This is a great point about making sensible assumptions. Too often I see evidence that people think that data analysis should be devoid of assumptions, and any

This is a great point about making sensible assumptions. Too often I see evidence that people think that data analysis should be devoid of assumptions, and any assumptions completely invalidate the analysis. In reality, almost all analysis will have some assumptions, and the good questions to ask is how plausible are the assumptions and what is the plausible impact of what happens if the assumptions are wrong.