Culturally Equipped for Socio‐Scientific Issues? A comparative study on how teachers and students in mono‐ and multiethnic schools handle work with complex issues

Abstract

Socio‐scientific issues (SSI) are not only said to increase students’ interest in science, but they also strengthen the generic skills of teamwork, problem‐solving, and media literacy. At the same time, these skills are prerequisites for successful work with SSI. The aim of the study is to analyze what happens when SSI are implemented in science classrooms with various degrees of ethnic diversity and socio‐cultural status. We are also interested in knowing how teachers structure the SSI work from discourses on what suits different students. Quantitative and qualitative methods are combined, for example, questionnaires and ethnographic fieldwork, presented through partial least squares analysis and thick descriptions. We can notice discursive differences between ‘Us’ and ‘The Other’ and between mono‐ and multiethnic schools. In an earlier research, images of differences between the different student groups emerged, and we can find these in the results from the questionnaires. In an observation study, another pattern appeared that indicated similarities rather than differences between mono‐ and multiethnic classrooms. The students are first of all inside the discourse of ‘the successful student.’ Noteworthy is that the teachers’ roles correspond better with the discourse than with how students actually act. The study also shows that SSI articulate a collision between different discourses on education: a discourse on differences between students in multi‐ and monoethnic classrooms; a discourse on how to become a successful student; and a discourse on the school’s mission to educate participating citizens. It is suggested that schools should relate to, expose, and articulate discursive clashes that emerge when introducing new work forms.

Keywords:

• Discourse
• Multicultural
• Science education

Acknowledgements

The research project SISC is supported by Swedish Research Council (Grant No 2006‐12555). The authors wish to thank Eva Silfver and Margareta Ekborg for valuable feedback.

Notes

1. The project (2007–2010) is funded by the Swedish Research Council. The research group is called SISC (Science in Social Contexts) consists of Britt Lindahl and Maria Rosberg at Kristianstad University, Christina Ottander, Eva Silfver and Mikael Winberg at Umeå University, and Margareta Ekborg, Malin Ideland, Claes Malmberg and Agneta Rehn at Malmö University.

2. Residuals were checked for normality, hotellings were calculated (displayed as an elipse in the loading plots) to assess any undue leverage from observations with extreme positions in the direction of the components. Moderate outliers were detected by measuring the DmodX (i.e., the distance between observations and the model hyperplane in the multivariate variable space). Model validity was tested by response permutation procedures, that is, the predictor variables were kept intact, while responses (in this case, the degree of multiculturality) were permutated 20 times. After each permutation, a new model was calculated and compared to the original PLS model with respect to its descriptive and predictive ability.

3. A VIP value higher than 1 means the variable is more important than average, while a VIP below 0.5 indicates that the variable probably is unimportant.