Generally speaking, in psychometric assessments, interpretations are best left to the respondent; e.g. it is what ’significant amount’ means to them that is important.
Some terms will mean slightly different things to different people, but all are validated against a large sample of how the broad population responds to them, not only individually, but in combination.
This statistical validation, and the way we combine responses over multiple statements into a single scale, effectively accounts for differences in how individuals respond slightly differently to particular terms. No one question, phrase, or word is treated in isolation. If a user is unsure about an answer, they should pick the response that seems most appropriate and move on.
A single answer will not make a material difference to the overall assessment, and it can always be tweaked later if the user has misunderstood or reconsidered something. Provided users have understood the questions and answered them truthfully and instinctively, they should provide a stable and accurate measure.