Curricular materials used to teach children not only impart academic knowledge but also prepare children for citizenship by teaching them about societal values. As a result, it is vital that we understand the messages that are conveyed in the educational content we present to our children. This dissertation focuses on both applying and improving methods of computer-driven content analysis to measure how different topics such as evolution or slavery and different groups such as women or LGBTQIA+ individuals are portrayed in educational materials. I document temporal changes in curriculum content and compare differences across multiple sets of influential curricula including both formal curriculum such as textbooks as well as informal curriculum such as prominent children’s books. I also examine curricula used in non-traditional educational settings such as religious private schools and homeschooling families, in addition to state-adopted textbooks from both Texas and California. Using computer vision tools such as face detection and skin segmentation, I measure the representation of different genders, races, and skin colors in images. By applying natural language processing tools such as named entity recognition and word embeddings, I measure not only which groups and topics are represented but the contexts in which they are discussed. This work contributes to our knowledge of what messages are conveyed in the content we use to teach our children and expands the set of tools available for social scientists to measure representation in a variety of contexts.