One of the most widespread trends in government in past decades has been the diffusion of quantitative measures of performance. This trend is motivated partly by the hope that policymakers will use this data. However, we know little about individual's basic tendencies to incorporate and use performance data. This paper draws from interactive dialogue theory to argue that the connection between data and decisions will not be automatic, but will depend upon circumstances such as the nature of the data and how it is presented. Because observational studies are limited in their ability to identify such circumstances, this paper argues for an experimental approach that offers greater flexibility in designing theory-based treatments. A vignette experiment methodology approach is used, providing subjects with a variety of budget scenarios, varying the amount and type of information that they receive, and asking them to make budget decisions. The results provide evidence that conditions of goal ambiguity, expectancy disconfirmation, and advocacy alter the potential for performance data to influence resource decisions.2