PLOS ONE Undocumented: An examination of legal identity and education provision for children in Malaysia Table 1. Characteristics of the study participants (n = 33). Respondents’ primary role Label No. Community organiser CO 4 Former students1 FS 7 Education provider2 EP 11 Policymaker POL 4 Researcher RES 7 Sabah Overall Malaysia Total 33 Non-citizen type Peninsular Malaysia Overall–non-citizens 5 Refugees 15 Stateless 3 Migrant3 Total 18 4 2 3 1 7 8 1 All the former students interviewed were adult refugees. Of the 7 interviewed, 3 were also education providers. 2 Of the 11 education providers interviewed, 7 also identified themselves as community organisers. https://doi.org/10.1371/journal.pone.0263404.t001 by at least 2 researchers, with one researcher leading and the other observing and taking field notes. Interviews averaged from 1 to 1.5 hours in length and were conducted either in English or Bahasa Malaysia (Malay language) depending on the participants’ preference. All sessions were recorded and audios were transcribed verbatim. Transcripts were coded into emerging themes using NVivo 12 Plus (QSR International, Melbourne, Australia) and quotations were extracted into Microsoft1 Excel1 for Office 365 (Microsoft, Redmond, WA, USA). Interviews in Bahasa Malaysia were analysed in the same language, while quotes extracted were translated to English for publication. We conducted a framework analysis; a qualitative methodology suited to applied policy research. Findings from the desk review and in-depth interviews were analysed using five steps: familiarisation, identifying a thematic framework, indexing, charting, and interpretation [32]. This descriptive analysis allowed us to categorise and contextualise undocumented children at risk of education deprivation by location (Overall Malaysia, Peninsular Malaysia, and Sabah, East Malaysia) and legal identities, define the concept of being ‘undocumented’, map the types of education provision and provide an analysis of key issues and policies on education provision that are specific or overlapping for each group of children. We triangulated our findings with desk research, incorporated perspectives of informed stakeholders from the interviews, and carried out bi-weekly discussions with the research team. Emerging themes from earlier interviews were probed further and validated in subsequent interviews. Results are presented with minimum quotations. Ethics Informed consent was obtained before commencing interviews. Participants were informed that study participation was voluntary, and at any point would be free to refuse to answer questions or terminate interviews. All participants agreed to be audio recorded and quoted anonymously in publications. Audio recordings and electronic transcripts were stored in secure data servers. This study was approved by the Medical Ethics Committee, University Malaya Medical Center (Approval numbers: UM.TNC 2/UMREC). PLOS ONE | https://doi.org/10.1371/journal.pone.0263404 February 2, 2022 4 / 26