Multilevel Modeling

20. Author Biography

S V Subramanian, PhD, (‘Subu’ or ‘Subra’) has a PhD in geography with specialization in multilevel statistical methods. He also has a Masters in the field of development studies from the University of Delhi and was the recipient of the 1999-2000 MacArthur Leadership Program in Population and Development Studies based at the Harvard Center for Population and Development Studies.

The main focus of his research is on understanding how different contextual settings influence individual health outcomes and the population disparities in health achievements. He has specifically investigated the impact of income inequality and social capital on individual health outcomes. His work has demonstrated the need to explicitly consider a multilevel methodological framework while conceptualizing and estimating contextual effects on public health issues.

Currently, through a National Institutes of Health/National Heart, Lung, Blood Institute Career Development Award, he is investigating the role of neighborhood-level factors (neighborhood structural disadvantage, collective psychosocial characteristics, and physical environmental conditions) in explaining the occurrence and distribution of asthma. He is also conducting original collaborative research on the different ways in which health gets spatially and socially stratified in India, with particular focus on the socioeconomic distribution and determinants of nutritional status.

Besides his substantive interest in understanding place-effects on health, Subu is additionally conducting independent research on the methodological challenges to estimating causal effects of neighborhood factors on health.

Subu has published over 100 journal articles, book chapters, books, and working papers. He is an Assistant Editor for Social Science and Medicine; Editorial Consultant to The Lancet; Member of the Editorial Board for Health and Place, and BMC Public Health; and Book Review Editor for Economics and Human Biology.

Subu has extensive experience in teaching and training graduate students and post-doctoral fellows in multilevel statistical methods. He has lectured and conducted workshops on the concept and practical applications of multilevel models in Argentina, Australia, Austria, Belgium, Chile, India, Japan, Switzerland, the U.K., and the U.S. He and Professor Kelvyn Jones (University of Bristol) have co-developed a training manual to assist researchers in the concept and application of multilevel models using the MLwiN program.


The work presented here draws upon earlier as well as on-going works. I am indebted to my generous and wonderful collaborators, Tony Blakely, William Browne, Nina Bullen, Craig Duncan, Maria Glymour, Harvey Goldstein, Myles Gould, Kelvyn Jones, Ichiro Kawachi, Graham Moon, Jon Rasbash, and Liz Twigg, from whom I have learned and continue to learn much about multilevel models. My work is supported by the National Institutes of Health Career Development Award (NHLBI 1 K25 HL081275-01), and the Robert Wood Johnson Foundation through their RWJ Health and Society Scholars.