Project Summary Situating Big Data seeks to marry theories of situated cognition to the big data movement by connecting clickstream data from technologies in isolation to key forms of multimodal data available from their contexts of use. Contextual data include individual and group discourse (online and in-room), individual and curricular artifacts, classroom assessments, and school performance data (grades and test scores). We will study the learning of a large number of students across five popular STEM games using diverse datasets, data types, and analyses: Clickstream telemetry data, a shared online community forum, and multiple formal and informal learning environments. We hope to generate a more data-driven methodology for investigating situated cognition through STEM technologies and new models for data-driven design. Situating big data in real-world environments for learning could create radically new models for data-driven education more broadly. Intellectual Merit Big data research techniques are revolutionizing domains including government, healthcare, athletics, entertainment but they require rich data trails from which to make inferences. Current applications of big data techniques often rely on data generated from relatively constrained learning tasks that fail to meet many goals of science education, such as inquiry and argumentation skills or scientific habits of mind. Educators know how to engineer such environments -- increasingly with digital gaming technologies (Clark, 2013; Clark, Tanner- Smith,& Killingsworth, 2013; Sitzmann, 2011; Young et al, 2012) -- but struggle to study learning in complex learning environments at scale due to difficulties in capturing how learners interact with a broad range of tools and resources across time and beyond the game platform itself. Enriching current efforts in big data for learning with data collection and analytic techniques using a situated cognition approach, in a theoretically principled and empirically validated manner, could dramatically improve the discussion of big data in the context of learning technologies by enabling consideration of the full ecosystem of learning by taking into account context and not just device. Broad Impact Our project could potentially transform current discussion of big data in science education by demonstrating where and how data can be situated in robust learning activities and reciprocally demonstrating how to assess the effectiveness and impact of interventions. Progress in such areas could open new frontiers for analysis and discovery in STEM learning. Formative and summative assessments might be collapsed into an integrated learning activity. Assessments could be regular, routine, and ongoing rather than one-shot events, creating more valid claims about what students know. The proliferation of digital devices and distribution makes the rapid expansion of some such big data systems for learning probable; our goal is to better align such systems toward the full gamut of STEM learning goals and with more informed awareness of where and how kids learn. STEM engagement and understanding is accomplished not solely through digital devices but rather through technologies in close collaboration with people, peers, interaction, other media and texts. Our analyses and assessments can and should reflect this fact.
|Effective start/end date||9/1/14 → 8/31/17|
- National Science Foundation (NSF): $243,584.00