help
help > RE: Questions about setup & within subject design
Feb 14, 2019 06:02 PM | Alfonso Nieto-Castanon - Boston University
RE: Questions about setup & within subject design
Hi Aisling,
My impression is that the issue here may be that you may be specifying 2 as the number of sessions for every subject, irrespective of whether they truly were scanned twice or not, and then trying to simply leave empty the data for the missing sessions. Instead, what you need to do (sorry if this is a bit more complicated) is:
1) enter in the Setup.Basic "number of sessions" field the actual number of sessions/scans for each subject (if a subject was scanned twice then enter 2, if he/she was scanned only at T1 or only at T2 then enter 1, as you only have one functional dataset for these subjects)
2) enter in Setup.Functionals the actual data for each subject (no missing data there). If a subject was scanned twice, then enter in Session1 (his/her first acquisition) the T1 data and in Session2 (his/her second acquisition) the T2 data. If a subject was only scanned in T1 then enter in Session1 (his/her first and only acquisition) the T1 data, and if a subject was only scanned in T2 then enter in Session1 (his/her first and only acquisition) the T2 data instead.
3) define in Setup.Conditions two conditions, one called T1 and another called T2, and specify there the association between each subject "sessions" and these conditions. If a subject was scanned twice, then associate T1 with session1 and T2 with session2. If a subject was only scanned in T1, then associate session1 with T1 and leave T2 empty (missing data). If a subject was only scanned in T2, then associate session1 with T2 and leave T1 empty (missing data)
The logic behind all this is that, in CONN, "sessions" are absolutely independent between subjects (i.e. session1 from one subject does not need to be in any way associated with or be similar to session1 from another subject). It is only "conditions" which, in CONN, are going to be treated jointly across subjects (e.g. anything that you label as T1 will be used to compute T1-specific connectivity measures, and the same for T2). This has the advantage of allowing arbitrarily complex acquisition sequences (e.g. T1 might consist on different number of runs/sessions for different subjects), and it also facilitates adding new data to an existing dataset (e.g. if at some point you have some T3 or even T0 data for a combination of a subset of these subjects and other new subjects that information can easily be added to your project without having to modify or re-compute any of the existing data there). Hope this makes sense
Best
Alfonso
Originally posted by Aisling O'Neill:
My impression is that the issue here may be that you may be specifying 2 as the number of sessions for every subject, irrespective of whether they truly were scanned twice or not, and then trying to simply leave empty the data for the missing sessions. Instead, what you need to do (sorry if this is a bit more complicated) is:
1) enter in the Setup.Basic "number of sessions" field the actual number of sessions/scans for each subject (if a subject was scanned twice then enter 2, if he/she was scanned only at T1 or only at T2 then enter 1, as you only have one functional dataset for these subjects)
2) enter in Setup.Functionals the actual data for each subject (no missing data there). If a subject was scanned twice, then enter in Session1 (his/her first acquisition) the T1 data and in Session2 (his/her second acquisition) the T2 data. If a subject was only scanned in T1 then enter in Session1 (his/her first and only acquisition) the T1 data, and if a subject was only scanned in T2 then enter in Session1 (his/her first and only acquisition) the T2 data instead.
3) define in Setup.Conditions two conditions, one called T1 and another called T2, and specify there the association between each subject "sessions" and these conditions. If a subject was scanned twice, then associate T1 with session1 and T2 with session2. If a subject was only scanned in T1, then associate session1 with T1 and leave T2 empty (missing data). If a subject was only scanned in T2, then associate session1 with T2 and leave T1 empty (missing data)
The logic behind all this is that, in CONN, "sessions" are absolutely independent between subjects (i.e. session1 from one subject does not need to be in any way associated with or be similar to session1 from another subject). It is only "conditions" which, in CONN, are going to be treated jointly across subjects (e.g. anything that you label as T1 will be used to compute T1-specific connectivity measures, and the same for T2). This has the advantage of allowing arbitrarily complex acquisition sequences (e.g. T1 might consist on different number of runs/sessions for different subjects), and it also facilitates adding new data to an existing dataset (e.g. if at some point you have some T3 or even T0 data for a combination of a subset of these subjects and other new subjects that information can easily be added to your project without having to modify or re-compute any of the existing data there). Hope this makes sense
Best
Alfonso
Originally posted by Aisling O'Neill:
Hi,
Did you figure this out? I have a dataset where some participants are missing timepoint 2 or timepoint 1. In the conditions section, after selecting "allow missing data", the participants who are missing T1 appear fine, with a note in the design display saying that they have no data for T1. However, for the participants who are missing T2, I just get a message saying display not available for both timepoints.
Attached is the conditions section I get for a subject who has data for T1, but not T2.
Any help appreciated!
A
Did you figure this out? I have a dataset where some participants are missing timepoint 2 or timepoint 1. In the conditions section, after selecting "allow missing data", the participants who are missing T1 appear fine, with a note in the design display saying that they have no data for T1. However, for the participants who are missing T2, I just get a message saying display not available for both timepoints.
Attached is the conditions section I get for a subject who has data for T1, but not T2.
Any help appreciated!
A
Threaded View
| Title | Author | Date |
|---|---|---|
| Darren Gitelman | Mar 7, 2012 | |
| Tamara Sussman | Feb 23, 2018 | |
| Aisling O'Neill | Feb 14, 2019 | |
| Alfonso Nieto-Castanon | Feb 14, 2019 | |
| Aisling O'Neill | Feb 14, 2019 | |
| Alfonso Nieto-Castanon | Mar 7, 2012 | |
| Darren Gitelman | Mar 7, 2012 | |
| Alfonso Nieto-Castanon | Mar 8, 2012 | |