No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Who are the top contributors in a MOOC? Relating participants’ performance and contributions
Abstract
The role of social tools in MOOCs (Massive Open Online Courses) is essential as they
connect the participants. Of all the participants in a MOOC, top contributors are the ones who
more actively contribute via social tools, sometimes with posts to the emergent discussions,
sometimes answering their peers’ questions and concerns and sometimes even adding
complementary sources of information to the course. This paper collects, analyzes, and reports
empirical data from five different social tools pertaining to an actual MOOC to characterize top
contributors and provide some insights aimed at facilitating their early detection. The results of
this analysis show that top contributors have better final scores than the rest. In addition, there is
a moderate positive correlation between participants’ overall performance (measured in terms of
final scores) and the number of posts submitted to the five social tools. This paper also studies
the effect of participants’ gender and scores as factors that can be used for the early detection of
top contributors. The analysis shows that gender is not a good predictor, and that taking the
scores of the first assessment activities of each type (test and peer assessment in the case study)
results in a prediction that is not substantially improved by adding subsequent activities. Finally,
better predictions based on scores are obtained for aggregate contributions in the five social tools
than for individual contributions in each social tool.
... The classification formulation aims to predict the discrete grade using a machine-learning classifier, such as logic regression (Elbadrawy et al., 2014) and support vector machine (SVM) (Xu and Yang, 2016). The regression formulation is to predict the continued grade by using a regression model, such as linear regression (LR) (Alario-Hoyos et al., 2016) and neural networks (Oladokun et al., 2008). Besides, many studies transfer continuous scores into discrete grades (Shahiri and Husain, 2015). ...
... A way to model the problem of grade prediction is to take into account the academic degree program. Degree program always requires students to take a set of courses in order, due to the knowledge provided by the previous courses being essential for subsequent courses (Tabandeh and Sami, 2010;Wang et al., 2015;Alario-Hoyos et al., 2016;Ren et al., 2016;Morsy and Karypis, 2017). With this idea, Polyzou et al. developed course-specific ...
... Decision trees Classification Safavian and Landgrebe, 1991;Al-Radaideh et al., 2006;Nghe et al., 2007;Dekker et al., 2009;Thai-Nghe et al., 2009;Bunkar et al., 2012;Shovon et al., 2012;Koprinska et al., 2015;Al-Barrak and Al-Razgan, 2016;Saa, 2016 10 Linear regression Regression Tabandeh and Sami, 2010;Elbadrawy et al., 2014;Kennedy et al., 2015;Meier et al., 2015;Wang et al., 2015;Alario-Hoyos et al., 2016;De Barba et al., 2016;Ren et al., 2016;Hu et al., 2017;Lorenzen et al., 2017;Morsy and Karypis, 2017 13 Support vector machines Classification Kentli and Sahin, 2011;Bydžovská, 2016;Xu and Yang, 2016 4 Matrix factorization Regression / Clustering Lee and Seung, 2001;Thai-Nghe et al., 2010a, 2011aToscher and Jahrer, 2010;Bokde et al., 2015;Hwang and Su, 2015;Sweeney et al., 2015;Thai-Nghe and Schmidt-Thieme, 2015;Hu et al., 2017;Lorenzen et al., 2017;Ren et al., 2018;Zhang et al., 2020c 15 Collaborative filtering Classification / Clustering Sheena et al., 2000;Schafer et al., 2007;Li and Zaman, 2014;Bydžovská, 2015;Meier et al., 2015;Cakmak, 2017;Jyoti and Walia, 2017 7 Artificial neural network Classification / Clustering / Regression Andrews et al., 1995;Oladokun et al., 2008;Sorour et al., 2014;Luo et al., 2015;Shahiri and Husain, 2015;Młynarska et al., 2016;Zacharis, 2016;Yang et al., 2017;Su et al., 2018 9 Deep learning Classification / Clustering / Regression Guo et al., 2015;Yang et al., 2017;Kim et al., 2018;Hu and Rangwala, 2019a,b 5 Other methods Regression / Clustering Slim et al., 2014;Li et al., 2016Iqbal et al., 2017 regression (CSR) (Polyzou and Karypis, 2016), and predict student grades in a course using a sparse linear combination (Zhang and Liu, 2020). Following this way, there are many improved works (Polyzou and Karypis, 2016; Hu et al., 2017;Morsy and Karypis, 2017). ...
Student performance prediction (SPP) aims to evaluate the grade that a student will reach before enrolling in a course or taking an exam. This prediction problem is a kernel task toward personalized education and has attracted increasing attention in the field of artificial intelligence and educational data mining (EDM). This paper provides a systematic review of the SPP study from the perspective of machine learning and data mining. This review partitions SPP into five stages, i.e., data collection, problem formalization, model, prediction, and application. To have an intuition on these involved methods, we conducted experiments on a data set from our institute and a public data set. Our educational dataset composed of 1,325 students, and 832 courses was collected from the information system, which represents a typical higher education in China. With the experimental results, discussions on current shortcomings and interesting future works are finally summarized from data collections to practices. This work provides developments and challenges in the study task of SPP and facilitates the progress of personalized education.
... Άλλα εργαλεία που οι εκπαιδευόμενοι θεωρούν σημαντικά και βοηθητικά για τη συμμετοχή και τη μάθησή τους, είναι τα κουίζ και το φόρουμ συζητήσεων , η αξιοποίηση γραφικών (κινούμενα σχέδια, εικόνες) και διασκεδαστικών κουίζ (Oakley, Poole, & Nestor, 2016), η απόκτηση εμβλημάτων (Anderson, et al., 2014;Tomkin & Charlevoix, 2014;Ruipérez-Valiente, et al., 2016), τα e-mails που στέλνονταν, είτε για να συνοψίσουν τις δραστηριότητες της τρέχουσας εβδομάδας είτε για να ενημερώσουν για το επικείμενο περιεχόμενο των μαθημάτων και για σημαντικές ημερομηνίες, οι πληροφορίες για το πρόγραμμα σπουδών, οι στόχοι και τα κριτήρια αξιολόγησης, ειδικά στην αρχή των μαθημάτων , καθώς και οι δραστηριότητες στις οποίες οι εκπαιδευόμενοι καλούνταν να δημιουργήσουν ένα τεχνούργημα και να το μοιραστούν με τους ομότιμούς τους Fournier, Kop, & Durand, 2014;Κουτσοδήμου & Τζιμογιάννης, 2016), αλλά και οι δραστηριότητες παιχνιδοποίησης Vaibhav & Gupta, 2014;Borras-Gene, et al., 2016), οι διαδραστικές δραστηριότητες, δηλαδή δραστηριότητες που παρέχουν υποδείξεις και ανατροφοδότηση Το φόρουμ συζητήσεων αποτελεί το κύριο και πιο δημοφιλές (Alario-Hoyos, et al., 2016) κοινωνικό εργαλείο, αν και εναλλακτικά, μπορεί να χρησιμοποιηθούν διάφορα εργαλεία κοινωνικής δικτύωσης Cross, 2013), όπως Blogs, Twitter, Facebook, Google+. Αποτελεί χώρο συζήτησης, ανατροφοδότησης, ανταλλαγής απόψεων, ιδεών, γνώσεων και εμπειριών, τόσο ανάμεσα στους εκπαιδευόμενους μεταξύ τους όσο και μεταξύ του εκπαιδευτικού προσωπικού, αλλά και ένα εργαλείο άντλησης πληροφοριών, διεύρυνσης τού κοινωνικού τους δικτύου, δημιουργίας της αίσθησης της κοινότητας, ακόμα και χώρο άσκησης κριτικής και έκφρασης συναισθημάτων. ...
... Πολλοί ψηφοφορίας στο φόρουμ (Coetzee, et al., 2014) ή υποδείξεων για χρήση μαθησιακών στρατηγικών αυτορρύθμισης , αλλά από διάφορους παράγοντες που σχετίζονται και με τους ίδιους τους εκπαιδευόμενους. Ειδικότερα, η επίδοση επηρεάζεται θετικά από την κατάστρωση ενός πλάνου μελέτης και τον αναστοχασμό της εφαρμογής του (Davis, Chen, Van der Zee, et al., 2016), την ενεργή συμμετοχή σε κάθε αυτορρυθμιζόμενη φάση, ιδιαίτερα σε αυτήν του αναστοχασμού (Min & Jingyan, 2017), τη συμμετοχή στο φόρουμ (Coetzee, et al., 2014;Comer, et al., 2014;Diver & Martinez, 2015;Alario-Hoyos, et al., 2016;Phan, et al., 2016), τη χρήση του μαθησιακού υλικού σε μεγάλο βαθμό και την ενεργή συμμετοχή στα μαθήματα (Guo & Reinecke, 2014;Diver & Martinez, 2015;Koedinger, et al., 2015;Ruipérez-Valiente, et al., 2016;Tseng, et al., 2016), την έγκαιρη υποβολή των αξιολογήσεων (Diver & Martinez, 2015), τη συμμετοχή στις ομότιμες αξιολογήσεις (Admiraal, et al., 2014) και την αποδοχή της αξιολόγησης από τους άλλους εκπαιδευόμενους (ομότιμη αξιολόγηση) (Comer, et al., 2014), την επανεξέταση και την επανυποβολή των εργασιών τους (Kennedy, et al., 2015), το γνωστικό υπόβαθρο για το μαθησιακό αντικείμενο, που κατέχουν ήδη οι εκπαιδευόμενοι Engle, et al., 2015;Phan, et al., 2016), το βαθμό αυτορρύθμισης (Min & Jingyan, 2017) και τα κίνητρά τους , όπως είναι η απόκτηση γνώσεων και δεξιοτήτων και η επίσημη αναγνώριση τους, η επαγγελματική τους ανάπτυξη και η συνεργασία με άλλους (Phan, et al., 2016). Επηρεάζεται ακόμα από τον καλό παιδαγωγικό σχεδιασμό (Castaño, et al., 2015) και την ενσωμάτωση δραστηριοτήτων παιχνιδοποίησης και εικονικών κοινοτήτων μέσω κοινωνικών δικτύων . ...
... Γενικά, μεγαλύτερη συμμετοχή παρατηρείται τις πρώτες εβδομάδες των μαθημάτων , ενώ στη συνέχεια μειώνεται. Όμως, σημαντικό ρόλο στις υψηλές επιδόσεις είχαν, όπως έχει προκύψει, η θετική τους συμβολή και από τη βιβλιογραφία, ο καλός παιδαγωγικός σχεδιασμός (Castaño, et al., 2015), ο βαθμός αυτορρύθμισης τους (Min & Jingyan, 2017), τα εξωτερικά αλλά και τα εσωτερικά, κυρίως, κίνητρά τους , το γνωστικό τους υπόβαθρο για το μαθησιακό αντικείμενο και η σχετική τους εμπειρία από την καθημερινότητα τους στα σχολεία Engle, et al., 2015;Phan, et al., 2016), η συμμετοχή στις ομότιμες αξιολογήσεις (Admiraal, et al., 2014) και η αποδοχή της αξιολόγησης τους από τους άλλους εκπαιδευόμενους (ομότιμη αξιολόγηση) (Comer, et al., 2014), η χρήση του μαθησιακού υλικού σε μεγάλο βαθμό και η ενεργή συμμετοχή τους στις δραστηριότητες των μαθημάτων (Guo & Reinecke, 2014;Diver & Martinez, 2015;Koedinger, et al., 2015;Barba, et al., 2016;Ruipérez-Valiente, et al., 2016;Tseng, et al., 2016) και στο φόρουμ ή τις δραστηριότητες αφόρμησης και τις προαιρετικές δραστηριότητες (Coetzee, et al., 2014;Comer, et al., 2014;Diver & Martinez, 2015;Alario-Hoyos, et al., 2016;Phan, et al., 2016) , όπως η απόκτηση γνώσεων και δεξιοτήτων και η επίσημη αναγνώριση τους, η επαγγελματική τους ανάπτυξη και η συνεργασία με άλλους (Phan, et al., 2016), από τον καλό παιδαγωγικό σχεδιασμό (Castaño, et al., 2015), το γνωστικό υπόβαθρο για το μαθησιακό αντικείμενο που κατέχουν, ήδη, οι εκπαιδευόμενοι Engle, et al., 2015;Phan, et al., 2016), τη συμμετοχή τους στις ομότιμες αξιολογήσεις (Admiraal, et al., 2014), την έγκαιρη υποβολή των αξιολογήσεων (Diver & Martinez, 2015), την ενεργή συμμετοχή στα μαθήματα (Guo & Reinecke, 2014;Diver & Martinez, 2015;Koedinger, et al., 2015;Barba, et al., 2016;Ruipérez-Valiente, et al., 2016;Tseng, et al., 2016), την κατάστρωση ενός πλάνου μελέτης και τον αναστοχασμό της εφαρμογής του (Davis, Chen, Van der Zee, et al., 2016), και τη συμμετοχή στο φόρουμ (Coetzee, et al., 2014;Comer, et al., 2014;Diver & Martinez, 2015;Alario-Hoyos, et al., 2016;Phan, et al., 2016) (Coetzee, et al., 2014;Comer, et al., 2014;Diver & Martinez, 2015;Alario-Hoyos, et al., 2016;Phan, et al., 2016). ...
Το Σεπτέμβριο του 2015, η UNESCO διαμόρφωσε την παγκόσμια ατζέντα με τους στόχους για την αειφόρο ανάπτυξη, που θα πρέπει να υλοποιηθούν μέχρι το 2030. Όσον αφορά στην εκπαίδευση, αναγνωρίστηκε η σημασία της για την κοινωνική και οικονομική ανάπτυξη των λαών, ενώ τέθηκε ο στόχος της διασφάλισης της δίκαιης και χωρίς αποκλεισμούς ποιότητας της εκπαίδευσης και της προώθησης ευκαιριών δια βίου εκπαίδευσης για όλους.
Αυτός ο στόχος φαίνεται ότι μπορεί να υλοποιηθεί με τα Ανοικτά Μαζικά Διαδικτυακά Μαθήματα ή MOOCs, όπως είναι ευρέως γνωστά, καθώς προσφέρει τριτοβάθμιου επιπέδου εκπαίδευση σε εκατομμύρια ανθρώπους, σε όλο τον κόσμο. Παρά ταύτα, ένα μεγάλο ποσοστό όσων εγγράφονται στα μαθήματα δεν τα ολοκληρώνουν ποτέ, εμφανίζοντας πολύ μικρά ποσοστά ολοκλήρωσης που κυμαίνονται από 5-15%.
Σκοπός της παρούσας διατριβής είναι να διερευνηθεί κατά πόσο ο εκπαιδευτικός σχεδιασμός, η γενικότερη οργάνωση του προγράμματος και η χρήση της μικροεφαρμογής που αναπτύχθηκε (MCII+), μπορεί να βοηθήσει τους εκπαιδευόμενους να αναπτύξουν το βαθμό της αυτορρύθμισής τους επιτυγχάνοντας υψηλές επιδόσεις και μεγαλύτερα ποσοστά ολοκλήρωσης από τα καθιερωμένα.
Για το λόγο αυτό σχεδιάστηκε και υλοποιήθηκε επιμορφωτικό πρόγραμμα MOOC με θέμα την «Ενδοσχολική βία και τον εκφοβισμό» το οποίο φιλοξενήθηκε σε πλατφόρμα OpenEdx που εγκαταστήσαμε και παραμετροποιήσαμε σε virtual server του Πανεπιστημίου Αιγαίου . Το επιμορφωτικό πρόγραμμα διάρκειας 8 εβδομάδων, και κατ’ επέκταση και η έρευνα, διεξήχθη από τις 3/2 έως τις 29/3 του 2020. Στο πρόγραμμα συμμετείχαν, κυρίως, εκπαιδευτικοί από όλες τις βαθμίδες εκπαίδευσης, αλλά και στελέχη εκπαίδευσης, φοιτητές και ιδιώτες από όλη την Ελλάδα.
Τα ερευνητικά ερωτήματα αφορούσαν τη διερεύνηση των παραγόντων που επηρεάζουν την αυτορρύθμιση των εκπαιδευομένων πριν την εγγραφή τους στο πρόγραμμα, κατά τη διάρκειά του και αφού ολοκληρωθεί. Τέθηκαν, επίσης, ερωτήματα σχετικά με τη συμβολή του εκπαιδευτικού υλικού στην κινητοποίηση των εκπαιδευομένων.
Για τη συλλογή των ερευνητικών δεδομένων χρησιμοποιήθηκαν τρία διαφορετικά ερωτηματολόγια που απαντήθηκαν σε τρεις φάσεις του προγράμματος, αρχή, μέσο και λήξη, καθώς και δεδομένα από την ίδια την πλατφόρμα των μαθημάτων (επίδοση), την υπηρεσία Google Analytics (επισκεψιμότητα) και τη μικροεφαρμογή που ενσωματώθηκε στην πλατφόρμα των μαθημάτων και χρησιμοποιήθηκε από τη μία ερευνητική ομάδα (Πειραματική).
Τα ποσοτικά και ποιοτικά ερευνητικά δεδομένα που αναλύθηκαν, έδειξαν μια γενική αποδοχή του εκπαιδευτικού σχεδιασμού του προγράμματος, του εκπαιδευτικού του υλικού και του τρόπου που υλοποιήθηκε, συμβάλλοντας στην αυτορρύθμιση όλων των συμμετεχόντων, επιτυγχάνοντας υψηλές επιδόσεις και ποσοστά επιτυχούς ολοκλήρωσης του, ανεξαρτήτως ερευνητικής ομάδας (Ελέγχου, Πειραματικής) στην οποία ανήκαν. Περισσότερο, όμως, βοηθήθηκε, ως προς την αυτορρύθμισή της, η Πειραματική ομάδα η οποία χρησιμοποίησε την ερευνητική εφαρμογή MCII+ που τους υποβοήθησε να εφαρμόσουν την αυτορρυθμιστική στρατηγική τη Ψυχικής αντίθεσης με προθέσεις υλοποίησης (Mental Contrasting with Implementation Intentions-MCII), καθώς και μια σειρά άλλων αυτορρυθμιστικών διεργασιών της 1ης (Πρόνοια) και 3ης (Αναστοχασμός) φάσης του μοντέλου αυτορρύθμισης του Zimmerman.
... Moreover, Manning and Sanders [26] matched learners' final grade in 23 Coursera MOOCs with the percentage of posts that these learners had submitted to the discussion forum, concluding that between 20% and 80% of the learners who obtained at least 60% of the final grade contributed through the forum. Alario-Hoyos et al. [30] analyzed the role of "top contributors" (1% of MOOC learners with more posts submitted) to try to detect them early and assign them special permissions as community teaching assistants in the MOOC, finding a moderate positive correlation (r = 0.343) between the number of posts published and final grade in the MOOC. Huang et al. [31] also studied the highest-volume forum contributors, named "superposters", concluding that "superposters" obtained higher grades than the average forum participants as well as replied faster and received more upvotes. ...
... Similarly Breslow et al. [25] concluded that 52% of the total number of learners who obtained a completion certificate in the first edX MOOCs were active contributors in the forum Moreover, Manning and Sanders [26] matched learners' final grade in 23 Coursera MOOCs with the percentage of posts that these learners had submitted to the discussion forum, concluding that between 20% and 80% of the learners who obtained at least 60% of the final grade contributed through the forum. Alario-Hoyos et al. [30] analyzed the role of "top contributors" (1% of MOOC learners with more posts submitted) to try to detect them early and assign them special permissions as community teaching assistants in the MOOC, finding a moderate positive correlation (r = 0.343) between the number o posts published and final grade in the MOOC. Huang et al. [31] also studied the highest volume forum contributors, named "superposters", concluding that "superposters" ob tained higher grades than the average forum participants as well as replied faster and received more upvotes. ...
... In addition, each learner is also represented with a point on the graph, increasing the size of that point if several learners get the same value for P(x,y). For example, Figure 7 shows a large point in the lower right corner representing a high number of learners who only sent one single not very relevant message (typically opened a new thread in the forum to post a presentation message); this graph does not include the possible extra points defined in Table 3 General Statistics also contain a histogram with grades distribution (either in linear or logarithmic scale), and a summary table with basic statistics: average, median, mode, quartiles, or the grade to be a top contributor (best 1%) [30], as it can be seen with an example in Figure 8. ., f(x) for Grade 50 and other lines for grades b etween 0 and 100 in scales of 10) (see Figure 7). In addition, each learner is also represented with a point on the graph, increasing the size of that point if several l earners get the same value for P(x,y). ...
MOOCs (massive open online courses) have a built-in forum where learners can share experiences as well as ask questions and get answers. Nevertheless, the work of the learners in the MOOC forum is usually not taken into account when calculating their grade in the course, due to the difficulty of automating the calculation of that grade in a context with a very large number of learners. In some situations, discussion forums might even be the only available evidence to grade learners. In other situations, forum interactions could serve as a complement for calculating the grade in addition to traditional summative assessment activities. This paper proposes an algorithm to automatically calculate learners’ grades in the MOOC forum, considering both the quantitative dimension and the relevance in their contributions. In addition, the algorithm has been implemented within a web application, providing instructors with a visual and a numerical representation of the grade for each learner. An exploratory analysis is carried out to assess the algorithm and the tool with a MOOC on programming, obtaining a moderate positive correlation between the forum grades provided by the algorithm and the grades obtained through the summative assessment activities. Nevertheless, the complementary analysis conducted indicates that this correlation may not be enough to use the forum grades as predictors of the grades obtained through summative assessment activities.
... Students' overall performance was positively correlated with the number of posts on social tools. Top students of MOOCs were those who learned assisted with social tools (Alario-Hoyos et al., 2016). By studying three surveys during learning based on three MOOCs, it was concluded that badges could obtain more acknowledgement of student performance compared with certificates (Leach & Hadi, 2016). ...
... Professional development motivation, rather than general interest, could improve success in learning via professional MOOCs (Brooker et al., 2018). However, no significant gender differences were revealed in learning outcomes and student behaviors (Alario-Hoyos et al., 2016). ...
With the rapid development of information technologies, the new decade has been witnessing an advancement of massive open online courses (MOOCs)-based learning. However, MOOCs are infamous for the lower engagement and completion rates and very few studies have systematically reviewed student performance, motivation, engagement and interactions in MOOCs-based learning in order to provide constructive suggestions for researchers and practitioners. Through content analysis, this study firstly identified top 10 cited works and their major concerns and then discussed student performance, motivation, engagement, and interactions, as well as methods to improve the effectiveness of MOOCs-based learning. It also provides constructive suggestions for future design of MOOCs, and complements for the missing link in literature. Future research could focus on the measurements of variables in MOOCs-based learning in order to improve the quality of MOOCs and help students achieve success in MOOCs.
... Because of the enormous number of learners in MOOCs, it is difficult for instructors to give sufficient personalized support to students with questions. This is one major reason why asynchronous discussion is the key element of MOOCs, as it creates a space for crowd learning (Alario-Hoyos et al. 2016). Cheng and colleagues (2011) found that the quantities of discussion forum postings in an online course were associated with student performance. ...
... On the one hand, instructors could design more collaborative tasks to stimulate interaction among learners and encourage them to establish study groups. Assigning top contributors to be 'community teaching assistants' in the discussion forums may also improve support for MOOC participants and promote discussion (Alario-Hoyos et al. 2016). On the other hand, natural language-based intelligent tutoring systems such as AutoTutor (Graesser et al. 2014) could be embedded into MOOCs to cultivate a discussion-based learning style. ...
This study explores the relationship between asynchronous discussion and satisfaction with massive open online courses (MOOCs). We collected data from a large MOOC community in China (https://mooc.guokr.com/), which included 11 platforms, 321 courses, and over 13,000 ratings. Hierarchical multiple regression was used to analyze the relationship among the number of asynchronous discussion postings, disciplines, and satisfaction levels. The results indicated that asynchronous discussion significantly predicted learners’ satisfaction with MOOCs and that discipline moderated the relationship between asynchronous discussion and satisfaction. Specifically, science and technology courses showed a significantly different slope when compared with humanities courses. These results imply that asynchronous discussion plays an important role in predicting satisfaction with MOOCs in China. Asynchronous discussion may have diverse effects on course satisfaction in different disciplines. Therefore, instructors should pay attention to the characteristics of their specific disciplines when organizing and monitoring asynchronous discussions.
... While prior research on MOOCs has explored learners' motivation [12,13], behaviour [14], and dropout rates [15,16], limited attention has been given to investigating the success factors crucial for reducing dropout rates and enhancing motivation and quality in MOOCs. Identifying the factors that contribute to retaining learners helps in designing courses that are engaging and conducive to learning. ...
Purpose
This paper seeks to explore the influence of success factors, specifically motivation and course quality, on MOOC retention intention. Going beyond a mere examination of these motivational and quality factors, the study investigates students’ motivation, considering needs, interests, course system, content, and service quality. Methodologically, a questionnaire survey was conducted, collecting data from 311 students enrolled in online courses. To ascertain the impact of interest or need-based motivation on students’ retention rates, a Structural Equation Model (SEM) was employed. Subsequently, Necessary Condition Analysis (NCA) was utilized to identify the essential factors and components in each area. SEM results revealed a positive influence of motivational factors and quality issues on students’ behavior. Retention behavior was notably affected by academic and professional needs, along with personal interests. Furthermore, course content and service quality demonstrated a significant effect on students’ perseverance behavior. NCA results identified academic motivation and system quality as having a substantial impact on retention behavior, while personal motivation and technological motivation had a comparatively smaller effect. Practically, the findings suggest that course developers should consider students’ academic and personal requirements when designing online courses. Additionally, providing students with the ability to customize course and system content according to their needs is crucial. Timely problem-solving attitudes from service providers are essential for ensuring student retention.
... The interview instruments were designed based on previous studies (Alario-Hoyos et al., 2016;Leon et al., 2015). ...
Lay Description
What is already known about this topic
There are a large number of K12 MOOC implementations, which are (i) STEM‐related and computer science courses, and (ii) from Europe and the USA.
Limited studies provided different learning strategies for MOOC learning from the perspective of K12 schools, which are not MOOC developers.
There is a need to explore MOOC initiatives encompassing a wide variety of subjects, including computer science MOOCs and non‐computer science MOOCs, in regions beyond Europe and the USA.
There is also a need for further scrutiny of the effectiveness of learning strategies provided by the K12 school.
What this paper adds
Our study launched the first MOOC implementation in K12, encompassing several subject areas beyond computer science in the region of Asia.
Our study provided different learning strategies adopted by the K12 school, and the analysis revealed that MOOC learners took positive attitudes towards overall school support with mentoring enjoying great popularity with these students.
Our study examined the effectiveness of school support and identified that mentoring and reimbursement are more effective than university training and the learning guide.
Implications for practitioners
The insights gained from the study enable other K12 schools and educators to implement MOOC into existing school infrastructures successfully.
These strategies provided in the MOOC implementation program can be helpful in many different online learning contexts.
... By the end of 2021, the number of universities that offer massive open online courses (MOOCs) had exceeded 950, the number of MOOC learners had exceeded 220 million, and the number of MOOCs had exceeded 19,000 (Shah, 2022). Early MOOC research tended to emphasise outcome measures valued in traditional higher education settings, particularly academic achievement (Alario-Hoyos et al., 2016;Kennedy et al., 2015) and retention (Greene et al., 2015;Pursel et al., 2016). The last three years have witnessed a shift from enhancing learning performance and retention to providing a more engaging learning environment for MOOC participants (Alemayehu & Chen, 2021;Romero-Rodríguez et al., 2019). ...
Video lectures in massive open online courses (MOOCs) provide an opportunity to not only deliver instructional content but also engage learners. While there are many different styles of video lectures, it is not clear how video styles affect learner engagement. This study analysed and critiqued different typologies of video styles and classified MOOC video styles on a speaker-centric to media-centric spectrum. A total of 1372 survey responses were used for data analysis. The findings indicated that the ‘media-centric’ and ‘balanced’ video styles enhanced learner engagement to varying degrees in MOOCs of different study areas. In contrast, the ‘speaker-centric’ video style offered no advantages for promoting engagement in any MOOC study area. Effect sizes ranged from .03 to .07, indicating that video styles had a small to medium effect on engagement. These findings can provide new insights into the design of video lectures for different study areas in MOOCs.
... La experiencia realizada por Martínez y Pulido (2015) en la Universidad Autónoma de Madrid, manifiesta que el uso de los SPOC en el contexto universitario favorece el rendimiento académico y el nivel de satisfacción del alumnado. En este trabajo el grupo manifiesta interés por formar parte de una comunidad virtual, por lo que el uso de las denominadas herramientas sociales en estos contextos favorece el curso (Alario-Hoyos et al., 2016). ...
Evolución del rol docente de la presencialidad a la virtualidad
... To understand MOOC learners' motivations, several studies have explored why learners sign up for MOOCs and have identified factors that keep them engaged during the course (Alario-Hoyos, Muñoz-Merino, Pérez-Sanagustín, Kloos, & Parada G, 2016;de Barba, Kennedy, & Ainley, 2016;Hew & Cheung, 2014; Author 2). Hew and Cheung (2014), for example, found four reasons why learners signed up for MOOCs: (a) to learn about a new topic or to extend current knowledge, (b) curiosity about MOOCs, (c) a desire for a personal challenge, and (d) to obtain completion certificates. ...
... In addition, learners' interaction also affects retention rates in MOOCs. Multiple studies have found correlations between students' interaction in MOOCs and retention and performance (e.g., Alario-Hoyos, Muñoz-Merino, Pérez-Sanagustín, Delgado Kloos, & Parada, 2016;Barba et al., 2016;Phan al., 2016;Pursel et al., 2016). Therefore, research in MOOCs should not only account for individual learners' characteristics but also for the interdependency that takes place among learners through interaction. ...
Interaction is a principle of high-quality course design in online learning. Previous research shows that interaction in Massive Open Online Courses is crucial for learner retention and course completion. Using panel network data of 386 MOOC learners, this study explored the mechanisms that drive learner-learner interaction over time, specifically, the patterns and evolution of learner-learner interaction in a MOOC through a stochastic-actor-oriented model. The results contradicted previous evidence that learners reciprocate open communication (i.e., replies) in discussion forums and tend to interact with those to whom their direct connections reply. The extent to which learners interact with others similar to themselves (i.e., homophily) was not a statistically significant predictor of learner-learner interaction over time. Popularity, as measured by open communication, suggested preferential attachment in MOOC learners. High levels of affective communication received (i.e., likes) reduced open communication over time. Implications for practice are discussed, and future research that analyzes the quality of open communication over time is recommended.
... pour l'exploration des appréciations. . . . . 34 2.8 Arbre de décision explicatif pour des apprenants [50] Introduction avec leurs pairs [5][6][7][8]. Ce sont ces mêmes forums qui sont les sources préférées des instructeurs pour évaluer l'activité générale du cours [9]. Ils rejoignent ainsi, dans la pratique, les théories socio-constructivistes [10,11] et le courant connectiviste [12] ( À partir d'un certain nombre de messages, les apprenants et les instructeurs éprouvent des difficultés, à la fois, pour se sentir appartenir à la communauté [13], et pour suivre les dynamiques du cours [14]. ...
Les formations à distance en ligne, en particulier les MOOC, voient leurs effectifs augmenter depuis la démocratisation d'Internet. Malgré leur popularité croissante ces cours manquent encore d'outils permettant aux instructeurs et aux chercheurs de guider et d'analyser finement les apprentissages qui s'y passent. Des tableaux de bord récapitulant l'activité des étudiants sont régulièrement proposés aux instructeurs, mais ils ne leur permettent pas d'appréhender les activités collectives, or du point vue socio-constructiviste, les échanges et les interactions que les instructeurs cherchent généralement dans les forums sont essentiels pour les apprentissages (Stephens, 2014). Jusqu'à présent, les études ont analysé les interactions soit sémantiquement mais à petite échelle, soit statistiquement et à grande échelle mais en ignorant la qualité des interactions. La proposition de cette thèse est une nouvelle approche de détection interactive des activités collectives qui prend en compte à la fois leurs dimensions temporelles, sémantiques et sociales. Nous cherchons un moyen de permettre aux instructeurs d'intervenir et d'encourager les dynamiques collectives qui sont favorables pour les apprentissages. Ce que nous entendons par "dynamique collective", c'est l'évolution des interactions à la fois qualitatives et quantitatives, des apprenants dans des forums. Nous nous appuyons sur des études (Boroujeni 2017, Dascalu 2017) qui proposent d'associer l'analyse statistique des interactions et le traitement automatique de la langue, pour étudier les flux d'informations dans les forums. Mais, à la différence des études précédentes, notre approche ne se limite pas à une analyse globale ou centrée sur un individu. Nous proposons une méthode de conception d’indicateurs et de tableaux de bord permettant les changements d'échelles et la personnalisation des vues afin de soutenir les instructeurs et les chercheurs dans leur tâche de détection, d'observation et d'analyse des dynamiques collectives de sous-groupes d'apprenants.
... Regarding the performance, the Experimental group shows higher percentages of individuals in the highest scale (90% -100%) compared to the control group, but without statistically significant differences between them. Therefore, in addition to the self-regulation of the learners, the good instructional design of the program (Castaño, Maiz, & Garay, 2015), their cognitive background for the learning object and their relevant experience from their daily life in schools (DeBoer, Stump, Seaton, & Breslow, 2013;Engle et al., 2015;Phan, McNeil, & Robin, 2016), participation in peer reviews (Admiraal, Huisman, & Van de Ven, 2014), their active participation in course activities (Guo & Reinecke, 2014;Diver & Martinez, 2015;Koedinger et al., 2015;de Barba, Kennedy, & Ainley, 2016;Ruipérez-Valiente et al., 2016;Tseng et al., 2016) and in the forum or start-up activities and optional activities (Coetzee, Fox, Hearst, & Hartmann, 2014;Comer, Clark, & Canelas, 2014;Diver & Martinez, 2015;Alario-Hoyos, Muñoz-Merino, Pérez-Sanagustín, Delgado Kloos, & Parada, 2016;Phan et al., 2016), where they exchanged views and/or were informed of the views of others. ...
MOOCs were created to change the way universities provide education. If to some extent they have succeeded, since they enable many people to attend them without prerequisites and conditions, it is observed that a very small percentage of those who participate finally manage to complete them. In the present study, which is part of the first researcher's doctoral research, we examine the extent to which helping learners apply the Mental Contrasting with Implementation Intentions (MCII) self-regulatory strategy in conjunction with a number of other self-regulatory processes in Zimmerman's model, contributed to the increase of self-regulation, performance and completion rates of those who participated in the first MOOC program of the University of the Aegean (Greece) on "Violence and bullying in schools". 1309 people started the program and completed it, 1050. The two research groups into which they were divided, showed statistically significant differences in their self-regulation, but not in the completion rates of the program and their performance. Nevertheless, a very high percentage managed to complete it (80.2%), achieving at the same time very high performance. This result shows that self-regulation is not the only factor that contributes to the successful completion of programs and high performance. The instructional design of the program, its organization, and the quality of the instructional material play also an important role. These results can be useful in the design of future MOOCs programs.
... One of the main reasons applying analytic is to monitor and predict student's performance to provide appropriate intervention program for students. Various factors can affect student's performance in MOOC such as student's frequency accessing activities, student's interaction level with peers or instructor, or student's time management [3][4][5]. The factors become the basis for the features constructed for prediction. ...
Increasing data recorded in massive open online course (MOOC) requires more automated analysis. The analysis, which includes making student’s prediction requires better strategy to produce good features and reduces prediction error. This paper presents the process of feature engineering for predicting MOOC student’s performance utilizing deep feature synthesis (DFS) method. The experiment produces features which all the top features selected using principal component analysis (PCA) are the features that are generated from method. In terms of prediction comparing both based features and generated features, the result shows better accuracy for generated features proposed using k-nearest neighbours technique which shows the method potential to be used for future prediction model.
... In order to alleviate these problems, it is possible to find learners in many MOOCs who altruistically devote their time to helping their peers. The identification of these "community teaching assistants" is essential at an early stage in order to provide them with special permissions to better help teachers in managing the MOOC forum [41]. In addition, methodologies and tools, such as 3S and LATƎS [42], have been proposed in the related literature to help teachers and learners digest and understand the large number of messages posted in the forum by detecting the most important topics discussed. ...
Tools are an essential support in any human activity. As the technology advances, we are able to design more advanced tools that help us in doing the activities more efficiently. Recently, we have seen breakthroughs in the two main components of tools, namely the interface and the computing engine behind. Natural interfaces allow us to communicate with tools in a way better adapted for humans. In relation to the engine, we are shifting from a computing paradigm to another one based on artificial intelligence, which learns as it is used. In this paper, we examine how these technological advances have an impact on education, leading to smart learning environments.
... Regarding the MOOC, the current bibliography points out the importance of the course design as one of the related variables which may condition the behavior of the students, such as: the type of video-lectures (Guo & Reinecke, 2014), the formative or summative type of evaluation activities (Alario-Hoyos, Muñoz-Merino, Pérez-Sanagustin, Delgado-Kloos, & Parada, 2016;Freitas et al., 2015), if the course offers certification (Hew & Cheung, 2014), the length of the course (duration of four or more weeks), and the nature of the proposed tasks (collaborative or individual) (Margaryan, Bianco, & Littlejohn, 2015). ...
Massive Open Online Courses (MOOCs) have become a source of digital content anytime
and anywhere. MOOCs offer quality content to millions of learners around the world,
providing new opportunities for learning. However, only a fraction of those who initiate a
MOOC complete it, leaving thousands of committed students without achieving their
goals. Recent research suggests that one of the reasons why students find it difficult to
complete a MOOC is that they have problems planning, executing, and monitoring their
learning process autonomously; that is, they do not effectively self-regulate their learning
(SRL). In this thesis, we will explore the possibilities that Learning Analytics (LA) offers
to investigate the learning strategies that students use when self-regulate their learning in
online environments such as MOOCs. Particularly, the main objective of this research is to
develop instruments and methods for measuring students’ SRL strategies (cognitive, metacognitive
and resource management) in MOOCs, and to analyze their relationship with
students’ learning outcomes. As a methodological approach, this thesis uses mixed
methods as a baseline for organizing and planning the research, combining trace-data with
self-reported data to better understand SRL in MOOCs. The main contribution of the thesis
is threefold. First, it proposes an instrument to measure learners’ SRL profiles in MOOCs.
This instrument was validated with an exploratory and confirmatory factorial analysis with
4,627 responses collected in three MOOCs. Second, it presents a methodology based on
data mining and process mining techniques to extract learners’ SRL patterns in MOOCs.
The methodology was applied in three self-paced Coursera MOOCs with data from 3,458
learners where six patterns of interaction were identified. Then, this methodology was
adapted and applied in an effort of replication for analyzing a synchronous edX MOOC
with data from 50,776 learners where twelve patterns of interaction we identified. The
third contribution is a set of empirical studies that show the relationship between SRL
strategies and academic performance, using data from six self-paced MOOCs in Coursera
and two synchronous MOOCs in Open edX. These empirical studies led us to identify selfreported
learners’ variables (i.e., gender, prior knowledge and occupation) and selfreported
SRL strategies (i.e., goal setting, strategic planning) that were identified as the
most relevant to predict academic
... Studies have found a correlation between the number of forum posts and the performance of students. Although it is not always the case, the students who tend to be more active in online discussions tend to perform better in the course [1]. Therefore, in this study, we are trying to understand how adding a human element to an online course, by providing personalized feedback emails, can affect students' performance. ...
The instructors in conventional classes play a crucial role in moti-vating students to participate in class activities. However, in asyn-chronous online courses, such a relationship where the instructor actas the observer and the motivator is missing. In this paper, we per-formed an experiment on an online introductory course in computerscience to understand how personalized feedback emails can addressthis limitation. For this purpose, we designed a forecasting systemto analyze the progress of students towards the end of the semesterand predict their final grades. Our quantitative and qualitative dataanalysis shows how such a feedback system can improve both theperformance and the level of satisfaction of students in the onlineclasses. More than~42% of students ended up getting grades betterthan their expected scores, and~78% of students confirmed that thefeedback emails motivated them to enhance their engagement in theclass discussions.
... La investigación relacionada también ha destacado la necesidad de que se produzcan acciones por parte del profesorado, como el envío de emails masivos periódicamente, para fomentar la interacción social como reacción a dichas acciones (Alario-Hoyos, Pérez-Sanagustín, Delgado Kloos, Parada G., & Muñoz-Organero, 2014b). También se ha analizado el perfil de los estudiantes más activos, los cuales actúan como líderes de la interacción social en el MOOC, y pueden apoyar a los docentes en la apropiada gestión de los foros, asignándose permisos especiales para la edición y/o borrado de mensajes inapropiados a estos "líderes de la comunidad de estudiantes" (Alario-Hoyos, Muñoz-Merino, Pérez-Sanagustín, Delgado Kloos, & Parada G., 2016). ...
MOOCs (Massive Open Online Courses) have been a major revolution in education, especially in higher education, and have served as a catalyst for educational institutions to reflect (beyond the discussion of whether or not to offer MOOCs) on their digital education strategy. Likewise, MOOCs have been an important source of data for large-scale research on educational technology. This article reflects on some of the MOOC-related research advances that have occurred in recent years. These advances are organized into six categories: 1) instructional design of MOOCs, 2) integration of external tools in MOOCs; 3) social interaction in MOOCs; 4) self-regulation of learning in MOOCs; 5) prediction of student behavior in MOOCs; and 6) reuse of MOOCs to support teaching on campus. In addition, this article discusses some of the current opportunities to do research in the field of MOOCs, taking as a reference the data that can be collected to improve teaching and learning processes, grouping most of these opportunities under the field of learning analytics. On the seventh anniversary of the publication of the famous article "The year of the MOOC", there are more challenges than ever about the future of education, and especially about the future of higher education. Only those who know how to make good use of the existing opportunities will survive in an area of increasing competition.
... For example, there have been research studies who detected that the most appropriate tool to manage social interaction in courses with a very large number of students is the built-in forum provided by the learning platforms (Alario-Hoyos, Pérez-Sanagustín, Delgado-Kloos, Parada G., & Muñoz-Organero, 2014). Some other research studies focused on the identification of leaders within the community of learners, characterizing these leaders as the most active students in the course forum (AlarioHoyos, Muñoz-Merino, Pérez-Sanagustín, Delgado Kloos, & Parada G., 2016); this identification of leaders is important to facilitate teachers' work, as leaders can act as a bridge between faculty and the rest of the students, even receiving special roles to be able to curate forum messages. Some other research studies focused on analyzing the overall class mood from social interactions, calculating the polarity of messages (positive, neutral, negative) posted by students in the course forum; the polarity of messages was calculated by applying word dictionaries and syntax rules, and the aim was to detect parts of the course in which the overall class mood was more positive or more negative to take corrective measures in the second case (Moreno-Marcos, Alario-Hoyos, Muñoz-Merino, Estévez-Ayres, & Delgado Kloos, 2018). ...
Innovation in education in general, and innovation in engineering education in particular, must be supported by data, properly collected and analyzed to guide decision-making processes. Today it is possible to collect data from many more stakeholders (not just students), and also to collect much more data from each stakeholder. Nevertheless, low-level data collected by monitoring the interactions of the multiple stakeholders with learning platforms and other computing systems must be transformed into meaningful high-level indicators and visualizations that guide decision-making processes. The aim of this paper is to discuss some notable trends in data-driven innovation in engineering education, including: 1) improvement of educational content; 2) improvement of learners' social interactions; 3) improvement of learners' self-regulated learning skills; and 4) prediction of learners' behavior. However, there are also significant risks associated with data collection and processing, including privacy, transparency, biases, misinterpretations, etc., which must also be taken into account, and that require creating specialized units, and training the personnel in data management.
... Regarding the MOOC, the current bibliography points out the importance of the course design as one of the related variables which may condition the behavior of the students, such as: the type of video-lectures (Guo & Reinecke, 2014), the formative or summative type of evaluation activities (Alario-Hoyos, Muñoz-Merino, Pérez-Sanagustin, Delgado-Kloos, & Parada, 2016;Freitas et al., 2015), if the course offers certification (Hew & Cheung, 2014), the length of the course (duration of four or more weeks), and the nature of the proposed tasks (collaborative or individual) (Margaryan, Bianco, & Littlejohn, 2015). ...
... To understand MOOC learners' motivations, several studies have explored why learners sign up for MOOCs and have identified factors that keep them engaged during the course (Alario-Hoyos et al. 2016;de Barba et al. 2016;Hew and Cheung 2014;Li 2019;Liu et al. 2014). Hew and Cheung (2014), for example, found four reasons why learners signed up for MOOCs: (a) to learn about a new topic or to extend current knowledge, (b) curiosity about MOOCs, (c) a desire for a personal challenge, and (d) to obtain completion certificates. ...
Literature on MOOCs has shown that understanding learners’ perspectives in taking MOOCs is critical if a MOOC needs to be successful. Now that MOOCs have been in wide use, in this study we took an updated look at learners’ perspective of taking MOOCs designed for working professionals and course aspects that these MOOC participants found beneficial. General interest in the topic, personal growth and enrichment, relevance to job, and career change were the top reasons for working professionals to enroll in MOOCs. First-time MOOC takers were more likely to seek a certificate, while MOOC veterans may complete most assignments but did not seek for a certificate. Quality materials from a reputable provider remains an important reason for working professionals to enroll in a MOOC. Offering meaningful ways for MOOC participants to interact with instructors and with each other calls for innovative designs than the current discussion forums in a learning management system can offer. This remains to be a big challenge for MOOC designers.
... Andrew (2017) [3] based on the characteristics of chemistry courses, studies how to enable more effective communication among learners and improve their learning in MOOC. Alario-Hoyos (2016) [4] analyzed MOOC learning behavior and its effectiveness as a means of assessing MOOC. Hone et al. (2016) [5] analyzed the factors affecting the retention of MOOC learning through the questionnaire. ...
MOOCs1 are rapidly developing. More and more universities are participating. These MOOCs differ in the teaching mode. The content-based MOOC is currently the mainstream. We do not think that this model is best suited for online learning, but that teachers are more willing to accept it. Its rapid development is due to the rapid conversion of traditional teaching resources. This article systematically analyzes various MOOCs operating models. We also compared major MOOCs in terms of authentication methods, profitability, and evaluation methods. By comparing the advantages of the four important MOOCs, we found out that they can learn from each other in many ways. Therefore, our view is that MOOC is still far from the goal of an excellent online learning platform. MOOCs need to be constantly reformed, making them an important way of education.
... This paper reviews issues relating to learning analytics in MOOC contexts, considering published data on MOOC learning analytics and discussing factors implicated in previous studies as being related to self-regulated learning [17,18]. The free nature of MOOCs is said to be behind the reasons for profound risk of dropout [11,12], and the students ability to self-regulated their learning habits [13]. While other studies point out personal reasons as a factor of learners' high dropout rate [1,14]. ...
Nowadays, the digital learning environment has revolutionized the vision of distance learning course delivery and drastically transformed the online educational system. The emergence of MOOCs (Massive Open Online courses) has exposed web technology used in education in a more advanced revolution ushering a new generation of learning environments. The digital learning environment is expected to augment the real world conventional education setting. The educational pedagogy are tailored with the standard practice which has been noticed to increase student success in MOOCs and provide a revolutionary way of self-regulated learning. However, there are still unresolved questions relating to the understanding of learning analytics data and how this could be implemented in educational contexts to support individual learning. One of the major issue in MOOCs is the consistent high dropout rate which over time has seen courses recorded less than 20% completion rate. This paper explores learning analytics from different perspectives in a MOOC context. Firstly, we review existing literature relating to learning analytics in MOOCs, bringing together findings and analyses from several courses. We explore meta-analysis of the basic factors that correlate to learning analytics and the significant in improving education. Secondly, using themes emerging from the previous study, we propose a preliminary model consisting of four factors of learning analytics. Finally, we provide a framework of learning analytics based on the following dimensions: descriptive, diagnostic, predictive and prescriptive, suggesting how the factors could be applied in a MOOC context. Our exploratory framework indicates the need for engaging learners and providing the understanding of how to support and help participants at risk of dropping out of the course.
... Andrew (2017) [3] based on the characteristics of chemistry courses, studies how to enable more effective communication among learners and improve their learning in MOOC. Alario-Hoyos (2016) [4] analyzed MOOC learning behaviours and its effectiveness as a means of assessing MOOC. Hone et al. (2016) [5] analyzed the factors affecting the retention of MOOC learning through the questionnaire. ...
... Andrew (2017) [3] based on the characteristics of chemistry courses, studies how to enable more effective communication among learners and improve their learning in MOOC. Alario-Hoyos (2016) [4] analysed MOOC learning behaviours and its effectiveness as a means of assessing MOOC. Hone et al. [7] analysed MOOC teaching quality and its influencing factors. ...
... Andrew (2017) [3] based on the characteristics of chemistry courses, studies how to enable more effective communication among learners and improve their learning in MOOC. Alario-Hoyos (2016) [4] analysed MOOC learning behaviours and its effectiveness as a means of assessing MOOC. Hone et al. (2016) [5] analysed the factors affecting the retention of MOOC learning through the questionnaire. ...
... Other approaches in the social area have tried to iden- tify the people who contribute the most in the forums and their possible relationship with course completion [47], the participation of learners in peer reviews [48], and the per- sonality of students, which was classified according to the Big Five personality dimensions [49] (openness, extraver- sion, conscientiousness, agreeableness, and neuroticism) by Chen et al. [50]. ...
This paper surveys the state of the art on prediction in MOOCs through a Systematic Literature Review (SLR). The main objectives are: (1) to identify the characteristics of the MOOCs used for prediction, (2) to describe the prediction outcomes, (3) to classify the prediction features, (4) to determine the techniques used to predict the variables, and (5) to identify the metrics used to evaluate the predictive models. Results show there is strong interest in predicting dropouts in MOOCs. A variety of predictive models are used, though regression and Support Vector Machines stand out. There is also wide variety in the choice of prediction features, but clickstream data about platform use stands out. Future research should focus on developing and applying predictive models that can be used in more heterogeneous contexts (in terms of platforms, thematic areas, and course durations), on predicting new outcomes and making connections among them (e.g., predicting learners' expectancies), on enhancing the predictive power of current models by improving algorithms or adding novel higher-order features (e.g., efficiency, constancy, etc.).
... Most MOOCs tend to replicate passive learning strategies, such as direct instruction, although in this case through canned video lectures, due to the difficulty of applying active learning pedagogies with thousands of learners. Teachers need to reflect on the fact that interaction is the key to learn in MOOCs, either the interaction with learning contents or the interaction with the course mates, which together form the MOOC community [32]. There are lots of existing resources which can be integrated in MOOCs to promote interaction, including simulators or animations. ...
... Regarding social interactions in MOOCs, there have been approaches that identified the top contributors in the forums (Alario-Hoyos et al. 2016), the personality of learners ( Chen et al. 2016), user's confusions (Yang, Kraut, and Rose 2016), the participation of learners in peer reviews ( Er et al. 2017) or if there will be interven- tion from an instructor in a forum message or not (Chaturvedi, Goldwasser, and Daumé III 2014). All of them also have in common the use of forum data, which have also yielded some contributions to classify forum messages. ...
The learning process in a MOOC (Massive Open Online Course) can be improved from knowing in advance learners’ grades on different assignments. This would be very useful to detect problems with enough time to take corrective measures. In this work, the aim is to analyse how different course scores can be predicted, what elements or variables affect the predictions and how much and in which way it is possible to anticipate scores. To do that, data from a MOOC about Java programming have been used. Results show the importance of indicators over the algorithms and that forum-related variables do not add power to predict grades, unlike previous scores. Furthermore, the type of task can vary the results. Regarding the anticipation, it was possible to use data from previous topics but with worse performance, although values were better than those obtained in the first seven days of the current topic.
... For instance, it has been found that student motivation decreases significantly following the first weeks of a MOOC (Lackner, Ebner, & Khalil, 2015). This factor has created an abundance of research questions with respect to patterns of engagement and categorization of students in MOOCs (Kizilcec, Piech, & Schneider, 2013;Alario-Hoyos et al., 2016;Khalil & Ebner, 2015a). Furthermore, the lack of interaction between learners and instructor(s), and the controversial argument regarding MOOCs' pedagogical approach act as roadblocks to the advancement of MOOCs. ...
Educational technology has obtained great importance over the last fifteen years. At present, the umbrella of educational technology incorporates multitudes of engaging online environments and fields. Learning analytics and Massive Open Online Courses (MOOCs) are two of the most relevant emerging topics in this domain. Since they are open to everyone at no cost, MOOCs excel in attracting numerous participants that can reach hundreds and hundreds of thousands. Experts from different disciplines have shown significant interest in MOOCs as the phenomenon has rapidly grown. In fact, MOOCs have been proven to scale education in disparate areas. Their benefits are crystallized in the improvement of educational outcomes, reduction of costs and accessibility expansion. Due to their unusual massiveness, the large datasets of MOOC platforms require advanced tools and methodologies for further examination. The key importance of learning analytics is reflected here. MOOCs offer diverse challenges and practices for learning analytics to tackle. In view of that, this thesis combines both fields in order to investigate further steps in the learning analytics capabilities in MOOCs. The primary research of this dissertation focuses on the integration of learning analytics in MOOCs, and thereafter looks into examining students' behavior on one side and bridging MOOC issues on the other side. The research was done on the Austrian iMooX xMOOC platform. We followed the prototyping and case studies research methodology to carry out the research questions of this dissertation. The main contributions incorporate designing a general learning analytics framework, learning analytics prototype, records of students' behavior in nearly every MOOC's variables (discussion forums, interactions in videos, self-assessment quizzes, login frequency), a cluster of student engagement...
... Liyanagunawardena, Williams, & Adams, 2013), a large number of studies continued to focus on MOOC learners, which accounted for 30.4% of the articles in the reviewed pool. Among the scholarly articles that studied learners' perspectives, the most researched topics were learner motivation (e.g., Bulger, Bright, & Cobo, 2015;Salmon, Pechenkina, Chase, & Ross, 2016;Durksen, Chu, Ahmad, Radil, & Daniels, 2016), engagement (e.g., Hew, 2016;Moskal, Thompson, & Futch, 2015;Rodrigues, Ramos, Silva, & Gomes, 2016;Sinclair & Kalvala, 2016), course performance (e.g., Alario-Hoyos, Muñoz-Merino, Pérez-Sanagustín, Delgado Kloos, & Parada, 2016;De Barba, Kennedy, & Ainley, 2016), and retention/dropout/persistence (e.g., Gomez-Zermeno & Aleman De La Garza, 2016;Kim, Yang, Bae, Min, Lee, & Kim, 2017;Xing, Chen, Stein, & Marcinkowski, 2016). ...
Massive Open Online Courses (MOOCs) have existed as a disruptive educational phenomenon for nine years. Grounded in the roots of distance education, open education, Open Educational Resources, and OpenCourseWare, MOOCs have now survived various critics and have continued growing globally. Reports about MOOCs in both the press and scholarly publications began to grow significantly in 2013 (Sánchez-Vera, Leon Urrutia, & Davis, 2015; Zancanaro & Domingues, 2017) and, since then, more and more researchers have joined the discussions, developing them to explore various new topics. To contribute to the literature of MOOC studies, this doctoral thesis begins with an in-depth analysis of the background, history, growth, and vision, and proposes a tentative definition of MOOCs. Meanwhile, by conducting bibliometric research to review MOOC studies conducted between 2015 and 2017, this thesis fills in the gap that has existed due to a lack of systematic reviews of MOOC literature since 2015.
The results of the bibliometric research summarised the relevant MOOC research into nine categories, including learner focused, commentary and concepts, case reports or evaluations, pedagogy, curriculum and design, course object focused, provider focused, technology, systematic review of literature, and learning analytics and big data. They also suggested a limited amount of provider focused research, which became the research interest and focus of this thesis.
In the centre of the Europe, Swiss universities have marched forward in the MOOC movement, together with other over 550 universities (Shah, 2016) around the world. Università della Svizzera italiana (USI; Lugano, Switzerland), a Swiss public university, became a MOOC provider in 2015 and offered the first MOOC in the topic of eTourism: eTourism: Communication Perspectives. This doctoral thesis is closely related to this university-level initiative, which was dedicated to producing the first pilot MOOC at USI. Therefore, the cases chosen by this thesis are positioned in the discipline of tourism and hospitality.
The first MOOC with a large audience taught artificial intelligence in 2011 (Zancanaro & Domingues, 2017). Nowadays, MOOCs have broken the barrier of space and time to educate the masses in a wide range of subjects. However, the provision of MOOCs in the subject of tourism and hospitality did not appear until 2013, when two MOOCs from two American universities became available. In the past four years since these MOOCs were launched, the number of tourism and hospitality MOOCs available in the market has remained limited (Tracey, Murphy, & Horton-Tognazzini, 2016). This scarcity contradicts the fact that tourism and hospitality is the field that contributes the most to the employment of the global workforce. Pressing problems, such as high turnover, Massive Open Online Courses (MOOCs) have existed as a disruptive educational phenomenon for nine years. Grounded in the roots of distance education, open education, Open Educational Resources, and OpenCourseWare, MOOCs have now survived various critics and have continued growing globally. Reports about MOOCs in both the press and scholarly publications began to grow significantly in 2013 (Sánchez-Vera, Leon Urrutia, & Davis, 2015; Zancanaro & Domingues, 2017) and, since then, more and more researchers have joined the discussions, developing them to explore various new topics. To contribute to the literature of MOOC studies, this doctoral thesis begins with an in-depth analysis of the background, history, growth, and vision, and proposes a tentative definition of MOOCs. Meanwhile, by conducting bibliometric research to review MOOC studies conducted between 2015 and 2017, this thesis fills in the gap that has existed due to a lack of systematic reviews of MOOC literature since 2015.
The results of the bibliometric research summarised the relevant MOOC research into nine categories, including learner focused, commentary and concepts, case reports or evaluations, pedagogy, curriculum and design, course object focused, provider focused, technology, systematic review of literature, and learning analytics and big data. They also suggested a limited amount of provider focused research, which became the research interest and focus of this thesis.
In the centre of the Europe, Swiss universities have marched forward in the MOOC movement, together with other over 550 universities (Shah, 2016) around the world. Università della Svizzera italiana (USI; Lugano, Switzerland), a Swiss public university, became a MOOC provider in 2015 and offered the first MOOC in the topic of eTourism: eTourism: Communication Perspectives. This doctoral thesis is closely related to this university-level initiative, which was dedicated to producing the first pilot MOOC at USI. Therefore, the cases chosen by this thesis are positioned in the discipline of tourism and hospitality.
The first MOOC with a large audience taught artificial intelligence in 2011 (Zancanaro & Domingues, 2017). Nowadays, MOOCs have broken the barrier of space and time to educate the masses in a wide range of subjects. However, the provision of MOOCs in the subject of tourism and hospitality did not appear until 2013, when two MOOCs from two American universities became available. In the past four years since these MOOCs were launched, the number of tourism and hospitality MOOCs available in the market has remained limited (Tracey, Murphy, & Horton-Tognazzini, 2016). This scarcity contradicts the fact that tourism and hospitality is the field that contributes the most to the employment of the global workforce. Pressing problems, such as high turnover, seasonality, and new global challenges have urged for solutions to quickly training people working in this area to become available (Cantoni, Kalbaska, & Inversini, 2009). A call for more studies about tourism and hospitality MOOCs has emerged.
The combined reality of the lack of studies regarding MOOC providers, opportunities for first-hand experience of producing a tourism MOOC in a university, and the deficiency in both the research and practises of tourism and hospitality MOOCs has inspired the direction of this thesis in regard to exploring MOOC instructors’ experiences, using cases in the field of tourism and hospitality. It cumulates six studies, using a mixed methods approach, to tackle the two main research objectives:
To investigate at large the tourism and hospitality MOOC provisions between 2008 and 2015;
To report the experiences of Università della Svizzera italiana (USI) when producing the eTourism MOOC.
In order, the first two studies in Chapter 3 of this thesis focus on tourism and hospitality MOOCs in general and produce a big picture context for the other four studies in Chapter 4. The first study proposes a conceptual framework through which to describe and analyse the course design of a MOOC and applies it to 18 tourism and hospitality MOOCs produced between 2008 and 2015. The second study then continues to interview six tourism and hospitality MOOC instructors, to describe their experiences and perspectives of teaching MOOCs.
After exploring a holistic view of the overall development of MOOCs in tourism and hospitality and gaining a deep understanding of the instructors behind these offerings, this thesis introduces the experiences of one single MOOC provider: Università della Svizzera italiana (USI) in Chapter 4. It first introduces its overall implementation process (Study 3), and further elaborates three phases of this process: how it selected a suitable MOOC platform at the beginning (Study 4); how it assessed learner engagement in the MOOC (Study 5); and, eventually, how it evaluated the performance of the MOOC (Study 6).
This thesis was written mainly from the perspective of eLearning, with the intention of benefiting its community of scholars and practitioners. It has contributed to the literature by developing a framework with which to review MOOCs (in Study 1), the implementation process of producing MOOCs (in Study 2), practical review schema of MOOC platforms (in Study 4), the MOOC Learner Engagement Online Survey (in Study 5), and how to use the Kirkpatrick model to evaluate MOOCs (in Study 6). These conceptual frameworks and experiential tools can benefit future researchers and practitioners. Meanwhile, due to its intimate connection with the field of tourism and hospitality, by directly using its cases, the research outputs of the six studies can also benefit the tourism and hospitality education and training sector as a reference for further action.
... Although participation is, of course, about more than just 'quantity,' the number of messages or posts contributed in a CSCL activity is telling, as previous research has noted that more messages are indicative of higher quality conversations (Author B & Author C, 2009;Hou & Wu, 2011;Schellens & Valcke, 2005). For instance, in research areas such as Massive Open Online Courses (MOOCs), a clear link has been found between the number of posts submitted to online collaboration forums and overall performance (Alario-Hoyos, Muñoz-Merino, Pérez-Sanagustín, Delgado Kloos, & Parada G, 2016). These variations in participation levels are also important from a student experience perspective; one consequence of unequal participation is negative perceptions of working with peers from different countries. ...
Computer-supported collaborative learning (CSCL) has been highlighted as a beneficial learning experience for students in blended and online settings. In highly diverse and international contexts, CSCL also allows students the opportunity to encounter new ideas and values from peers with different backgrounds. However, previous research has highlighted that there are wide variations in student participation levels in CSCL activities and that many students experience social and cultural tensions when working with diverse peers. These issues may damage the quality of online collaboration and limit the potential gains of CSCL. In this study, we explored one evidence-based solution for encouraging CSCL participation: the internationalisation of the online academic content used for collaborative activities. Using a randomised control trial method with 428 undergraduate students in an introductory statistics course, we compared individual and group-level participation in an online collaborative task when students used content from the local context compared to content from international contexts. Our findings suggest that internationalisation of online content can encourage individual-level participation and decrease the disparity of participation within small groups when the content is situated in countries that are personally relevant to students’ own backgrounds. At the same time, participation was influenced by individual demographics and group dynamics.
In the Massive Online Open Course (MOOC) forum, learning engagement encompasses three fundamental dimensions—cognitive, emotional, and behavioral engagement—that intricately interact to jointly influence students’ learning achievements. However, the interplay between multiple engagement dimensions and their correlations with learning achievement remain understudied, particularly across different academic disciplines. This study adopts an automated configurational approach that integrates bidirectional encoder representation from transformers (BERT) and fuzzy set qualitative comparative analysis (fsQCA) to explore the configurations of learning engagement, their connections with learning achievement, and variations across disciplines. Our analysis reveals a nuanced profile of learners' learning engagement, indicating the high-achieving individuals demonstrated more frequent posting and commenting behaviors and the high-level cognitive engagement than low-achieving individuals. Second, our analysis revealed multiple configurations where the coexistence or absence of factors at different levels of the cognitive, behavioral, and emotional dimensions significantly impacted learning achievement. Learners who conducted posting and replying behaviors, expressed positive emotions, and engaged in deep cognitive engagement tended to achieve superior learning outcomes. Third, there were significant differences in behavioral and emotional engagement among learners across different academic disciplines. Specifically, pure discipline learners were more inclined to engage in posting behaviors than the applied discipline learners. Across academic disciplines, positive emotions correlated strongly with higher achievement. These findings deepen our understanding of the multifaceted characteristics of learning engagement in MOOCs and highlight the importance of disciplinary distinctions, providing a foundation for educators and designers to optimize learners’ MOOC effects and tailor learning experiences in diverse disciplinary contexts.
Background
Forums in massive open online courses (MOOCs) enable written exchanges on course content; hence, they can potentially facilitate learners' cognitive engagement. Given the myriad of MOOC forum messages, this engagement is commonly analysed automatically through the linguistic features of the messages. Assessing linguistic features of learners' forum messages involves consideration of the learning tasks. MOOC forum discussion tasks, however, have not been previously considered.
Objective and Method
This study explores the effects of MOOC forum discussion tasks on learners' cognitive engagement. Based on the structure of observed learning outcomes (SOLO) taxonomy, we manually annotate distinct levels of cognitive engagement encouraged in forum discussion tasks and displayed by learners in messages starting discussions (i.e., thread starters). We study the linguistic features of thread starters in relation to the pedagogical design of the discussion tasks. Additionally, we use random‐forest modelling to identify the linguistic and task‐related features that help to categorise learners' cognitive engagement according to SOLO levels.
Results
Manual analysis showed that learners' thread starters mainly reflect surface SOLO levels and include few academic words and cohesive language. Random‐forest modelling showed that these linguistic features, together with the SOLO levels encouraged in the discussion tasks, played an important role in identification of learners' cognitive engagement.
Major Takeaways
Our results highlight the importance of the pedagogical design of MOOC forum tasks in helping learners engage cognitively. Our study also contributes to the empirical evidence that learners' linguistic choices can afford insights into the quality of their cognitive engagement.
This study constructed an online learning behavior framework consisting of two dimensions: knowledge processing and learning context. A C program practice APP named D-quiz was used to track the learning behavior of 543 freshmen over one semester. The collected data were then transformed into seven indicators that corresponded to the online learning model. Using five machine learning algorithms, this study analyzed the factors that impact online learning. Results demonstrated that both knowledge processing and learning context significantly impact online learning, with the Random Forest algorithm achieving the highest F1 Score of 0.743. Furthermore, the feature importance ranking indicated that learning context exerts a stronger influence on online learning effect than knowledge processing.KeywordsOnline learningLearning contextKnowledge processingLearning behaviorMachine learning
High dropout rates have always been a problem for massive open online courses (MOOCs) to address. Dropout prediction has been used as an important method to address this situation. Learners' online learning behaviour is a good indicator of MOOC learning status. However, traditional prediction models treat each learning behaviour as an independent variable and ignore the correlation between learning behaviours. To address this issue, we propose a fusion deep dropout prediction model. The convolutional neural network (CNN) is used to obtain local feature representations based on learners' behaviour data, and the self-attention mechanism is used to learn correlations between different feature representations. At the same time, a bidirectional long short-term memory network (Bi-LSTM) is used to obtain a time-series feature vector representation and input the vector representation into the sigmoid function for prediction. We use the fusion deep model to predict behaviour data of 55,045 learners on the KDD Cup 2015 dataset. The experimental results show that the accuracy of the fusion deep model is 3% higher than that of the traditional method, which effectively improves the accuracy of dropout prediction.
To examine the similarities and differences between two closely related yet distinct fields – Educational Data Mining (EDM) and Learning Analytics (LA) – this study conducted a literature review of the empirical studies published in both fields. We synthesized 492 LA and 194 EDM articles published during 2015–2019. We compared the similarities and differences in research across the two fields by examining data analysis tools, common keywords, theories, and definitions listed. We found that most studies in both fields did not clearly identify a theoretical framework. For both fields, theories of self-regulated learning are most frequently used. We found, through keyword analysis, that both fields are closely related to each other as “learning analytics” is most frequently listed keyword for EDM and vice versa for LA. However, one notable difference relates to how LA studies listed social-related keywords whereas EDM studies listed keywords related to technical methods. The tools used for data analysis overlap largely but some of the LA studies listed tools for qualitative data analysis and social network analysis whereas EDM studies did not. Finally, the distinction of the two fields is defined differently by authors as some demarcate the differences whereas some address them interchangeably.
Το 2008 οι καθηγητές George Siemens και Stephen Downes δημιούργησαν το πρώτο MOOC, έναν νέο τύπο online μαθημάτων που έμελλε να αλλάξει το τοπίο της online εξ αποστάσεως εκπαίδευσης, προσφέροντας τη δυνατότητα σε χιλιάδες άτομα να τα παρακολουθήσουν. Παρά τη θετική τους συμβολή στην προσφορά τριτοβάθμιου επιπέδου εκπαίδευσης σε χιλιάδες ενδιαφερόμενους από όλο τον κόσμο χωρίς περιορισμούς, υπάρχουν ερωτήματα όσον αφορά την αποτελεσματικότητά τους και
τον υψηλό βαθμό εγκατάλειψης που παρουσιάζουν. Με στόχο τη βελτίωσή τους, οι ερευνητές επιχειρούν διάφορες προσεγγίσεις. Μια από αυτές είναι η τροποποίηση του βασικού σχεδιασμού των μαθημάτων με σκοπό να βοηθήσουν την αυτο-ρύθμιση των συμμετεχόντων, η οποία θεωρείται αναγκαία σε αυτού του τύπου τα εκπαιδευτικά περιβάλλοντα. Στην παρούσα εργασία που αποτελεί τμήμα ευρύτερης βιβλιογραφικής ανασκόπησης σχετικά με τα MOOCs, θα παρουσιαστούν τα ευρήματα σχετικά με τις παρεμβάσεις που έχουν γίνει στον σχεδιασμό των μαθημάτων με στόχο να διευκολυνθεί η αυτο-ρύθμιση όσων τα παρακολουθούν.
This study evaluates the validity of an algorithm for measuring the efficiency of social learning networks in discussion forums accompanying MOOCs of conventional format, which consist of video lectures and problem assignments. The algorithm models social learning networks as a function of users’ knowledge seeking and knowledge disseminating tendencies across course topics and offers a means to optimize social learning networks by connecting users seeking and disseminating information on specific topics. We use the algorithm to analyze the social learning network manifest in the discussion format of a MOOC forum incorporating video lectures and problem assignments. As a measure of the degree that knowledge seekers and knowledge disseminators are connected in the network, we observe a very sparse network with few discussion participants and a limited range of topics. Hence, only small gains are available through optimization, since for a very sparse network, few connections can be made. The development of a metric for the analysis of social learning networks would provide instructors and researchers with a means to optimize online learning environments for empowering social learning. Finally, we discuss our findings with respect to the potential of self-optimizing discussion forums for supporting social learning online.
Recent research shows that learners who are able to self-regulate their learning show greater levels of engagement with Massive Open Online Course (MOOC) content. To improve support for learners in their self-regulatory processes, researchers have proposed technological solutions to transform recorded MOOC data into actionable knowledge. However, studies providing empirical evidence on how these solutions impact learners' engagement with the course and their self-regulatory behavior remain scarce. In this paper, we present the results of an observational case study in which NoteMyProgress (NMP), a web-based tool designed to support learners' self-regulation in MOOCs, is applied as an intervention in two MOOCs. The main aim of this study is to provide insights into how the support of learners' Self-regulated Learning (SRL) strategies correlates with course engagement. We performed the evaluation using a sample of 263 learners and utilized distinct data sources in order to propose indicators for learners' engagement with the course and NMP. Results show a positive correlation between learners' final grades with NMP functionalities that support goal setting, organization (note-taking), and self-reflection (social comparison) SRL strategies. Furthermore, we found no significant behavioral differences in how learners with low SRL and high SRL profiles engage with the course or NMP. Finally, we discuss how these results relate to prior work and the implications for future technological solutions that seek to promote engagement in MOOCs.
Nowadays, the digital learning environment has revolutionized the vision of distance learning course delivery and drastically transformed the online educational system. The emergence of Massive Open Online Courses (MOOCs) has exposed web technology used in education in a more advanced revolution ushering a new generation of learning environments. The digital learning environment is expected to augment the real-world conventional education setting. The educational pedagogy is tailored with the standard practice which has been noticed to increase student success in MOOCs and provide a revolutionary way of self-regulated learning. However, there are still unresolved questions relating to the understanding of learning analytics data and how this could be implemented in educational contexts to support individual learning. One of the major issues in MOOCs is the consistent high dropout rate which over time has seen courses recorded less than 20% completion rate. This paper explores learning analytics from different perspectives in a MOOC context. First, we review existing literature relating to learning analytics in MOOCs, bringing together findings and analyses from several courses. We explore meta-analysis of the basic factors that correlate to learning analytics and the significant in improving education. Second, using themes emerging from the previous study, we propose a preliminary model consisting of four factors of learning analytics. Finally, we provide a framework of learning analytics based on the following dimensions: descriptive, diagnostic, predictive and prescriptive, suggesting how the factors could be applied in a MOOC context. Our exploratory framework indicates the need for engaging learners and providing the understanding of how to support and help participants at risk of dropping out of the course.
This study evaluates a two-week MOOC delivered on FutureLearn as part of an MSc in Nursing accredited by Coventry University to establish whether learners are demonstrating transformative learning. Evaluation is in the form of a rubric which is designed using Mezirow’s theory of Transformative Learning as a theoretical framework, alongside the activity types used to inform design of FutureLearn courses. The literature review finds that there is a gap in the research in evaluating for-credit MOOCs against the intended educational aims of the accrediting institution. The rubric created for this research attempts to fill that gap, by providing a means to evaluate both student learning and learning design. The rubric identifies that the learners on this course are demonstrating elements of transformative learning at the lower levels of Mezirow’s seven stages of critical reflection. Although the rubric was designed with the aim of evaluating MOOCs, it can be applied to any online learning experience that includes student engagement, either written or spoken. The paper makes recommendations for future developments and further research.
In MOOCs, course reviews serve as a new interactive tool that has not yet been sufficiently exploited. Therefore, this study investigated learners' explicit behaviors and implicit discourse content derived from reviews by using a mixed approach of text mining and statistical analysis. We proposed an improved topic model called Behavior–Emotion Topic Model (BETM) to detect reviews' semantic content between two achievement groups (completers and non-completers). Then we performed statistical analysis to investigate differences in the two groups' discourse behaviors and content. Results showed significant differences in discourse behaviors and focused topics between completers and non-completers. Specifically, posting reviews was a significant behavior for completers, while replying and giving peers' reviews “thumps up” were significant behaviors for non-completers. Furthermore, completers tended to express appreciation of course-related content by posting reviews and, afterward, showing certificates, whereas non-completers tended to hold negative attitudes toward the platform construction's technical issues by replying. Finally, we conducted an evolutionary analysis to explore the dynamics of learners' focused content throughout the course over 3 years; this can provide instructors new insights for the development of online course, thus meeting future learners' needs by adjusting the teaching process.
Vivimos años en los que las tendencias emergentes en educación están tomando un papel relevante en las discusiones sobre la deriva de las instituciones de enseñanza. La inclusión de las
TIC y la proliferación de lo que se conoce bajo el nombre de m-learning conlleva repensar las formas en las que los estudiantes aprenden y los docentes enseñan. Este es un escenario convulso y
en constante cambio, pero que abre puertas a nuevas formas de hacer en educación. Un camino
hacia una educación más abierta, democrática y eficaz.
No es un camino sencillo. Con frecuencia, la rapidez con la que se producen los avances tecnológicos es más rápida que nuestra capacidad de integrar estas nuevas posibilidades en esquemas teóricos significativos y en propuestas de investigación novedosas. Las buenas prácticas y la
investigación conforman los dos ejes de esta publicación.
Relacionado con las buenas prácticas, encontraremos varios capítulos sobre diferentes experiencias innovadoras basadas en la utilización de la tecnología en los diversos niveles educativos
que conforman nuestro sistema educativo actual. El lector podrá leer y reflexionar sobre la variedad de usos que se le pueden dar a la tecnología en diversos contextos educativos y relacionado
con temas básicos y transversales que se desarrollan o se deberían trabajar en las aulas actuales y
del futuro.
Unido a esto último tampoco podemos olvidarnos de la investigación educativa con TIC, aspecto que se recoge desde diversas perspectivas en los diferentes capítulos de esta publicación.
Para poder realizar propuestas innovadoras y significativas, debemos analizar, comparar y reflexionar, por lo que la investigación rigurosa y académica se convierte en pilar fundamental para
el cambio.
Nowadays, the digital learning environment has revolutionized the vision of distance learning course delivery and drastically transformed the online educational system. The emergence of Massive Open Online Courses (MOOCs) has exposed web technology used in education in a more advanced revolution ushering a new generation of learning environments. The digital learning environment is expected to augment the real-world conventional education setting. The educational pedagogy is tailored with the standard practice which has been noticed to increase student success in MOOCs and provide a revolutionary way of self-regulated learning. However, there are still unresolved questions relating to the understanding of learning analytics data and how this could be implemented in educational contexts to support individual learning. One of the major issues in MOOCs is the consistent high dropout rate which over time has seen courses recorded less than 20% completion rate. This paper explores learning analytics from different perspectives in a MOOC context. First, we review existing literature relating to learning analytics in MOOCs, bringing together findings and analyses from several courses. We explore meta-analysis of the basic factors that correlate to learning analytics and the significant in improving education. Second, using themes emerging from the previous study, we propose a preliminary model consisting of four factors of learning analytics. Finally, we provide a framework of learning analytics based on the following dimensions: descriptive, diagnostic, predictive and prescriptive, suggesting how the factors could be applied in a MOOC context. Our exploratory framework indicates the need for engaging learners and providing the understanding of how to support and help participants at risk of dropping out of the course.
In this paper, we analyse the role of the student’s digital reputation as a motivational factor for successfully completing Massive Open Online Courses (MOOCs). After a review of the academic literature on the role that the student’s reputation plays in community learning and to understand the role of involving the student that new techniques of gamification have taken in virtual learning, an empirical analysis is performed on the basis of data from a pioneer MOOC of Social Entrepreneurship (5016 registered students, 3250 of whom actually started). Using Classification and Regression Trees (CARTs) and Principal Component Analysis (PCA), we conclude that social reputation is a key factor for the student in successfully completing the course, and that the student’s retention is, along with his e-reputation (“karma”), directly related to his degree of participation, the rewards received, and his correspondence with peers. This shows that one of the factors that explain the student’s MOOC completion is his degree of interaction with other students, and particularly his reputation among them.
The rapid growth of technology has transformed education from conventional classroom teaching to learner centred, flexible learning anytime anywhere resulting in the popularity of Massive Open Online Courses (MOOCs). It is evident that many MOOCs have adopted social media tools to engage their learners through information sharing, participation, and collaboration to create and consume information. In addition to this, social media is the best place to understand crowd sentiments considering the massive amount of data being generated. This study through text mining of MOOC providers related Twitter data discovers the sentiments of users towards MOOCs, association rules, and the engagement of MOOC providers in Twitter. This article also analyses the traits of the influencers in seven MOOC providers and discovers that the influencers are popular users with higher numbers of followers count. The implications are discussed for MOOC industry, academic institutions, and the government.
Online learning has evolved over the years, becoming increasingly accessible and gaining credibility as a learning medium. Universities are reshaping education on the Web by means of Massive Open Online Courses (MOOC). The learners can access educational content anytime, from anyplace and participatory culture using social networks seems to be the biggest advantage that MOOC has brought to the world of education especially in the higher education arena. MOOC has rapidly developed from a small section of specialist courses to millions of registered users on major online platforms offering hundreds of courses. As MOOC expands, attrition and accreditation is a problem. Though MOOC is expanding very fast, the literature on MOOC research is limited. This paper presents a survey on MOOC providers which aims to capture the state of MOOC, how they evolved and to identify issues and challenges that are important for MOOC providers.
“Internetworking with TCP/IP” is a Massive Open Online Course, held in German at openHPI end of 2012, that attracted a large audience that has not been in contact with higher education before. The course followed the xMOOC model based on a well-defined sequence of learning content, mainly video lectures and interactive self-tests, and with heavy reliance on social collaboration features. From 2726 active participants, 38% have participated in a survey at the end of the course. This paper presents an analysis of the survey responses with respect to the following questions: 1) How can a MOOC accommodate different learning styles and 2) What recommendations for the design and organization of a MOOC can be concluded from the responses? We finally give an outlook on challenges for the further development of openHPI. Those challenges are based on didactical and technical affordances for a better support of the different learning styles. We propose an evolution of the xMOOC, that bridges the gap to the cMOOC model by developing tools that allow users to create diverging paths through the learning material, involve the user personally in the problem domain with (group) hands-on exercises and reward user contributions by means of gamification.
We present a method for assessing science inquiry performance, specifically for the inquiry skill of designing and conducting experiments, using educational data mining on students' log data from online microworlds in the Inq-ITS system (Inquiry Intelligent Tutoring System; www.inq-its.org). In our approach, we use a 2-step process: First we use text replay tagging, a type of rapid protocol analysis in which categories are developed and, in turn, used to hand-score students' log data. In the second step, educational data mining is conducted using a combination of the text replay data and machine-distilled features of student interactions in order to produce an automated means of assessing the inquiry skill in question; this is referred to as a detector. Once this detector is appropriately validated, it can be applied to students' log files for auto-assessment and, in the future, to drive scaffolding in real time. Furthermore, we present evidence that this detector developed in 1 scientific domain, phase change, can be used—with no modification or retraining—to effectively detect science inquiry skill in another scientific domain, density.
Nowadays, the Web has revolutionized our vision as to how deliver courses in a radically transformed and enhanced way. Boosted by Cloud computing, the use of the Web in education has revealed new challenges and looks forward to new aspirations such as MOOCs (Massive Open Online Courses) as a technology-led revolution ushering in a new generation of learning environments. Expected to deliver effective education strategies, pedagogies and practices, which lead to student success, the massive open online courses, considered as the “linux of education”, are increasingly developed by elite US institutions such MIT, Harvard and Stanford by supplying open/distance learning for large online community without paying any fees, MOOCs have the potential to enable free university-level education on an enormous scale. Nevertheless, a concern often is raised about MOOCs is that a very small proportion of learners complete the course while thousands enrol for courses. In this paper, we present LASyM, a learning analytics system for massive open online courses. The system is a Hadoop based one whose main objective is to assure Learning Analytics for MOOCs’ communities as a mean to help them investigate massive raw data, generated by MOOC platforms around learning outcomes and assessments, and reveal any useful information to be used in designing learning-optimized MOOCs. To evaluate the effectiveness of the proposed system we developed a method to identify, with low latency, online learners more likely to drop out
Learning analytics offers higher education valuable insights that can inform strategic decision-making regarding resource allocation for educational excellence. Research demonstrates that learning management systems (LMSs) can increase student sense of community, support learning communities and enhance student engagement and success, and LMSs have therefore become core enterprise component in many universities. We were invited to undertake a current state analysis of enterprise LMS use in a large research-intensive university, to provide data to inform and guide an LMS review and strategic planning process. Using a new e-learning analytics platform, combined with data visualization and participant observation, we prepared a detailed snapshot of current LMS use patterns and trends and their relationship to student learning outcomes. This paper presents selected data from this "current state analysis" and comments on what it reveals about the comparative effectiveness of this institution's LMS integration in the service of learning and teaching. More critically, it discusses the reality that the institutional planning process was nonetheless dominated by technical concerns, and made little use of the intelligence revealed by the analytics process. To explain this phenomenon we consider theories of change management and resistance to innovation, and argue that to have meaningful impact, learning analytics proponents must also delve into the socio-technical sphere to ensure that learning analytics data are presented to those involved in strategic institutional planning in ways that have the power to motivate organizational adoption and cultural change.
This paper presents an in-depth empirical analysis of a nine-week MOOC. This analysis provides novel results regarding participants’ profiles and use of built-in and external social tools. The results served to detect seven participants’ patterns and conclude that the forum was the social tool preferred to contribute to the MOOC.
The Khan Academy platform enables powerful on-line courses in which students can watch videos, solve exercises or earn badges. This platform provides an advanced learning analytics module with useful visualizations for teachers and students. Nevertheless, this learning analytics support can be improved with recommendations and new useful higher level visualizations in order to try to improve the learning process. In this paper, we describe our architecture for processing data from the Khan Academy platform in order to show new higher level learning visualizations and recommendations. The different involved elements of the architecture are presented and the different decisions are justified. In addition, we explain some initial examples of new useful visualizations and recommendations for teachers and students as part of our extension of the learning analytics module for the Khan Academy platform. These examples use data from an undergraduate Physics course developed at Universidad Carlos III de Madrid with more than 100 students using the Khan Academy system.
To process low level educational data in the form of user events and interactions and convert them into information about the learning process that is both meaningful and interesting presents a challenge. In this paper, we propose a set of high level learning parameters relating to total use, efficient use, activity time distribution, gamification habits, or exercise-making habits, and provide the measures to calculate them as a result of processing low level data. We apply these parameters and measures in a real physics course with more than 100 students using the Khan Academy platform at Universidad Carlos III de Madrid. We show how these parameters can be meaningful and useful for the learning process based on the results from this experience.
As MOOCs grow in popularity, the relatively low completion rates of learners has been a central criticism. This focus on completion rates, however, reflects a monolithic view of disengagement that does not allow MOOC designers to target interventions or develop adaptive course features for particular subpopulations of learners. To address this, we present a simple, scalable, and informative classification method that identifies a small number of longitudinal engagement trajectories in MOOCs. Learners are classified based on their patterns of interaction with video lectures and assessments, the primary features of most MOOCs to date.
In an analysis of three computer science MOOCs, the classifier consistently identifies four prototypical trajectories of engagement. The most notable of these is the learners who stay engaged through the course without taking assessments. These trajectories are also a useful framework for the comparison of learner engagement between different course structures or instructional approaches. We compare learners in each trajectory and course across demographics, forum participation, video access, and reports of overall experience. These results inform a discussion of future interventions, research, and design directions for MOOCs. Potential improvements to the classification mechanism are also discussed, including the introduction of more fine-grained analytics.
Nell’era di Internet, delle tecnologie mobili e dell’istruzione aperta, la necessità di interventi per migliorare l’efficienza e la qualità dell’istruzione superiore è diventata pressante. I big data e il Learning Analytics possono contribuire a condurre questi interventi, e a ridisegnare il futuro dell’istruzione superiore. Basare le decisioni su dati e sulle evidenze empiriche sembra incredibilmente ovvio. Tuttavia, l’istruzione superiore, un campo che raccoglie una quantità enorme di dati sui propri “clienti”, è stata tradizionalmente inefficiente nell’utilizzo dei dati, spesso operando con notevole ritardo nell’analizzarli, pur essendo questi immediatamente disponibili.
In questo articolo, viene evidenziato il valore delle tecniche di analisi dei dati per l’istruzione superiore, e presentato un modello di sviluppo per i dati legati all’apprendimento. Ovviamente, l’apprendimento è un fenomeno complesso, e la sua descrizione attraverso strumenti di analisi non è semplice; pertanto, l’articolo presenta anche le principali problematiche etiche e pedagogiche connesse all’utilizzo delle tecniche di analisi dei dati in ambito educativo. Cionondimeno, il Learning Analytics può penetrare la nebbia di incertezza che avvolge il futuro dell’istruzione superiore, e rendere più evidente come allocare le risorse, come sviluppare vantaggi competitivi e, soprattutto, come migliorare la qualità e il valore dell’esperienza di apprendimento.
The Open University has published the second in its influential series of Innovating Pedagogy reports that explore new forms of teaching, learning and assessment, to guide educators and policy makers. The 2013 report updates four previous areas of innovation and introduces six new ones: Crowd Learning, Learning from Gaming, Maker Culture, Geo-Learning, Digital Scholarship and Citizen Inquiry. The report can be downloaded from www.open.ac.uk/innovating.
Growing interest in data and analytics in education, teaching, and learning raises the priority for increased, high-quality research into the models, methods, technologies, and impact of analytics. Two research communities -- Educational Data Mining (EDM) and Learning Analytics and Knowledge (LAK) have developed separately to address this need. This paper argues for increased and formal communication and collaboration between these communities in order to share research, methods, and tools for data mining and analysis in the service of developing both LAK and EDM fields.
“Circuits and Electronics” (6.002x), which began in March 2012, was the first MOOC developed by edX, the consortium led by MIT and Harvard. Over 155,000 students initially registered for 6.002x, which was composed of video lectures, interactive problems, online laboratories, and a discussion forum. As the course ended in June 2012, researchers began to analyze the rich sources of data it generated. This article describes both the first stage of this research, which examined the students’ use of resources by time spent on each, and a second stage that is producing an in-depth picture of who the 6.002x students were, how their own background and capabilities related to their achievement and persistence, and how their interactions with 6.002x’s curricular and pedagogical components contributed to their level of success in the course.
In recent years, increasing attention has been devoted to virtual learning. In the last decade, a large number of studies in Computer-Supported Collaborative Learning (CSCL) have assessed how social interaction, learning processes and outcomes in virtual settings are intertwined. Although recent research findings indicate that learners differ with respect to the amount and type of discourse contributed in virtual settings, little is known about the causes of these differences. The research presented here looks into the effects of motivation of learners on their contribution to discourse using the Deci and Ryan framework of (intrinsic/extrinsic) motivation. This study of 100 participants who were randomly distributed in six groups of 14 members collaborated in a virtual setting to remediate deficiencies in economics indicates that individuals differed with respect to the amount of discourse activity. Furthermore, an integrated multi-method approach (Content Analysis, Social Network Analysis and Academic Motivation Scale) was used in order to examine the impact of academic motivation on the type of discourse activity contributed and on the position of the learner in the social network. The results indicate that highly intrinsically motivated learners become central and prominent contributors to cognitive discourse. In contrast, extrinsically motivated learners contribute on average and are positioned throughout the social network. The research results reveal that differences in academic motivation influences the type of contributions to discourse as well as the position a learner takes within the social network.
Group work is widespread in education. The growing use of online tools supporting group work generates huge amounts of data. We aim to exploit this data to support mirroring: presenting useful high-level views of information about the group, together with desired patterns characterizing the behaviour of strong groups. The goal is to enable the groups and their facilitators to see relevant aspects of the group's operation and provide feedback if these are more likely to be associated with positive or negative outcomes and where the problems are. We explore how useful mirror information can be extracted via a theory-driven approach and a range of clustering and sequential pattern mining. The context is a senior software development project where students use the collaboration tool TRAC. We extract patterns distinguishing the better from the weaker groups and get insights in the success factors. The results point to the importance of leadership and group interaction, and give promising indications if they are occurring. Patterns indicating good individual practices were also identified. We found that some key measures can be mined from early data. The results are promising for advising groups at the start and early identification of effective and poor practices, in time for remediation.
We propose that the design and implementation of effective Social Learning Analytics presents significant challenges and opportunities for both research and enterprise, in three important respects. The first is the challenge of implementing analytics that have pedagogical and ethical integrity, in a context where power and control over data is now of primary importance. The second challenge is that the educational landscape is extraordinarily turbulent at present, in no small part due to technological drivers. Online social learning is emerging as a significant phenomenon for a variety of reasons, which we review, in order to motivate the concept of social learning, and ways of conceiving social learning environments as distinct from other social platforms. This sets the context for the third challenge, namely, to understand different types of Social Learning Analytic, each of which has specific technical and pedagogical challenges. We propose an initial taxonomy of five types. We conclude by considering potential futures for Social Learning Analytics, if the drivers and trends reviewed continue, and the prospect of solutions to some of the concerns that institution-centric learning analytics may provoke.
The issue of leadership in virtual teams is an increasingly important one for many modern organizations, because these teams – typically project-based with a finite lifespan and specific deliverables – are frequently self-managed, having no designated leader within the team (Yukl 1998). While such teams may have increased flexibility in setting goals and achieving them, the virtual context may hinder team members' abilities to coordinate activities and influence others (Manz and Sims 1993). Longitudinal data were collected over the course of a semester from virtual teams, comprised of students from three North American universities. Our results suggest that high performing self-managed virtual teams displayed significantly more leadership behaviors over time compared to their low performing counterparts. Specifically, these teams displayed significantly more concentrated leadership behavior focused on performance (i.e. “Producer” behavior) and shared leadership behavior focused on keeping track of group work (i.e. “Monitor” behavior) than the lower performing teams. Interestingly, these behaviors emerged strongly during the first half of the groups' lifespan, and stayed throughout the life of the groups, but steadily dissipated in strength over time.
Designing effective CSCL processes is a complex task that can be supported by existing good practices formulated as pedagogical patterns or script. Over the past years the TEE research has shown that CSCL script acts as Mediating Artifacts (MA) designing educational scenarios and structuring and prescribing roles and activities. This work proposes an approach, based on Social Network Analysis and Semantic Web, in order to improve definition and instantiation phases of IMS-LD scripts.
The ASSISTment system was used by over 600 students in 2004-05 school year as part of their math class. While in (7) we reported student learn- ing within the ASSISTment system, in this paper we focus on the assessment aspect. Our approach is to use data that the system collected through a year to tracking student learning and thus estimate their performance on a high-stake state test (MCAS) at the end of the year. Because our system is an intelligent tutoring system, we are able to log how much assistance students needed to solve problems (how many hints students requested and how many attempts they had to make). In this paper, our goal is to determine if the models we built by taking the assistance information into account could predict students' test scores better. We present some positive evidence that shows our goal is achieved.
Accurate prediction of demographic attributes from social media and other informal online content is valuable for marketing, personalization, and legal investigation. This paper describes the construction of a large, multilingual dataset labeled with gender, and investigates statistical models for determining the gender of uncharacterized Twitter users. We explore several different classifier types on this dataset. We show the degree to which classifier accuracy varies based on tweet volumes as well as when various kinds of profile metadata are included in the models. We also perform a large-scale human assessment using Amazon Mechanical Turk. Our methods significantly out-perform both baseline models and almost all humans on the same task.
The opportunity to learn from the best educators of the top universities at no cost has captured the imagination of millions who have registered for MOOCs across the United States and abroad. As MOOCs have gained momentum, critical debate is brewing on the pedagogical effectiveness of MOOCs. The high attrition rate of students who register at the beginning of a MOOC is a major cause of concern regarding the long-term success, impact, and sustainability of MOOCs. Having conducted many experiments and studies on the application of adaptive learning in personalized online learning, Synaptic Global Learning set out to develop the first adaptive MOOC (aMOOC) platform, providing a strong pedagogical framework and a personalized learning experience in a MOOC learning environment. In partnership with the Center for Innovation and Excellence in eLearning of the College of Advancing and Continued Studies, University of Massachusetts Boston, Synaptic Global Learning developed the first aMOOC in the area of computational molecular dynamics (CMD). This article describes the design, development, and deployment of this aMOOC. Part I of this case-study is focused on the pedagogical framework that allows four dimensions of learning and dynamic rendering of the content to accommodate five distinct learning strategies using the adaptive mobile learning (AMOL) platform. The AMOL platform is then implemented on the Amazon Web Services (AWS) cloud architecture and load-tested for the use of aMOOCs. The results indicate that even with dynamic rendering of the fine learning strategies and real-time feedback, the AWS-based scalable architecture is sufficient to provide high throughput and acceptable response time.
Many courses are currently delivered using Course Management Systems (CMS). Discussion forums within these systems provide the basis for collaborative learning. In this chapter, we present the use of Social Network Analysis (SNA) to analyze the structure of interactions between the students in these forums. Various metrics are introduced for ranking and determining roles, while clustering and temporal analysis techniques are applied to study the student communications, the forming of groups, the role changes, as well as scrutinizing the content of the exchanged messages. Our approach provides the instructor with better means to assess the participation of students by (1) identification of participants' roles; (2) dynamic visualization of interactions between the participants and the groups they formed; (3) presenting hierarchy of the discussed topics; and (4) tracking the evolution and growth of these patterns and roles over time. The applicability of the proposed analyses are illustrated through several case studies.
We propose that the design and implementation of effective Social Learning Analytics (SLA) present significant challenges and opportunities for both research and enterprise, in three important respects. The first is that the learning landscape is extraordinarily turbulent at present, in no small part due to technological drivers. Online social learning is emerging as a significant phenomenon for a variety of reasons, which we review, in order to motivate the concept of social learning. The second challenge is to identify different types of SLA and their associated technologies and uses. We discuss five categories of analytic in relation to online social learning; these analytics are either inherently social or can be socialised. This sets the scene for a third challenge, that of implementing analytics that have pedagogical and ethical integrity in a context where power and control over data are now of primary importance. We consider some of the concerns that learning analytics provoke, and suggest that Social Learning Analytics may provide ways forward. We conclude by revisiting the drivers andtrends, and consider future scenarios that we may see unfold as SLA tools and services mature. © International Forum of Educational Technology & Society (IFETS).
HarvardX and MITx are collaborative institutional efforts between Harvard University and MIT to enhance campus-based education, advance educational research, and increase access to online learning opportunities worldwide. Over the year from the fall of 2012 to the summer of 2013, HarvardX and MITx launched 17 courses on edX, a jointly founded platform for delivering massive open online courses (MOOCs). In that year, 43,196 registrants earned certificates of completion. Another 35,937 registrants explored half or more of course content without certification. An additional 469,702 registrants viewed less than half of the content. And 292,852 registrants never engaged with the online content. In total, there were 841,687 registrations from 597,692 unique users across the first year of HarvardX and MITx courses. This report is a joint effort by institutional units at Harvard and MIT to describe the registrant and course data provided by edX in the context of the diverse efforts and intentions of HarvardX and MITx instructor teams.
A lack of student motivation is a problem in many courses in electrical engineering. Introducing competition between students can enhance their motivation, but it can also generate negative emotions. This paper presents an empirical study of students in a telecommunications engineering degree; it measured their level of motivation, and their emotions about competition in education in general, and in particular about the Information System for Competition based on pRoblem solving in Education (ISCARE). ISCARE makes adaptive matches, pairing students who share a similar knowledge level. The study results show that students had moderate levels of motivation and experienced no negative emotions while using ISCARE. Under certain circumstances, therefore, competition in education, using technology systems that provide competition, can produce motivation without generating negative emotions. Female students and students with low ISCARE scores were less motivated and experienced more negative emotions. In addition, motivation and emotions toward ISCARE were more positive than they had been toward competition in general.
This paper describes two diagnostic tools to predict students are at risk of dropping out from an online class. While thousands of students have been attracted to large online classes, keeping them motivated has been challenging. Experiments on a large, online HCI class suggest that the tools these paper introduces can help identify students who will not complete assignments, with an F1 score of 0.46 and 0.73 three days before the assignment due date.
As college tuitions soar, various online models vie to educate college students worldwide---at no cost.
The constant development of new Internet platforms is shifting the users’ role of such platforms, from
viewers to main actors. In the field of education, faculty can take advantage of these new technologies
for the design of pedagogical contents. The face-to-face observation of behavioral patterns allows faculty
to detect and track new problems, and to apply possible corrections which would improve the learning/
teaching process. However, with a distance methodology, these observations are not possible. When forums
are created they are intended to discuss particular topics. It is relevant to monitor that the topics
discussed are the intended ones in order to achieve course objectives. To tackle this shortcoming, our
work studies the dynamics of relevant topics in on-line asynchronous discussion forums, and this is done
by analyzing the large amount of students’ interactions generated in the forums of our Learning Management
System (LMS). In particular, we analyze the students’ behavior patterns in the forums of a distance
subject, and characterize the relevant topics and subtopics from the forums’ messages belonging to two
academic years. From the statistical and graphical results obtained, a set of valuable recommendations
are also given.
We study user behavior in the courses offered by a major Massive Online Open
Course (MOOC) provider during the summer of 2013. Since social learning is a
key element of scalable education in MOOCs and is done via online discussion
forums, our main focus is in understanding forum activities. Two salient
features of MOOC forum activities drive our research: 1. High decline rate: for
all courses studied, the volume of discussions in the forum declines
continuously throughout the duration of the course. 2. High-volume, noisy
discussions: at least 30% of the courses produce new discussion threads at
rates that are infeasible for students or teaching staff to read through.
Furthermore, a substantial portion of the discussions are not directly
course-related.
We investigate factors that correlate with the decline of activity in the
online discussion forums and find effective strategies to classify threads and
rank their relevance. Specifically, we use linear regression models to analyze
the time series of the count data for the forum activities and make a number of
observations, e.g., the teaching staff's active participation in the discussion
increases the discussion volume but does not slow down the decline rate. We
then propose a unified generative model for the discussion threads, which
allows us both to choose efficient thread classifiers and design an effective
algorithm for ranking thread relevance. Our ranking algorithm is further
compared against two baseline algorithms, using human evaluation from Amazon
Mechanical Turk.
Learning analytics are rapidly being implemented in different educational settings, often without the guidance of a research base. Vendors incorporate analytics practices, models, and algorithms from datamining, business intelligence, and the emerging "big data" fields. Researchers, in contrast, have built up a substantial base of techniques for analyzing discourse, social networks, sentiments, predictive models, and in semantic content (i.e., "intelligent" curriculum). In spite of the currently limited knowledge exchange and dialogue between researchers, vendors, and practitioners, existing learning analytics implementations indicate significant potential for generating novel insight into learning and vital educational practices. This paper presents an integrated and holistic vision for advancing learning analytics as a research discipline and a domain of practices. Potential areas of collaboration and overlap are presented with the intent of increasing the impact of analytics on teaching, learning, and the education system.
Handbook of Educational Data Mining (EDM) provides a thorough overview of the current state of knowledge in this area. The first part of the book includes nine surveys and tutorials on the principal data mining techniques that have been applied in education. The second part presents a set of 25 case studies that give a rich overview of the problems that EDM has addressed. Researchers at the Forefront of the Field Discuss Essential Topics and the Latest Advances With contributions by well-known researchers from a variety of fields, the book reflects the multidisciplinary nature of the EDM community. It brings the educational and data mining communities together, helping education experts understand what types of questions EDM can address and helping data miners understand what types of questions are important to educational design and educational decision making. Encouraging readers to integrate EDM into their research and practice, this timely handbook offers a broad, accessible treatment of essential EDM techniques and applications. It provides an excellent first step for newcomers to the EDM community and for active researchers to keep abreast of recent developments in the field.
The trend to adopt more online technologies continues unabated in the higher education sector. This paper elaborates the means by which such technologies can be employed for pedagogical purposes beyond simply providing virtual spaces for bringing learners together. It shows how data about student ‘movement’ within and across a learning community can be captured and analysed for the purposes of making strategic interventions in the learning of ‘at risk’ students in particular, through the application of social network analysis to the engagement data. The study that is set out in the paper indicates that online technologies bring with them an unprecedented opportunity for educators to visualise changes in student behaviour and their learning network composition, including the interventions teachers make in those networks over time. To date, these evaluative opportunities have been beyond the reach of the everyday practitioner—they can now be integrated into every teaching and learning plan.
The role concept has attracted a lot of attention as a construct for facilitating and analysing interactions in the context of computer-supported collaborative learning (CSCL). So far much of this research has been carried out in isolation and the focus on roles lacks cohesion. In this article we present a conceptual framework to synthesise the contemporary conceptualisation of roles, by discerning three levels of the role concept: micro (role as task), meso (role as pattern) and macro (role as stance). As a first step to further conceptualise ‘role as a stance’, we present a framework of eight participative stances defined along three dimensions: group size, orientation and effort. The participative stances – Captain, Over-rider, Free-rider, Ghost, Pillar, Generator, Hanger-on and Lurker – were scrutinised on two data sets using qualitative analysis. The stances aim to facilitate meaningful description of student behaviour, stimulate both teacher and student awareness of roles at the macro-level in terms of participative stances, and evaluate or possibly change the participation to collaborative learning on all levels.