100 Chapter 4 Data analysis Subsequently, we performed an inductive content analysis (23,24) using the online software program Dedoose version 8.3.17. The first and last author developed a preliminary coding scheme based on the data of three, randomly chosen interviews and the structure of the Attitude, Social influence, SelfEfficacy model (17,18). The final coding scheme emerged during further analysis based on consensus. We grouped the codes into subthemes and themes by examining the commonalities, differences, and relationships within and among the interviews. After eleven interviews, we reached saturation on the level of themes and subthemes, but we analyzed two additional interviews for confirmation. We read the remaining seven interviews to check whether any codes or themes had been missed, and confirmed the stated themes. During the analysis, a theory emerged on midwifery styles towards interventions and was supported by the themes and subthemes (23). Findings All participating midwives worked in primary care in the Netherlands and they varied in age, place of education, years of experience, and midwifery practice characteristics (see Table 2). Five midwives were educated in Belgium, Switzerland or the United Kingdom, reflecting the overall educational background of midwives in the Netherlands (25). The following two main themes were evident in the data: attitude towards interventions and influences on midwives’ attitudes. Within the theme influences on midwives’ attitudes we found three subthemes: experiences in collaboration, trust and fear, and woman centeredness. Finally, an emerging theory on midwifery styles towards childbirth interventions is presented.
RkJQdWJsaXNoZXIy MTk4NDMw