Data literacy has become a key learning objective in K-12 education, but it remains an ambiguous concept as teachers interpret it differently. When creating assessments, teachers turn broad ideas about "working with data" into concrete decisions about what materials to include. Since working with data visualizations is a core component of data literacy, teachers' decisions about how to include them on assessments offer insight into how they interpret data literacy more broadly. Drawing on interviews with 13 teachers, we identify four challenges in enacting data literacy in assessments: (1) conceptual ambiguity between data visualization and data literacy, (2) tradeoffs between using real-world or synthetic data, (3) difficulty finding and adapting domain-appropriate visual representations and data visualizations, and (4) balancing assessing data literacy and domain-specific learning goals. Drawing on lessons from data visualization, human-computer interaction, and the learning sciences, we discuss opportunities to better support teachers in assessing data literacy.

翻译：数据素养已成为K-12教育的关键学习目标，但教师对其诠释各异，使其仍属模糊概念。在创建评估时，教师需将关于"处理数据"的宽泛理念转化为选择具体评估材料的决策。由于数据可视化是数据素养的核心组成部分，教师如何将其纳入评估的决策，可揭示其对数据素养更广义的理解。基于对13位教师的访谈，我们识别出实施数据素养评估面临的四重挑战：（1）数据可视化与数据素养之间的概念模糊性；（2）使用真实数据与合成数据的权衡；（3）难以获取并适配特定学科领域的可视化表征与数据可视化；（4）在评估数据素养与学科特定学习目标之间寻求平衡。借鉴数据可视化、人机交互及学习科学领域的研究成果，我们探讨了为教师评估数据素养提供更好支持的潜在机遇。