VAINE: Visualization and AI for Natural Experiments

Guo, Grace; Glenski, Maria; Shaw, ZhuanYi; Saldanha, Emily; Endert, Alex; Volkova, Svitlana; Arendt, Dustin

doi:10.1109/vis49827.2021.9623285

Cited by 3 publications

(3 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These distinctions mean that visualizations designed for the PO framework must support entirely different user tasks and analysis processes. To this end, we identified only three visualization packages developed for the PO framework -VAINE [23], causallib [53], and Cobalt [22]. VAINE and causallib are both designed for use with the Jupyter notebook environment.…”

Section: Causal Visualizationmentioning

confidence: 99%

“…Within data visualization, a range of interactive notebook widgets have been developed for use in domains such as biology [13], machine learning [4,44], data comparison [59], and programming education [28]. Specific to causal inference, both VAINE [23] and causallib [53] are designed to be used in Jupyter Notebook. Inspired by these tools, we implemented Causalvis as a Python package for computational notebooks so that users can easily collaborate with experts and iterate through their causal inference workflow rapidly.…”

Section: Visualizations In Computational Environmentsmentioning

confidence: 99%

“…Analysts often rely on visualizations to inspect and resolve errors in each step of the process. However, while there are existing visualization libraries used by causal inference experts, they are often not designed specifically for causal inference [29,66], or support only limited analysis tasks [5,23,53]. This is further complicated by the existence of different causal inference approaches that adopt incompatible assumptions and processes [33,61,62], resulting in visualization tools that provide only fragmented support for the analysis workflow, and are not designed to work with one another.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Causalvis: Visualizations for Causal Inference

Guo

Karavani

Endert

et al. 2023

Proceedings of the 2023 CHI Conference on Human Factors in Computing Systems

Self Cite

View full text Add to dashboard Cite

Causal inference is a statistical paradigm for quantifying causal effects using observational data. It is a complex process, requiring multiple steps, iterations, and collaborations with domain experts. Analysts often rely on visualizations to evaluate the accuracy of each step. However, existing visualization toolkits are not designed to support the entire causal inference process within computational environments familiar to analysts. In this paper, we address this gap with Causalvis, a Python visualization package for causal inference. Working closely with causal inference experts, we adopted an iterative design process to develop four interactive visualization modules to support causal inference analysis tasks. The modules are then presented back to the experts for feedback and evaluation. We found that Causalvis effectively supported the iterative causal inference process. We discuss the implications of our findings for designing visualizations for causal inference, particularly for tasks of communication and collaboration.

show abstract

Section: Causal Visualizationmentioning

confidence: 99%