Hiring presents an important opportunity for districts to find educators with values and beliefs that align with district goals. Yet beliefs about how to best address persistent challenges like inequality are difficult to measure. We use administrative data from over 10,000 applications to certificated positions in an urban California school district and structural topic modeling (STM) to code applicant essays about closing achievement gaps, identify their beliefs, and test how they influence application behavior and hiring. Compared to white applicants, Hispanic and African American applicants are more likely to identify structural causes for inequities and discuss educators' responsibilities for addressing inequality by challenging biases, as well as experiences with cultural and linguistic diversity. Similar themes emerge among applicants applying to and being hired in schools with more marginalized students. Techniques like STM can decipher hard-to-measure beliefs from administrative data, providing valuable information for hiring and decision-making. [This paper was published in "RSF: The Russell Sage Foundation Journal of the Social Sciences" v5 n3 p103-127 2019.]