The same hypothesis can reverse sign depending on the case population chosen

Evidence

Quite notably the sign for the power variable flips from positive to negative. They do not discuss this much, but it is very directly related to how one chooses the population of cases. — Gary Goertz, Social Science Concepts, 1512

Categorization

• SHARE: Worth sending to anyone running dyad-level IR regressions who treats the sample frame as a footnote.
• QUESTION: How many published, replicated findings in comparative politics would flip if the polity population were redrawn?