Dataset for “From Simulation to Repeated Indigenous Community Engagement: A Longitudinal Mixed-Methods Evaluation of Medical Students’ Intercultural Sensitivity”
Description
This dataset contains de-identified quantitative data from a longitudinal mixed-methods educational study evaluating intercultural sensitivity and perceived preparedness among post-baccalaureate medical students participating in a staged curriculum integrating simulation-based preparation with repeated engagement in an Indigenous community in Taiwan. The complete educational cohort comprised 12 students. Phase 1 included a locally developed nine-item process-evaluation questionnaire administered at baseline, immediately after the initial simulation and community-learning sequence, and approximately three months later. Phase 2 included paired pre- and post-experience scores from the 24-item Chinese Intercultural Sensitivity Scale (ISS), including overall scores and five domains: interaction engagement, respect for cultural differences, interaction confidence, interaction enjoyment, and interaction attentiveness. The dataset is intended to support transparency and reproducibility of the quantitative analyses reported in the associated manuscript. Qualitative reflections are not included because of the potential risk of participant identification.
Files
Steps to reproduce
Data were analyzed using de-identified participant-level records. For Phase 1, calculate each participant’s mean score across the nine 5-point Likert items at baseline, immediately after the initial educational activity, and approximately three months later. Restrict paired comparisons to participants with complete data at all three time points (n = 11). Compare baseline versus immediate post-intervention, baseline versus three-month follow-up, and immediate post-intervention versus three-month follow-up using exact Wilcoxon signed-rank tests. For Phase 2, reverse-score negatively worded items of the 24-item Intercultural Sensitivity Scale (ISS) according to the established scoring framework. Calculate overall ISS and five domain mean-item scores (interaction engagement, respect for cultural differences, interaction confidence, interaction enjoyment, and interaction attentiveness) for each participant before and after the second community experience. Conduct paired pre-post comparisons using Wilcoxon signed-rank tests and calculate rank-biserial correlations as effect-size estimates. Calculate percentile bootstrap 95% confidence intervals for mean paired changes. Apply Holm adjustment across the five exploratory domain comparisons. Summarize participant-level changes as improved, unchanged, or declined. Changes of ≥0.25 and ≥0.50 mean-item points may additionally be summarized descriptively but should not be interpreted as validated minimal important differences. Qualitative reflection data are not included in the deposited dataset because of the potential risk of participant identification.
Institutions
- National Chung Hsing UniversityTaiwan, Taichung