Examining potential predictors for completion of the gardasil vaccine sequence based on data gathered at clinics of Johns Hopkins medical institutions

This paper presents categorical data that were gathered at two urban clinics and two suburban clinics of Johns Hopkins in an effort to identify characteristics of young female patients who successfully complete the three-injection sequence of the Gardasil quadrivalent human papillomavirus vaccine (HPV4). Available categorical correlates included patient age group, patient race, clinic location type, type of insurance provider, and clinic practice type. The data may be used to illustrate graphical display techniques and inference procedures for categorical data, as well as illuminate ways in which relationships between categorical variables may be hidden or behave differently than expected in the presence of confounding variables. The methods used to gather the data may also serve to illustrate the limitations of drawing conclusions from observational studies.