D. Leann Long, PhD, Assistant Professor in the Department of Biostatistics, has published a new statistical model for count data that contain a greater than expected number of zeroes, commonly known in biostatistics as "zero-inflated" count data. Dr. Long's work is now available online in Statistics in Medicine, one of the most influential journals in biostatistics. For many fields of research, the new marginalized zero-inflated Poisson model provides more meaningful interpretations than traditional methods for count data with excess zeroes.
Traditional zero-inflated models used in public health studies produce parameters without direct interpretation on the entire population. Dr. Long and colleagues created the marginalized zero-inflated Poisson model to bridge the gap between complicated statistical models and conclusions that directly answer underlying research questions applicable to the entire population. The article “A marginalized zero-inflated Poisson regression model with overall exposure effects” is available here (http://onlinelibrary.wiley.com/doi/10.1002/sim.6293/abstract).