Three groups setup

[Anna Kannonier]

Hi Frank,

I have three groups (two intervention groups and one control group) and three scan time points (pre-, mid- and post-intervention). I have already created three separate databases for comparing the different time points.

What would be the best way to handle the three groups for pairwise comparisons?

Create separate databases again? or how would you set up the demographics?  

Thank you!

Anna

[Frank Yeh]

Hi Anna,

For the cross-sectional analysis, the best approach is to create a single database that includes all subjects. You can then use the cohort selection feature to include or exclude specific groups as needed for your comparisons.

Regarding the pre- and post-intervention comparisons, I recommend first computing the paired differences using the [Step C2] interface. This will allow you to generate the necessary difference maps before proceeding with further statistical analysis.

I hope this helps clarify the setup. Please let me know if you have any other questions.

Best regards,

Frank

[Anna Kannonier]

Thank you for your response!

I computed the difference maps and compared groups (group 1 vs. group 2) using correlational tractography, with "group" as the study variable. Does this calculate the longitudinal change correlated with group 1 relative to group 2, and vice versa, independently for each group and for both directions (increase/decrease)? We are not sure whether the findings reflect decreases/increases relative to which group.

Best wishes,
Anna

[Frank Yeh]

I computed the difference maps

I assume you get a new database storing the longitudinal changes here.

 

and compared groups (group 1 vs. group 2) using correlational tractography, with "group" as the study variable. Does this calculate the longitudinal change correlated with group 1 relative to group 2, and vice versa, independently for each group and for both directions (increase/decrease)?

I am not sure what "independently for each group" means here.

Here you are correlating subjects' longitudinal changes with the group variable.

 

We are not sure whether the findings reflect decreases/increases relative to which group.

Best wishes,
Anna

Frank Yeh schrieb am Mittwoch, 8. April 2026 um 14:58:00 UTC+2:

Hi Anna,

For the cross-sectional analysis, the best approach is to create a single database that includes all subjects. You can then use the cohort selection feature to include or exclude specific groups as needed for your comparisons.

Regarding the pre- and post-intervention comparisons, I recommend first computing the paired differences using the [Step C2] interface. This will allow you to generate the necessary difference maps before proceeding with further statistical analysis.

I hope this helps clarify the setup. Please let me know if you have any other questions.

Best regards,

Frank

--
You received this message because you are subscribed to the Google Groups "DSI Studio" group.
To unsubscribe from this group and stop receiving emails from it, send an email to dsi-studio+...@googlegroups.com.
To view this discussion visit https://groups.google.com/d/msgid/dsi-studio/16dceefe-7a20-4db7-83d9-f70a379ab52an%40googlegroups.com.