Differences between revisions 5 and 6
Deletions are marked like this. Additions are marked like this.
Line 6: Line 6:
For the example described [wiki:Self:FsTutorial/CorrectFsgdFile here] there were two classes (male, female) and one variable (age). The 3 regressors in the design matrix, when using DOSS, will be:
Line 7: Line 8:
The design matrix generated from the FSGD file consists of the last three columns of:
{{{
SubjectID Male Female Age
1. Intercept of Male
2. Intercept of Female
3. Age Slope
Line 11: Line 12:
   004 1 0 22
   008 0 1 88
   017 1 0 25
   021 1 0 22
   032 1 0 79
   039 0 1 69
   040 0 1 24
   045 0 1 78
   049 1 0 19
   067 1 0 71
   073 1 0 77
   074 1 0 85
   080 1 0 23
   084 1 0 21
   091 0 1 23
   092 1 0 29
   093 0 1 22
   095 0 1 70
   097 1 0 74
   099 0 1 93
   102 0 1 82
   103 0 1 66
   106 1 0 27
   108 1 0 28
   111 0 1 81
   114 0 1 87
   123 0 1 69
   124 0 1 30
   128 1 0 90
   129 0 1 65
   130 1 0 68
   131 0 1 80
   133 0 1 73
   136 0 1 78
   138 0 1 26
   140 0 1 18
   141 0 1 20
   144 1 0 78
   145 1 0 80
   149 1 0 77
}}}
Your contrast vector can be set up, using these regressors, to make any comparisons you'd want. If you wanted to compare the thickness differences between Males and Females, while regressing out age you could use a contrast vector [1 -1 0]. If you wanted to make a direct comparison between thickness and age, regardless of gender, you could use the contrast vector [0 0 1].
Line 53: Line 14:
There are three regressors. The first regressor codes membership in the Male group. The second codes for membership in the Female group. The last is the age for the subject. The contrast vector for testing the given subjects for straight correlation between thickness and age can be specified as:
{{{
[0 0 1]
}}}
The contrast vector has three components because there are three regressors.
If you are using DODS there will be 4 regressors in the design matrix:

1. Intercept of Male
2. Intercept of Female
3. Age Slope (male)
4. Age Slope (female)

Your contrast vector can be set up, using these regressors, to make any comparisons you'd want. If you wanted to compare the thickness differences between Males and Females, while regressing out age you could use a contrast vector [1 -1 0 0]. If you wanted to make a direct comparison between thickness and age between class, you could use the contrast vector [0 0 1 -1].

[wiki:FsTutorial top] | [wiki:FsTutorial/GroupAnalysis previous]

  • To follow this exercise exactly be sure you've downloaded the [wiki:FsTutorial/Data tutorial data set] before you begin. If you choose not to download the data set you can follow these instructions on your own data, but you will have to substitute your own specific paths and subject names.

Specify contrast vectors to test hypotheses

For the example described [wiki:FsTutorial/CorrectFsgdFile here] there were two classes (male, female) and one variable (age). The 3 regressors in the design matrix, when using DOSS, will be:

1. Intercept of Male 2. Intercept of Female 3. Age Slope

Your contrast vector can be set up, using these regressors, to make any comparisons you'd want. If you wanted to compare the thickness differences between Males and Females, while regressing out age you could use a contrast vector [1 -1 0]. If you wanted to make a direct comparison between thickness and age, regardless of gender, you could use the contrast vector [0 0 1].

If you are using DODS there will be 4 regressors in the design matrix:

1. Intercept of Male 2. Intercept of Female 3. Age Slope (male) 4. Age Slope (female)

Your contrast vector can be set up, using these regressors, to make any comparisons you'd want. If you wanted to compare the thickness differences between Males and Females, while regressing out age you could use a contrast vector [1 -1 0 0]. If you wanted to make a direct comparison between thickness and age between class, you could use the contrast vector [0 0 1 -1].

To use this in your mri_glmfit command create a file that contains just the one line of your contrast vector using the text editor of your choice. Again, using an appropriate naming convention is a good idea. Make sure this contrast vector has been created for you and is called age.mat:

cd $FREESURFER_HOME/subjects/buckner_data/tutorial_subjs/group_analysis_tutorial/stats
cat age.mat

Create age.mat if it does not exist. It should contain this line:

0 0 1

As an additional exercise, try constructing the contrast vector for testing the same data for difference between males and females independent of age. [wiki:FsTutorial/MaleFemaleContrastSolution Click here for the answer.]

FsTutorial/CreateContrastVectors (last edited 2008-04-29 11:46:15 by localhost)