not a problem to fix

plurality is not a problem to fix as much as it's a space to synthesise.

The problems to fix are far less important than the challenge of synthesisation.

there is no one model to rule them all. There is no one analytic framing to rule them all. There are many facets of the same problem. Many analytic framings that all inform. Each of which may or may not have been implemented correctly to some degree but that is not our focus, because the point is to challenge the one question == one model assumption.

Modelling, like metrics, like eda is just a tool. It's a tool that can be used in many ways. But when answering a domain question we need to consider an analytic space that encompasses data collection, data validation and exploration, models plural, and reporting. All of these are going to be influenced by analysts' experiences and opinions and that's okay.

Heterogeneity is not necessarily a problem to fix in fact it is just how science works!