What represents meaningful (as distinct from statistically significant) improvement in teaching? Using data from the standardized observation tool PLATO, the talk will consider how criterion-referenced standards are transformed when used in norm-referenced contexts.