continuous subgroup analysis

[Jennifer May]

Hi all,

I have run the following regression:
regress attendance i.allocation attendance_BL fsm numstudents, vce(cluster name)

Due to the variables I have access to, I want to conduct a continuous subgroup analysis and be able to conclude "in the intervention group, for each additional % of students eligible for FSM, attendance increases/decreases by ... "

Is it correct to add the continuous variable as a covariate to my regression or will I just be adjusting for the variable? Do I include an interaction term between allocation and the subgroup variable?

Any help appreciated.

[Bruce Weaver]

Hello Jennifer May. What is "continuous subgroup analysis"? Can you direct me to any good resources that explain it? Thanks.

[Jennifer May]

sorry continuous subgroup analysis may have been the wrong wording, I'd like to see if/how the effect of allocation on attendance varies by the % of students eligible for FSM (continuous variable)

[Jennifer May]

I think I need to include an interaction term between allocation and % FSM?

regress attendance i.allocation##c.fsm attendance_BL fsm numstudents, vce(cluster name)

[Bruce Weaver]

Ah, okay. In that case, you need to include the allocation x FSM interaction in your model. See the example(s) on these UCLA pages:

• https://stats.oarc.ucla.edu/stata/fa...tion-stata-12/
• https://stats.oarc.ucla.edu/stata/se...actions-stata/

Cheers,
Bruce

[Jennifer May]

Thank you for your help!