Publications:
Dey, T., Goli, S., 2025. Examining the Role of Modernization and Urbanization in Family Changes in India: Evidence from Panel Data Analyses. Journal of Family History, https://doi.org/10.1177/03631990251376548.
Abstract: India's family structure is rapidly transforming, yet limited research examines its connection with demographic transition, urbanization, and modernization. Using macro-level panel data (1991–2021), we analyze how urbanization and modernization drive the shift from traditional to nuclear families. Fixed-effects models reveal a significant positive relationship: each unit increase in urbanization and modernization increases nuclear families by 0.29–1.32 percent and 0.21–0.49 percent, respectively. Factors like rising literacy, economic shifts, and women's autonomy further accelerate the nucleation of families. As India's rural-urban transition continues, sustained urbanization and modernization will likely deepen family nucleation, signaling an impending second demographic transition in the near future.
Dey, T., Bansal, A., Dwivedi, L.K., 2021. “Population Projection of Metropolitan Cities in India: An Approach to Bayesian Hierarchical Model”. Demography India, Vol. 50, No. 2, pp. 16-32.
Abstract: Population projection based on the cohort-component method fails to incorporate the uncertainty component. So, to overcome this issue, we have used the probabilistic approach using the Monte Carlo Markov Chain (MCMC) method. We obtained data from previous censuses from 1901 to 2011, and applied the Bayes rule to get the posterior distributions. A four-parameter logistic growth model was used to project the population of eight metropolitan cities in India by taking the sample nodes for 2021-2071. The results observed smooth curves of the posterior density of the nodes, and the curves are bell-shaped, which indicates asymptotically normal. The four-parameter logistic model shows a close fit to observed data for the years 1901 to 2011. The dotted line provides the 95% highest posterior density region, which is not large for the different population estimates. The projected population of Mumbai, Delhi, Kolkata, Nagpur, Indore, and Meerut will be stagnant after 2051, and Chennai and Hyderabad will be stabilized after 2061. The study shows that the logistic growth model using the MCMC technique is suitable for the population projection, and holds a significant importance in the absence of the latest Census of India 2021. The projection of these metro cities will help to formulate the future strategies that need to be changed with population growth. The approach can be used as an extension to the classical approach to predict future population, as it yields more accurate estimates to measure uncertainty.
Dwivedi, L.K., Rai, B., Shukla, A., Dey, T., Ram, U., Shekhar, C., Dhillon, P., Yadav, S., Unisa, S., 2020. “Accessing the Impact of Complete Lockdown on COVID-19 Infection in India and its Burden on Public Health Facilities”. Demography India, Vol. 49, Special Issue, pp. 37-50.
Abstract: COVID-19 has emerged as a global public health concern due to a large-scale community-based outbreak across countries. The number of confirmed cases has also increased in India in the past few weeks. The predictions for COVID-19 can provide insights into the epidemiology of the disease that help policymakers assess the health system's preparedness. We obtained data on daily confirmed, recovered, and deaths for 21 days and have implemented the exponential growth model to predict future cases for all three components. The mathematical model was used to calculate the average reproduction number and herd immunity. We estimated the number of active cases till the 3rd of May, 2020, and tried to analyze the burden on public health in combating COVID-19 in India. Further, an attempt is also made to study the role of lockdown in reducing the number of confirmed COVID-19 cases and deaths. If the exponential growth in the number of cases continues then the total number of active cases would have been around 2,49,635 by the April end. The reproduction number for COVID-19 in India was found to be 2.56, and herd immunity was 61%. The total number of confirmed cases till 23rd April was 23,039 as opposed to 86,373 if the cases had grown exponentially. The estimated number of COVID-19 cases in a government hospital would be in the range of around 8 to 15 until 3rd May 2020, but this figure has been significantly reduced to 0.82 based on the observed cases till 23rd April, 2020. Results show that preventive measures like a complete national lockdown have resulted in curtailing the exponential growth in the number of confirmed cases. The government of India has taken many preventive measures at the early stage of the disease, such as a complete lockdown for three weeks, case isolation, and deferring national and international travel. The most important one is that a large number of healthcare workers are visiting households in the hotspot zones across the country to trace and isolate people who might have had contact with those having COVID-19 symptoms, etc., to curtail the spread of the disease. Results indicate that these preventive measures have successfully helped in preventing a large number of deaths and infected cases in India.
Previous Research:
MPhil Thesis Report 2021: Construction of Complete Life Tables and Abridged Life Tables Using the Elandt-Johnson Method for Males and Females of West Bengal.
The study aims to construct complete life tables for males and females in West Bengal using the Elandt–Johnson Method, based on abridged life tables derived from the Sample Registration System (SRS) data spanning 1995–99 to 2014–18. Due to inadequacies in India’s death registration system, indirect estimation techniques were employed. Initially, regression models were developed to estimate key life table functions (eₓ and qₓ) from life expectancy at birth (e₀) for various ages. The best-fitting relationships—primarily linear, with some quadratic fits—were selected based on the highest coefficients of determination (R²). The Elandt–Johnson interpolation approach, along with the Gompertz model for ages above 75, was then applied to expand the abridged tables into complete life tables. The results show close agreement between the generated and SRS-based life tables, indicating the reliability of the method. Findings reveal that female child mortality (ages 0–6) is slightly higher than male child mortality, while overall life expectancy is higher among females. The study concludes that the Elandt–Johnson method provides an effective and robust framework for constructing complete life tables for regions where direct mortality data are limited.
MSc Thesis Report 2020: Population Projection of Metropolitan Cities in India: Using Bayesian Hierarchical Model and Dynamic Logistic Model.
The study projects the future population of India’s major metropolitan cities—Mumbai, Delhi, Kolkata, and Chennai—using two advanced statistical frameworks: the Bayesian Hierarchical Model and the Dynamic Logistic Model. Population forecasting is crucial for urban planning and policy formulation; however, traditional methods, such as the cohort-component model, face challenges in estimating uncertainty and addressing data limitations. This research employs a Bayesian inference approach, utilizing Markov Chain Monte Carlo (MCMC) simulations via WinBUGS to estimate parameters and credible intervals for the logistic growth model. Additionally, a Dynamic Logistic Model is developed, accounting for time-varying carrying capacity to reflect technological and infrastructural changes in urban areas. Using census data from 1901 to 2011, the study produces probabilistic projections up to 2071, demonstrating a close model fit to historical data and credible future estimates. Results indicate population stabilization or slow decline after 2061, aligning with national demographic transition trends. The findings highlight the suitability of Bayesian and dynamic modeling techniques for urban demographic forecasting in data-limited contexts, providing valuable insights for planners and policymakers.
Summer Internship Report 2019: Determination of the Series of Vaccination Coverage of Children in North-East India: A Statistical Approach.
This study examines vaccination coverage among children aged 12–23 months across eight northeastern states in India, using data from the National Family Health Survey (NFHS-4). Despite national progress in immunization, the region continues to exhibit disparities in vaccine uptake due to socio-demographic and household factors. The study examines the coverage of various vaccine doses—BCG, Polio, DPT, Hepatitis B, and Measles—and investigates their relationships with maternal education, caste, religion, economic status, place of residence, birth order, child sex, and breastfeeding duration. Using binary logistic regression, the analysis identifies key determinants that influence vaccination access. Findings reveal that maternal education, household wealth, and urban residence significantly increase the likelihood of full immunization, whereas higher birth order and lower socioeconomic status are associated with incomplete vaccination. The study emphasizes the need for targeted interventions that focus on rural and socioeconomically disadvantaged populations to achieve equitable immunization coverage in the region.
Talks:
“Changing Family Patterns in India: A Path to Convergence? Understanding through the Roles of Fundamental Socio-Demographic Components”
Tapas Dey, Srinivas Goli
Royal Statistical Society (RSS) International Conference, Edinburgh, United Kingdom, September 2025.
31st UK Stata Conference, United Kingdom
University of Westminster, London, September 2025.
“Examining the Role of Modernization and Urbanization in Family Changes in India: Evidence from Panel Data Analyses”
Tapas Dey, Srinivas Goli
Asian Population Association (APA) 2024 Conference, Kathmandu, Nepal, November 2024.
“Examining the Role of Modernization and Urbanization in Family Changes in India: Evidence from Panel Data Analyses”
Tapas Dey, Srinivas Goli
Work and Family Researchers Network Conference (WFRN) 2024 Annual Meeting, Concordia University, Canada, June 2024.
“Examining the Role of Modernization and Urbanization in Family Changes in India: Evidence from Panel Data Analyses”
Tapas Dey, Srinivas Goli
Population Association of Singapore (PAS) 2024 Annual Meeting, National University of Singapore (NUS), Singapore, May 2024.
Professional Experience:
International Institute for Population Sciences (IIPS), Mumbai, India
Demographic and Health Surveys and its Quality in India: A Programme to Develop Survey Research, Worked as a Research Officer, November 2021 - June 2022, and Consultant, September 2021 - October 2021.
Funded by the Bill & Melinda Gates Foundation via the Population Council, New York.
Workshops:
United States Census Data: An Exploration Using the Integrated Public Use Microdata Series
International Institute for Population Sciences, Mumbai, January 2025.
Utilizing and Interpreting Nutrition Data in Research
The Harvard T.H. Chan School of Public Health - India Research Centre, Mumbai, India, April 2024.
10th KOSTAT-UNFPA Summer Seminar on Population
Statistics Korea and UNFPA, July - August 2023.
Qualitative Methods in Social Sciences: Ethnography, Narratives, and Social Stories
Council for Social Development, New Delhi, and G.B. Pant Social Science Institute, Prayagraj, November - December 2022, at New Delhi.
Analytical Techniques for Statistical Analysis in Mental Health
AIIMS (New Delhi), DST-SERB-Accelerate Vigyan, July 2021.
Methods and Approaches for Health Research in Social Sciences
ICMR-NIRTH, VIT-AP University, and the Central University of Karnataka, June 2021.
Sample Size and Power Analysis in Medical Research
Department of Biostatistics, National Institute of Mental Health & Neuro Sciences (NIMHANS), Bengaluru, September 2019.
Clinical Research Methodology
Tata Memorial Hospital, Clinical Research Secretariat of Tata Memorial Hospital (TMH) in association with the Department of Atomic Energy (DAE) – Clinical Trials Centre, July 2019.