Annotation: This DataSpeak program is the second in a three-part series on the use of contextual analysis, an approach for assessing the effect of contextual, or neighborhood, characteristics along with individual-level factors in explaining disparities in health outcomes. Each program in the series features one of several university-based researchers funded by the Maternal and Child Health Bureau (MCHB) in the Health Resources and Services Administration to explore the effect of neighborhoods on our country’s relatively high infant mortality rate as compared to other industrialized countries and wide racial disparities in infant mortality and preterm birth.
This series is intended to provide public health professionals with background and knowledge of concepts and statistical analysis techniques to begin developing and adapting the approach to their specific States and communities. The first program in the series, broadcast on May 16th, presented an overview of contextual analysis, including discussion of how neighborhoods are defined, what sources of data are available at the neighborhood level, and how neighborhood conditions can affect health.
This second program will describe several different multilevel analysis techniques, the advantages and disadvantages of these different approaches, examples of their use for the analysis of preterm birth data, and the interpretation of statistical results.
Learning Objectives: • Understand the terminology used when describing multilevel models.
• Understand the use of random-effect models, as well as mixed models that include random and fixed effects.
• Learn how to interpret various multilevel models.
Special Instructions: DataSpeak uses a number of different technologies. To get the most out of the information, please review the technical requirements at http://hrsa.gov/archive/mchb/dataspeak/techreq/index.html