Canadian provinces administer large-scale assessments (LSAs) of educational achievement in order to collect data for evaluating the efficacy of school systems, guiding policymaking, making decisions regarding improving student learning, and—in the case of provincial secondary end-of-course examinations—playing a role in the certification of graduating students (Crundwell, 2005; Earl, 1999; Ercikan & Barclay-McKeown, 2007; Taylor & Tubianosa, 2001). The majority of these LSAs contain both closed-response and extended open-response items. The usefulness of the data from these assessments requires that scores be accurate indicators of the targeted learning outcomes (Watermann & Klieme, 2002). At the core of such accuracy are the procedures used for scoring examinee responses and performances.