longitudinal binary data question

[sgthomson99]

Hi everyone,

I am trying to find out if SPSS can analyze longitudinal binary data.  We
have two patient groups, and we are following them over 4 time points - and
the physician seeing each subject codes whether or not they are symptomatic
(yes or no) at each time point.  I know I can't do repeated measures ANOVA
on this because the data are binary.  Would this work in mixed models in
SPSS? I've tried clicking on mixed models, but I'm getting lost.

Any helpful hints greatly appreciated.

Thanks!

Susan Thomson, M.Sc.
Research Coordinator
Alberta Children's Hospital

_________________________________________________________________
Your Space. Your Friends. Your Stories. Share your world with Windows Live
Spaces. http://spaces.live.com/?mkt=en-ca

[Maguin, Eugene]

Susan,

14 can not do what you need. Perhaps somebody with 15 can say if it can.

Gene Maguin

[peter link]

Hi -

The MIXED procedure in SPSS can not handle binary outcomes.  You would need
to use different software such as HLM or MLwiN.  SAS or R may be able to
handle binary outcomes, as well, although I am not 100% sure they can.

Peter Link
VA San Diego Healthcare System

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of
S Crawford
Sent: Friday, March 30, 2007 2:12 PM
To: [hidden email]
Subject: longitudinal binary data question


Hi everyone,

I am trying to find out if SPSS can analyze longitudinal binary data.  We
have two patient groups, and we are following them over 4 time points - and
the physician seeing each subject codes whether or not they are symptomatic
(yes or no) at each time point.  I know I can't do repeated measures ANOVA
on this because the data are binary.  Would this work in mixed models in
SPSS? I've tried clicking on mixed models, but I'm getting lost.

Any helpful hints greatly appreciated.

Thanks!

Susan Thomson, M.Sc.
Research Coordinator
Alberta Children's Hospital

_________________________________________________________________
Your Space. Your Friends. Your Stories. Share your world with Windows Live
Spaces. http://spaces.live.com/?mkt=en-ca

[Dale Glaser]

However, with V 15.0  I believe the GEE option  is equipped to handle such models.......dale

peter link <[hidden email]> wrote:  Hi -

The MIXED procedure in SPSS can not handle binary outcomes. You would need
to use different software such as HLM or MLwiN. SAS or R may be able to
handle binary outcomes, as well, although I am not 100% sure they can.

Peter Link
VA San Diego Healthcare System

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of
S Crawford
Sent: Friday, March 30, 2007 2:12 PM
To: [hidden email]
Subject: longitudinal binary data question


Hi everyone,

I am trying to find out if SPSS can analyze longitudinal binary data. We
have two patient groups, and we are following them over 4 time points - and
the physician seeing each subject codes whether or not they are symptomatic
(yes or no) at each time point. I know I can't do repeated measures ANOVA
on this because the data are binary. Would this work in mixed models in
SPSS? I've tried clicking on mixed models, but I'm getting lost.

Any helpful hints greatly appreciated.

Thanks!

Susan Thomson, M.Sc.
Research Coordinator
Alberta Children's Hospital

_________________________________________________________________
Your Space. Your Friends. Your Stories. Share your world with Windows Live
Spaces. http://spaces.live.com/?mkt=en-ca



Dale Glaser, Ph.D.
Principal--Glaser Consulting
Lecturer/Adjunct Faculty--SDSU/USD/AIU
President-Elect, San Diego Chapter of
American Statistical Association
3115 4th Avenue
San Diego, CA 92103
phone: [phone]
fax: [phone]
email: [hidden email]
website: www.glaserconsult.com

[Kornbrot, Diana]

As far as I can make out mixed is only for continuous data, won¹t do binary
or ordinal
This is very serious omission in SPSS.

Try putting data in as for mixed with time as a categorical variable and
repeat subject numbers for each  time poitn
The run binary logit with both participant and time point as factors
Not ideal, as assumes ONLY binomial variance, not participant variance. So
you may get spurious significant effects as variance estimate is too small
BUT
If you save probabilities for each participant and then convert them back to
logits, you can estimate the extra binomial variance

Realy clutsy, I tried telling SPSS at their UL user conference, fell on deaf
ears
Best

diana


On 30/3/07 22:12, "S Crawford" <[hidden email]> wrote:

> Hi everyone,
>
> I am trying to find out if SPSS can analyze longitudinal binary data.  We
> have two patient groups, and we are following them over 4 time points - and
> the physician seeing each subject codes whether or not they are symptomatic
> (yes or no) at each time point.  I know I can't do repeated measures ANOVA
> on this because the data are binary.  Would this work in mixed models in
> SPSS? I've tried clicking on mixed models, but I'm getting lost.
>
> Any helpful hints greatly appreciated.
>
> Thanks!
>
> Susan Thomson, M.Sc.
> Research Coordinator
> Alberta Children's Hospital
>
> _________________________________________________________________
> Your Space. Your Friends. Your Stories. Share your world with Windows Live
> Spaces. http://spaces.live.com/?mkt=en-ca
>

Professor Diana Kornbrot
Evaluation Co-ordinator, Blended Learning Unit
University of Hertfordshire
College Lane, Hatfield, Hertfordshire AL10 9AB, UK

email:  [hidden email]
web:    http://web.mac.com/kornbrot/iweb/KornbrotHome.html
         
Blended Learning Unit
    voice [phone]
    fax    [phone]
Psychology        
    voice [phone]
    fax    [phone]
Home
19 Elmhurst Avenue
London N2 0LT, UK
   voice [phone]

[Reutter, Alex]

Dale is correct; in SPSS 15 from the menus, choose: Analyze > Generalized Linear Models > Generalized Estimating Equations...  (This runs GENLIN command syntax)

The dialog has a lot of options; while on the Repeated tab, click the Help button and then click the "Show Me" link in the help topic that pops up.  This opens the associated "Case Study" material, which briefly runs through an analysis of longitudinal binary data.

Alex


-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of Dale Glaser
Sent: Friday, March 30, 2007 4:40 PM
To: [hidden email]
Subject: Re: longitudinal binary data question

However, with V 15.0  I believe the GEE option  is equipped to handle such models.......dale

peter link <[hidden email]> wrote:  Hi -

The MIXED procedure in SPSS can not handle binary outcomes. You would need
to use different software such as HLM or MLwiN. SAS or R may be able to
handle binary outcomes, as well, although I am not 100% sure they can.

Peter Link
VA San Diego Healthcare System

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of
S Crawford
Sent: Friday, March 30, 2007 2:12 PM
To: [hidden email]
Subject: longitudinal binary data question


Hi everyone,

I am trying to find out if SPSS can analyze longitudinal binary data. We
have two patient groups, and we are following them over 4 time points - and
the physician seeing each subject codes whether or not they are symptomatic
(yes or no) at each time point. I know I can't do repeated measures ANOVA
on this because the data are binary. Would this work in mixed models in
SPSS? I've tried clicking on mixed models, but I'm getting lost.

Any helpful hints greatly appreciated.

Thanks!

Susan Thomson, M.Sc.
Research Coordinator
Alberta Children's Hospital

_________________________________________________________________
Your Space. Your Friends. Your Stories. Share your world with Windows Live
Spaces. http://spaces.live.com/?mkt=en-ca



Dale Glaser, Ph.D.
Principal--Glaser Consulting
Lecturer/Adjunct Faculty--SDSU/USD/AIU
President-Elect, San Diego Chapter of
American Statistical Association
3115 4th Avenue
San Diego, CA 92103
phone: [phone]
fax: [phone]
email: [hidden email]
website: www.glaserconsult.com

[Mark A Davenport MADAVENP]

HLM 6 allows non-linear models with a variety of outcomes (dichotomous,
counts, multinomial and ordinal).

***************************************************************************************************************************************************************
Mark A. Davenport Ph.D.
Senior Research Analyst
Office of Institutional Research
The University of North Carolina at Greensboro
336.256.0395
[hidden email]

'An approximate answer to the right question is worth a good deal more
than an exact answer to an approximate question.' --a paraphrase of J. W.
Tukey (1962)






Dale Glaser <[hidden email]>
Sent by: "SPSSX(r) Discussion" <[hidden email]>
03/30/2007 04:39 PM
Please respond to
Dale Glaser <[hidden email]>


To
[hidden email]
cc

Subject
Re: longitudinal binary data question






However, with V 15.0  I believe the GEE option  is equipped to handle such
models.......dale

peter link <[hidden email]> wrote:  Hi -

The MIXED procedure in SPSS can not handle binary outcomes. You would need
to use different software such as HLM or MLwiN. SAS or R may be able to
handle binary outcomes, as well, although I am not 100% sure they can.

Peter Link
VA San Diego Healthcare System

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]]On Behalf Of
S Crawford
Sent: Friday, March 30, 2007 2:12 PM
To: [hidden email]
Subject: longitudinal binary data question


Hi everyone,

I am trying to find out if SPSS can analyze longitudinal binary data. We
have two patient groups, and we are following them over 4 time points -
and
the physician seeing each subject codes whether or not they are
symptomatic
(yes or no) at each time point. I know I can't do repeated measures ANOVA
on this because the data are binary. Would this work in mixed models in
SPSS? I've tried clicking on mixed models, but I'm getting lost.

Any helpful hints greatly appreciated.

Thanks!

Susan Thomson, M.Sc.
Research Coordinator
Alberta Children's Hospital

_________________________________________________________________
Your Space. Your Friends. Your Stories. Share your world with Windows Live
Spaces. http://spaces.live.com/?mkt=en-ca



Dale Glaser, Ph.D.
Principal--Glaser Consulting
Lecturer/Adjunct Faculty--SDSU/USD/AIU
President-Elect, San Diego Chapter of
American Statistical Association
3115 4th Avenue
San Diego, CA 92103
phone: [phone]
fax: [phone]
email: [hidden email]
website: www.glaserconsult.com