Dear Anderson,
I hope this email receives you well.
I hope you can enlighten me on whether my added design matrix is
correct? I am kind of getting stuck at this point, and some deadlines
are coming up...
Best wishes,
Brigitte
On 2018-11-20 11:50, Dr A.B.C. Dijkshoorn wrote:
> Dear Anderson,
>
> Thank you very much for your clear description! I included your tip of
> adding the behavioural component in an additional EV for the follow-up
> analysis. Could you please maybe check if I included this correctly (I
> added the file).
>
> With regards to the independence of the variables. We expect that the
> behavioural measurement should , according to our hypothesis, change
> over time because of an intervention between T1 and T2. We did correct
> for multiple testing (from control data) and in a seperate control
> group we saw that the behavioural measurements stayed constant over
> the same period of time.
>
>
> Best wishes,
> Brigitte
>
> On 2018-11-16 12:27, Anderson M. Winkler wrote:
>> Hi Aisha, hi Brigitte,
>>
>> That's more clear, thank you. For the structural analysis the design
>> would indeed use a paired t-test. You could also compute the
>> difference between after and before treatment and use that in a
>> one-sample t-test.
>>
>> The follow up analysis using would use the differences (computed as in
>> the option above) in FA vs differences in the behavioural measurement.
>> The design could also be akin to the paired t-test, with the
>> additional variable included for testing.
>>
>> However, if there are differences in the behavioural measurements over
>> time, this the whole analysis will fall into the "voodoo" category,
>> that is, it becomes a circular, invalid analysis.
>>
>> To make this more clear, you must not correlate a set of quantities A
>> (like FA) and a variable B (like time), then take the best
>> correlations in A and correlate these with C (like the behavioural
>> measurements), when C is not independent of B.
>>
>> All the best,
>>
>> Anderson
>>
>> On Wed, 14 Nov 2018 at 07:01, Dr A.B.C. Dijkshoorn <[log in to unmask]>
>> wrote:
>>
>>> Dear Andersson,
>>>
>>> Thank you very much for your response.
>>>
>>> With the structural analysis I mean whether we can observe a change
>>> in
>>> FA before-after intervention.
>>> With the structural-behavioral analysis I mean whether the change
>>> in FA
>>> (before-after intervention) is correlated to change in performance
>>> on a
>>> cognition test.
>>>
>>> In essence, are the paired t-test design matrixes different in the
>>> structural analysis (10 EVs, eye-shaped design matrix) from the
>>> structural - cognition analysis (put cognition data in 1EV column
>>> or use
>>> the eye-shaped design matrix?)
>>>
>>> Does this clarify your question?
>>>
>>> Best wishes,
>>> Aicha
>>>
>>> On 2018-11-14 11:47, Anderson M. Winkler wrote:
>>>> Hi Brigitte,
>>>>
>>>> I don't really understand what you mean by "struct-behav"
>>> analysis and
>>>> how that would differ from the "struct" analysis. Also I don't
>>>> understand what you mean by the "this" in your first paragraph
>>> (see
>>>> highlighted below). Could you clarify?
>>>>
>>>> Thanks!
>>>>
>>>> All the best,
>>>>
>>>> Anderson
>>>>
>>>> On Thu, 8 Nov 2018 at 07:13, Brigitte Evers <[log in to unmask]>
>>> wrote:
>>>>
>>>>> Dear FSLusers,
>>>>>
>>>>> I am fairly new to TBSS and would like to check if I am doing
>>> the
>>>>> correct thing! I am trying to do a paired t-test to look at
>>>>> structural changes in FA before and after an intervention in 9
>>>>> lesion patients. I then want to correlate THIS to performance on
>>> an
>>>>> cognition task. I am aware that I have 2 options (1) subtracting
>>> FA
>>>>> images post-intervention from pre-intervention with fslmaths (2)
>>>>> create design and contrast files according to FEAT paired t-test
>>>>> manual (we chose for the latter for visualisation purposes)
>>>>>
>>>>> My question is do I set up the struct-behav analysis matrix file
>>>>> the same as the structural analysis part (e.g. 10 EVs/columns)
>>> or do
>>>>> I just create 1 EV/column with cognition data (+additional
>>> columns
>>>>> for covariates) with the struct-behav analysis (and how do I
>>> know
>>>>> FSL "remembers" the within subjects group design of the
>>> structural
>>>>> part)?
>>>>>
>>>>> Any help you could offer would be highly appreciated!
>>>>>
>>>>> Best wishes,
>>>>> Brigitte
>>>>>
>>>>> For the structural analysis I have created contrast and matrix
>>>>> files according tp the FEAT manual:
>>>>> /NumWaves 10
>>>>> /NumPoints 18
>>>>> /PPheights 2.000000e+00 1.000000e+00
>>>>> 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
>>>>> 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
>>>>>
>>>>> /Matrix
>>>>> /NumWaves 10
>>>>> /NumPoints 18
>>>>> /PPheights 2.000000e+00 1.000000e+00
>>>>> 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
>>>>> 1.000000e+00 1.000000e+00 1.000000e+00 1.000000e+00
>>>>>
>>>>> /Matrix
>>>>> 1.000000e+00 1.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 1.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 1.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 1.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 1.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 1.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 1.000000e+00
>>>>> -1.000000e+00 1.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 1.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 1.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 1.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 1.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 1.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 1.000000e+00
>>>>>
>>>>> /NumWaves 1
>>>>> /NumPoints 18
>>>>>
>>>>> /Matrix
>>>>> 1
>>>>> 2
>>>>> 3
>>>>> 4
>>>>> 5
>>>>> 6
>>>>> 7
>>>>> 8
>>>>> 9
>>>>> 1
>>>>> 2
>>>>> 3
>>>>> 4
>>>>> 5
>>>>> 6
>>>>> 7
>>>>> 8
>>>>> 9
>>>>>
>>>>> /ContrastName1 preint - postint
>>>>> /ContrastName2 postint - preint
>>>>> /NumWaves 10
>>>>> /NumContrasts 2
>>>>> /PPheights 2.000000e+00 2.000000e+00
>>>>> /RequiredEffect 5.446 5.446
>>>>>
>>>>> /Matrix
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>> 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>>
>>>>> For the functional analysis I am doubting between these 2
>>> contrast
>>>>> matrixes (I used the same .grp file)
>>>>> (1)
>>>>> /NumWaves 1
>>>>> /NumPoints 18
>>>>> /PPheights 2.000000e+01
>>>>>
>>>>> /Matrix
>>>>> 1.900000e+01
>>>>> 1.800000e+01
>>>>> 2.000000e+01
>>>>> 1.700000e+01
>>>>> 1.900000e+01
>>>>> 1.900000e+01
>>>>> 1.800000e+01
>>>>> 2.000000e+01
>>>>> 1.600000e+01
>>>>> 5.000000e+00
>>>>> 7.000000e+00
>>>>> 4.000000e+00
>>>>> 7.000000e+00
>>>>> 8.000000e+00
>>>>> 5.000000e+00
>>>>> 7.000000e+00
>>>>> 8.000000e+00
>>>>> 2.000000e+00
>>>>>
>>>>> (2)
>>>>> /NumWaves 10
>>>>> /NumPoints 18
>>>>> /PPheights 2.000000e+00 5.000000e+00
>>>>> 6.000000e+00 6.000000e+00 6.000000e+00 6.000000e+00
>>>>> 6.000000e+00 6.000000e+00 6.000000e+00 5.000000e+00
>>>>>
>>>>> /Matrix
>>>>> 1.000000e+00 5.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 6.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 5.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 6.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 6.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 6.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 6.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 4.000000e+00 0.000000e+00
>>>>> 1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 5.000000e+00
>>>>> -1.000000e+00 4.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 6.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 6.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 5.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 6.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 6.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 6.000000e+00
>>>>> 0.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 6.000000e+00 0.000000e+00
>>>>> -1.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 0.000000e+00 0.000000e+00 0.000000e+00
>>>>> 0.000000e+00 5.000000e+00
>>>>>
>>>>>
>>>>
>>>
>> ########################################################################
>>>>>
>>>>> To unsubscribe from the FSL list, click the following link:
>>>>> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 [1]
>>> [1]
>>>>
>>>> -------------------------
>>>>
>>>> To unsubscribe from the FSL list, click the following link:
>>>> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 [1]
>>> [2]
>>>>
>>>> Links:
>>>> ------
>>>> [1]
>>> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 [2]
>>>> [2] https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
>>> [1]
>>>
>>>
>> ########################################################################
>>>
>>> To unsubscribe from the FSL list, click the following link:
>>> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 [1]
>>
>> -------------------------
>>
>> To unsubscribe from the FSL list, click the following link:
>> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1 [3]
>>
>> Links:
>> ------
>> [1] https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
>> [2] https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&amp;A=1
>> [3] https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
>
> ########################################################################
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/webadmin?SUBED1=FSL&A=1
|