Hello all,
I am undertaking a matched case control study with 1 case matched to 5 controls, and in order to perform the conditional logistic regression, I need to make a group ID that identifies the cases and their matched controls. However, I performed the matching with replacement, and therefore a control can be assigned to multiple cases.
For instance, if case "one" and its controls are assigned group_id “1”, and case two and its controls are assigned group_id “2”, but there is a case that is assigned to both cases 1 and 2, I would have to choose which case that control will go to and miss out on its contribution to the other case.
As far as I can tell, the variable referenced for the "group" component of the clogit syntax needs to be a unique identifier, that does not allow for what I'm trying to do. How do I accomplish this? Is there something I'm getting wrong?
Please help! I've never used the Statalist forum before and really hope you all can point me in the right direction.
Thanks,
Nick
I am undertaking a matched case control study with 1 case matched to 5 controls, and in order to perform the conditional logistic regression, I need to make a group ID that identifies the cases and their matched controls. However, I performed the matching with replacement, and therefore a control can be assigned to multiple cases.
For instance, if case "one" and its controls are assigned group_id “1”, and case two and its controls are assigned group_id “2”, but there is a case that is assigned to both cases 1 and 2, I would have to choose which case that control will go to and miss out on its contribution to the other case.
As far as I can tell, the variable referenced for the "group" component of the clogit syntax needs to be a unique identifier, that does not allow for what I'm trying to do. How do I accomplish this? Is there something I'm getting wrong?
Please help! I've never used the Statalist forum before and really hope you all can point me in the right direction.
Thanks,
Nick
Comment