The 7th International Learning Analytics and Knowledge (LAK) Conference is being held in Vancouver. Here Professor Järvelä's keynote presentation slides from the conference.
The Learning and Educational Technology (LET) research unit of the University of Oulu, where the SLAM research takes place, has a systematic research collaboration relationship with the Welten Institute of Open Universiteit in Heerlen, the Netherlands. Both groups share an interest in advancing our understanding of learning and the role of technology in supporting technology enhanced learning innovations.
As part of this ongoing collaboration, a PhD student in the SLAM project, Héctor J. Pijeira Díaz, visited the Welten Institute in the week of February 6th to 10th. There, Héctor had the opportunity to present the SLAM research to the overall audience, to the WEKIT project (http://wekit.eu/) team, and more specifically, the line of his PhD studies to the Technology Enhanced Learning Innovations (TELI) group at Welten Institute. These three presentations gave Héctor the opportunity to interact with the PhDs and staff there, be exposed to different viewpoints on his and other research, and receive constructive feedback on how to progress in a direction meaningful to the community.
Apart from these group presentations, Héctor was honoured to have 1:1 insightful meetings with researchers at different stages in their careers, from early stage PhD students to accomplished professors. The expertise of these researchers varied from learning sciences to learning technology to teaching as they belong to the three different groups the Welten Institute is comprised of, Fostering Efficient, Effective and Enjoyable Learning (FEEEL), TELI, and Teachers and Teachers’ Profesionalisation (T2P).
SLAM’s PhD student Héctor had the chance to learn about other relevant projects such as the abovementioned Wekit, the MOOQ project dealing with the quality of MOOC courses, and projects dealing with the biopsychology of learning and the impact of lifestyle on learning.
In a hands-on approach, Héctor was invited to an eye-tracking training session by the expert in the field Dr. Halszka Jarodzka (the SLAM first experiment data collection included eye-tracking); and also to try the latest version of the award-winning Presentation Trainer, which uses Microsoft Kinect Sensor technology to provide immediate feedback based on experts’ recommendations, on presentations aspects such as posture, hand position and movement, voice pitch and cadence, and more. The presentation trainer is developed by Dr. Jan Schneider.
Overall, the visit was a fruitful exchange of ideas from the SLAM project research and a variety of projects being carried out at the Welten Institute, all with the common goal of advancing our understanding of learning and the role of technology to support learning research and practice.
SLAM project is lucky to have Assistant Professor Omid Noroozi joining the team with his strong expertise. Below you can read his introduction:
"I am currently an assistant professor of Educational Technology at Wageningen University in the Netherlands. In my works, I design, implement and evaluate various Computer-Supported Collaborative Learning (CSCL) environments for argumentation-based learning. Specifically, I design various types of instructional interventions, e.g. scaffolding and scripting approaches, and test their effects on a variety of learning process and outcome aspects in both real educational and control-based laboratory settings. My projects covers a wide range of advanced qualitative and quantitative methods to analyze various aspects of learning processes and outcomes in CSCL environments.
It is an honour to be part of the SLAM project and work with a great team. Working on SLAM project provides me with the opportunity to be involved in a cutting-edge research domain in the learning sciences. What inspires me most about this project is the application of advanced technologies and the learning analytics to enhance regulation of individual and collaborative learning. I will start on the SLAM project with redrafting a paper on designing learning analytics dashboard feedback. In this paper the aim is to provide a conceptual framework for dashboards, by grounding feedback theories in the learning sciences and more precisely on regulatory mechanisms underlying learning processes. I hope to bring added value for this great project."
Collaboration with the Oulu University Center for Machine Vision and Signal Processing (CMVS) experts Dr. Ilkka Juuso and M.Sc. Iman Alikhani have produced powerful physiological data visualizations for the project. The next steps are to get hands-on and dive deeper into the data while at the same time further developing some of the technical aspects like quality of the data and types of visualizations.
We have also been exploring facial expression analysis opportunities in our collaborative learning session video data with Assoc. Prof. Guoying Zhao and Dr. Xiaohua Huang. This machine learning based method holds great potential for us to study and work with the data. The first experiments with parts of the video data are promising and the next step is to refine and apply the method to a larger set of data.
Assoc. Prof. Zhao will also present a keynote lecture at the forthcoming EARLI Special Interest Group 27, Online Measures of Learning Processes conference held in Oulu from November 29 through December 1. Members of the SLAM team are responsible for the local arrangements of the conference. We are also looking forward to presenting our first results and to contact with other researchers working with similar kinds of data-sets.
This year EARLI Metacognition Sig 16 was held in Nijmegen Netherlands. The SLAM team presented a poster in a workshop Using Data Visualizations to Understand and Reason about Self-Regulated Learning organized by Prof. Roger Azevedo, NC State university.
The SLAM team members, Dr Jonna Malmberg and Prof. Sanna Järvelä also presented their initial findings of using multichannel data to understand regulated learning in the context of collaboration in the symposiums 'Advances in scaffolding metacognition with advanced learning technologies' and E-CIR invited Symposium: Measuring and supporting students' self-regulated learning in adaptive educational technologies.
Encouraging feedback was received and many questions arised. Our discussant Prof. Phil Winne concluded that symposium represents state of the art in SRL research, but also reminded that it is important to consider what are the standards for new methodology.
Spring was a hectic period for the SLAM team due to intense data collection. The team collected data from two high school Advanced Physics courses involving a total of 43 students and 36 sessions. During the courses the students solved physics related problems using experimental equipment including lasers, lenses and online simulations. Twelve of the sessions were held in LeaForum research laboratory equipped with the MORE system  for high quality audio and video recording and twelve Empatica E4 sensors  were used to track students’ physiological signals. This yielded 101 hours of video, 266,216,000 (more the a quarter of a billion) data points of physiological data and 236,000 (almost a quarter of a million) log events from the virtual learning environment.
Virtual learning environment – Open edX
During the courses, Open edX served as the virtual learning environment (VLE). Taking advantage of its open source character, the SLAM team installed and hosted its own instance of the platform on a dedicated server. One of the main reasons why the team decided to use edX was its built-in event tracking system. It made it possible to track students’ navigation and interactions with the platform.
In practice, the VLE included task related materials and the script for collaboration. The script followed the same form in most of the sessions. Students started by doing a retrospective evaluation considering their last lesson. A collaborative learning dashboard was provided to support this phase (see the gauges below). After this, students continued with answering orientation questions, and then moved on to collaboratively working on the task. At the end of the lessons, students reflected on their performance by answering a short survey.
Values for the dashboard and evaluation came from the survey students filled out during the previous session. In addition, the history of measures was provided for students to see the overall evolution of their group according to their own self-reports. The dashboard as a tool was evaluated using a questionnaire on usability, usefulness and user experience. A quick look at the results suggests that many of the students were interested in the gauges and considered them useful. Further analysis might reveal how it affected their learning process.
Left: Example of an optics task with embedded simulation Right: The Learning dashboard with gauges
The next challenge for the SLAM team is to continue the analysis with the extensive, multimodal dataset. Combining very different modalities of data isn’t easy and each data type also has its own challenges. The pilot study carried out during the spring in 2015 gave some useful ideas on where to start, which will be developed and enriched with the vast amount of newly collected data.
Over one hundred hours of video data will be compressed and coded in order to gain an insight into how students collaborated. The physiological data, recorded during collaborative learning situations, is being synchronized at both individual and group levels. Further signal-specific data processing and analysis will be carried out in collaboration with the Oulu University Biosignal Processing Team, led by Prof. Tapio Seppänen and with extensive expertise in this area. Events from the Open edX VLE are also to be synchronized with the physiological data, as an indispensable context-related complement. To this purpose, the raw data in the databases from the platform and text files with log data are being processed.
To sum it up, extensive, multimodal data has been collected from the February to June 2016 SLAM experiment. An exciting period is starting, as we start to focus on data analysis. We will keep you posted developments
 Keskinarkaus, A., Huttunen, S., Siipo, A., Holappa, J., Laszlo, M., Juuso, I., Väyrynen, E., Heikkilä, J., Lehtihalmes, M., Seppänen, T., & Laukka, S. (2015). MORE–a multimodal observation and analysis system for social interaction research. Multimedia Tools and Applications, 1-25.
 Garbarino, M., Lai, M., Bender, D., Picard, R. W., & Tognetti, S. (2014, November). Empatica E3—A wearable wireless multi-sensor device for real-time computerized biofeedback and data acquisition. In Wireless Mobile Communication and Healthcare (Mobihealth), 2014 EAI 4th International Conference on (pp. 39-42). IEEE.
Three SLAM researchers – Paul A. Kirschner, Hendrik Drachsler and Héctor J. Pijeira Díaz – participated in the Sixth Learning Analytics and Knowledge Conference (LAK16). The conference was held in Edinburgh, the inspiring capital of Scotland, from April 25 to 29.
Professor Paul Kirschner, SLAM principal researcher, engaged the audience to reflect on utopias and dystopias in the field of learning analytics with his keynote opening the Thursday session. The talk made a strong impact on the audience and Professor Sir Timothy O'Shea, first in the question round, expressed: “that's one of the best keynotes I've ever seen”. As a summary, following are Prof. Kirschner’s dystopias and utopias:
Associate Professor Hendrik Drachsler was one of the conference program chairs for the research track. His full paper “Privacy and Analytics – it’s a DELICATE Issue. A Checklist for Trusted Learning Analytics” co-authored with Wolfgang Greller was awarded Best Paper. Congratulations Hendrik on behalf of your colleagues from the SLAM team!
Opening the collaborative learning session, PhD student Héctor J. Pijeira Díaz presented the physiological approach to the session topic being explored in the SLAM project. The full paper, co-authored with Assoc. Prof. Hendrik Drachsler, Prof. Sanna Järvelä and Prof. Paul Kirschner, is entitled “Investigating collaborative learning success with physiological coupling indices based on electrodermal activity”, and published in the Conference Proceedings.
Héctor was also one of the ten PhD students accepted to the Conference Doctoral Consortium (DocCon). Being a full day event of engaging and constructive feedback oriented academic discussions, Héctor had the opportunity to present his research line within the SLAM project to fellow PhD students and the DocCon chairs as well. In the Demo & Poster session, Héctor showed to interested delegates the current version of the dashboard under development in SLAM and the direction it is moving towards.
116 papers were submitted to the conference, out of which 36 were accepted, for an acceptance rate of 31%. This was also a LAK record breaking edition according to the number of participants – over 450 delegates from over 35 countries.
The SLAM data collection is underway. Working together with the physics teachers from the Oulu University Teacher Training School, the SLAM team developed the lesson plans for an advanced physics course.
Altogether 43 students are participating in this data collection. The lessons include students collaborating to solve complex tasks using an Open edX based learning environment, from which log data is collected. Physiological data is collected using Empatica sensors. Part of the data collection is taking place in the LeaForum, which will provide additional video data.
Since SLAM project is using advanced technologies, we also need advanced expertise to provide support for this side of the study. That's why we are lucky to have Abdul Moiz in our team to take care of some of the technical aspects.
Here Abdul answers on couple of questions about himself:
1. What’s your background, expertise and interests?
I am a currently a Doctoral student in University of Oulu and my previous educational background is MS in wireless communication engineering. My area of expertise are Software development and wireless communication engineering. I also have quite extensive experience in fixed network planning, design and implementation. My interest is frontend as well as backend software development.
2. What’s your role/roles on SLAM project?
I am working in SLAM as an application developer as well as backend server support. My main task is to transfer the real-time sensors data from Empatica E4 wrist band to the backend server via a tablet PC as an intermediate node. My responsibilities also include support to edX server.
3. What you expect to learn from this project?
I love new challenges and want to learn more and more new things. This project has exposed me to many more new technologies and I have learned a lot of things in this project. I expect to learn more in detail about the setup and management of edX server and how to make it more robust and fail proof.