In this presentation, I present novel quantitative methods for analyzing language attitudes using data from 1.3% of the adult population on Pohnpei in the Federated States of Micronesia. Previous methods for language attitudes analyses typically utilized pre-defined essentialized categories that are heavily influenced by non-localized contexts. To counter this method, I propose new quantitative analyses that allow groups and categories to emerge from the data. I present results both from analyses that utilize pre-defined categories and those that do not to demonstrate the benefits of the new analyses. The results show how the new analyses allow for voices in the data to be see that would otherwise have been ignored.