Baseline Exposure to Common Data Visualization Types Among the U.S. Adult Population

• Familiarity with a visualization can improve interpretation and make graphics seem easier to understand, motivating the question of how much exposure people have.Source: §1 · p. 1
• Because literacy studies often focus on simple, common charts, low performance on less familiar formats may reflect limited exposure rather than general inability.Source: §1 · p. 1
• The study measures exposure to 16 chart types among more than 1,000 U.S. adults using a probability-based sample.Source: §1 · p. 1
• Its nationally representative estimates provide context for studying literacy across complex charts and groups, and for designing visualizations for different audiences.Source: §1 · p. 2

• Visualization literacy spans tasks from locating values to evaluating conclusions, reflecting that visualizations serve different goals and audiences.Source: §2 · p. 2
• Assessments range from the 53-item VLAT across 12 chart types to shortened and adaptive variants, with substantial variation in measurement approaches [23] [8].Source: §2 · p. 2
• Chart coverage is inconsistent across studies, and practitioner-favored alternatives and uncertainty formats remain insufficiently studied.Source: §2 · p. 2
• Research has also examined targeted populations and population-level literacy, including differences in interpretation by education and chart type [24].Source: §2 · p. 2
• Lower graph literacy has been associated with greater difficulty interpreting health graphics, including among Medicaid-eligible populations.Source: §2 · p. 2
• Familiarity is needed to interpret literacy findings because low performance on less familiar formats may reflect limited exposure; this study measures exposure across chart types and groups.Source: §2 · pp. 2–3
• Prior studies often used small or convenience samples, while population-level studies have concentrated on common formats such as bar, pie, and line charts.Source: §2 · p. 3

• The 16 tested formats include common charts, practitioner-favored alternatives, and charts that represent uncertainty, with the set limited to reduce respondent burden.Source: §3 · p. 3
• Participants viewed four icon-and-name items per screen and answered whether they had seen each chart, choosing Yes, No, or Unsure.Source: §3 · pp. 3–4, Figure 2 · p. 3
• Icons accompanied chart names because formats can have overlapping names, such as boxplot for box and whisker plot and dumbbell plot for barbell chart.Source: §3 · p. 4

• AmeriSpeak draws from U.S. households using a national sample frame covering over 97% of households and surveys about 1,000 adults per wave.Source: §4 · p. 4
• Sampling used 48 demographic strata, and final weights were calibrated to Census Bureau CPS benchmarks across demographic and geographic characteristics.Source: §4 · p. 4
• Visualization questions were administered only online; web-mode participants represented 96.3% of the panel’s weighted population representation.Source: §4 · p. 4
• After five respondents who skipped all 16 items were removed, the sample contained 1,168 adults; item-level omissions were included when responses were available.Source: §4 · p. 4

5.1 Exposure Across Types

• The median respondent reported familiarity with 10 chart types; the weighted mean was 9.87, and 11.4% reported exposure to all sixteen.More than 40% were unsure about at least one chart, but only 2.2% were unsure about nine or more.Source: §5.1 · p. 4
• Education and age were the strongest predictors of overall exposure: exposure rose with postsecondary education and declined across age groups.Respondents with postgraduate study or a professional degree reported up to a 28 percentage-point gain, while those aged 60 and older reported 22 percentage points fewer graphics than younger adults.Source: §5.1 · p. 4
• Income, gender, and race showed smaller overall associations; metropolitan versus non-metropolitan residence was not significantly associated with exposure after adjustment.Respondents with incomes above $100,000 reported exposure to approximately 7 percentage points more graphics than those with incomes under $30,000.Source: §5.1 · pp. 4–5

5.2 Exposure by Chart Type

• More than 85% reported exposure to bar, pie, line, and grouped bar charts, compared with less than half for five chart types, including heat maps and bubble charts.Reported exposure for the six chart types between these groups ranged from 52.8% to 67.7%; 75.7% reported exposure to donut charts.Source: §5.2 · p. 5, Figure 4 · p. 6

5.3 Chart Exposure by Demographic Group

• Across most chart types, higher education was associated with greater reported exposure, while older respondents generally reported less exposure.Gender and race or ethnicity showed no systematic differences after adjustment, although some chart-specific differences remained.Source: §5.3 · p. 6

5.4 Relationships Between Chart Types

• Exposure to bar, pie, line, and grouped bar charts was highly correlated, but familiarity with these common types was not strongly related to exposure to less common charts.Some less common types were also paired: 86% of respondents exposed to box-and-whisker plots reported exposure to dot plots with error bars.Source: §5.4 · p. 7

6 DISCUSSION

• The median respondent had seen 10 of 16 chart types, and 94% had seen at least four.Overall exposure nevertheless varied significantly across demographic groups.Source: §6 · p. 7
• Higher education is associated with exposure to more chart types, while older adults report fewer; education and age are the strongest predictors.Women, Black non-Hispanic adults, and high-income households also show smaller overall differences in exposure.Source: §6 · pp. 7–8
• Lower exposure does not inherently mean lower literacy, but familiarity may improve literacy for a given visualization mode.Exposure patterns can therefore complicate interpretation of literacy differences between groups.Source: §6 · p. 8
• Bar, pie, and line charts are widely seen, but fewer than half of adults have seen five tested types, including heat maps and bubble charts.The five less-seen types are dot plots with error bars, heat maps, barbell charts, box-and-whisker plots, and bubble charts.Source: §6 · p. 8
• Exposure gaps differ by chart type: for example, age differences were not significant for either familiar bar charts or uncommon barbell charts.Other differences vary across age, education, race and ethnicity, gender, income, and metropolitan residence.Source: §6 · p. 8
• Exposure to common chart types does not necessarily indicate broad exposure; familiarity with rarer types is associated with higher exposure across other types.Box-and-whisker plots and dot plots with error bars also frequently co-occur.Source: §6 · p. 8

6.1 Limitations

• The study omits maps, complex network diagrams, treemaps, and icon arrays, so its estimates do not cover all public-facing or assessed visual formats.Treemaps and icon arrays are used in some key visualization-literacy assessments.Source: §6.1 · p. 8
• Exposure may be undercounted because participants judged a simple chart icon paired with a name, which may not match what they recognize from real-world encounters.The authors also note that some chart types have multiple names or may be difficult to distinguish.Source: §6.1 · p. 8

7 CONCLUSIONS AND FUTURE WORK

• The study establishes population-level exposure estimates for sixteen common visualization types among U.S. adults, finding variation within the population and across chart types.Exposure occurs through both formal and informal pathways.Source: §7 · p. 8
• Visualization-literacy researchers should account for exposure differences when developing literacy tests and interpreting their results.Underlying exposure patterns may affect findings about overall literacy.Source: §7 · p. 8
• Educators could introduce less-common chart types earlier, since formal education is strongly related to exposure and earlier inclusion could increase baseline familiarity.The authors identify a lack of exposure to many chart types in primary and secondary education.Source: §7 · p. 8
• Practitioners should consider audience-specific exposure when selecting charts for important communications and may provide guidance for rarer graphics.Lack of reported exposure does not itself establish lack of interpretive ability, though unfamiliar charts may take longer to process or lead to misunderstanding.Source: §7 · pp. 8–9
• Future work could compare exposure measures, use real examples or alternate chart names, and link exposure measures with literacy assessments.The authors propose measuring direct relationships between exposure and understanding.Source: §7 · p. 9