Machine learning technology has become ubiquitous, but, unfortunately, often exhibits bias. As a consequence, disparate stakeholders need to interact with and make informed decisions about using machine learning models in everyday systems. Visualization technology can support stakeholders in understanding and evaluating trade-offs between, for example, accuracy and fairness of models. This paper aims to empirically answer "Can visualization design choices affect a stakeholder's perception of model bias, trust in a model, and willingness to adopt a model?" Through a series of controlled, crowd-sourced experiments with more than 1,500 participants, we identify a set of strategies people follow in deciding which models to trust. Our results show that men and women prioritize fairness and performance differently and that visual design choices significantly affect that prioritization. For example, women trust fairer models more often than men do, participants value fairness more when it is explained using text than as a bar chart, and being explicitly told a model is biased has a bigger impact than showing past biased performance. We test the generalizability of our results by comparing the effect of multiple textual and visual design choices and offer potential explanations of the cognitive mechanisms behind the difference in fairness perception and trust. Our research guides design considerations to support future work developing visualization systems for machine learning.

A viewer's existing beliefs can prevent accurate reasoning with data visualizations. In particular, confirmation bias can cause people to overweigh information that confirms their beliefs, and dismiss information that disconfirms them. We tested whether confirmation bias exists when people reason with visualized data and whether certain visualization designs can elicit less biased reasoning strategies. We asked crowdworkers to solve reasoning problems that had the potential to evoke both poor reasoning strategies and confirmation bias. We created two scenarios, one in which we primed people with a belief before asking them to make a decision, and another in which people held pre-existing beliefs. The data was presented as either a table, a bar table, or a bar chart. To correctly solve the problem, participants should use a complex reasoning strategy to compare two ratios, each between two pairs of values. But participants could also be tempted to use simpler, superficial heuristics, shortcuts, or biased strategies to reason about the problem. Presenting the data in a table format helped participants reason with the correct ratio strategy while showing the data as a bar table or a bar chart led participants towards incorrect heuristics. Confirmation bias was not significantly present when beliefs were primed, but it was present when beliefs were pre-existing. Additionally, the table presentation format was more likely to afford the ratio reasoning strategy, and the use of ratio strategy was more likely to lead to the correct answer. These findings suggest that data presentation formats can affect affordances for reasoning.

The language for expressing comparisons is often complex and nuanced, making supporting natural language-based visual comparison a non-trivial task. To better understand how people reason about comparisons in natural language, we explore a design space of utterances for comparing data entities. We identified different parameters of comparison utterances that indicate what is being compared (i.e., data variables and attributes) as well as how these parameters are specified (i.e., explicitly or implicitly). We conducted a user study with sixteen data visualization experts and non-experts to investigate how they designed visualizations for comparisons in our design space. Based on the rich set of visualization techniques observed, we extracted key design features from the visualizations and synthesized them into a subset of sixteen representative visualization designs. We then conducted a follow-up study to validate user preferences for the sixteen representative visualizations corresponding to utterances in our design space. Findings from these studies suggest guidelines and future directions for designing natural language interfaces and recommendation tools to better support natural language comparisons in visual analytics.