Research Publications
Explore research publications, policy briefs, and academic papers from the DREAM Network community.
From Evidence to Policy: A Study of Research Uptake in Bangladeshi Ministries
Interviews with 32 senior officials across 6 ministries reveal that research uptake is driven less by methodological rigour than by timing, accessibility of executive summaries and the presence of internal research champions. We propose a 'policy-research liaison' role to bridge the evidence-to-policy gap.
Climate Vulnerability Assessment of Coastal Communities in Bangladesh
This study maps the climate vulnerability of 12 coastal districts in Bangladesh using a composite index combining exposure, sensitivity and adaptive capacity indicators. Findings reveal that saline intrusion and cyclone frequency are the strongest predictors of community displacement, and that women-headed households exhibit 1.8x lower adaptive capacity than male-headed ones.
Citizen Science for Water Quality Monitoring: A Feasibility Study in Dhaka
We piloted a citizen-science water quality monitoring program across 40 schools in Dhaka, training students to collect pH, turbidity and dissolved-oxygen readings. Comparison with lab-verified samples showed 94% agreement, demonstrating that trained school networks can produce research-grade data at a fraction of conventional cost.
A Mixed-Methods Evaluation of In-Service Teacher Training Programs in Rural Bangladesh
This mixed-methods study evaluates three in-service teacher training programs across 18 upazilas, combining classroom observation rubrics with student learning gains. Results indicate that programs emphasising sustained mentoring outperform one-off workshops by 2.3 standard deviations in teacher practice improvement.
Barriers to Reporting Gender-Based Violence Among University Students in Bangladesh
Through 48 in-depth interviews and a survey of 620 university students, this study identifies stigma, fear of academic retaliation and lack of confidential reporting channels as the three primary barriers to GBV reporting. We propose a tiered, peer-supported reporting model currently being piloted at two universities.
A Spatio-Temporal Biostatistical Analysis of the 2023 Dengue Outbreak in Dhaka
Using GIS-linked case data from 98 wards, we modelled the spatio-temporal spread of the 2023 dengue outbreak in Dhaka City. Our negative-binomial regression identifies rainfall lag, population density and standing-water index as significant predictors, and we propose a ward-level early-warning threshold.