What are the established best practices for reporting the results of a linear model in journals in the social sciences?
I’m not sure if there are established best practices across social science as a whole - different fields have different professional associations that may have different approaches. Regardless, the reporting wholly depends on things like what the model is answering, the journal, etc. But I guess here are some things I personally like to see:
- Actual interpretation of coefficients in the units of the variables included (rather that “and this coefficient is X”)
- A discussion about the relative magnitude of coefficients compared to what we think is a “practically” significant change in y
- Acknowledgement of model fit statistics and what they mean for interpretation of results