Reliability of the findings presented here was strengthened through the
use of a repeated coding process. This process involved the author coding
each interview transcript on three separate occasions to ensure that each
efficacy influence had been coded to accurately represented the type of
experience (e.g. mastery, vicarious, etc.) being described in the interview
transcript. When discrepancies arose in the recoding process, Bandura’s
(1997) description of the sources of efficacy was referenced to allow for the
most accurate code to be assigned.