Workshop on ecologic inference: 28-30 November 2007, DIMACS, Rutgers University
The workshop is part of the DIMACS Special Focus on Computational and Mathematical Epidemiology.
- Rather than focus solely on traditional ecologic studies, the workshop concentrated primarily on studies that combine individual and group-level data.
- Confirmed speakers included Adam Glynn (Harvard University), Sebastien Haneuse (Center for Health Studies, Seattle), Kosuke Imai (Princeton University), Anna Oudin (Lund University, Sweden), Tyler VanderWeele (University of Chicago), Pia Verkasalo (National Public Health Institute, Finland), Tom Webster (Boston University School of Public Health), Rob Eisinga (Radboud University Nijmegen, Netherlands), Dan Wartenberg (Rutgers University, USA)
- More information on the workshop
Studies that combine individual and ecologic data
There are a number of interesting issues surrounding studies that use combinations of individual and group-level data:
- Consider a study that uses a group-level exposure variable but measures outcome and covariates individually. Is such a study subject to ecologic bias?
- Suppose we can supplement aggregate data with a sample of individual-level data. Can we reduce the amount of ecologic bias?
- How do we best measure contextual effects?
To learn more, attend the DIMACS workshop
Ecologic bias
For a primer on ecologic bias, look here
Literature on ecologic bias
There is a vast literature on ecologic bias. A few references include:
- Greenland S: Divergent biases in ecologic and individual-level studies. Statistics in Medicine 1992, 11:1209-1223. [full text]
- King G: A Solution to the Ecological Inference Problem: Reconstructing Individual Behavior from Aggregate Data. Princeton, NJ: Princeton University Press; 1997.
- Morgenstern H: Ecologic studies. In Modern Epidemiology. Second edition. Edited by: Rothman KJ, Greenland S. Philadelphia, PA: Lippincott-Raven; 1998:459-480.
- Wakefield J: Ecological inference for 2 × 2 tables.Journal of the Royal Statistical Society A 2004, 167:385-445. [full text]
- Wakefield J: Ecologic studies revisited. Annu Rev Public Health 2007 (online 3 October) full text
- Webster TF. Bias magnification in ecologic studies: a methodological investigation. Environmental Health 2007, 6:17. [full text].
Literature on studies that combine individual and group-level data
The literature on combining individual and group data is smaller (although there is a boom in applications of 'hierarchical models'); here are a few:
- Björk J, Strömberg U: Effects of systematic exposure assessment errors in partially ecologic case-control studies. International Journal of Epidemiology 2002, 31:154-160. full text]
- Greenland S: Ecologic versus individual-level sources of bias in ecologic estimates of contextual health effects. International Journal of Epidemiology 2001, 30:1343-1350. [full text]
- Haneuse S, Wakefield J. Geographic-based ecological correlation studies using supplemental case-control data. Statistics in Medicine (in press 2007)
- Jackson C, Best N, Richardson S: Improving ecological inference using individual-level data. Statistics in Medicine 2006, 25:2136-2159. [full text]
- Künzli N, Tager IB: The semi-individual study in air pollution epidemiology: a valid design as compared to ecologic studies. Environmental Health Perspectectives 1997, 105:1078-1083. [full text]
- Subramanian SV, Glymour MM, Kawachi I. Identifying causal ecologic effects on health: a methodological assessment (forthcoming).
- Wakefield J: Ecologic studies revisited. Annu Rev Public Health 2007 (online 3 October) full text
- Webster T: Cross-level bias in partially ecologic studies. Proceedings of Spatial Epidemiology Conference, London 2006. 23–25 May London. Small Area Health Statistics Unit, Imperial College; 2006, 127-132. [full text]
- Webster T: Commentary: Does the spectre of ecologic bias haunt epidemiology? International Journal of Epidemiology 2002, 31:161-162. [full text]
- Webster T. Bias in ecologic and semi-individual studies. DSc dissertation. Boston University School of Publoic Health, 2000.
return to spatial epidemiology
return to Tom Webster