Selection-Bias-Corrected Visualization via Dynamic Reweighting

TL;DR

This paper introduces Dynamic Reweighting, a new computational method to mitigate selection bias in large-scale, high-dimensional data visualizations, enhancing the validity of insights in complex system analyses.

Contribution

It presents a novel dynamic reweighting approach, including workflow, visualization designs, and statistical methods, to help users create bias-corrected visualizations in complex data analysis.

Findings

Effective bias mitigation demonstrated in medical data case studies.

User interviews indicate improved understanding of bias effects.

Workflow and visualization tools support practical application of the method.

Abstract

The collection and visual analysis of large-scale data from complex systems, such as electronic health records or clickstream data, has become increasingly common across a wide range of industries. This type of retrospective visual analysis, however, is prone to a variety of selection bias effects, especially for high-dimensional data where only a subset of dimensions is visualized at any given time. The risk of selection bias is even higher when analysts dynamically apply filters or perform grouping operations during ad hoc analyses. These bias effects threatens the validity and generalizability of insights discovered during visual analysis as the basis for decision making. Past work has focused on bias transparency, helping users understand when selection bias may have occurred. However, countering the effects of selection bias via bias mitigation is typically left for the user to accomplish as a separate process. Dynamic reweighting (DR) is a novel computational approach to selection bias mitigation that helps users craft bias-corrected visualizations. This paper describes the DR workflow, introduces key DR visualization designs, and presents statistical methods that support the DR process. Use cases from the medical domain, as well as findings from domain expert user interviews, are also reported.