schools was ensured by use of pseudonyms on all documentation and by
ensuring safe and confidential storage of all project artifacts and data.
The approach to analysis followed a similar cycle of processes within each of
the research teams for the six science domains. Interviews were transcribed,
open-coded by pairs of researchers using a grounded-theory approach (Stauss
and Corbin, 1994) from which emerging themes were examined, tested for
validity across data sets by blind comparisons and finally used to write
detailed vignettes or case studies of each of the teachers.
For the blind comparisons, a sample of at least half the interviews was selected
randomly. These were independently coded and then analyses compared.
Coding accuracy was improved by writing theme code desciptors and, in some
cases, further coding through paired discussion and this resulted in raising
coding agreement rate above 70%.
Audio and videotapes of the CPD teacher meetings were not transcribed
always in full. Instead, they were listened to and summarized independently
by at least two researchers in each domain. This decision was partly due to the
large amount of such data and also because some of the quality, particularly on
the audiotapes made it difficult to hear every word spoken. Critical incidents
or noteworthy sections were identified and these were then transcribed and
added to the summaries. In some domains where the main data source for a
particular line of investigation involved teacher talk, researchers elected to do
full transcriptions and analyses of videotapes. This is evident in the
Knowledge Integration paper in this set.
Classroom observations were recorded through field notes and video. The
video was summarized and critical incidents and noteworthy sections added in
a similar process to that used for the recordings of the teacher meetings.
Coding of the teacher meeting summaries and interviews resulted in a number
of themes emerging. Analysis has involved cross-referencing and triangulation
of themes in data sets from audio- and video-recordings of the teacher
discussions in the CPD sessions, teacher interviews and the portfolio
documents and has involved comparisons of data sets across the domains.
Discussion of findings
This introductory paper is intended to map out the ground for the papers that follow
where more specific findings will be reported. The findings that we report are
therefore in more general terms across the six domains of the project and describe the
constraints and affordances that we encountered.
Finding 1: The CPD programmes provided opportunities for teacher reflection and
learning.
I really enjoyed turning up for the CPD sessions. The activities were good for getting
you thinking and working with like-minded teachers was a real bonus. You were
10