SPSSX Discussion

Is there a way to adjust for...

Classic

List

Threaded

5 messages Options

ishaybarat

Is there a way to adjust for...

Is there a way to adjust for...

I study a population of 900 hip fracture elderly. This type of patient has extreme high mortality (10% at 30 days and 30% at 1 year). The strongest predictors for death are age and gender. These two overshadow any other possible predictor (like co-morbidity, complications etc).
Is there a way to adjust for these two predictors without the need of stratifying the data according to age and gender?

I used the SPSS logistic regression were the dependent variable is dead/alive and the rest are covariates. Method is Enter.

Thanks

*¸..• ´¨)) -:¦:- *
¸.•´ .
(( -:¦:- * Ishay * -:¦:-
´•.. ..•´
((¸¸.•´* -:¦:-

lori.andersen

Re: Is there a way to adjust for...

Enter the predictors in separate levels such as age and gender in level 1, then other predictors in a subsequent level. Then you can look at the effect of the other predictors beyond the other two, in effect "controlling" for them.

On Sat, Jan 21, 2012 at 1:00 PM, ishaybarat [via SPSSX Discussion] <[hidden email]> wrote:

Is there a way to adjust for...

I study a population of 900 hip fracture elderly. This type of patient has extreme high mortality (10% at 30 days and 30% at 1 year). The strongest predictors for death are age and gender. These two overshadow any other possible predictor (like co-morbidity, complications etc).
Is there a way to adjust for these two predictors without the need of stratifying the data according to age and gender?

I used the SPSS logistic regression were the dependent variable is dead/alive and the rest are covariates. Method is Enter.

Thanks

*¸..• ´¨)) -:¦:- *
¸.•´ .
(( -:¦:- * Ishay * -:¦:-
´•.. ..•´
((¸¸.•´* -:¦:-

If you reply to this email, your message will be added to the discussion below:
http://spssx-discussion.1045642.n5.nabble.com/Is-there-a-way-to-adjust-for-tp5163060p5163060.html

To start a new topic under SPSSX Discussion, email [hidden email]
To unsubscribe from SPSSX Discussion, click here.
NAML

statisticsdoc

Re: Is there a way to adjust for...

In reply to this post by ishaybarat

Ishay,

Enter the covariates in a hierarchical order using multiple Method Enter lines. On the first line, enter gender and age (I am assume that you have coded age as a continuous variable, otherwise enter the dummy codes that represent age categories).
Best,
Stephen Brand

www.StatisticsDoc.com

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ishaybarat
Sent: Saturday, January 21, 2012 1:00 PM
To: [hidden email]
Subject: Is there a way to adjust for...

Is there a way to adjust for...

I study a population of 900 hip fracture elderly. This type of patient has
extreme high mortality (10% at 30 days and 30% at 1 year). The strongest
predictors for death are age and gender. These two overshadow any other
possible predictor (like co-morbidity, complications etc).
Is there a way to adjust for these two predictors without the need of
stratifying the data according to age and gender?

I used the SPSS logistic regression were the dependent variable is
dead/alive and the rest are covariates. Method is Enter.

Thanks

-----

*¸..• ´¨)) -:¦:- *
¸.•´ .
(( -:¦:- * Ishay * -:¦:-
´•.. ..•´
((¸¸.•´* -:¦:-

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Is-there-a-way-to-adjust-for-tp5163060p5163060.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

Raffe, Sydelle, SSA

Re: Is there a way to adjust for...

I've been out of the statistical world for a while, so perhaps someone can re-educate me. Aren't these data right censored? Is Survival Analysis (:) relevant here?

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of StatisticsDoc
Sent: Saturday, January 21, 2012 5:37 PM
To: [hidden email]
Subject: Re: Is there a way to adjust for...

Ishay,

Enter the covariates in a hierarchical order using multiple Method Enter lines. On the first line, enter gender and age (I am assume that you have coded age as a continuous variable, otherwise enter the dummy codes that represent age categories).
Best,
Stephen Brand

www.StatisticsDoc.com

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of ishaybarat
Sent: Saturday, January 21, 2012 1:00 PM
To: [hidden email]
Subject: Is there a way to adjust for...

Is there a way to adjust for...

I study a population of 900 hip fracture elderly. This type of patient has
extreme high mortality (10% at 30 days and 30% at 1 year). The strongest
predictors for death are age and gender. These two overshadow any other
possible predictor (like co-morbidity, complications etc).
Is there a way to adjust for these two predictors without the need of
stratifying the data according to age and gender?

I used the SPSS logistic regression were the dependent variable is
dead/alive and the rest are covariates. Method is Enter.

Thanks

-----

*¸... ´¨)) -:¦:- *
¸..´ .
(( -:¦:- * Ishay * -:¦:-
´... ...´
((¸¸..´* -:¦:-

--
View this message in context: http://spssx-discussion.1045642.n5.nabble.com/Is-there-a-way-to-adjust-for-tp5163060p5163060.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

statisticsdoc

Re: Is there a way to adjust for...

Sydelle,

I wondered about survival analysis as well. The choice between survival
analysis and logistic regression depends on the nature of the question Ishay
is asking. Reading between the lines, I assumed he was only interested in
predicting death over two specific time frames, but I agree that using
Survival Analysis might allow for a richer set of analyses if he is
interested in applying Cox Regression and Kaplan-Meier methods.

Best,

Stephen Brand

www.StatisticsDoc.com

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Raffe, Sydelle, SSA
Sent: Monday, January 23, 2012 12:50 PM
To: [hidden email]
Subject: Re: Is there a way to adjust for...

I've been out of the statistical world for a while, so perhaps someone can
re-educate me. Aren't these data right censored? Is Survival Analysis (:)
relevant here?

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
StatisticsDoc
Sent: Saturday, January 21, 2012 5:37 PM
To: [hidden email]
Subject: Re: Is there a way to adjust for...

Ishay,

Enter the covariates in a hierarchical order using multiple Method Enter
lines. On the first line, enter gender and age (I am assume that you have
coded age as a continuous variable, otherwise enter the dummy codes that
represent age categories).
Best,
Stephen Brand

www.StatisticsDoc.com

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
ishaybarat
Sent: Saturday, January 21, 2012 1:00 PM
To: [hidden email]
Subject: Is there a way to adjust for...

Is there a way to adjust for...

I study a population of 900 hip fracture elderly. This type of patient has
extreme high mortality (10% at 30 days and 30% at 1 year). The strongest
predictors for death are age and gender. These two overshadow any other
possible predictor (like co-morbidity, complications etc).
Is there a way to adjust for these two predictors without the need of
stratifying the data according to age and gender?

I used the SPSS logistic regression were the dependent variable is
dead/alive and the rest are covariates. Method is Enter.

Thanks

-----

*¸... ´¨)) -:¦:- *
¸..´ .
(( -:¦:- * Ishay * -:¦:-
´... ...´
((¸¸..´* -:¦:-

--
View this message in context:
http://spssx-discussion.1045642.n5.nabble.com/Is-there-a-way-to-adjust-for-t
p5163060p5163060.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