127 – Applications and Administrative Practices of Statistical Consulting and Collaborations
Much Ado About Almost Nothing: Methods for Dealing with Limited Data
Stephen Looney
Georgia Health Sciences University
Courtney E. McCracken
Emory University
Applied statisticians are often confronted with statistical inference problems dealing with situations in which there appear to be no data, or data of only limited usefulness. For example, when attempting to find a confidence interval for a binomial proportion, the sample may contain no successes. Such a scenario could be encountered when attempting to estimate the incidence of an extremely rare side effect associated with the administration of a newly developed drug. Other statistical inference situations in which there may be no or only limited data include estimating an odds ratio when one of the cells in a 2x2 table is empty, estimating a risk ratio when one of the groups experiences none of the outcome of interest, and incorporating observations below the limit of detection into a statistical analysis. In this presentation, we illustrate each of these scenarios with real data and describe the preferred methods for handling them.