As a result of the growing use of state and national testing of literacy and numeracy among school students, there are increasing demands for teachers to interpret assessment data. In light of this, there is a need to provide benchmarks or a framework that identifies critical aspects of teachers’ understanding that are needed to interpret data effectively. In this study, 24 items from the Attitudes and Statistical Literacy Instrument are used with 704 teachers to provide a hierarchical scale of teacher ability to interpret these assessment data. Using an item response theory model for partial credit scoring, three levels of ability are identified, related to reading values, comparing values, and analyzing a data set as a single entity. Teacher ability is then compared across various demographic variables, such as number of years of teaching, main teaching grade levels, previous professional learning experience, last time statistics was studied, and gender. Implications are drawn for professional learning for teachers and for further research.