Dirty work is socially constructed as tainted on one or more domains (physical: dangerous, dirty, or associated with death; moral: underhanded or in contradiction to prevailing norms; social: in association with stigmatized others or done in subservience), and it shapes dirty workers’ perceptions and experiences of their identities. The processes through which the perception of taint shapes identities and associated outcomes (e.g., identity ambivalence, isolation) and the effects of the magnitude of dirt are not fully understood. To understand these processes, the present study describes the development of a tool to measure the dirt of dirty work. First, the author developed a series of item to assess the content domain of dirty work based on a literature review supported by open-ended responses describing work perceptions from dirty workers. In the subsequent studies, the author reduced the item pool by a series of exploratory factor analyses (EFA). Then, the author tested the overall model fit across two separate samples via confirmatory factor analysis (CFA) and identified a three-factor model. Finally, the author gathered validity evidence through convergent and discriminatory validity analyses: the pattern of correlations generally provided convergent validity evidence with the respective covariates, and the data tentatively supported the measure’s ability to discriminate among forms of taint by occupation in a one-way MANOVA.