Using Big Data and Visual Analytics to Investigate the Long-term, Cascading Effects of Informal STEM Learning

为了配合推进非正式STEM学习（AISL）计划的总体战略，推进非正式环境中STEM学习的设计和开发的新方法和基于证据的理解，来自俄勒冈州立大学、爱达荷大学和德克萨斯大学达拉斯分校的首席研究员将研究在线社交网络中的一系列数据，以确定非正式STEM教育长期影响的证据。长期跟踪非正式学习者以了解非正式学习经历的影响一直是一项长期、艰巨和难以捉摸的挑战。现在，随着大量数据在网上共享和存储，教育研究人员有了一个前所未有的机会来研究这些数据，并应用数据分析和可视化技术来确定非正式STEM学习的长期级联效应。研究结果将为数据分析工具的设计和开发提供信息，供教育从业者使用，以通过电视和电影以及非正式教育机构改善STEM在线学习体验。一个由学习科学家、计算机科学家、工程师、非正式STEM教育从业者、参与合作伙伴、广播媒体专业人士和政策制定者组成的独立外部批判性审查委员会，将确保对数据分析工具的研究、有效性和实用性进行强有力的评估，以改进实践。该领域将撰写一份总结报告，内容涉及研究和数据分析模型的科学和实践可靠性和有效性，以及该工作对非正式STEM教育从业者和政策制定者以外的受众的价值。该研究同时具有相关性，促进了数据挖掘和数据分析过程的创新使用，以更好地了解非正式学习者如何在一系列社交网络中交流STEM学习经验并与STEM内容进行互动。调查人员将研究：1)参加非正式STEM教育的学习者是否通过在社交媒体网络中讨论和分享信息来进一步学习；2)社交媒体中存在哪些类型的数据与理解随着时间的推移学习的级联影响相关；3)学习如何在共享的社交网络中独立发展，如果发现，可以提供一个预测计算模型，对正规和非正规教育产生重大影响。调查人员将使用现有的和修改的数据爬虫来搜索关键术语和短语，评估帖子，查询和博客中的峰值和变形，并使用他们的测试数据进行实验，以找到哪种类型或配置的关键字或搜索术语提供最可靠和准确的结果。将探索各种形式，与参与的合作伙伴和从业者一起测试各种策略。数据将被可视化，以表示以下学习维度：a)兴趣/影响，b)建议，c)理解/寻求知识，d)更深入的参与。

In concert with the overall strategy of the Advancing Informal STEM Learning (AISL) program to advance new approaches to, and evidence-based understanding of, the design and development of STEM learning in informal environments, Principal Investigators from Oregon State University, University of Idaho, and University of Texas at Dallas, will study a range of data in online social networks to identify evidence of the long-term impact of informal STEM education. Tracking informal learners over time to understand the impact of informal learning experiences has been a longstanding, daunting, and elusive challenge. Now, with massive amounts of data being shared and stored online, education researchers have an unprecedented opportunity to study such data and apply data analytics and visualization technologies to identify the long-term, cascading effects of informal STEM learning. Research findings will inform the design and development of a data-analysis tool for use by education practitioners to improve STEM learning experiences online, through television and film, and at informal education institutions. An independent external critical review board of learning scientists, computer scientists, engineers, informal STEM education practitioners, participating partners, broadcast media professionals, and policymakers, will ensure a robust evaluation of the research and effectiveness and utility of the data analysis tool to improve practice. A summary report for the field will be written on the scientific and practical reliability and validity of the research and data-analysis model, and the value of the work for audiences beyond informal STEM education practitioners and policymakers. The research is contemporaneously relevant, advancing innovative use of data-mining and data-analysis processes to better understand how informal learners communicate STEM learning experiences and interact with STEM content over time, across a range of social networks. Investigators will research: 1) whether learners who engage in informal STEM education experiences further their learning through discussions and sharing of information in social media networks, 2) which types of data are present in social media that are relevant for understanding the cascading impacts of learning over time, and 3) how learning may evolve independently within shared social networks, which, if discovered, could provide a predictive computational model with implications for significant impact across both formal and informal education. Investigators will employ existing and modified data crawlers to search for key terms and phrases, assess spikes and deformations in posts, queries, and blogs, and experiment with their test data to find which types or configurations of keywords or search terms deliver the most reliable and accurate results. A variety of formats will be explored to test various strategies with participating partners and practitioners. Data will be visualized to represent the following dimensions of learning: a) Interest/Affect, b) Recommendations, c) Understanding/Knowledge-Seeking, and d) Deeper Engagement.