Hi Lara,
If time can be considered as a categorical variable (e.g., baseline, visit after 3 months and visit after one year), you could create a specific intercept for each combination of time category, group and gender. For example, one of them would be an intercept for male in Group 1 at baseline.
It time must be considered as a continuous variable, you can have a model with, first, a specific intercept for each pair of group and gender (e.g., male in Group 1, female in Group 1, male in Group 2 and female in Group 2). Then, for the interaction with time, you can multiply the time covariate by each of the specific group/gender intercept to model the interaction. Nevertheless, please note that if the time covariate is varying across subjects, it is also recommended to first split it into two components, one capturing the between-subject variation and one capturing the within-subject variation before multiplying these two components by the intercepts.
Hope this helps,
Bryan
> On 9 Jul 2020, at 22:27, Lara Foland-Ross <[log in to unmask]> wrote:
>
> Hi Bryan,
>
> Could you tell me how to conduct a 3 way interaction (group by time by sex) using SWE?
>
> thanks,
> Lara
>
> ########################################################################
>
> To unsubscribe from the FSL list, click the following link:
> https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1
>
> This message was issued to members of www.jiscmail.ac.uk/FSL, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
########################################################################
To unsubscribe from the FSL list, click the following link:
https://www.jiscmail.ac.uk/cgi-bin/WA-JISC.exe?SUBED1=FSL&A=1
This message was issued to members of www.jiscmail.ac.uk/FSL, a mailing list hosted by www.jiscmail.ac.uk, terms & conditions are available at https://www.jiscmail.ac.uk/policyandsecurity/
|