Measuring college sailing teams ability: an application of the many-facet Rasch model to ordinal data.

For those who look at typical approaches to sports ranking, sailing seems an almost impossible challenge, thereby making the evidence supporting Rasch measurement in this study even more intriguing. This article describes our application of MFRM and the results from our analysis of one year's data from the North American college sailing competitions. We discuss the following issues for inclusion in the development of a Rasch model useful to college sailing team ability: 1. The level of data quality (as described by Stevens, 1946) 2. The connectedness of the contests 3. Empty cells (missing data) 4. Interpreting outliers, trends, or unusual results 5. Judges' or polls' bias Our results indicate the utility and precision of MFRM as a tool generally appropriate for ordinal ranking applications and sailing ability specifically.