# SVLAT: Scientific Visualization Literacy Assessment Test

arXiv: https://arxiv.org/abs/2603.19000 · HTML: https://mlsanity.com/paper/2603.19000/

## Key facts

- Cited by 6 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 19 Mar 2026.

## Built on by (most cited first, 6 of 6)

- [SciVisAgentBench: A Benchmark for Evaluating Scientific Data Analysis and Visualization Agents](https://mlsanity.com/paper/2603.29139/) (2026)
- [Toward AI VIS Co-Scientists: A General and End-to-End Agent Harness for Solving Complex Data Visualization Tasks](https://mlsanity.com/paper/2605.21825/) (2026)
- [Exploring LLM Agent Designs and Interaction Modalities for Scientific Visualization](https://mlsanity.com/paper/2604.27996/) (2026)
- [HiLSVA: Design and Evaluation of a Human-in-the-Loop Agentic System for Scientific Visualization](https://mlsanity.com/paper/2606.26614/) (2026)
- [SciVisAgentSkills: Design and Evaluation of Agent Skills for Scientific Data Analysis and Visualization](https://mlsanity.com/paper/2606.05525/) (2026)
- [Benchmarking Multimodal Large Language Models for Scientific Visualization Literacy](https://mlsanity.com/paper/2607.15176/) (2026)

## Abstract (arXiv)

Scientific visualization (SciVis) has become an essential means for exploring, understanding, and communicating complex scientific phenomena. However, the field still lacks a validated instrument assessing how well people read, understand, and interpret them. We present a scientific visualization literacy assessment test (SVLAT) that measures the general public's SciVis literacy. Covering a range of visualization forms and interpretation demands, SVLAT comprises 49 items grounded in 18 scientific visualizations and illustrations spanning eight visualization techniques and 11 tasks. Instrument development followed a staged, psychometrically grounded pipeline. We defined the construct and blueprint, followed by item generation, and expert review with five SciVis experts using the content validity ratio (mean CVR = 0.79). We subsequently administered a pilot test (30 participants) and a large-scale test tryout (485 participants) to evaluate the instrument's psychometric properties. For validation, we performed item analysis and refinement using both classical test theory (CTT) and item response theory (IRT) to examine item functioning and overall test quality. SVLAT demonstrates high reliability in the tryout sample (McDonald's omega_t = 0.82, Cronbach's alpha = 0.81). The assessment materials are available at https://osf.io/hr3nw/.