Do Tourism Preferences Differ Across Self-Reported MBTI Personality Types? Evidence from Thai Travellers

Drawing on person–environment fit and self-congruity theory, this study examines whether tourism preferences and travel-related outcomes differ across 16 self-reported personality types and evaluates the potential contribution of personality information to offering-level market segmentation. A cross-sectional survey generated 502 valid responses from Thai travellers recruited through online and on-site convenience sampling. Pearson’s chi-square tests examined differences across personality types in interest in 18 tourism offerings, reported travel-expenditure categories, and preferred travel companions. The Benjamini–Hochberg procedure was applied to control the false discovery rate across the tourism-offering tests. Significant associations, with small-to-moderate effect sizes, were identified for selected nature-based offerings, comprising ecotourism, geotourism, agritourism, and astrotourism, and selected special-interest offerings, comprising ethnic tourism, adventure tourism, homestay and farm-stay tourism, and meetings, incentives, conferences, and exhibitions tourism. No significant associations were found for the cultural tourism offerings, travel-expenditure categories, or preferred travel companions. These findings indicate that differences across self-reported personality types are offering-specific and are not consistently evident across broader travel-related outcomes. Voluntarily provided personality information may therefore serve as a supplementary input for initial tourism recommendations when behavioural histories are unavailable. However, the results do not establish individual-level predictive accuracy, incremental recommendation value, or stable personality-based market segments. The study clarifies the conditions under which categorical personality information may contribute to tourism personalisation and provides an exploratory foundation for future research using validated personality measures, multi-item preference scales, and out-of-sample predictive modelling.