Office Phone: (732) 932-1941 FAX: (732) 932-1945 > > ===================== > 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 |
Re: Logistic regression with nested data
|
|
The Advance Regression modules of both v15 and v16
offer both generalized linear models and general estimating equations. Both allow for binary DVs and nesting/multilevel models. --- Gary S. Rosin Professor of Law South Texas College of Law 1303 San Jacinto Houston, TX 77002 <[hidden email]> 713-646-1854 ----- Original Message ----- From: Christian Deindl <[hidden email]> Date: Saturday, November 3, 2007 7:25 pm Subject: Re: Logistic regression with nested data To: [hidden email] > as far as I know it is not possible to do a multilevel analysis > with a > binary dependent variable in spss. > you might consider using mlwin, hlm, stata (version 10, or > gllamm), sas or > r. > > please correct me if I'm wrong. perhaps there is a new possibility > in spss > 16? > > christian > > > On Fri, 2 Nov 2007 19:57:15 -0400 > "Pablo A. Mora" <[hidden email]> wrote: > > You may want to consider hierarchical generalized linear models. > I am not > > sure whether SPSS can do this though. > > The GEE idea is also a good possibility. > > > > > > On 11/2/07, James C. Whanger <[hidden email]> wrote: > >> > >> Gene, > >> > >> I'm not familiar enough with the different versions of SPSS to > answer>> your question about when this module became available, > but I do know > >> it is available in SPSS 16. > >> > >> You might take a look at this website for a more precise > explanation>> of using SPSS Logistic Regression for nested models: > >> > >> http://www2.chass.ncsu.edu/garson/PA765/logistic.htm > >> > >> Hope this helps, > >> > >> Jim > >> > >> On Nov 2, 2007 1:30 PM, Gene Maguin <[hidden email]> wrote: > >> > James, > >> > > >> > >>You can use either the Logistic Regression procedure or the > >> > Multinomial Logistic Regression procedure available in the > Regression>> > add-on module. > >> > > >> > What is the Regression add-on module? What version does this > apply to? > >> Will > >> > this module truly do a multilevel logistic model? When Graham > said '... > >> > people being nested within > >> > > counties.' I had understood that to imply a multilevel model. > >> > > >> > Thanks, Gene Maguin > >> > > >> > > >> > ===================== > >> > 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 > >> > > > > > > > > -- > > Pablo Mora, Ph.D. > > 30 College Avenue > > The Institute for Health > > Rutgers University > > New Brunswick, NJ 08901-1293 > > Office Phone: (732) 932-1941 FAX: (732) 932-1945 > > > > ===================== > > 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 |
Re: Logistic regression with nested data
|
|
Hi Gary Rosin,
I'm very interested in the possibility of doing multilevel logistic regression on SPSS. Have you got som reference to instructions for doing that kind of analysis with the procedure for 'general estimating equations'? Best, Henrik ************************************************************ Henrik Lolle Department of Economics, Politics and Public Administration Aalborg University Fibigerstraede 1 9200 Aalborg East Phone: (+45) 9635 8184 ************************************************************ Quoting Gary Rosin <[hidden email]>: > The Advance Regression modules of both v15 and v16 > offer both generalized linear models and general > estimating equations. Both allow for binary DVs > and nesting/multilevel models. > > --- > > Gary S. Rosin > Professor of Law > South Texas College of Law > 1303 San Jacinto > Houston, TX 77002 > > <[hidden email]> > 713-646-1854 > > ----- Original Message ----- > From: Christian Deindl <[hidden email]> > Date: Saturday, November 3, 2007 7:25 pm > Subject: Re: Logistic regression with nested data > To: [hidden email] > >> as far as I know it is not possible to do a multilevel analysis >> with a >> binary dependent variable in spss. >> you might consider using mlwin, hlm, stata (version 10, or >> gllamm), sas or >> r. >> >> please correct me if I'm wrong. perhaps there is a new possibility >> in spss >> 16? >> >> christian >> >> >> On Fri, 2 Nov 2007 19:57:15 -0400 >> "Pablo A. Mora" <[hidden email]> wrote: >> > You may want to consider hierarchical generalized linear models. >> I am not >> > sure whether SPSS can do this though. >> > The GEE idea is also a good possibility. >> > >> > >> > On 11/2/07, James C. Whanger <[hidden email]> wrote: >> >> >> >> Gene, >> >> >> >> I'm not familiar enough with the different versions of SPSS to >> answer>> your question about when this module became available, >> but I do know >> >> it is available in SPSS 16. >> >> >> >> You might take a look at this website for a more precise >> explanation>> of using SPSS Logistic Regression for nested models: >> >> >> >> http://www2.chass.ncsu.edu/garson/PA765/logistic.htm >> >> >> >> Hope this helps, >> >> >> >> Jim >> >> >> >> On Nov 2, 2007 1:30 PM, Gene Maguin <[hidden email]> wrote: >> >> > James, >> >> > >> >> > >>You can use either the Logistic Regression procedure or the >> >> > Multinomial Logistic Regression procedure available in the >> Regression>> > add-on module. >> >> > >> >> > What is the Regression add-on module? What version does this >> apply to? >> >> Will >> >> > this module truly do a multilevel logistic model? When Graham >> said '... >> >> > people being nested within >> >> > > counties.' I had understood that to imply a multilevel model. >> >> > >> >> > Thanks, Gene Maguin >> >> > >> >> > >> >> > ===================== >> >> > 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 >> >> >> > >> > >> > >> > -- >> > Pablo Mora, Ph.D. >> > 30 College Avenue >> > The Institute for Health >> > Rutgers University >> > New Brunswick, NJ 08901-1293 >> > Office Phone: (732) 932-1941 FAX: (732) 932-1945 >> > >> > ===================== >> > 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 |
|
Check this out:
http://www2.chass.ncsu.edu/garson/PA765/logistic.htm |I'm very interested in the possibility of doing multilevel |logistic regression on SPSS. Have you got som reference to |instructions for doing that kind of analysis with the |procedure for 'general estimating equations'? | ===================== 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 |
«
Return to SPSSX Discussion
|
1 view|%1 views
| Free forum by Nabble | Edit this page |