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
|