In a changing society, schools have a great responsibility to help students to develop into critical thinkers. Given the importance of this mission, there is still a surprising amount of uncertainty about how to successfully teach critical thinking. Recurrently, reviews of research on critical thinking teaching interventions have pointed out shortcomings regarding 1) linking the development of students’ critical thinking to theories of teaching (McMillan, 1987), 2) creating intervention designs enabling causal inference (Behar-Horenstein & Niu, 2011; El Soufi & See, 2019), 3) using measurement that captures the broad skill of critical thinking (Niu et al., 2013; Tsui, 1998). In this study, we target these shortcomings by presenting an intervention study on a particular aspect of critical thinking, i.e. students’ ability to alternate points of view, asking:  What are the effects of a variation theory-based model for teaching civics, on 9^th grade students’ ability to alternate points of view?

Departing from the Delphi definition and the standard Ennis definition (1993), where critical thinking is defined as “reasonable reflective thinking focused on deciding what to believe or do”, we particularly targeted one feature that reoccurs in many of the most renowned definitions, the ability to alternate between points of view.

In order to understand students’ thoughts and arguments regarding such a critical thinking feature, there has been a call for using more essay-question material instead of prefabricated standardized multiple-choice tests (Niu et al., 2013; Tsui, 1998).      

The Swedish national test in social studies, funded by the Swedish National Agency for Education, uses essay questions to test students’ ability to change perspectives. As the test is developed and executed by the University of Gothenburg, the research team for this study had access to a nationwide sample of student tests, selected for documentation. The database includes 600 student tests per year, randomly selected. Consequently, the database creates a yearly cross-section sample of student answers to the national social studies questions, from the population of Swedish 15- to 16-year-olds. These student answers were the foundation for building the variation-theory teaching model.

One of the questions used was the following: In Sweden we have had two referendums in the last 20 years. What are the advantages and disadvantages of having a referendum? Reason and argue. To analyze and categorize student answers, we used the variation theory framework, identifying so-called, critical aspects of perspective change (Marton, 2015). For instance, a recurrent perspective regarding referendums among students was to make an argument that it is good for the individual citizen to be able to influence political decisions. However, some answers also included another perspective, noting that the government can benefit from understanding the public opinion on certain political issues. A third perspective included seeing the referendum from the perspective of people’s opinions, making mobilization possible. Another example of a question used in the study regarded voting age. The students answered a question about why the voting age in Sweden is 18 and not 21 or 16. In a similar way to the referendum question, students’ answers varied with regard to whether or not they were able to outline one a perspective and then move on to another.

The answers were categorized depending on how many of these changes in perspectives (0-3) were identified. A sample of answers (type values) of each category were selected. When developing the teaching material, these extracts from students’ answers (changing perspectives) were key components.

The design of the intervention was experimental, using an experiment group and a control group, and pre- and post-tests (Shadish et al., 2002). A regular medium performing 1-9 primary school in the area of Gothenburg, Sweden, partnered the project. All ninth-grade students, five classes of year 2020/2021, and all five social studies teacher, participated in the intervention. The key element of the experimental design was the random assignment of students to new classes. The students were informed that they were to have a “social studies theme week” that, during the first days (and seven lessons), included a teaching study developed in cooperation between their teachers and researchers from the university, and that they, during these days, would be split in to new classes. The random assignment of students into new classes followed standard procedure. Two of the five teachers were to take on the variation-theory-based teaching material (experimental teaching method) and was invited to three two-hour teacher-researcher meetings during the spring. The researchers also met the three teachers implementing the control teaching method to briefly discuss the content and arrangement of the seven lessons. The control teaching method, was designed to represent a normal, “business-as-usual” set up. Regarding measurements, the pre-test was the voting age task used in the intervention (described above). All student answered this question before the intervention began. In the post-test, the voting age task was asked again, i.e. targeting the students’ post intervention ability to change perspectives discussing the same issue as used in the teaching. The key issue measured when assessing the student responses to the task was the amount of alternations of point of view, i.e. perspective changes (0-3), each student did in their written responses. Yet, two additional essay tasks, not used in neither of the intervention groups, were added in the post-test, targeting generalization/transfer to the civic content. The two additional tasks dealt with i) pros and cons with using electronic ankle bracelet as penalty instead of prison and ii) what might influence a person’s life choices. The same type of analyses, as were made for the other tasks used in the study (voting age and referendums), were made for the tasks in the post-test.

The results from the intervention show significant mean differences in student ability to change perspectives, in favor of the experimental teaching (variation theory teaching). Students answered the essay question voting age before and after the teaching sequences. A 2 group (experimental n= 39, control n= 53) × 2 time (pre-, post-test) mixed design analysis of variance (ANOVA), with group as a between-subject factor and time as a within-group, was carried out. The result show that the students participating in the variation theory based teaching made on average 1.85 more perspective changes after the intervention compared to before. The control group students only increased by 0.44. There is a significant main effect of the treatment group intervention on students’ ability to change perspectives F (1, 90)= 27.58, p < 0.001, ηp2= 0.24 and a main effect of time F (1, 90)= 150, p < 0.001, ηp2= 0.62. These main effects was qualified by a group × time interaction, F (1, 90)= 57.53, p < 0.001, ηp2= 0.39 showing the effect of the intervention relative to the control condition from pre-to post-test. The other essay tasks electronic ankle bracelet, and influence on life choices were only post-tested. A one-way independent ANOVA analysis of variance was carried out and the results show smaller but significant effects of the treatment group intervention on students’ perspective change on these questions also F (1, 90)= 7.38, p < 0.008, ηp2= 0.08; F (1, 90)= 7.28, p < 0.001, ηp2= 0.07. Using Cohen’s (1988) interpretation of effect sizes (partial eta square), the results show a large effect of variation theory teaching compared to control group teaching on students ability to change perspective in the voting age essay task. On the other two post-test essay tasks, the effects were smaller, however in favor of the variation theory teaching.

Behar-Horenstein, L. S., & Niu, L. (2011). Teaching critical thinking skills in higher education: A review of the literature. Journal of College Teaching & Learning (TLC), 8(2). Cohen, J. (1988). Statistical Power Analysis for the Behavioral Sciences (2nd ed.) New York: Academic Press. Ennis, R. H. (1993). Critical thinking assessment. Theory into practice, 32(3), 179-186. El Soufi, N., & See, B. H. (2019). Does explicit teaching of critical thinking improve critical thinking skills of English language learners in higher education? A critical review of causal evidence. Studies in educational evaluation, 60, 140-162. Marton, F. (2014). Necessary conditions of learning. New York: Routledge. McMillan, J. H. (1987). Enhancing college students' critical thinking: A review of studies. Research in higher education, 26(1), 3-29. Niu, L., Behar-Horenstein, L. S., & Garvan, C. W. (2013). Do instructional interventions influence college students’ critical thinking skills? A meta-analysis. Educational research review, 9, 114-128. Lai, E. R. (2011). Critical thinking: A literature review. Pearson's Research Reports, 6, 40-41. Shadish, W., Cook, T. & Campbell, D. (2002) Experimental and Quasi- experimental Designs for Generalized Causal Inference. Wadsworth Cengage Learning. Tsui, L. (1998). Fostering critical thinking in college students: A mixed-methods study of influences inside and outside of the classroom. University of California, Los Angeles.