Instrument

To explore science teachers’ conceptions of NOS, we employed an instrument called the Myths of Science Questionnaire (MOSQ). The MOSQ was successfully employed to explore pre-service science teachers’ conceptions of NOS in an Asian context (Buaraphan & Sung-Ong, 2009); we therefore assumed that it would also be applicable to Bangladeshi respondents. The MOSQ (see the Appendix) consists of 14 items and addresses three NOS aspects considered in this research: (a) scientific knowledge (8 items—Items 1, 2, 3, 4, 8, 9, 10, 11); (b) scientific inquiry (3 items—Items 5, 6, 7); and (c) scientific enterprise (3 items—Items 12, 13, 14). In the MOSQ, respondents are required to select which of three responses, i.e., “agree”, “uncertain”, or “disagree”, best fits their opinion on the item statement. They were also asked to provide an additional written explanation to support their selection. The MOSQ, therefore, collects the explanatory data in conjunction with the responses.

We discussed the relevance and suitability of the MOSQ instrument with two Bangladeshi science educators in order to maintain validity. A translated version (in Bengali) of the MOSQ was used as we anticipated that participants will feel more comfortable both in understanding the item statements and explaining the reason for choice in their own language. In order to resolve any translation issue, we at first translated the questionnaire individually, and then discussed it as a group. After a thorough discussion, we came to a consensus on our translation. This translated version was then piloted with 10 science teachers to determine whether they understood the items. Any ambiguities found during this piloting were clarified for the respondents and recorded for further revision of the MOSQ. This final revised version of the MOSQ was given to the participants. It had a Cronbach’s alpha reliability coefficient of 0.79, which suggested that the MOSQ items have relatively high internal consistency.

Participants

The first author of this paper arranged five workshops in different regions in Bangladesh for his PhD research. The workshop participants were science teachers. We invited these workshop participants to participate in this study as well. A total of 145 teachers, including 129 males and 16 females, voluntarily agreed to participate in this study.

Data collection and analysis

The volunteer participants were asked to respond to the final version of the MOSQ. They took 30-40 minutes to complete the questionnaire. All data was collected from February to March 2010.

We analysed the quantitative data using the Statistical Package for the Social Sciences (SPSS), version 15.0 for Windows. Frequencies of the responses (“agree”, “uncertain” and “disagree”) were counted and then presented in terms of percentage to interpret the results. For analysing the explanatory data that was qualitative in nature, we employed the data transformation procedure (Creswell, 2009). In this procedure, themes were identified from the qualitative data, and the frequency of the themes was counted and then presented in percentages.