Multidimensional poverty and caste-based exclusion: a mixed-methods study of Dalit deprivation in northern Bangladesh

The study examines the multifaceted nature of poverty among Dalit community members in Rangpur, Bangladesh. The study employed a convergent parallel mixed-methods design, integrating structured surveys (n = 400) with a focus group discussion (FGD), from August to October 2024. The adapted Multidimensional Poverty Index (MPI) indicates an overall score of 0.397, with a headcount ratio of 70.25% (281 of 400 respondents identified as multidimensionally deprived) and an average deprivation intensity of 0.566 among low-income individuals. Dimension-specific analysis identifies education as the most critically deprived dimension (57.3% of its maximum), followed by governance (54.5%) and social inclusion (52.6%). Statistical analyses show that educational attainment (F (3, 396) = 28.47, p < 0.001) and caste-based discrimination (t (398) = 6.03, p < 0.001) are the strongest predictors of multidimensional poverty. Conversely, income level (F (2, 397) = 1.71, p = 0.183) and geographic location (F (2, 397) = 1.45, p = 0.236) did not reach statistical significance. The lack of significance for income suggests that, within this highly marginalized population, caste-based structural barriers supplant the protective influence of higher-income categories. Qualitative narratives from the FGD shed light on how educational deprivation and caste discrimination interact to perpetuate poverty.