Subscribe Bookmark RSS Feed

Dunnett’s tests in repeated measures

jaykim

Community Trekker

Joined:

Nov 1, 2011

Hi,

I have data collected from a repeated meaures study.  Each subject's physiological data were measured eight times and 18 subjects participated in the study. The first measurement will be used as a reference value and compared with other 7 measures. So, I would like to use Dunnett's test instead of using Tukey HSD. Is there any simple way to do it?

Thanks,

Jay

1 ACCEPTED SOLUTION

Accepted Solutions
Solution

You can use the Fit Y by X platform. Set measured variable as Y, sampling (1-8, Nominal or Ordinal) as X and Subject as Block. Then under the red triangle pick "Campare Means --> With Control, Dunnett's". You will be asked to specify which level that is treated as control.

Script example:

dt = Open( "$ENGLISH_SAMPLE_DATA/Blood Pressure.jmp" );

dt << Oneway(

          Y( :BP 8M ),

          X( :Dose ),

          Block( :Subject ),

          With Control( 1, {"Control"} ),

          Comparison Circles( 1 )

);


2 REPLIES
Solution

You can use the Fit Y by X platform. Set measured variable as Y, sampling (1-8, Nominal or Ordinal) as X and Subject as Block. Then under the red triangle pick "Campare Means --> With Control, Dunnett's". You will be asked to specify which level that is treated as control.

Script example:

dt = Open( "$ENGLISH_SAMPLE_DATA/Blood Pressure.jmp" );

dt << Oneway(

          Y( :BP 8M ),

          X( :Dose ),

          Block( :Subject ),

          With Control( 1, {"Control"} ),

          Comparison Circles( 1 )

);


jaykim

Community Trekker

Joined:

Nov 1, 2011

Thanks a lot!!