This paper evaluates empirical methods for detecting the focus-sensitivity of clause-embedding predicates, with the objective of making them applicable to predicates within and across languages. Our positive proposal is a test based on a combination of inferential judgments and truth-value judgments. We compare this method with three alternatives: one based on truth-value judgments alone; another based on judgments of coherence or contradiction; and a third one based on inferential judgments alone. We conclude that our combined test is more successful than its alternatives in terms of cross-linguistic uniformity (i.e., it is applicable to different languages with as few language-specific elements involved as possible) and item generalizability (i.e., there is a clear recipe for how to carry out the test for any clause-embedding predicate).