In educational assessment, rubrics are a key tool because they define clear criteria for evaluating learner responses and specify the evidence required. Yet, in research on automatic short-answer scoring, rubrics are seldom employed as explicit scoring references and, when they are, they are typically treated only as supplementary inputs. This study contrasts this by exploring the usage of rubrics as explicit scoring anchors in automatic short-answer scoring. It introduces a task definition for rubric-based automatic short-answer scoring, a version of short-answer scoring that uses scoring rubrics as scoring references. As an approach for implementing rubric-based short answer scoring, we introduce the idea of guided rubric answer alignment and provide two concrete architectures based on this, GRAASP and ToLeGRAA. Both are novel transformer-based architectures that use attention mechanisms to predict the alignment between student answers and rubric criteria. We compare them to multiple baselines using ALICE-LP, a novel German short-answer scoring dataset collected from formative assessments in authentic German school contexts, and the widely used ASAP-SAS dataset, collected at American schools. For both datasets, GRAASP and ToLeGRAA achieve highly competitive performance. By explicitly aligning learner responses with rubric criteria when assigning scores, these models demonstrate the feasibility of rubric-based short-answer scoring. This finding underscores the high potential of integrating rubric-driven scoring models into educational assessment.

In recent years, Learning Analytics (LA) has become a very heterogeneous research field due to the diversity in the data generated by the Learning Management Systems (LMS) as well as the researchers in a variety of disciplines, who analyze this data from a range of perspectives. In this paper, we present the evaluation of a LA tool that helps course designers, teachers, students and educational researchers to make informed decisions about the selection of learning activities and LA indicators for their course design or LA dashboard. The aim of this paper is to present Open Learning Analytics Indicator Repository (OpenLAIR) and provide a first evaluation with key stakeholders (N=41). Moreover, it presents the results of the prevalence of indicators that have been used over the past ten years in LA. Our results show that OpenLAIR can support course designers in designing LA-based learning activities and courses. Furthermore, we found a significant difference between the relevance and usage of LA indicators between educators and learners. The top rated LA indicators by researchers and educators were not perceived as equally important from students' perspectives.

This paper introduces the Visual Inspection Tool (VIT) which supports researchers in the annotation of multimodal data as well as the processing and exploitation for learning purposes. While most of the existing Multimodal Learning Analytics (MMLA) solutions are tailor-made for specific learning tasks and sensors, the VIT addresses the data annotation for different types of learning tasks that can be captured with a customisable set of sensors in a flexible way. The VIT supports MMLA researchers in 1) triangulating multimodal data with video recordings; 2) segmenting the multimodal data into time-intervals and adding annotations to the time-intervals; 3) downloading the annotated dataset and using it for multimodal data analysis. The VIT is a crucial component that was so far missing in the available tools for MMLA research. By filling this gap we also identified an integrated workflow that characterises current MMLA research. We call this workflow the Multimodal Learning Analytics Pipeline, a toolkit for orchestration, the use and application of various MMLA tools.

Learning Pulse explores whether using a machine learning approach on multimodal data such as heart rate, step count, weather condition and learning activity can be used to predict learning performance in self-regulated learning settings. An experiment was carried out lasting eight weeks involving PhD students as participants, each of them wearing a Fitbit HR wristband and having their application on their computer recorded during their learning and working activities throughout the day. A software infrastructure for collecting multimodal learning experiences was implemented. As part of this infrastructure a Data Processing Application was developed to pre-process, analyse and generate predictions to provide feedback to the users about their learning performance. Data from different sources were stored using the xAPI standard into a cloud-based Learning Record Store. The participants of the experiment were asked to rate their learning experience through an Activity Rating Tool indicating their perceived level of productivity, stress, challenge and abilities. These self-reported performance indicators were used as markers to train a Linear Mixed Effect Model to generate learner-specific predictions of the learning performance. We discuss the advantages and the limitations of the used approach, highlighting further development points.