I used a content analysis approach to analyzing
the data (Marshall & Rossman, 2014), examining the
participants’ responses as telling their story with respect
to their participation in the SERSG. I developed a set
of a priori codes to analyze the data but left open the
possibility for emergent themes. The process of coding
the responses involved sharing the codes with a graduate
research assistant and having the assistant code the data.
To establish inter-rater reliability I also coded a portion
of the data to compare to the codings of the graduate
student. Given the nature of the responses and coding of
the data, I determined at least two rounds of comparison
were necessary to ensure coding consistency. At the
end of the second round of coding, we established an
acceptable level of inter-rater reliability of approximately
85% (Stemler, 2004), at which time the graduate research
assistant completed the data coding. It is important to
note that it was possible for a participant to share multiple
ideas in an item response that aligned to multiple
codes, therefore, frequencies of codes do not sum to the
number of participants, but rather the number of aligned
responses.