Hanna de Jong-Markus

Chapter 4 74 several authors (Afdal, 2006; De Ruyter & Kole, 2001; Versteegt, 2010). The interview guide was discussed in the research group and was adapted after three pilot interviews. The questions were mainly about the meaning of the school’s mono-religious characteristic, the teachers’ professional ideals, the teachers and pupils’ experiences with religious others, and the teachers’ opinions about this (see Appendix [IX]). The interviews were recorded and transcribed verbatim. This provided 128–228 min of interview data per participant (173 min on average). 4.6 Data Analysis Our analysis followed Elliott and Timulak’s (2005) general framework for descriptive/ interpretive qualitative research. This form of thematic analysis (Braun & Clarke, 2006) starts with initial reading and editing, followed by dividing the data intomeaningful units and finding an overall organizational structure by assigning domains. Our domains were distinguished on the basis of the units’ objects.39 Part of the domain headings were categories of the topic list (e.g., “Tolerance” and “Religious Others”), but another part was new (e.g., “Refugees”). Then, the meaning units within a specific domain were coded with descriptive coding (Miles et al., 2014) and categorized. Because of our current research question, we then had three strategies. First, we focused on looking for similarities and regularities within the subcategory “School Choice” of the domain “Teacher.” Doing so, we assigned more abstract categories (Elliott &Timulak, 2005). Second, inorder tobetter interpret the outcomes of this first strategy, we examined the meaning units from the domains “Christian Schools” and “Homogenous Population.” Third, when we came to certain interpretations, we checked whether the data contained contradicting elements. If so, we mentioned this. We used the software program ATLAS.ti7,40 and we did the coding in Dutch (Van Nes et al., 2010). For the sake of reliability, the coding and further analyses were discussed several times within the research group and with other researchers. If statements from participants were used in this article, two authors translated these into intelligible English sentences (Van Nes et al., 2010). 39 We used the following domain headings (translated into English): Bible (44), Citizenship (Education) (84), Christian Schools (67), School Context (3), Didactics (97), Principal (5), Questioning/ inquisitiveness [Doorvragen] (15), Faith (400), Homogeneous Population (2), Religious Education (95), (Professional) Ideals (258), Interview (38), Class/Pupils (179), Teacher (900), Social Media (11), Religious Others (463), Parents (50), Government (7), Society (87), School (236), Teacher Team (101), Tolerance (196), and Refugees (21). The numbers in parentheses indicate how many units are within this domain; some units are assigned to more than one domain (counting performed on January 20, 2017). 40 www.atlasti.com