Researchers have introduced the Cognitively-Ordered Visualization Literacy Test, or CoViLT, a task-grounded framework designed to operationalize Bloom’s revised taxonomy through observable visualization tasks. Developed by Furkan Aydin, Seyfullah Gokoglu, and Fatma Rumeysa Ucar and published in the International Journal of Data Science and Analytics, the instrument addresses a recognized gap in existing assessment methods. Although visualization literacy is increasingly critical for interpreting and making decisions from data-driven visual representations, earlier testing tools have often emphasized low-level decoding and simple value lookup while undersampling mid- to high-level visual-analytic reasoning.
To bridge this gap, the CoViLT framework maps Bloom’s revised cognitive taxonomy directly onto visualization tasks aligned with Brehmer and Munzner’s task typology. The resulting assessment integrates four distinct task categories: Retrieve/Identify, Compute/Derive, Compare/Detect, and Judge/Decide. Structurally, CoViLT consists of 48 dichotomously scored multiple-choice items, producing a possible total score range from 0 to 48. These 48 test items are systematically balanced across four cognitive levels as well as 12 different visualization types to ensure comprehensive coverage across both visual formats and analytical demands.
Prior to administering the assessment, the researchers evaluated the content validity of the 48 multiple-choice items with a panel of 20 experts. In this validation phase, the established Content Validity Ratio threshold was set at 0.42, and all 48 individual items successfully met this threshold. Following expert confirmation, the authors conducted a pilot study with 187 undergraduate students. From this initial testing cohort, six students failed to complete the examination, while an additional 14 participants were removed from the dataset. These 14 exclusions occurred because the respondents combined rapid item-response times of less than 10 seconds with an invariant response-position pattern, specifically selecting either the first or the fifth answer choice consistently across the test. These screening procedures produced a final analytic sample of 167 undergraduates.
Analysis of the pilot performance data revealed that item difficulty across the test spanned from 0.16 to 1.00, with an overall mean difficulty of 0.76. Item discrimination values ranged from 0.000 to 0.449, resulting in a mean discrimination value of 0.224. Evaluation of the assessment’s internal psychometric properties demonstrated a coefficient alpha of .735, alongside a standard error of measurement of 2.47. Based on these statistical outcomes, the authors concluded that the pilot findings offer preliminary empirical support for utilizing CoViLT’s overall score in group-level research focused on visual-analytic reasoning. However, the researchers emphasized that additional validation studies remain necessary before the test can be recommended for individual-level diagnostic evaluations or high-stakes testing applications.
Subject of Research: Visualization literacy assessment and visual-analytic reasoning
Article Title: Operationalizing visualization literacy through Bloom’s cognitive levels and visualization task typologies: CoViLT
Article References: Aydin, F., Gokoglu, S., & Ucar, F. R. (2026). Operationalizing visualization literacy through Bloom’s cognitive levels and visualization task typologies: CoViLT. International Journal of Data Science and Analytics, 22(1), 378. https://doi.org/10.1007/s41060-026-01356-7
Image Credits: AI Generated
DOI: 10.1007/s41060-026-01356-7
Keywords: visualization literacy, CoViLT, Bloom's revised taxonomy, data visualization, visual-analytic reasoning, psychometrics, pilot study, task typologies, Operationalizing, visualization, literacy, Bloom
News Source: Denise Maddox. (November 7, 2026). New CoViLT Test Assesses Higher Cognitive Levels in Visualization Literacy. Scienmag.
