I'm not an expert on the IDI but I think the potential is there. The right variables would have to be in the data, but it should be possible to explore associational evidence of potential outcomes. For example, to see whether people who received a certain social service have better employment or health outcomes than those who didn’t. Not direct evidence of causality but potentially useful in combination with a theory of change, stakeholder interviews, Bradford Hill's criteria, Jane Davidson's 8 strategies (etc)
Hi Julian, how do you see the govt/SIA being able to track intervention outcomes via the IDI, when IDI data is fully anonymized?
I'm not an expert on the IDI but I think the potential is there. The right variables would have to be in the data, but it should be possible to explore associational evidence of potential outcomes. For example, to see whether people who received a certain social service have better employment or health outcomes than those who didn’t. Not direct evidence of causality but potentially useful in combination with a theory of change, stakeholder interviews, Bradford Hill's criteria, Jane Davidson's 8 strategies (etc)