[TINCAN-XAPI statements] correctResponsesPattern for Numeric and Fill-in

Hi there, Bruno...welcome to Heroes! 

Thanks for your question.  For content published for Tin Can API, you're right; we only communicate that metadata for some question types.  This article explains in detail (with examples) of the data that is available for your LMS.  

The impact for you is that for question types where we don’t send correctResponsePattern, the LMS learns (1) what the question was, (2) what the learner answered, (3) whether the learner answered correctly, but not (4) what the correct answer was.

So while we're meeting xAPI standards, it sounds like you might have a need beyond that.  Would you be interested in sharing your thoughts with our development team on what that implementation would mean to you?  Thanks again for reaching out!