The Design and Impact of Introducing Humanoid Robots
for Analyzing University Classroom Interactions
| Number | Topic | Journal | Page | Year | URL | Rank | Authors | Category | Experiment | Participants | Measurement | Results | Main Points | Notes |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Work in Progress: A Constructivist Didactic Methodology for a Humanoid Robotics Workshop | Frontiers in Education Conference Proceedings | pp.1-3 | 2012 | Unranked | Miranda, A; Bolea, Y; Grau, A; Sanfeliu, A | Robotics in Education | 45-hour workshop, lesson is conducted using constructivist methods by splitting into small groups, hands-on experimentation, promoting discussion, ask students to develop own ideas as in project-based learning. | 20 engineering degree students in Industrial School of Robotics, Technical University of Catalonia UPC. | Qualitative | Claimed to be positive, students showed more interest in learning. Work-in-progress, more workshops to be carried out in future. | Use of humanoid robots through constructivist methods enhances learning and teaching interest. | I put this one in because this is the only higher education research I can find, but the article doesn't actually provide any data to be seen. | |
| 2 | Impact of Using an Educational Robot-Based Learning System on Students' Motivation in Elementary Education | IEEE Transactions On Learning Technologies | Vol.7(4), pp.333-345 | 2014 | Q3 | Chin, KY; Hong, ZW; Chen, YL | Robotics in Education | Students were assigned to use either the proposed robot-based learning system or a PowerPoint-based learning system. The robot is used as an assistant; it performs gestures according to the instruction materials presented in slides, such as when the "Guess What am I?" slide is shown, the robot waves. The teacher also asks students to answer to the robot, promoting interaction with the robot. | 1 homeroom teacher and 52 second-grade students from two classes at an elementary school in Taiwan. | Mixed | The pre-test and post-test results show that the proposed learning system improved student performance more than the PowerPoint-based learning system. A questionnaire based on the Instructional Materials Motivation Survey (IMMS) was employed to measure four motivational factors (attention, relevance, confidence, and satisfaction), showing relevance and satisfaction as highest rated for the proposed robot-based learning system. | Social interaction with humanoid robots have positive results on motivation and performance. | Why is this kind of social experimentation only done on children? | |
| 3 | Kindergarten Social Assistive Robot (KindSAR) for children's geometric thinking and metacognitive development in preschool education: A pilot study | Computers in Human Behavior | Vol.35, pp.400-412 | 2014 | Q1 | Keren, G; Fridin, M | Robotics in Education | NAO robot introduced in class following a set of procedures. First procedure is cognitive stage, robot teaches children. Second procedure is metacognitive stage, child teach another child on how to interact with robot. Post hoc analysis of video footage to derive velocity of cognitive learning (Keren, Ben-David, & Fridin, 2012), an observational checklist for metacognition by Whitebread et al. (2009), and child-robot interaction measurement index (Fridin, 2014) used to evaluate the quality of child-robot interaction, which combines eye contact and emotional expressions, such as voice, face, and body expressions (with Cronbach's alpha of 0.686). | 3 groups of children (Israeli born, 10 boys and 7 girls), 1 technician, 1 staff member. | Mixed | Data revealed significant improvement in the children's metacognitive abilities in the second session of the experiment compared to the first. KindSAR can provide children psychology development data in real-time for teachers to regulate. | Social interaction with humanoid robots show improvement in geometric thinking and metacognitive performance. | Robots like NAO can be used to collect data as part of the research/usage. Analysis methods used in this research can be tried out on university students. | |
| 4 | Employing Humanoid Robots for Teaching English Language in Iranian Junior High-Schools | International Journal of Humanoid Robotics | Vol.11(3) | 2014 | Q3 | Alemi, M; Meghdari, A; Ghazisaedy, M | Robotics in Education | 45-item vocabulary test (pre-test) taken from their textbook (Prospect-1) to 60 participants assessed vocabulary recognition and production. 46 students who were at the beginner's levels were selected for the study, 30 for NAO robot-assisted language learning (RALL) and 16 for non-RALL groups. The students who had achieved a score of two standard deviations above and below the mean were chosen for the study totaling a number of 46 students. At the end of the study, the same test was administered as an immediate post-test and the questions were counter balanced and administered as a delayed post-test two weeks after the treatment process. This test enjoyed the reliability of Cronbach alpha of 0.89. Thus, the high reliability obtained is a clear evidence of the consistency of scores over time and that the vocabulary test is a reliable source of data. | 46 Middle School Iranian students | Quantitative | The results obtained from the descriptive statistics show that the mean of the pretest is 13.53, that of the post-test is 39.76, and mean of the delayed post-test is 39.50. The results clearly show that the students' improvement has been tripled and remained in the delayed post-test, thus providing evidence of the effectiveness of acquiring rudimentary linguistic skills through the treatment. The scores obtained in the delayed post-test help show improved knowledge retention over non-RALL groups. | Uses social interaction to aid learning. Teacher starts conversation with NAO robot, and also asks students to answer/interact with NAO. NAO will perform gestures, such as when teacher says policeman, NAO will act like a policeman with gun, etc. | Cronbach's Alpha | |
| 6 | We, robot: Using robotics to promote collaborative and mathematics learning in a middle school classroom | Middle Grades Research Journal | Vol.9(3), pp.73(16) | 2014 | Unranked | Ardito, G; Mosley, P; Scollins, L | Robotics in Education | The students were engaged in robot challenges that required them to work together over the course of 1 semester to enhance problem solving capabilities. Textual analysis was done on students writings in a class blog, and their mathematics subject grades were assessed. | Approximately 1600 students | Mixed | Analysis of the end of year state-mandated mathematics exam showed that students with the most involvement in the robotics program achieved higher scores on concepts associated with algebra, measurement, and probability, all skills related to the group problem solving with which they were engaged. Textual analyses (forming a word cloud) of student writing via a class blog demonstrated the development of student experiences and perceptions of collaboration in important and interesting ways, most frequent words used are "we" (696 times in total) and "our". While certainly these terms are consistent with the writing prompts themselves, they are too prevalent to be found in the students' writing by sheer coincidence. | Introducing a robotics subject have positive impact in problem solving skills and promotes collaboration even if it is not directly used to aid in a particular subject. | Supports problem-based learning and constructivist theory | |
| 7 | Collaboration by Design: Using Robotics to Foster Social Interaction in Kindergarten | Computers in the Schools | Vol.30(3), pp.271-281 | 2013 | Unranked | Lee, KTH; Sullivan, A; Bers, MU | Robotics in Education | Each child was randomly placed in one of two groups: (a) In the first classroom, children were part of an instructionist environment in which they learned how to program their robot by participating in pre-designed teacher-guided challenges (labeled the structured curriculum group); or (b) in the second classroom, a constructionist approach was followed. Children did not have structured experiences, and instead were given free time to explore interesting ideas and concepts on their own (labeled the unstructured curriculum group). Assessment on number of reported interactions triangulated with the number of interactions observed across the three videos taken per group. A total of six video cameras (three per group) recorded the entirety of the camp at different angles. Any interactions that were self-reported but not observed in the video were taken out of the analysis. | 19 students | Mixed | The unstructured group reflected higher mean numbers of interactions. Findings from this study indicated that to foster peer collaboration, a less structured learn-by-doing approach might be useful for teachers when integrating technology. | The main point supports constructionism, and it suggests that for robots to be integrated in classrooms, the lesson needs to be less structured. Don't follow a set itenary or lesson plan; instead, focus on what constructionism does best. | How to design a framework to introduce NAO into classrooms without a big change in lesson structure or teacher skillset? | |
| 9 | Informing Pedagogical Action: Aligning Learning Analytics With Learning Design | American Behavioral Scientist | Vol.57(10), pp.1439-1459 | 2013 | Q2 | Lockyer, L; Heathcote, E; Dawson, S | Learning Analytics | Uses a learning design drawn from a repository established through an Australian project that identified, reviewed, and documented examples of university courses that effectively used technology to facilitate flexible learning (Agostinho et al., 2008). Applies analytics on a case-by-case basis comprising of individual, small group, and large group learning tasks and use of online resources and discussion forums. | - | Qualitative |
Common elements of learning design:
|
Learning analytics allows a learning design to be evaluated in light of its pedagogical intent, using a rich set of real-time, behavior-based data on learner interaction within the learning environment. Diagrams portraying social networks/interactions formed in a collaborative group can be mapped out and analyzed. | Mapping of interactions is nice. Can use concepts of learning design here and the points made in order to measure and improve pedagogical intent. | |
| 10 | Teacher regulation of multiple computer-supported collaborating groups | Computers in Human Behavior | Vol.52, pp.233-242 | 2015 | Q1 | van Leeuwen, A; Janssen, J; Erkens, G; Brekelmans, M | Learning Analytics | The classes of students and their teachers made use of the CSCL environment called Virtual Collaborative Research Institute (VCRI) which logs communications during student collaboration. All students had their own computer in the classroom, and synchronously communicated with their group members through chatroom. Teachers make use of learning analytics information provided to them via this tool to try and regulate the class, deciding when to intervene, etc. Assessment using interviews for teachers' reports about their strategies for diagnosing and intervening, and the associated challenges and opportunities. | 2 male history teachers, 51 secondary school students (mean age of 14) | Qualitative | Teachers must continuously choose which level (individual, group and class) to monitor and how to divide their attention. In this study, with seven collaborating groups, found that as a result of high information load the teachers' priority was to detect students' needs and offer support where needed, rather than the question whether it was better to let students work it out themselves before intervening. The manageability of the available information decreased as a result of the number of groups, and the teachers were not always able to maintain an overview of all student activities. | Must decrease information load when trying to apply learning analytics for teachers: e.g. supporting tools, balance between number of groups and group size. May have a need for tools that analyze the texts students write and indicate weak or strong points. | Can and should NAO provide real-time learning analytics? When teachers had access to learning analytics, they were not better at detecting problematic groups, but they did offer more support in general, and more specifically targeted groups that experienced problems. This could indicate that learning analytics increase teachers' confidence to act. | |
| 11 | Learning Analytics at "Small" Scale: Exploring a Complexity-Grounded Model for Assessment Automation | Journal of Universal Computer Science | Vol.21(1), pp.66-92 | 2015 | Q4 | Goggins, S; Xing, W; Chen, X; Chen, B; Wadholm, B | Learning Analytics | Through analysis of existing analysis models, decided that TAN model has demonstrated remarkable accuracy [Keogh 99][Friedman, 97]. A set of words and actions were analyzed e.g. since the task is to draw a triangle, conversations with the word "triangle" or usage of the "segment" tool indicate collaboration towards a goal, "?" indicates not moving (status-quo maintained) as they are questioning back and forth, and using the wrong tool such as "perpendicular" indicates moving away from the goal. | - | Quantitative | An assessment model was developed, achieving the highest accuracy (95.8%) as compared to baseline models. A web-based tool developed to visualize time-series activities, assesses small group learning automatically, and also offers actionable intelligence for teachers to provide real-time support and intervention to students. | Group dynamics and interaction can be modeled via theoretically grounded, simple rules. "Many studies of small group learning use summative assessment methods (e.g. final solution, grade) to measure performance. These assessment approaches usually overlook the collaborative process and the affordances of technology in contribution to group learning." Many quantitative explorations are based on ad-hoc guesswork to build their measures and do not systematically address complex small group dynamics and interactions [Mirriahi et al., 13]. | The points highlighted in Main Points is exactly my concern. | |
| 12 | A visual recommender tool in a collaborative learning experience | Expert Systems with Applications | Vol.45, pp.248-259 | 2016 | Q1 | Anaya, AR; Luque, M; Peinado | Learning Analytics | Survey to obtain feedback on the developed tool which applied this quantitative method of data mining, results are positive. The data mining method was formed through vigorous literature reviews and analysis. | 23 students | Qualitative | This research resulted in a quantitative analysis method based on a set of statistical indicators derived from the forum interactions such as initiative, activity, constancy and regularity, and acknowledgment by their peers (Anaya & Boticario, 2011a). Two approach to assess student collabration: clustering (group students according to collaboration) and metric (calculate metrics based on student collaboration indicators). Uses influence diagrams (Howard & Matheson, 1984) based on Bayseian network to solve uncertainties in collaboration, this is used to suggest decisions if the tutor should intervene or not based on decision variables and the decision maker's preferences. | "In educational environments, where the student is responsible for the collaboration process, or where there are a large number of students, regular and frequent processing of collaboration poses serious challenges if tutors have to address this task. That's why need an autonoumous tool for data mining. | The result of this research is a quantitative method, but carrying out this research is qualitative as it uses surveys. | |
| 15 | Expressive Robots in Education - Varying the Degree of Social Supportive Behavior of a Robotic Tutor | Proceedings of the 28th ACM Conference on Human Factors in Computing Systems (CHI2010) | pp. 1613-1622 | 2010 | Unranked | Saerbeck, M; Schut, T; Bartneck, C; Janse, M | Robotics in Education | Use of a robotic tutor capable of doing gestures, speech and emotional expressions. iCat from Philips Research was used in five behavior dimensions to vary the social supportiveness for its role as tutor. Participants were randomly assigned between one of two modes implemented in the robot, Neutral or Socially Supportive. Video recordings were analyzed, survey on first impressions, a language test (to grade performance), a questionnaire to assess motivation and finally a final interview was administered. | 9 girls and 7 boys in the Primary International School of Eindhoven (10-11 years old) | Mixed | Participants in the social supportive condition were significantly more motivated. The neutral condition is solely on knowledge transfer while for the social supportive condition the focus is on active dialog and positive social supportive behaviors. |
"Currently, most learning materials and educational technology focus solely on a mere knowledge transfer and hardly on the dialogue and social supportiveness aspects."
Five behavior dimensions used to vary social supportiveness of a robotic tutor:
|
This test also takes into account the proximity of robot and student, and the number of touches on the robot. Social interactions like this can be important in my data. | |
| 17 | Exploring the educational potential of robotics in schools: A systematic review | Computers and Education | Vol.58(3), pp. 978-988 | 2012 | Q1 | Benitti, FBV | Robotics in Education | Analyzes the purpose of the study, the content to be taught with the aid of robotics, the type of robot used, the research method used, and the sample characteristics (sample size, age range of students and/or level of education) and the results observed for 10 selected articles. | - | Review |
|
For an experiment to be classified as a true experimental design, it must fit all of the following criteria (Shuttleworth, 2008):
|
Decide what to test and how to obtain reliable data. | |
| 20 | A Discourse Analytic Approach to Video Analysis of Teaching | Journal of Teacher Education | Vol.66(3), pp.245-260 | 2015 | Q1 | Schieble, M; Vetter, A; Meacham, M | Learning Analytics | During the first phase, the authors organized all data sources generated by individual participants and treated each individual as an intrinsic case within the larger qualitative study to analyze the complexities of each teacher candidate as an individual with particular histories of participation. Second phase is to re-examined teacher reflections, discourse analysis charts, interviews, and videos for patterns related to overarching themes that indicated her identity work. | 30 preservice English teachers | Qualitative | Without data of classroom interactions through modes including video, preservice teachers rely on memory and their construction of events to engage in reflection of teaching and learning. Discourse analytic approaches using positioning theory allow candidates to focus specifically on how their linguistic and non-verbal choices impact the enactment of identities related to teaching and learning. This type of framework helps candidates notice how they build relationships with students and scaffold academic language in the classroom in ways that invite students to see themselves as capable literacy learners and take more ownership over their learning. |
Discourse analytics need to show:
|
Positionality is defined as "the discursive process whereby selves are located in conversations as observably and subjectively coherent participants in jointly produced storylines" (Davies & Harre, 1990, p. 91). Positioning, then, is how people engage in conversation to "situate themselves and others with particular rights and obligations [and] take up or resist positions others create for them" (Rex & Schiller, 2009, p. 9). | |
| 25 | Diagnostic Robotic Agent in the RoboEduc environment for educational robotics | Latin American Robotics Symposium | pp.131-136 | 2008 | Unranked | Silva, AF; Barros, RP; Azevedo, SO; Silva, A; G. Goncalves, LM | Educational Theory | Propose the use of Vygotsky theory into a Diagnostics Robotic Agent (DRA), to be used in the RoboEduc, a software environment for Educational Robotics, running in a public school for 4 years. The interaction of each student with the DRA software records the mistakes, how long the student(s) leads to perform a task and how many times requested help from a more capable partner. When students finish a task, the system updates the zpd of the student based on information obtained. Each time the student controls, teaches or programs a robot, the DRA compares the plan generated with the five best results. For each plan, it is counted the number of operations and the time spent to accomplish the task. DRA calculates the minimum and maximum time to complete a given task. After determining which plans are more or less shares, the DRA draws a pattern of performance of the task. The agent analyzes the plans generated by the students every time they try to improve the performance of the robot to perform a mission. When the student finishes a mission, the DRA sends the information of the best plans and performance of the student to a central database. By using this information, the module of the GDR Author creates a bank of plans developed for each task and determines the zpd of each student. | Children up to 10 years old in a public school | Qualitative | Emergence of the ZPD of children and of identifying the potential of them. An individual who has a skill in some zpd can interact with other one more capable, more enabled in the decision-making, on to intervene in the process of construction of the concepts discussed in the workshops. | To increase success in e-learning implementation for the nursing programme, lecturers should be educated regarding proper instructional design so that their content delivery blends well with the technology and pedagogy. | I don't like how this research quantifies ZPD, but it is something to discuss about. Also, what can we do with this ZPD information? This article does not discuss that at all, other than as a measure of learning potential. | |
| 27 | Experiential Learning in the Development of a DARwIn-HP Humanoid Educational Robot | Journal of Intelligent & Robotic Systems | Vol.81(1), pp.41-49 | 2016 | Q3 | Yi, H; Knabe, C; Pesek, T; Hong, DW | Robotics in Education | Using a Dynamic Anthropomorphic Robot with Intelligence(DARwIn)- High Performance(HP) as an educational tool in robotics undergraduate classes. The impact is experimented on number of activities, number of extracurricular activies related to robotics and outreach activities (e.g. competitions, free-time, etc.) before and after participating in development with the DARwIn-HP robot. | 65 undergraduates | Quantitative | This study shows that undergraduates who attended the DARwIn-HP development are likely to feel strongly the necessity for studying STEM curriculum than before. | This is done in university level. | Just another example of robotics being used as a tool in STEM subjects, but at least this is in university level. | |
| 28 | Teaching humanoid robotics by means of human teleoperation through RGB-D sensors | Robotics and Autonomous Systems | Vol.75, pp.671-678 | 2016 | Q3 | Michieletto, S; Tosello, E; Pagello, E; Menegatti, E | Robotics in Education | A graduate course project on humanoid robotics is presented using a project-based learning constructivist approach. The task combines teleoperation of NAO robot using Kinect with an integrated programming framework, analysis on how students autonomously solve problems in groups. Quantitative data such as project marks, course marks and student questionnaire. | About 20 postgraduate students in University of Padova | Quantitative | Students were asked to control the robot motion and stability by means of human teleoperation instead of analytically solving the robot inverse kinematics and dynamic. The approach makes students able to face the problem from a more natural point of view. The correct resolution of the assigned problems and the positive students' feedback give instructors the certainty that combining constructivism with a gradual increase of the level of difficulty is effective in teaching robotics. | Combining a constructivist approach with the assignation of tasks of increasing complexity leads to the desired results. | Another example of robots used as appratus in STEM subjects. Also, this analysis does not actually cover the interaction of students in a group, but rather the end-result - the grades. | |
| 31 | Constructive Learning for Human-Robot Interaction | IEEE Potentials | pp.13-19 | 2013 | Unranked | Singh, A; Karanam, S; Kumar, D | Human-Robot Interaction | Maintaining a positive learning rate of a student being taught in a classroom using facial expression recognition and tree augmented naive (TAN) Bayes classifier on a biped robot platform. The TAN is a probability measurement to determine what emotion is being expressed. Measures learning rate of students through affective emotions expressed by students while the robot mimicks teacher's actions. | A group of students, no specific number | Quantitative | Actions can be repeated if emotions detected are confused or frustrated. If emotions such as sad are detected, the robot tries a different set of actions instead. This type of response helped maintained a positive learning rate in this study. | - | Maybe one way of analyzing the classroom. | |
| 33 | Why Robots Should Be Social: Enhancing Machine Learning through Social Human-Robot Interaction | PLoS One | Vol.10(9), p.e0138061 | 2015 | Q1 | de Greeff, J; Belpaeme, T | Learning Analytics | The interaction between human participants as teachers and the robot as a learner is modelled through a language game. Participants are randomly assigned to social and non-social robot group. Assessment is done on robot learning performance, participants' choice of topic, participants' gazing behaviour, and questionnaire on subjective experience. | 38 participants recruited from around a British university campus | Mixed | The findings illustrate how a robot might positively influence an interaction with a person through using social cues that are generally perceived as natural. The social learning shows a clear gender effect with female participants being responsive to the robot's bids, while male teachers appear to be less receptive. This work shows how additional social cues in social machine learning can result in people offering better quality learning input to artificial systems, resulting in improved learning performance. |
Challenges of machine social learning:
|
Talk about how NAO's social behaviour allows for better input for learning analytics. | |
| 34 | Learning analytics should not promote one size fits all: The effects of instructional conditions in predicting academic success | The Internet and Higher Education | Vol.28, pp.68-84 | 2016 | Q1 | Gasevic, D [Dragan]; Dawson, S; Rogers, T; Gasevic, D [Danijela] | Learning Analytics | The study followed a correlational (non-experimental) design (Field & Hole, 2003) and statistical analysis, as it investigated the effects of the variables derived from the trace data and the data from the institutional student information system on the prediction of students' academic success. The data for the study were extracted from a public research-intensive university in Australia. 9 undergraduate courses were selected: accounting, communications, computer science, economics, graphic design, marketing, mathematics and 2 biology classes. | 4134 undergraduate students | Quantitative | There is a need to create models for academic success prediction for individual courses, incorporating instructional conditions into the analysis model. There must be careful consideration in any interpretation of any predictive model of academic success, if these models do not incorporate instructional conditions. In such cases, several threats to the validity of the results may emerge such as overestimation or underestimation of certain predictors. |
|
Indeed this research may result in a general learning analytics model, but maybe NAO's direct interaction in the classroom can tailor to specific conditions and individuals. | |
| 35 | AMOEBA: Designing for collaboration in computer science classrooms through live learning analytics | International Journal of Computer-Supported Collaborative Learning | Vol.10(4), pp.425-447 | 2015 | Q1 | Berland, M; Davis, D; Smith, CP | Learning Analytics | Students were placed in pairs on the basis of predictive CS-ZPD as indicated by AMOEBA. During the course of single programming sessions, most commonly approximately 45 min in length, students were asked to collaborate based on the predictive CS-ZPD score. As programming data was generated and analyzed, some students were re-paired with others with a greater shared CS-ZPD score. Students' program data were analyzed to explore how pairing with AMOEBA impacted: the complexity of students' code; program novelty; and program quality. | 95 students (junior and high school) | Quantitative | Teachers in this study paired their students far more on average than a typical CS class. Students were paired spontaneously on the basis of their approximated CS-ZPD and then demonstrated the improvement suggested by such an approach geared towards leveraging ZPD. Students, after having been paired on the recommendations provided by AMOEBA, evidenced more proficient program development. AMOEBA utilizes a "surprising similarity" metric to identify uniqueness in students' programs to come up with a ZPD metric to recommend student pairings for different tasks. | In order to support teachers while allowing flexibility, tools to support teachers' orchestration of programming collaboration should perhaps initially focus on providing data mined metrics in real time that determine whether learning has occurred, particularly as evident through transfer, and that measure student participation (Dillenbourg et al. 2009). | See Main Points. This is also an example of real-time learning analytics being applied. | |
| 36 | Problems in Big Data Analytics in Learning | Journal of Engineering Education | Vol.105(1), pp.6-14 | 2016 | Q1 | Madhavan, K; Richey, MC | Learning Analytics | A review of Big Data learning analytics problems in education. | - | Review | Decision-making systems need to provide students with the ability to make informed decisions at a course or content level (micro-level decisions). At the same time, these systems need to include capabilities to help learners navigate institutional requirements (macro-level decisions). Work on even understanding the types of models that could be used for predictive efforts in spanning analytic boundaries between micro and macro-level decisions is still very much in its infancy. | The latency between collecting data from students and the ability to translate these data into actionable intelligence is a significant barrier. A new class of algorithms based on variations of techniques known as Kalman filtering (Kalman, 1960) and ensemble Kalman filtering (Evensen, 2003) are starting to emerge to deal with data noise and sparseness. | Can NAO provide real-time analytics that can lead to actionable decisions in a classroom? | |
| 37 | Applying social learning analytics to message boards in online distance learning: A case study | Computers in Human Behavior | Vol.47, pp.68-80 | 2015 | Q1 | Hernandez-Garcia, A; Gonzalez-Gonzalez, I; Jimenez-Zarco, AI, Chapparo-Pelaez, J | Learning Analytics | All the interaction between students and teachers - and among students - in the "Introduction to financial information" course at University of Catalonia takes place in the message board. For the SNA and data visualization, we used Gephi 0.8.2, an open-source software for network visualization and analysis. Network parameters provide a quantitative interpretation of data, while data visualization facilitates qualitative explanation. | 656 students | Mixed | Although SNA parameters may be useful to describe social interactions and unveil abnormal behaviors - disconnected students, passive learners - their use alone may not provide valid information for predictive purposes and need to be complemented with other variables. Teachers' activity has a big effect on network parameters and behaviors. Students who got more replies from consultant teachers tended to get higher grades. | Social network analysis (SNA) parameters are related to academic performance only in some cases, not all. Instructors and coordinators have to receive adequate training in social learning analytics and visualization tools to better understand the meaning of SNA results and their visualizations, in order to benefit the most from the possibilities of these techniques. There must exist a campus-wide IT strategy plan so that these technologies can be implemented at a relative low cost and as seamlessly as possible. | Use of SNA parameters for predictive purposes might not be adequate, and that additional research is needed in order to effectively design predictive systems based on SNA - such as additional variables or conditions under which SNA can be a valid predictor of academic performance. | |
| 39 | Investigating student motivation in the context of a learning analytics intervention during a summer bridge program | Computers in Human Behavior | Vol.47, pp.90-97 | 2015 | Q1 | Lonn, S; Aguilar, SJ; Teasley, SD | Learning Analytics | An early warning system (EWS) utilizes analytics information from the institutional learning management system (LMS) to inform just-in-time advisor interventions designed to identify student's maladaptive academic behaviors and study habits before students fall further into academic jeopardy. It uses Achievement Goal Theory to measure motivation, and Patterns of Adaptive Learning Scales to measure achivement goal orientations. | 216 students | Mixed | Students' perceptions of their goals and performance were strongly related to changes in their motivation. The next generation of learning analytics interventions must resolve the tension between ease of scalability in current data sources (e.g., grades) and the richness of measures such as students' intentionality, goals, and motivations that provide direction about how to tailor learning environments to learners' needs. | Showing students their own data within the context of the EWS may have contributed to decrease in mastery orientation. Student perceptions, rather than advisor' actual utilization of EWS are the strongest predictors of changes in student motivation. Any "student-facing" learning analytics intervention will need to be developed and deployed with care. | Presentation of learning analytics data to students can affect motivation negatively. | |
| 40 | Using Learning Analytics to improve teamwork assessment | Computers in Human Behavior | Vol.47, pp.149-156 | 2015 | Q1 | Fidalgo-Blanco, A; Sein-Echaluse, ML; Garcia-Penalvo, FJ; Conde, MA | Learning Analytics | Implements a learning analytics system which makes use of the CTMTC method (Comprehensive Training Model of the Teamwork Competence) (Leris, Fidalgo, & Sein-Echaluce, 2014) to analyze total number of team messages, team views, members in the team, average number of messages for each member, creation dates of the first and last thread, and views and messages for each thread throug the LMS Moodle forum. | 110 first-year Biotechnology degree students from the Technical University of Madrid, separated into 19 groups | Quantitative | The 19 work teams created 5136 messages in the forum, with a total of 37,930 message views. Active student-student interactions have a greater relation with the individual performance in teamwork contexts than passive student-student interactions. The relation between message views and individual final grade is only significant in roughly more than half of the cases, so that is inconclusive. | There is strong positive correlation between grades and number of student-student interactions. | Active participation is suggested to be one of the factors for academic performance. | |
| 41 | A multivariate approach to predicting student outcomes in web-enabled blended learning courses | The Internet and Higher Education | Vol.27, pp.44-53 | 2015 | Q1 | Zacharis, NZ | Learning Analytics | Data from an introductory Java programming course, taught in a blended mode, was gathered using the Moodle LMS activity tracking feature. A bivariate correlation between 29 online activities with student grade resulted in 14 variables with strong impact on student final achievement, which were then used as the input in a regression analysis. Given the fact that exploratory univariate regression analyses for student age, gender, previous grades, working status and ethnicity revealed that none of these variables had a significant effect on course grade, they were dropped from further analysis. The data is used to predict student risk of failure. | 134 university freshmen students of Computer Science and Computer Engineering courses | Quantitative | Overall accuracy of the prediction of student risk of failing is 81.3%. Only 4 variables: reading and posting messages, content creation contribution, quiz efforts and number of files viewed - predicted 52% of the variance in the final student grade. | Asynchronous learning activities correspond with greater levels of interaction (Greenland, 2011). Three variables that have been floating around in the online learning literature - Total time online, Total LMS hits and number of Online sessions - and are considered as measures of student effort, engagement and participation, were found to have a weak (univariate) correlation with performance grade possibly due to blended learning course design. | Again, active participation is what shines here. Passive participation (lurkers) have low significance on grades. | |
| 42 | Learning Analytics: Ethical Issues and Dilemmas | American Behavioral Scientist | Vol.57(10), pp.1510-1529 | 2013 | Q2 | Slade, S; Prinsloo, P | Learning Analytics | A sociocritical perspective on the use of learning analytics. | - | Review | The inherent peril and promise of having access to and analyzing "big data" (Bollier, 2010) necessitate a careful consideration of the ethical dimensions and challenges of learning analytics. The proposed principles and considerations included within this article provide an ethical framework for higher education institutions to offer context-appropriate solutions and strategies to increase the quality and effectiveness of teaching and learning. |
|
Not having an ethical framework increases chance of institution to misuse data, for sole purpose of profits rather than education. Also talks about the approach, to be transparent and let the students decide if the data is helping them or not. | |
| 43 | Supporting Human-Robot Interaction Based on the Level of Visual Focus of Attention | IEEE Transactions on Human-Machine Systems | Vol.45(6), pp.664-675 | 2015 | Q2 | Das, D; Rashed, MG; Kobayashi, Y; Kuno, Y | Human-Robot Interaction | A human-robot interaction approach for social robots that attracts and controls the attention of a target person based on his/her current visual focus of attention. It estimates "task-related contextual cues" and "gaze pattern" to determine a suitable time to interact. Questionnaires were used to assess the performance. | 36 students from Saitama University | Quantitative | Among 72 interactions, the system was able to detect 66 times the gaze point of visitors in a museum and make a successful interaction at a rate of 91.7%. | Visual Focus of Attention can be detected through visual cues such as head pose, head movement and overlapping face window; as well as gaze pattern by movement of the iris. Context of the task must also be known so that we know what type of gaze/action contributes to high attention. The robot needs to also make the person notice that it is looking at her/him by establishing a communications channel, in this case: blinking 3 times. | Apart from aiding interaction, this can also be good for learning analytics. | |
| 45 | The Effect of a Robot's Social Character on Children's Task Engagement: Peer Versus Tutor | Lecture Notes in Artificial Intelligence | Vol.9388, pp.704-713 | 2015 | Q4 | Zaga, C; Lohse, M; Truong, KP; Evers, V | Human-Robot Interaction | A study on the effect of two different social characters of a robot (peer vs. tutor) on children's task engagement. Engagement is measured at the cognitive (attention to the task and the robot), affective (emotional response to the task), and behavioral (performance) level by frequency and duration of these attributes. Peer robot is more high-pitched, empathic style of words, works together. Tutor robot is more low-ptiched, focused on questions and answers, points and directs. Video recording is analyzed and a questionnaire is given. | 20 children, 6-9 years old | Mixed | In the peer character condition, children paid attention to the robot and the task for a longer period of time and solved the puzzles quicker and better than in the tutor character condition. | Peer-like interactions is more engaging, for children at least. | Social robots are more suitable as peers, and human tutors are still relevant. | |
| 47 | An Empathic Robotic Tutor for School Classrooms: Considering Expectation and Satisfaction of Children as End-Users | Lecture Notes in Artificial Intelligence | Vol.9388, pp.21-30 | 2015 | Q4 | Alves-Oliveira, P; Ribeiro, T; Petisca, S; di Tullio, E; Melo, FS; Paiva, A | Human-Robot Interaction | Children are paired in groups to interact with a NAO robot tutor that guides them through a collaborative multiplayer game, EnerCities for 20 minutes. Questionnaires were given before and after the test. | 56 children, 14-16 years old | Quantitative | Majority of children expected the robotic tutor to be a good game companion (75.0%) and after the interaction almost all children revealed higher satisfaction levels towards the tutor's competence (94.6%). Majority of children expected the robotic tutor to play best in the collaborative serious learning game about sustainability (69.6%), showing a significant decrease in their satisfaction after the interaction (50.0%). | Children will detain high satisfaction levels when evaluating the capabilities of a robotic tutor, after having experienced it. The expectations regarding the fictional view of the robotic tutor are lower and remain lower after the interaction, which means that although children are exposed to sci-fi media, their expectations seem to be adapted to reality. | The interaction with the robot elicited a significant positive change in satisfaction, but not so much on expectations. This means that even if expectations were not met, the social interaction alone resulting from robots can have a net positive effect. | |
| 48 | Effects of the Robot's Role on Human-Robot Interaction in an Educational Scenario | Lecture Notes in Artificial Intelligence | Vol.9222, pp.391-402 | 2015 | Q4 | Blancas, M; Vouloutsi, V; Grechuta, K; Verschure, PFMJ | Human-Robot Interaction | This study assesses whether the role a robot plays in a classroom affects knowledge retrieval, subjective experience, and the perception of the learners. The NAO robot delivers the history class in either a teacher or peer condition with differences in posture, gestures and speech (formal/informal). Questionnaires, pre-test, post-test and video recordings were analyzed. | 28 adults | Mixed | There are no significant differences between the conditions in the amount of knowledge retrieved. Subjects in the Peer condition graded the robot as a tutor higher than in the other conditions. The perceived authority of the robot is positively correlated with the clarity with which it expresses itself. | Regardless of the statistical results, some subjects in the Peer condition reported that the robot was speaking to them, and one of the participants reported that the robot offered its hand to him, so he shook hands with the robot during the experiment. This perspective can be related to mindreading, that is, estimating the mental states of others by observing their behavior. Mindreading is necessary for reciprocal acts between humans and robots. | Supports robot as peer due to sociability aspect, which can improve learning experience. | |
| 50 | The architecture of children's use of language and tools when problem solving collaboratively with robotics | Australian Educational Researcher | Vol.40(3), pp.315-337 | 2013 | Q4 | Mills, KA; Chandra, V; Park, JY | Educational Theory | Analysed children's interactions during a series of problem solving experiments using Lego Mindstorms and Vygotsky theory with incrementally difficult challenges through students' speech interactions with tools, peers, and other experts, teacher interviews, and student focus group data. | 24 Year 4 students (aged 8.5-9.5) | Qualitative | There is a growing consensus in educational research that social and cultural dimensions of learning cannot be ignored. Language and the use of tools play a dynamic and interactive role in the learning process. Children use clarifying questions and directive statements to focus their peers' attention on certain solutions. The use of speech, gestures, and tools produces new relations with the problem-solving environment, providing a window into the students' mental organisation of behaviour. | The specific architecture of language-mediated problem solving in collaborative and formal contexts of learning begins with phases of interaction for each goal or sub-goal that typically begins with a statement or restatement of the problem, followed by the use of predictive questions and directive statements, and culminates in an emotive utterance of greater intensity upon realisation of a likely solution. | Collaborative interaction when solving problems with robots. This is an example of robots as apparatus and social constructivist application, but not social robotics. | |
| 51 | The Impact of Social Robotics on L2 Learners' Anxiety and Attitude in English Vocabulary Acquisition | International Journal of Social Robotics | Vol.7(4), pp.523-535 | 2015 | Q3 | Alemi, M; Meghdari, A; Ghazisaedy, M | Robotics in Education | This study aimed to examine the effect of robot assisted language learning (RALL) using NAO robot on the anxiety level and attitude in English vocabulary acquisition amongst Iranian EFL junior high school students. Two questionnaires of anxiety and attitude were utilized to measure the students' anxiety and attitude. | 46 female students (aged 12) | Quantitative | The results of descriptive and t tests indicated that there was lower anxiety and a more positive attitude towards English vocabulary acquisition in the RALL group compared with those in the non-RALL group. | The students in the RALL group proved to be less reserved, more engaged in the learning process, and less anxious towards judgment from their teacher and peers. | Social robots can reduce anxiety? | |
| 52 | The Social Construction of Creativity in Educational Robotics | Advances in Intelligent Systems and Computing | Vol.351, pp.329-338 | 2015 | Unranked | Zawieska, K; Duffy, BR | Robotics in Education | This paper argues that a new form of creativity concerns the meanings students make of anthropomorphic robots in the course of human-robot social interaction. This is based on the following assumptions: creativity is socially constructed and the main reason for students to be interested in robotics is a fascination with the illusion of life. In particular, this paper proposes to encourage the ability to create meanings through exploration of a mismatch between humanlike robot design and the human frame of reference. | - | Review | Educational robotics researchers often follow the constructionist educational approach because of its application in STEM subjects to build and program robots. But the role of anthropomorphic social robots is to engage as social actors, designing and manipulating robots as a form of creation is no longer a goal for educational robotics. |
The combination of anthromorphic robot design and social interaction results in new ways to foster creativity:
|
Very nice view on social robotics, because the robot is capable of socially interacting with humans but is clearly not human, it creates new meaning to motivate students curiousity and creativity. That is how it should aid the learning environment. | |
| 58 | Teachers' perceptions of the benefits and the challenges of integrating educational robots into primary/elementary curricula | European Journal of Engineering Education | pp.1-11 | 2015 | Unranked | Khanlari, A | Robotics in Education | Qualitative case study using online surveys on teachers' perceptions of the effects of using robotics on students' lifelong learning skills, teachers' perceptions of the barriers of using robotics and the support they need. | 11 primary/elementary teachers from Newfoundland and Labrador English Schools District | Qualitative | Teachers perceived that robotics has positive effects on scientific inquiry skills. Obstacles and challenges include inadequate access to supporting materials, inadequate technical and instructional support, the lack of preparation time and classroom time, teachers' lack of knowledge about robotics, and their lack of confidence in their technology skills. | Most of the participants are satisfied with administrative support and perceive they will receive adequate support from school administration if they want to integrate robotics into their teaching activities. | Teacher perception on introducing robots into classrooms. | |
| 59 | Teachers' views on the use of empathic robotic tutors in the classroom | 2014 RO-MAN: The 23rd IEEE International Symposium | pp.955-960 | 2014 | Unranked | Serholt, S; Barendregt, W; Leite, I; Hastie, H; Jones, A; Paiva, A; Vasalou, A; Castellano, G | Robotics in Education | An interview study conducted across several European countries on teachers' views on the use of empathic robotic tutors in the classroom. | 8 teachers (England, Scotland, Portugal, Sweden) | Qualitative | Robotic tutors should avoid creating administrative overhead for the teacher: should fit with existing classroom practices and social norms. | Robots assist in recording information that is later used for assessment: teachers emphasized the importance of accessing understanding and competencies. | Beyond learning analytics, teachers proposed robot tutors to support by providing data on students' learning. | |
| 60 | Robotics in the early childhood classroom: learning outcomes from an 8-week robotics curriculum in pre-kindergarten through second grade | International Journal of Technology and Design Education | Vol.26(1), pp.3-20 | 2016 | Q4 | Sullivan, A; Bers, MU | Robotics in Education | Quantitative data was collected from the participating children in the form of two assessments for robot and programming knowledge after the 8-week robotics curriculum in their classrooms using the KIWI robotics kit. | 60 children from pre-kindergarten to 2nd grade | Quantitative | Tests for statistical significance indicates that the kindergarten, first, and second grade classes performed equally well on advanced programming. | The pre-kindergartenclass struggled with mastery of the robotic parts and was not introduced to the KIWI sensors at all in their curriculum. They also worked at a slower pace. | No pre-test, only quantitative post-test. | |
| 61 | Learning Effects of Pedagogical Robots with Programming in Elementary School Environments in Korea | Indian Journal of Science and Technology | Vol.8(26) | 2015 | Unranked | Park, I; Kim, D; Oh, J; Jang, Y; Lim, K | Robotics in Education | Students participated in the robot learning curriculum in 3 different subjects (Korean, mathematics and music) for 12 weeks, a paired t-test was conducted with pre-tests and post-tests on creativity and class satisfaction. | 27 third-grade students in Korea | Quantitative | Fluency and originality were significantly improved. Class satisfaction was measured by descriptive statistics with mean of 4.45 out of 5. | Small sample. | Creativity was measured using The Korean Figural Creativity Test for Elementary school Students (K-FCTES), while satisfaction is a questionnaire Likert scale rating. | |
| 62 | Evaluating the impact of educational robotics on pupils' technical- and social-skills and science related attitudes | Robotics and Autonomous Systems | Vol.75, pp.679-685 | 2016 | Q3 | Kandlhofer, M; Steinbauer, G | Robotics in Education | The study relied on a quasi-experimental two-group design with two measurement points (pre- and post-test) and applied a multiple-choice questionnaire as assessment instrument. Using well-grounded statistical methods, the gathered data were analyzed around 14 different topics ('sub-scales') arranged in three main categories. | 148 pupils (mean age 14.9 years) | Quantitative | Educational robotics has a significant positive impact on some separate sub-scales (mathematics and scientific investigation, teamwork, social skills) but not all. Educational robotics has a positive impact on a group of related topics. | - | Purely statistical analysis using online surveys testing for significance and correlation. | |
| 63 | The positive effects of verbal encouragement in mathematics education using a social robot | Integrated STEM Education Conference (ISEC) | pp.1-5 | 2014 | Unranked | Brown, LN; Howard, AM | Robotics in Education | Integrates a socially interactive robotic tutor (DARwIn-OP) to engage students in the classroom environment. Two trials: one with college students and another with high school students. Students are randomly assigned to either a control group with no robot agent or a treatment group with the robot tutor present. Completion time, Likert-scale survey on experience and a freeform feedback were analyzed. | Trial 1: 24 college students, Trial 2: 20 high school students | Mixed | Verbal cues are able to increase and/or maintain student engagement regardless of student age and math content level. | The control group was less nervous with the robot tutor. Some students felt like the robotic platform was wasting space, while others enjoyed the robot's presence. | The social aspect of a humanoid robot may introduce more complex interactions and responses, such as fear of disappointing the robot tutor. | |
| 64 | Learning to Program with Personal Robots: Influences on Student Motivation | ACM Transactions on Computing Education | Vol.12(1), pp.1-32 | 2012 | Unranked | McGill, MM | Robotics in Education | Usage of Institute for Personal Robots in Education (IPRE) robot to study its motivational effects on non-computer science students in a CS0 introductory programming course. Uses Keller's Instructional Materials Motivation Survey to measure four components of motivation: attention, relevance, confidence, and satisfaction; then analyzed with statistical t-tests, ANOVA and MANOVA. | 35 non-computer science undergraduate students | Mixed | Little or no effect on relevance, confidence, or satisfaction; but significant effect on attention for non-computer science students to learn programming. | Gender, technical self-perception, and interest in software development had no bearing on student motivation. | Still produce positive effect on motivation (through attention), has some contradictions with #2. | |
| 65 | Instructional design using an in-house built teaching assistant robot to enhance elementary school English-as-a-foreign-language learning | Interactive Learning Environments | Vol.23(6), pp.696-714 | 2015 | Q1 | Wu, WCV; Wang, RJ; Chen, NS | Robotics in Education | An intelligent teaching assistant robot named Powerful English Tutor (PET) was created to assist in teaching English as a Foreign Language. Students were randomly assigned to either a control group without PET and a treatment group with PET. Post-test was used to assess performance, survey for motivation and interest, questionnaire for perception and video recordings for observation. Statistical t-tests were performed. | 64 3rd grade students in Yunlin County, Taiwan | Mixed | Treatment group had significantly better performance and motivation. Assistant robot was highly valued. | Pedagogical design should come first. Voice recognition is challenging in classroom environments which are normally noisy. | - | |
| 66 | Acceptance of socially assistive humanoid robot by preschool and elementary school teachers | Computers in Human Behavior | Vol.33, pp.23-31 | 2014 | Q1 | Fridin, M; Belokopytov, M | Robotics in Education | A modified Unified Theory of Acceptance and the Use of Technology model was applied using the questionnaires following interactions with Nao robot in a workshop (non-random sample). Results were analyzed using statistical tests for correlations and linear regressions to determine reliability of data. | 18 teachers | Quantitative | Positive reactions but lack of consolidated views and there is a need for an adaptation of the model. | - | - | |
| 67 | Learning Approaches to Applying Robotics in Science Education | Journal of Baltic Science Education | Vol.12(3), pp.365-377 | 2013 | Q3 | Altin, H; Pedaste, M | Educational Theory | Systematic review on robotics curricula for STEM subjects. | 8 research papers | Review | There is not enough quantitative evidence for applying robots in curricula to achieve educational goals. | The following approaches have been used in educational robotics: discovery learning, collaborative learning, problem solving, project-based learning, competition-based learning, and compulsory learning. Most robotics education approaches should not be used alone as these methodologies support and enhance each other. | - | |
| 68 | LEGO-based robotics in higher education: 15 years of student creativity | International Journal of Advanced Robotic Systems | Vol.11(1), p.1 | 2014 | Q4 | Danahy, E; Wang, E; Brockman, J; Carberry, A; Shapiro, B; Rogers, CB | Educational Theory | Reflect on the role LEGO robotics has played in college engineering education over the last 15 years, starting with the introduction of the RCX in 1998 and ending with the introduction of the EV3 in 2013. | 4 case studies | Review | The LEGO Mindstorms products have allowed engineers (and non-engineers) to grapple with questions of sensor accuracy, motor latency, response times and priorities without having to have extensive experience in circuit design, assembly-level programming or in artificial intelligence. Further, they allow students to easily explore topics in product design and prototyping. | - | - | |
| 69 | Assessment of engagement for intelligent educational agents: A pilot study with middle school students | Computers in Education Journal | Vol.5(4), pp.96-106 | 2014 | Unranked | Brown, LN; Howard, AM | Educational Theory | 15 question math test in the computer to assess total time, response accuracy and proper function execution with webcam to monitor eye gaze and pose. Exit survey and video observations were analyzed. | 13 participants, 10-14 years old middle school students in Atlanta, GA | Mixed | If a student is classified as being on-task, he or she is engaged (regardless of speed or response). Eye gaze and head pose technique is not an effective measure of engagement when high-level cognitive thinking is required. Proposed model of engagement uses physical events. | Most educational agents do not monitor engagement explicitly, but rather assume engagement and adapt their interaction based on the student's responses to questions and tasks. | - | |
| 70 | Educational Robots for Internet-of-Things Supported Collaborative Learning | Communications in Computer and Information Science | Vol.465, pp.346-358 | 2014 | Unranked | Plauska, I; Damasevicus, R | Educational Theory | Introduces Internet-of-Things Supported Collaborative Learning (IoTSCL) paradigm based on constructivism in a robotics course. Assessment done on student engagement using survey and four-phase interest model. | 22 university students in Lithuania | Mixed | Student engagement and feedback should be about physical things (in this case, robots) rather than virtual. Robots are part of the learning environment and must interact with their environment. | The robot serves both as the educational service that allows to visualize knowledge through explicit actions and behaviour as well as the enabler of learning and providing student engagement through immersion and instant feedback. | - | |
| 71 | Effects of educational robots on learning STEM and on students' attitude toward STEM | Engineering Education (ICEED), 2013 IEEE 5th Conference | pp.62-66 | 2013 | Unranked | Khanlari, A | Educational Theory | Interview on teachers' perceptions of the effects of robotics on students' learning experiences and on their interests towards STEM subjects. | 6 teachers with 2-7 years experience | Qualitative | Teachers said robots help learning because of its hands-on nature, effects on student self-confidence and "play and learn" nature. | Working with robots may make use of constructionism theory, and interaction with both humans and robots have constructivism features. | - | |
| 72 | Using peer assessment with educational robots | Lecture Notes in Computer Science | Vol.8699, pp.57-65 | 2014 | Q4 | Catlin, D | Educational Theory | A review on how peer and self-assessment (PASA) is applied in educational robotics. | - | Review | Black and Willaims' Assessment for Learning (AfL) strategies offers a way of structuring lessons while fostering essential intellectual freedom of the student. | - | This is about how technology such as robots is applied in structured curriculum without limiting creativity and imagination of students that pedagogy (constructionism) wants to promote. | |
| 73 | Computational thinking and tinkering: Exploration of an early childhood robotics curriculum | Computers and Education | Vol.72, pp.145-157 | 2014 | Q1 | Bers, MU; Flannery, L; Kazakoff, ER; Sullivan, A | Educational Theory | TangibleK curriculum based on constructionism was designed for robotics and programming course and is carried out with CHERP programming language and LEGO Mindstorms robots. Student performance is assessed. | 53 kindergarteners in Boston | Quantitative | When given age-appropriate technologies, curriculum and pedagogies, young children can actively engage in learning from computer programming as applied to the field of robotics. | Technology is becoming increasingly available for children as early as kindergarten level. | - | |
| 74 | Edutainment Robotics as Learning Tool | Lecture Notes in Computer Science | Vol.5940, pp.25-35 | 2009 | Q4 | Bilotta, E; Gabriele, L; Servidio, R; Tavernise, A | Educational Theory | An edutainment robotics program built based on constructivist theory and LEGO Mindstorms is assessed on work distribution in each student group, description of task resolution, correctness of programming strategies and number of tests completed before success. | 28 students in University of Calabria, Italy | Mixed | Constructionist approach of using robotic artefacts stimulates students to collaboratively analyze processes and experiment the consequences of their behavior. | - | - | |
| 75 | When a Classroom Is Not Just a Classroom: Building Digital Playgrounds in the Classroom | Turkish Online Journal of Educational Technology | Vol.11(1), pp.202-211 | 2012 | Unranked | Chen, GD; Chuang, CK; Nurkhamid; Liu, TC | Educational Theory | A look on the application of digital technology such as robots, projectors and computers to build a game-based learning environment for classrooms, called Digital Learning Playground (DLP) with the use of Total Scenario Response (TSR) learning design methods. | - | Review | Suggests that physical things have higher potential to engage and to support authentic and possibly experiential learning. | - | - | |
| 76 | Storytelling by a kindergarten social assistive robot: A tool for constructive learning in preschool education | Computers and Education | Vol.70, pp.53-64 | 2014 | Q1 | Fridin, M | Educational Theory | Derives a child-robot interaction level from eye contact and affective factor to analyze its relationship to cognitive and motor performance after Nao robot interaction. Experiment procedure is based on constructivist methods. Statistical analysis using repeated measures ANOVA and Pearson product–moment correlations. | 10 kindergarten children (5 boys & 5 girls, aged 3-3.6) in Israel | Quantitative | Children performance is positively correlated with interaction levels, and is not significantly affected by any of the between-subject factors. | Experiences that arouse emotions have proved to be more memorable than neutral experiences. A humanoid robot can act as an agent to convey these emotions through storytelling to children. | Interaction metric is derived from eye gaze and affective factors. #69 says this should not be the case when high-level cognitive thinking is involved. | |
| 77 | Constructivist Game-based Robotics Simulator in Engineering Education | International Journal of Engineering Education | Vol.29(4), pp.1024-1036 | 2013 | Q4 | Lee, JV; Taha, Z; Yap, HJ; Kinsheel, A | Educational Theory | Assesses the learning environment of a constructivist game-based robotics simulator compared to non game-based conventional robotics simulator on students perceptions via Constructivist Simulation-based Learning Environment Survey (CSLES) and the one scale of Test of Robotics Related Attitudes (TORRA) questionnaires with statistical tests for correlation and multiple regression analysis. | 114 undergraduate students (aged 22-25) | Quantitative | Game-based robotics simulator is more effective in terms of Negotiation, Inquiry Learning, Reflective Thinking and Challenge. There is positive but relatively weak relationship between the undergraduate students' enjoyment of robotics lessons and the game-based learning environment. | - | - | |
| 78 | Developing Cognition with Collaborative Robotic Activities | Educational Technology and Society | Vol.12(4), pp.317-330 | 2009 | Q2 | Mitnik, R; Nussbaum, M; Recabarren, M | Educational Theory | Students work in groups of three, using a robot and wirelessly interconnected Personal Digital Assistants (PDA) based on Feuersteins Mediated Learning theory. Statistical t-test is made for significance and video observations were analyzed. | 24 16-year-old students (10th grade) | Mixed | Statistically significant increase in performance. Students are highly motivated. | - | - | |
| 79 | Student privacy in learning analytics: An information ethics perspective | Information Society | Vol.32(2), pp.143-159 | 2016 | Q2 | Rubel, A; Jones, KML | Learning Analytics | Thoughts on information access concerns, the intrusive nature of information-gathering practices, whether or not learning analytics is justified given the potential distribution of consequences and benefits, and issues related to student autonomy. | N/A | Review | Learning analytics systems should provide controls for differential access to private student data, must be able to justify data collection, full accounting of how benefits are distributed between institutions and students, and students should be made aware of collection and use of their data and permitted reasonable choices regarding collection and use of that data. | Learning analytics presents significant student privacy problems for higher education institutions. | - | |
| 80 | Multimodal selfies: Designing a multimodal recording device for students in traditional classrooms | ACM International Conference on Multimodal Interaction, ICMI 2015 | pp. 567-574 | 2015 | Unranked | Dominguez, F; Chiluiza, K; Echeverria, V; Ochoa, X | Learning Analytics | Discusses and evaluates the design of a personal Multimodal Recording Device (MRD) to capture student actions during lectures, including its foreseeable costs, scalability, flexibility, intrusiveness and recording quality. | N/A | Review | Low-cost devices change the paradigm from centralized to distributed recording in order to establish student behavior in class with the potential for learning analytics research. | - | - | |
| 81 | The use of an online learning and teaching system for monitoring computer aided design student participation and predicting student success | International Journal of Technology and Design Education | pp. 1-20 | 2015 | Q4 | Akhtar, S; Warburton, S; Xu, W | Learning Analytics | Implements a Computer Supported Collaborative Learning (CSCL) environment to support lab-based CAD teaching. Student participation is monitored to identify predictors of success, analyzed using ANOVA, Pearson correlation and linear regression. | 331 undergraduate students from University of Surrey, UK | Quantitative | Attendance and average time-spent on task has a direct relation with the learning outcomes. Students who prefer to sit in groups or remain next to their fellow students tend to score better. | Suggests that learning analytics can be used to predict student outcomes and can ensure that timely and appropriate teaching interventions can be incorporated by tutors to improve class performance. | - | |
| 82 | A data preprocessing framework for students' outcome prediction by data mining techniques | 19th International Conference on System Theory, Control and Computing (ICSTCC) | pp. 836-841 | 2015 | Unranked | Danubianu, M | Learning Analytics | A case study for a data preprocessing framework for students' outcome prediction using data collected by Moodle system. It shows some methods of aggregating and extracting useful data from LMS for further analysis. | 960 university students in 130 courses of Faculty of Electrical Engineering and Computers Science in University of Suceava, Romania | Qualitative | Before further analysis, data may need to be preprocessed to establish the dependencies between courses and form association rules such as clustering. | - | - | |
| 83 | The Givenness of the Human Learning Experience and Its Incompatibility with Information Analytics | Educational Philosophy and Theory | pp. 1-14 | 2015 | Q4 | Lundie, D | Learning Analytics | Article which explores learning analytics in terms of computing philosophy and the information-theoretic account of knowledge. | - | Review | The human learning subject is not reducible to informational transactions. Human subjects experience and value their own information incommensurably with the ways in which computers measure and quantify information. | - | - | |
| 84 | The impact of learning design on student behaviour, satisfaction and performance: A cross-institutional comparison across 151 modules | Computers in Human Behavior | Vol.60, pp. 333-341 | 2016 | Q1 | Rienties, B; Toetenel, L | Learning Analytics | University modules were mapped to learning design categories which are then analyzed with data from Moodle. Learner satisfaction is assessed using Student Experience on a Module (SEaM) questionnaire. Correlation and multiple regression analyses were conducted using IBM SPSS 21 statistics software package. | 151 modules and 111.256 students with students' behaviour (<400 million minutes of online behaviour), satisfaction and performance at the Open University UK | Quantitative | The primary predictor of academic retention was the relative amount of communication activities. Learner satisfaction was strongly influenced by learning design. | - | - | |
| 85 | Assessing the suitability of student interactions from Moodle data logs as predictors of cross-curricular competencies | Computers in Human Behavior | Vol.47, pp. 81-89 | 2015 | Q1 | Iglesias-Pradas, S; Ruiz-de-Azcarate, C; Agudo-Peregrina, AF | Learning Analytics | Explores the applicability of learning analytics for prediction of development of two cross-curricular competencies: teamwork and commitment. "Interactions" Moodle plugin was used for interaction data extraction and categorization. Multiple regression analysis on total number of interactions of each category as independent variables and the total score of each competency as dependent variable. | 39 Masters degree students at Universidad a Distancia de Madrid, Spain | Quantitative | The results showed no relation whatsoever between interactions of any kind in the Learning Management System and the students' final level of teamwork competency. There is also no relation between any type of interaction and commitment levels. | - | - | |
| 86 | Social presence in online discussions as a process predictor of academic performance | Journal of Computer Assisted Learning | Vol.31(6), pp.638-654 | 2015 | Q1 | Joksimovic, S; Gasevic, D; Kovanovic, V; Riecke, BE; Hatala, M | Learning Analytics | Examines the relationship between indicators of social presence and academic performance. Investigation is done using minimal guidance for social interaction (control group) and one tailored for social and cognitive presence (treatment group). Pearson's correlation and multiple regression analysis was performed. | 81 students and 1747 student messages in online discussion from a public online university in Canada | Quantitative | Certain indicators of social presence were significant predictors of final grades in a master's level computer science online course. Course design that increased the level of meaningful interactions between students had a significant impact on the development of social presence and could positively affect students' academic performance. | - | - |