Beyond Internal Validity: Field Notes from the Methodological Borderlands