*2.2. Data Collection and Analysis*

Due to the characteristics of subjects, convenience sampling in schools for NK refugee children was conducted.

Participants were recruited from three schools in Seoul and Incheon, and the study was conducted with parents who volunteered to take part in the study. In-depth interviews were conducted between July and November 2018. As most of them were engaged in irregular economic activities, one-to-one or one-to-many interviews had to be done based on the participants' circumstances; each participant was interviewed three times. The qualitative interview was conducted in a comfortable atmosphere. The interview was conducted by one interviewer, but the data analysis process was cross-validated with co-author. Before the in-depth interview, the interviewer had enough time to form a rapport with the participants, and the in-depth interview questions were conducted with semi-structured questionnaires such as 'How is your relationship with your child' and 'What does your child mean to you?' The interviewer added meaningful questions to suit the interview situation. The transcript was prepared within 2–3 days of the interview, and nonverbal elements such as facial expressions and tone found during the interview were additionally recorded. IRB approval was obtained from the ethics committee of the Yonsei University Institute Review Board (IRB No. 7001988-201810-HR-408-04.

Qualitative case studies were selected as the research method. Qualitative case studies are a way of identifying relationships within specific contexts based on an understanding of cases (Stake, 1995). This method was considered appropriate for this study because it seeks to build an in-depth understanding of the experiences of NK refugees living in South Korea.

The data analysis employed the analytic strategies of 'Categorical Aggregation or Direct Interpretations' presented by [28] Stake, with meaningful statements from the recorded data sorted into common themes and aggregated, thereby creating similar categories based on links between them. Triangulation was conducted among the researchers to enhance the validity of the analysis, with expert advice consulted in an effort to increase the accuracy of the content.
