The research team combined a survey of 384 Dalit respondents with 30 in-depth interviews and one focus group discussion, an unusually comprehensive evidence base for a population that is often invisible in national statistics. The quantitative results are stark in their descriptive power. More than six in ten respondents, 61.2 percent, reported experiencing caste-based discrimination directly. Nearly half, 47.7 percent, described housing insecurity, and 34.9 percent reported a complete absence of political representation. Subjective freedom of choice—the sense that one’s life path is genuinely open—averaged just 2.26 on a five-point scale, among the lowest readings of any construct in the survey.

Bivariate analyses reinforced the picture of systemic exclusion. Respondents strongly endorsed statements describing institutional neglect, with a mean agreement score of 3.99, and capability deprivation, at 3.86. When the researchers stratified respondents by income, analysis of variance revealed a statistically significant difference in health outcomes across income strata, F(2, 381) = 3.44, p = 0.033, though the effect size was small, with eta squared of 0.018. Health, in other words, deteriorated measurably as income fell, consistent with a body of South Asian literature linking caste exclusion to unequal access to medical care, sanitation work and safe housing.

Yet the study is notable as much for its methodological honesty as for its findings. The eleven Likert-scale items used to measure attitudes toward structural inequality and capability deprivation showed very low internal reliability, with a Cronbach’s alpha of just 0.01, and inter-item and inter-construct correlations were weak, averaging an absolute correlation of roughly 0.04. An exploratory multivariate model seeking predictors of a composite Poverty Dynamics index found no significant predictors at all, explaining a mere 2 percent of variance. The authors are explicit that their theoretical contributions are interpretive rather than statistically demonstrated, a candour that is rare in development research and that shifts the evidentiary weight decisively onto the qualitative material.

And it is the qualitative material that carries the study’s most striking argument. Across interviews, respondents described practices of untouchability that persist in labour markets, educational institutions and spatially segregated settlements. Dalit workers reported being confined to occupations considered polluting—sanitation work, sweeping, the handling of carcasses—regardless of their skills or aspirations. Parents described children humiliated in classrooms and steered away from education by teachers and peers alike. Settlements were described as physically separated from mainstream villages, a segregation that reproduces itself in who gets visited by extension workers, who receives relief, and who is invited to community decision-making.

From these accounts the authors derive their central conceptual contribution: a form of entrapment they call bonded poverty. The term deliberately echoes bonded labour, because respondents described something beyond individual income constraint—a systemic condition in which caste identity itself forecloses exit routes. A family may acquire a small asset, a child may complete primary school, a worker may change employer, yet the underlying conversion rates remain so unfavourable that gains evaporate. In capability terms, the problem is not merely that Dalit households lack resources; it is that the social rules governing how resources translate into valued functionings—being educated, being healthy, being respected, participating in public life—are rigged against them.

This is where the study’s engagement with the capability approach becomes theoretically significant. Sen’s framework holds that development should be judged by the substantive freedoms people enjoy, and that commodities alone do not determine wellbeing; conversion factors—personal, social and environmental—determine what a given resource can become in a person’s life. Husain and colleagues argue that caste operates precisely as a social conversion factor in Bangladesh: the same taka of income, the same year of schooling, the same plot of land yields less capability for a Dalit household than for a dominant-caste neighbour. The framing connects the Rangpur findings to a wider literature on caste-based diminished returns in India, where education has been shown to translate into less wealth for marginalised groups, and to systematic reviews documenting caste exclusion and health discrimination across South Asia.

The study also draws on intersectionality, noting that disadvantages do not simply add up but multiply. Dalit women appear in the qualitative accounts as bearing a compounded burden—excluded by caste in the street, by gender within the household economy, and by class in the labour market. The authors situate their analysis alongside scholarship on structural violence, the idea that harm is embedded in institutional arrangements rather than delivered only by identifiable perpetrators. Institutional neglect, in this reading, is not an absence of action but an active mechanism: when a settlement is left off a sanitation programme or a school roster, the omission itself produces injury.

The policy implications the authors draw are correspondingly structural. They call for strengthening social protection so that Dalit households receive reliable buffers against shocks, for enforcing anti-discrimination measures in labour markets and schools rather than leaving constitutional equality on paper, and for designing capability-enhancing interventions—education access, health services, political representation—that directly target the conversion factors their interviews exposed. They caution, implicitly, that income transfers alone will not dissolve bonded poverty if the social rules that devalue Dalit effort remain untouched. The finding that political representation is absent for a third of respondents suggests that even well-designed programmes may fail without Dalit voices inside the institutions that allocate them.

The research, conducted under ethical approval from Begum Rokeya University and in line with Bangladesh Medical Research Council guidelines, used pseudonyms and trauma-informed interviewing throughout, and the authors acknowledge the Dalit community of Rangpur Division for its openness. Its limitations are acknowledged with equal clarity: the survey instruments did not cohere statistically, the cross-sectional design cannot establish causation, and the multivariate model was null. But the study’s value lies in what it demonstrates about method as much as about caste. Where numbers failed to capture the mechanism, lived testimony succeeded—and the result is a rare, rigorous account of how an ancient hierarchy reproduces itself in a modern state, converting every generation’s modest gains back into the poverty it inherited.

Subject of Research: Caste-based marginalisation and intergenerational poverty among Dalit communities in Rangpur Division, Bangladesh, analysed through structural inequality and capability deprivation frameworks.

Article Title: Caste and poverty in Bangladesh: a mixed-methods analysis of dalit marginalisation through structural inequality and capability deprivation lenses

Article References: Husain, T., Hossain, D., & Kabir, H. (2026). Caste and poverty in Bangladesh: a mixed-methods analysis of dalit marginalisation through structural inequality and capability deprivation lenses. SN Social Sciences, 6(10), Article 458. https://doi.org/10.1007/s43545-026-01747-4

Image Credits: AI Generated

DOI: 10.1007/s43545-026-01747-4

Keywords: Dalit, Bangladesh, caste discrimination, poverty, capability approach, structural inequality, mixed methods, social exclusion, Rangpur Division, bonded poverty, untouchability, social protection