SPSSX Discussion

fixed effects clarification

Classic

List

Threaded

4 messages Options

Paola Chivers-2

fixed effects clarification

HI,

Can anyone tell me what the '.' In the results table of fixed effects means
when running a mixed model (2nd last line). There are no reported errors in
my log output.

Type III Tests of Fixed
Effects(a)

Source

Numerator df

Denominator df

F

Sig.

Intercept

1

1580.041

602.651

.000

CatB

2

1353.582

1.666

.189

Sex

1

1344.267

2.709

.100

Age

1

1351.604

2.256

.133

Age2

1

1348.826

45.803

.000

AdvDis

1

1577.524

.036

.849

Age(Sex)

1

1351.898

1.931

.165

Age2(Sex)

1

1347.338

4.854

.028

Age(CatB)

2

1361.391

25.459

.000

Age2(CatB)

2

1357.819

136.505

.000

AdvDis(Sex)

1

1334.261

1.471

.225

AdvDis(CatB)

2

1344.145

1.126

.325

Age2 * AdvDis

0

.

.

.

Age * AdvDis

1

1348.308

.374

.541

Regards,

Paola

"Ours has become a time-poor society, fatigued by non-physical demands and
trying to compartmentalize daily living tasks. It is small wonder that
physical activity is discarded in this environment" p126 (Steinbeck, 2001)

P Please consider the environment before printing this email.

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

Swank, Paul R

Re: fixed effects clarification

Looks like that term has 0 degrees of freedom. I think we are going to
have to know what the model is and the nature of the variables.

Paul R. Swank, Ph.D.
Professor and Director of Research
Children's Learning Institute
University of Texas Health Science Center - Houston

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Paola Chivers
Sent: Monday, July 21, 2008 12:41 AM
To: [hidden email]
Subject: fixed effects clarification

HI,

Can anyone tell me what the '.' In the results table of fixed effects
means
when running a mixed model (2nd last line). There are no reported
errors in
my log output.

Type III Tests of Fixed
Effects(a)

Source

Numerator df

Denominator df

F

Sig.

Intercept

1

1580.041

602.651

.000

CatB

2

1353.582

1.666

.189

Sex

1

1344.267

2.709

.100

Age

1

1351.604

2.256

.133

Age2

1

1348.826

45.803

.000

AdvDis

1

1577.524

.036

.849

Age(Sex)

1

1351.898

1.931

.165

Age2(Sex)

1

1347.338

4.854

.028

Age(CatB)

2

1361.391

25.459

.000

Age2(CatB)

2

1357.819

136.505

.000

AdvDis(Sex)

1

1334.261

1.471

.225

AdvDis(CatB)

2

1344.145

1.126

.325

Age2 * AdvDis

0

.

.

.

Age * AdvDis

1

1348.308

.374

.541

Regards,

Paola

"Ours has become a time-poor society, fatigued by non-physical demands
and
trying to compartmentalize daily living tasks. It is small wonder that
physical activity is discarded in this environment" p126 (Steinbeck,
2001)

P Please consider the environment before printing this email.

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

Paola Chivers-2

Re: fixed effects clarification

HI,

Yes the problem is that it assigns 0 degrees of freedom when in fact it should be 1. In this particular model the interaction is NS, but I have had this occur a few times now with other covariates too, some of which are significant.

Here is further clarification.

The estimates of fixed effects for the age2*advdis which reported the '.' is:
Estimate 0.000000
Std error 0.000000
Df 1345.537
T=-1.325
Sig 0.185
Lower & upper bound 0.000000

MIXED
DV BY CatB Sex WITH Age Age2 AdvDis
/FIXED = CatB Sex Age Age2 AdvDis Age*sex Age2*Sex Age*CatB Age2*CatB AdvDis*Sex AdvDis*CatB Age2*AdvDis Age*AdvDis | SSTYPE(3)
/METHOD = REML
/PRINT = SOLUTION
/REPEATED=time | SUBJECT (studyno) COVTYPE (UN)
/SAVE=fixpred(pred2).

DV is continuous ranging from 10 - 40.
CAtB is 3 categorical groups: 1,2,3
Age is continuous in months ranging from 0-180
Age2 is age squared.
AdvDis is continuous 880-1207, mean 1021 sd 71

Regards,
Paola

“Ours has become a time-poor society, fatigued by non-physical demands and trying to compartmentalize daily living tasks. It is small wonder that physical activity is discarded in this environment” p126 (Steinbeck, 2001)

 Please consider the environment before printing this email.

-----Original Message-----
From: Swank, Paul R [mailto:[hidden email]]
Sent: Monday, 21 July 2008 11:35 PM
To: [hidden email]; [hidden email]
Subject: RE: fixed effects clarification

Looks like that term has 0 degrees of freedom. I think we are going to
have to know what the model is and the nature of the variables.

Paul R. Swank, Ph.D.
Professor and Director of Research
Children's Learning Institute
University of Texas Health Science Center - Houston

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Paola Chivers
Sent: Monday, July 21, 2008 12:41 AM
To: [hidden email]
Subject: fixed effects clarification

HI,

Can anyone tell me what the '.' In the results table of fixed effects
means
when running a mixed model (2nd last line). There are no reported
errors in
my log output.

Type III Tests of Fixed
Effects(a)

Source

Numerator df

Denominator df

F

Sig.

Intercept

1

1580.041

602.651

.000

CatB

2

1353.582

1.666

.189

Sex

1

1344.267

2.709

.100

Age

1

1351.604

2.256

.133

Age2

1

1348.826

45.803

.000

AdvDis

1

1577.524

.036

.849

Age(Sex)

1

1351.898

1.931

.165

Age2(Sex)

1

1347.338

4.854

.028

Age(CatB)

2

1361.391

25.459

.000

Age2(CatB)

2

1357.819

136.505

.000

AdvDis(Sex)

1

1334.261

1.471

.225

AdvDis(CatB)

2

1344.145

1.126

.325

Age2 * AdvDis

0

.

.

.

Age * AdvDis

1

1348.308

.374

.541

Regards,

Paola

"Ours has become a time-poor society, fatigued by non-physical demands
and
trying to compartmentalize daily living tasks. It is small wonder that
physical activity is discarded in this environment" p126 (Steinbeck,
2001)

P Please consider the environment before printing this email.

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

Matthew Pirritano

Re: fixed effects clarification

In reply to this post by Paola Chivers-2

I believe that this is due to too high of a correlation between variables in your model. I have seen this many times before in both linear regression and mixed models and that has always been the case. Two of your independent variables must be perfectly correlated. With all of your interaction effects you must either standardize or center the variables before you create the interaction effect or else the correlation between variables is potentially perfect.

matt
Matthew Pirritano, Ph.D.
Email: [hidden email]

----- Original Message ----
From: Paola Chivers <[hidden email]>
To: [hidden email]
Sent: Monday, July 21, 2008 4:56:48 PM
Subject: Re: fixed effects clarification

HI,

Yes the problem is that it assigns 0 degrees of freedom when in fact it should be 1. In this particular model the interaction is NS, but I have had this occur a few times now with other covariates too, some of which are significant.

Here is further clarification.

The estimates of fixed effects for the age2*advdis which reported the '.' is:
Estimate 0.000000
Std error 0.000000
Df 1345.537
T=-1.325
Sig 0.185
Lower & upper bound 0.000000

MIXED
DV BY CatB Sex WITH Age Age2 AdvDis
/FIXED = CatB Sex Age Age2 AdvDis Age*sex Age2*Sex Age*CatB Age2*CatB AdvDis*Sex AdvDis*CatB Age2*AdvDis Age*AdvDis | SSTYPE(3)
/METHOD = REML
/PRINT = SOLUTION
/REPEATED=time | SUBJECT (studyno) COVTYPE (UN)
/SAVE=fixpred(pred2).

DV is continuous ranging from 10 - 40.
CAtB is 3 categorical groups: 1,2,3
Age is continuous in months ranging from 0-180
Age2 is age squared.
AdvDis is continuous 880-1207, mean 1021 sd 71

Regards,
Paola

“Ours has become a time-poor society, fatigued by non-physical demands and trying to compartmentalize daily living tasks. It is small wonder that physical activity is discarded in this environment” p126 (Steinbeck, 2001)

 Please consider the environment before printing this email.

-----Original Message-----
From: Swank, Paul R [mailto:[hidden email]]
Sent: Monday, 21 July 2008 11:35 PM
To: [hidden email]; [hidden email]
Subject: RE: fixed effects clarification

Looks like that term has 0 degrees of freedom. I think we are going to
have to know what the model is and the nature of the variables.

Paul R. Swank, Ph.D.
Professor and Director of Research
Children's Learning Institute
University of Texas Health Science Center - Houston

-----Original Message-----
From: SPSSX(r) Discussion [mailto:[hidden email]] On Behalf Of
Paola Chivers
Sent: Monday, July 21, 2008 12:41 AM
To: [hidden email]
Subject: fixed effects clarification

HI,

Can anyone tell me what the '.' In the results table of fixed effects
means
when running a mixed model (2nd last line). There are no reported
errors in
my log output.

Type III Tests of Fixed
Effects(a)

Source

Numerator df

Denominator df

F

Sig.

Intercept

1

1580.041

602.651

.000

CatB

2

1353.582

1.666

.189

Sex

1

1344.267

2.709

.100

Age

1

1351.604

2.256

.133

Age2

1

1348.826

45.803

.000

AdvDis

1

1577.524

.036

.849

Age(Sex)

1

1351.898

1.931

.165

Age2(Sex)

1

1347.338

4.854

.028

Age(CatB)

2

1361.391

25.459

.000

Age2(CatB)

2

1357.819

136.505

.000

AdvDis(Sex)

1

1334.261

1.471

.225

AdvDis(CatB)

2

1344.145

1.126

.325

Age2 * AdvDis

0

.

.

.

Age * AdvDis

1

1348.308

.374

.541

Regards,

Paola

"Ours has become a time-poor society, fatigued by non-physical demands
and
trying to compartmentalize daily living tasks. It is small wonder that
physical activity is discarded in this environment" p126 (Steinbeck,
2001)

P Please consider the environment before printing this email.

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