You can check Cronbach's alpha for realizability test... It might help you.

Hi, if semi structured interview is done as a qualitative research then I think the approach should be different.

You can establish validity of your data by collection from multiple settings. Allowing participants to go through verbatim transcribed data and confirming their inputs will be helpful

Triangulation should also be done

It can be data triangulation, investigator triangulation, theory triangulation or multiple triangulation

I don't know if I understand your question exactly. But I believe that you should validate the information collected via interview using other methods. Perhaps a content analysis (as presented by Bardin) is a way to validate the evidence reported by the interviewees.