The study discussed the importance of test validity, often established when making decisions that may affect a student's future. The decisions made by policymakers and educators must not adversely affect any particular subgroups of students (i.e., year of administration, gender, ethnicity, level English proficiency, socioeconomic status, and disability status). The study discussed the testing of measurement invariance across subgroups on an assessment as a process of validation. Methods used to detect measurement invariance at the test, subtest, and item levels were reviewed and three of these methods were applied to a reading test for administrative, gender, and ethnic subgroups. The purpose of this study was to demonstrate how to detect measurement invariance using (1) hierarchical linear modeling at the test level, typically used by policymakers, (2) confirmatory factor analysis at the subtest level for instructional designers, and (3) Rasch item analysis at the item level for psychometricians. The results of the study provided validity evidence that supported the comparison across administration years at the test, subtest, and item levels. Validity evidence also supported the comparison of gender subgroups at the subtest level via partial scalar invariance and at the item level. Finally, the results provided evidence that supported the comparison of ethnic groups at the subtest level via partial scalar invariance.