When two answers look plausible, go back to the exact wording of the question and the evidence required to support each option. One choice often relies on an extra assumption.

Separate 'could be true' from 'is best supported by the data.' Multiple-choice assessments reward the latter.

If additional analysis is not producing new evidence, commit and move on rather than reopening the same debate repeatedly.

same issue. I started asking which option needs fewer assumptions.

if both can be true, q is usually asking which is BEST supported.