READ ME File For 'Dataset title' Dataset DOI: https://doi.org/10.5258/D3919 ReadMe Author: Shahad Alshehri, University of Southampton ORCID ID 0009-0007-4468-2807 This dataset supports the thesis entitled Evaluating Disaster Risk Management in Saudi Arabian Public Hospitals: Perceptions and Practices AWARDED BY: University of Southampton DATE OF AWARD: 2026 Date of data collection: 27-11-2022 / 02-02-2025 Information about geographic location of data collection: Saudi Arabia Licence: CC BY Related projects/Funders: PhD Thesis -------------------- DATA & FILE OVERVIEW -------------------- This dataset contains three files: Study One, Study Two, and Study Three. Each file corresponds to one study included in my thesis. File descriptions: Study One: Quantitative data, including participant information sheets (PIS), consent forms, external approvals from Saudi public hospitals, email templates, survey materials, ethics application, and data files in SPSS format. Study Two: Qualitative data, including IRB approvals from Saudi public hospitals, interview questions, consent forms, email templates, ethics application documents, participant information sheets, interview transcripts, and audit records for different Saudi regions. Study Three: Quantitative data, including PIS templates, combined PIS and consent forms, ethics application, survey materials, and data files in SPSS format. Relationship between files: These three files represent the three studies that form my thesis, with each file containing the data and supporting materials for one study. -------------------------- METHODOLOGICAL INFORMATION -------------------------- Description of methods used for collection/generation of data: This research investigates perceptions and practices of disaster risk management (DRM) in Saudi Arabian public hospitals. Data were collected across three studies conducted in four regions of Saudi Arabia (Southern, Western, Central, and Eastern). Study One and Study Three employed quantitative survey methods using Qualtrics to collect data from healthcare workers (Study One) and the general public (Study Three). Study Two used qualitative methods, involving semi-structured interviews with healthcare workers. All instruments (surveys and interview guides) were developed based on existing literature and research objectives. Surveys were pilot-tested prior to distribution. Ethical approval was obtained from the University of Southampton and relevant Saudi public hospitals (IRB approvals). Methods for processing the data: Quantitative survey data were exported from Qualtrics in CSV/Excel format and analysed using IBM SPSS Statistics. Data cleaning included removal of incomplete responses, consistency checks, and anonymisation. Qualitative interview data were transcribed verbatim and anonymised at the transcription stage. Thematic analysis was conducted using NVivo, with systematic coding and theme development. Only cleaned and final datasets were used for analysis and submission. Software- or instrument-specific information needed to interpret the data: Qualtrics (online survey platform) for data collection IBM SPSS Statistics (for quantitative data analysis) NVivo (for qualitative data coding and analysis) Microsoft Excel and Word (for data organisation and documentation) Standards and calibration information: Survey instruments were pilot-tested to ensure clarity and reliability. Translation of materials (where applicable) was verified by authorised translators to maintain accuracy. Environmental/experimental conditions: Data collection was conducted remotely. Surveys were distributed online to participants in Saudi public hospitals and the general public. Interviews were conducted virtually or via approved communication methods, depending on participant availability. Quality-assurance procedures: Several procedures were implemented to ensure data quality: Pilot testing of survey instruments Verification of translations Accuracy checks during transcription Systematic coding validation in NVivo Removal of incomplete or inconsistent survey responses Anonymisation of all datasets prior to analysis People involved with data collection, processing, and analysis: The research was conducted by Shahad (PhD researcher, University of Southampton), who was responsible for data collection, processing, analysis, and submission. Academic supervisors at the University of Southampton provided guidance and oversight throughout the research process. -------------------------- DATA-SPECIFIC INFORMATION -------------------------- Study One Number of variables: Approximately 45 variables (demographics + 40 Likert-scale items across DRM phases). Number of cases/rows: 542 responses from healthcare workers. Variable list: Demographics (age, role, experience, region) and DRM perception items across four phases: mitigation, preparedness, response, and recovery (1 = strongly disagree to 5 = strongly agree). Missing data codes: Incomplete responses removed; remaining missing values coded as system-missing. Specialised formats or abbreviations used: Likert scale (1–5); DRM = Disaster Risk Management; HCWs = Healthcare Workers. Date that the file was created: October 2022 Study Two Number of variables: Not applicable (qualitative data). Number of cases/rows: 24 interview transcripts. Variable list: Textual data covering themes such as organisational culture, preparedness, leadership, communication, and regional differences. Participants coded as P01–P24. Missing data codes: Not applicable. Specialised formats or abbreviations used: Participant codes (e.g., P01); thematic coding labels. Date that the file was created: December 2023 Study Three Number of variables: 5 main variables measured using Likert scales (each with multiple items). Number of cases/rows: 436 responses from the general public. Variable list: Perceptions of preparedness, awareness, communication effectiveness, satisfaction, and trust (1 = strongly disagree to 5 = strongly agree). Missing data codes: Missing responses coded as system-missing; incomplete responses excluded. Specialised formats or abbreviations used: Likert scale (1–5); reverse-coded items applied where necessary. Date that the file was created: February 2025