Tasso Partners worked over several years with the Knowledge Technologies group at Pearson and more recently with the University of Washington, Seattle, to develop scaling methods that produce oral reading scores that are relatively invariant across texts and that provide additional formative information about an individual’s reading processes. The project’s ultimate goal is to enable the development of a machine into which a teacher could enter any roughly grade-appropriate passage, have a child read that passage, and then produce a profile of standardized scores. The scores should reflect an estimate of that child’s oral reading skills in a way that is meaningful for the teacher and precise enough to track reading progress month to month. The SORSS project has investigated the feasibility of producing IRT-based scores that operate on words, phrases, and sentences in a passage, analyzing them as ‘items’. The details of passage-internal item timing can be automatically extracted from an oral reading, and this passage-internal timing information is used by an IRT analysis to calibrate diverse text elements. This means that sub-skills can be combined into a single scale score that covers the abilities that are currently reported as decoding and word accuracy, words correct per minute (WCPM), and vocabulary knowledge. The SORSS research program is based on the insight that passages and readers can be located on a shared scale and that oral reading data is the most stable link. SORSS methods enable ASR-based systems to return standardized scores and diagnoses of a student’s strengths and weaknesses as manifest in the student’s performance in passage and list reading.