The analysis focused on generating profiles of PSTs’ understandings of NOS before and
after the STS course followed the explicit project based science. The questionnaire and interview
data were analysed using the analytical inductive model of qualitative data analysis approach. All
the questionnaires and interviews were treated separately. The entire data analysis was repeated
two times to increase the reliability of analysis. It was found that there was almost 80 per cent
agreement between the two sets of analyses (Lederman et al., 2002).
To generate each participant’s NOS profile, the PSTs’ responses to the questionnaire
were holistically analysed because there is not a restrictive one-to-one correspondence between
an item on the questionnaire and a target NOS aspect (Lederman et al. 2002). The holistic
approach in data analysis leads to more accurate profile generation for NOS understanding.
When a person can maintain the same understanding in many different contexts, it is valid to
infer that this person actually has this understanding
For the inferential statistic test, each participant’s response to each item of the
questionnaire was classified under three categories, such as ‘informed’, ‘mixed’ and ‘naïve’
based on the following model. Based on this classification, the nonparametric Wilcoxon test was
used to compare the pre and post instruction results. This test is used when the data is categorical
and the sample is paired.
Informed: If the response included conceptions consistent with the standpoint of current
science education literature about the NOS.
Mixed: If the response included both partially informed and naïve conceptions.
Naïve: If the response did not include conceptions consistent with the standpoint of
current science education literature about the NOS