Modeling a Fragmented Archive: A Missing Data Case Study from Provenance Research
Matthew Lincoln, Sandra van Ginhoven · Figshare · 2020
Paper presented at the Alliance of Digital Humanities Organizations annual conference, 2018, Mexico City. Historians grapple with missing information constantly. While there are many statistical tools for gauging the impact of missing source data on quantitative results and conclusions, DH researchers have rarely deployed these tools in their work. This paper presents one implementation of data imputation used in the study of the New York City art dealer M. Knoedler & Co. Demonstrating the significant contribution imputation had on our study and its conclusions, this paper will discuss specific, practical rhetorical strategies, including static and interactive visualization, for explaining this methodology to an audience that does not specialize in quantitative methods.