The effect of nature of science teaching based on the reconceptualized family resemblance approach on students' nature of science views and pseudoscientific beliefs

The purpose of this research is to investigate the impact of Nature of Science (NOS) instruction, conducted within a science–pseudoscience context and grounded in the Reconceptualized Family Resemblance Approach (FRA), on students' views of NOS and their pseudoscientific beliefs. The research employed an embedded mixed-methods design, in which quantitative and qualitative data were examined concurrently. The research group consisted of a total of 46 randomly selected students (22 in the experimental group and 24 in the control group) enrolled in a public middle school in Adıyaman and its Kahta district. Quantitative data were collected through the "FRA Student Questionnaire" and the "Pseudoscience Beliefs Scale." Qualitative data were gathered via pre- and post-intervention interviews as well as students' activity worksheets administered during the instructional process. A statistical software package was used for the analysis of quantitative data, while descriptive and content analysis methods were employed for qualitative data. The findings revealed that the experimental group demonstrated a statistically significant improvement in their understanding of NOS both compared to their own pre-test scores and to the control group. At the subcategory level, significant differences favoring the experimental group were observed in "methods and methodological rules," "scientific practices," and "social-institutional systems." However, no significant differences emerged in the categories of "aims and values" or "scientific knowledge." Contrary to the quantitative results, qualitative findings suggested potential improvement through increased codes and frequency counts. With respect to pseudoscientific beliefs, quantitative analyses indicated that the experimental group significantly reduced their pseudoscientific beliefs overall, both compared to their pre-test scores and to the control group. Subcategory analyses showed that significant differences occurred in the domains of "divination" and "health," while no significant difference was observed in the "paranormal" subdomain. Qualitative findings further suggested that students' awareness of pseudoscientific beliefs increased throughout the process. In conclusion, the research demonstrates that NOS instruction based on the FRA framework is effective in both enhancing students' understanding of the nature of science and reducing their pseudoscientific beliefs.