We utilised semi-structured interviews as our primary
means of data collection (see the interview questions
in Appendix II and two interview excerpts in Appendix
III). In the original conceptualisation of the study we
also wanted samples of written communication from
teachers and from the schools or authorities. We did
find some private schools issuing interesting pieces
of correspondence. However, we quickly discovered
that in the public sector obtaining such pieces of
correspondence between homes and schools was
very problematic. Systematic access to samples of
written correspondence proved to be impossible. We
also discovered that in some private schools teachers
had not experienced the local context to a sufficient
degree for them to be able to adequately explain what
had been generated by the school.