Across the United States, Student Evaluations of Instruction (SEIs) are often the primary (if not only) metric used to evaluate the quality of an instructor. Although SEIs probably reflect student attitudes towards the class in some way, it is not clear to what extent SEI scores represent how much students learned in the class. This study looks at data from faculty volunteers who were recruited from a pool of recent attendees of the APS’s New Faculty Workshop. The study solicited numerous forms of class artifacts from these faculty including student evaluations of instruction and multiple-choice conceptual survey data. The data indicate that there is no correlation between SEI ratings and normalized learning gains on the FCI, or other instruments. Thus, it appears that faculty receiving high (or low) evaluations from their students has no connection to how much conceptual understanding their students developed throughout the semester.