Multimodal Learning Analytics is a field that studies how to process learning data from dissimilar sources in order to automatically find useful information to give feedback to the learning process. This work processes video, audio and pen strokes information included in the Math Data Corpus, a set of multimodal resources provided to the participants of the Second International Workshop on Multimodal Learning Analytics. The result of this processing is a set of simple features that could discriminate between experts and non-experts in groups of students solving mathematical problems. The main finding is that several of those simple features, namely the percentage of time that the students use the calculator, the speed at which the student writes or draws and the percentage of time that the student mentions numbers or mathematical terms, are good discriminators between experts and non-experts students. Precision levels of 63% are obtained for individual problems and up to 80% when full sessions (aggregation of 16 problems) are analyzed. While the results are specific for the recorded settings, the methodology used to obtain and analyze the features could be used to create discriminations models for other contexts.
The emergence of tools that support fast-and-easy visualization creation by non-experts has made the benefits of InfoVis widely accessible. Key features of these tools include attribute-level operations, automated mappings, and visualization templates. However, these features shield people from lower-level visualization design steps, such as the specific mapping of data points to visuals. In contrast, recent research promotes constructive visualization where individual data units and visuals are directly manipulated. We present a qualitative study comparing people's visualization processes using two visualization tools: one promoting a top-down approach to visualization construction (Tableau Desktop) and one implementing a bottom-up constructive visualization approach (iVoLVER). Our results show how the two approaches influence: 1) the visualization process, 2) decisions on the visualization design, 3) the feeling of control and authorship, and 4) the willingness to explore alternative designs. We discuss the complex trade-offs between the two approaches and outline considerations for designing better visualization tools.
Previous research has identified trade-offs when it comes to designing visualization tools. While constructive "bottomup" tools promote a hands-on, user-driven design process that enables a deep understanding and control of the visual mapping, automated tools are more efficient and allow people to rapidly explore complex alternative designs, often at the cost of transparency. We investigate how to design visualization tools that support a user-driven, transparent design process while enabling efficiency and automation, through a series of design workshops that looked at how both visualization experts and novices approach this problem. Participants produced a variety of solutions that range from example-based approaches expanding constructive visualization to solutions in which the visualization tool infers solutions on behalf of the designer, e.g., based on data attributes. On a higher level, these findings highlight agency and granularity as dimensions that can guide the design of visualization tools in this space.
Learning analytics has been as used a tool to improve the learning process mainly at the micro-level (courses and activities). However, another of the key promises of learning analytics research is to create tools that could help educational institutions at the meso-and macro-level to gain better insight into the inner workings of their programs in order to tune or correct them. This work presents a set of simple techniques that, if applied to readily available historical academic data, could provide such insights. The techniques described are real course difficulty estimation, course impact on the overall academic performance of students, curriculum coherence, dropout paths, and load/performance graph. The usefulness of these techniques is validated through their application to real academic data from a Computer Science program. The results of the analysis are used to obtain recommendations for curriculum redesign.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.