Multilevel Modeling

In 2009 and 2011 we sponsored five-day workshops to train participants in the concepts and applications of multilevel statistical modeling, particularly in a spatial and demographic context. Material from these workshops is archived below. However, we encourage those interested in learning about multilevel modeling to visit the website of The Centre for Multilevel Modelling (University of Bristol), which is dedicated to the education and use of multilevel modeling. The center offers software designed for multilevel modeling work. Many of these are free and all are available as trial versions. This list includes the MLMwin software package and a number of free packages for existing statistical software, including R, MATLAB, and STATA. All these softwares have associated learning tools that can be found at their portions of the website. The website also has a variety of learning and training resources, including a free online course, video and audio materials, reading lists, and much more. Finally, the center conducts regular training workshops in various locations in the U.K.

GISPopSci’s workshop was led by Kelvyn Jones (University of Bristol, U.K.) and S.V. (Subu) Subramanian (Harvard University) and was designed to give participants a training experience in the concepts and applications of multilevel statistical modeling, particularly in a spatial and demographic context. More specifically, the workshop had five major objectives, introducing participants to: (1) the complex hierarchical and non-hierarchical structures in terms of unit diagrams and classification diagrams (e.g., panel, repeated cross-sectional, multivariate, multistage survey, and spatial designs); (2) a thorough consideration of normal theory two-level models; (3) an appreciation of more advanced topics (3-levels structures, multilevel logit models, estimation (including maximum likelihood estimators [(R)IGLS] as well as MCMC); properties of shrinkage estimates; (4) the use of specialist MLwiN software; and (5), the application of multilevel modeling to a social science research problem. Throughout there was a strong emphasis on interpretation, not technical facility per se.

Presentations:

The slide presentations for this workshop are currently password protected for access by participants in prior Penn State and UCSB workshops offered as part of the Advanced Spatial Analysis (2008-2011) series. If you have questions about access, please contact us.

Readings:

The references listed below will be of particular interest to those seeking an understanding of applications of multilevel modeling in geographical context.

  • Duncan, C., K. Jones, and G. Moon. 1998. Context, composition and heterogeneity: Using multilevel models in health research. Social Science and Medicine 46(1): 97117.
  • Jones, K., and N. Bullen. 1994. Contextual Models of Urban House Prices: A Comparison of Fixed- and Random-Coefficient Models Developed by Expansion. Economic Geography 70(3): 252272.
  • Subramanian, S.V.  K. Jones, A. Kaddour, and N. Krieger. 2009. Revisiting Robinson: The perils of individualistic and ecologic fallacy. International Journal of Epidemiology 38(2): 342360.
  • Subramanian, S.V. 2004. The relevance of multilevel statistical methods for identifying causal neighborhood effects. Social Science and Medicine (1982) 58(10): 1961–1967.
  • Subramanian, S.V. C. Duncan, and K. Jones. 2001. Multilevel perspectives on modeling census data. Environment and Planning A 33(3): 399417.
  • Blakely, T. and S.V. Subramanian. 2006. Multilevel Studies. In Methods in social epidemiology, edited by M.J. Oakes and J.S. Kaufman. San Francisco, CA: Jossey-Bass, pp. 316340.
  • Anderson, N.B. 2004. Multilevel Methods, Theory, and Analysis. In Encyclopedia of health & behavior. Thousand Oaks, CA: Sage Publications, pp. 602609.
  • Subramanian, S.V.  M.M. Glymour, and I. Kawachi. 2010. Identifying causal ecological effects on health: A methodological assessment. In Macrosocial Determinants of Population Health, edited by S. Galea. Gardners Books.
  • Subramanian, S.V.  J.T. Chen, D.H. Rehkopf, P.D. Waterman, and N. Krieger. 2006. Subramanian et al. Respond to “Think Conceptually, Act Cautiously.” American Journal of Epidemiology 164(9): 841844.
  • Subramanian S.V., J.T. Chen, D.H. Rehkopf, P.D. Waterman, and N. Krieger. 2006. Comparing individual- and area-based socioeconomic measures for the surveillance of health disparities: A multilevel analysis of Massachusetts births, 19891991. American Journal of Epidemiology 164(9): 823–834.
  • Subramanian S.V., D.J. Kim, and I. Kawachi. 2002. Social trust and self-rated health in U.S. communities: A multilevel analysis. Journal of Urban Health: Bulletin of the New York Academy of Medicine 79(4): 2134.
  • Subramanian, S.V, I. Kawachi, and B.P. Kennedy. 2001. Does the state you live in make a difference? Multilevel analysis of self-rated health in the U.S. Social Science & Medicine 53(1): 919.
  • Duncan, Craig, and K. Jones. 2010. Using Multilevel Models to Model Heterogeneity: Potential and Pitfalls. Geographical Analysis 32(4): 279305.
  • Soobader, M., C. Cubbin, G.C. Gee, A. Rosenbaum, and J. Laurenson. 2006. Levels of analysis for the study of environmental health disparities. Environmental Research 102(2): 172180.
  • Diez Roux A.V. 2002. A glossary for multilevel analysis. Journal of Epidemiology and Community Health 56(8): 588–594.