State-wide tests are designed to measure student overall ability on grade-level standards. School leaders want fine-grained information on student performance to inform curriculum and instruction. One currently used target scoring method, which compares student scores to expected values is currently used to give this feedback to schools, but there is little research on the accuracy of these scores. This dissertation examines the generating processes for student and school abilities. The generating model impacts whether target scores can be accurately estimated and are meaningful. The use of the target scoring method is compared to a model-based score. While target scores are not direct estimates of school effects, they can be well correlated with generating effects. Multilevel multidimensional item response models are challenging to estimate, but with certain constraints will converge and provide accurate estimates of school effects. Ability estimates after a computer adaptive test are comparable to estimates from a large complete data set. Finally the scoring methods are applied to Connecticut State testing data, showing that statistics and probability content in grade 6 may support target scoring. [The dissertation citations contained here are published with the permission of ProQuest LLC. Further reproduction is prohibited without permission. Copies of dissertations may be obtained by Telephone (800) 1-800-521-0600. Web page: http://bibliotheek.ehb.be:2222/en-US/products/dissertations/individuals.shtml.]