The citation and interpretation of precedent and the development and utilization of legal doctrine are distinct concepts. Whereas previous literature has focused on the use of precedent, this study makes a theoretical argument for the importance of distinguishing between precedent and doctrine and applies automated content analysis tools to the measurement of legal doctrine in court opinions. These tools are used to study doctrinal utilization by the Supreme Court and the circuit courts in the United States judicial system. From a theoretical perspective, this study leverages a qualitative case study of the development and application of the Lemon test in Establishment Clause jurisprudence to illustrate the importance of carefully distinguishing between the concepts of precedent and doctrine. The case study exposes potential weaknesses in dependence upon Shepard’s Citations as a tool for understanding the development of legal doctrine. The concept of doctrinal vitality is proposed as a way to measure the impact of legal doctrine across time. Given the difficulties that are inherent in measuring a qualitative concept (language) quantitatively, a careful examination of various automated text analysis methodologies was conducted. The programming language Python was used to analyze the doctrinal composition of court opinions through unsupervised topic modeling and supervised sentence counting. Utilization of doctrinal language was modeled as a function of variables that impact judicial behavior on the Supreme Court and circuit courts, including doctrine age, judicial ideology, doctrinal vitality, opinion characteristics, and hierarchical effects. While the substantive findings are mixed, this dissertation considers important theoretical implications regarding the development and utilization of legal doctrine and explores the potential benefits and challenges related to the use of automated text analysis in the study of legal doctrine.