Learning Feedback Based on Dispositional Learning Analytics

Dirk Tempelaar, Quan Nguyen, Bart Rienties

Research output: Chapter in Book/Report/Conference proceedingChapterAcademic

Abstract

The combination of trace data captured from technology-enhanced learning support systems, formative assessment data and learning disposition data based on self-report surveys, offers a very rich context for learning analytics applications. In previous research, we have demonstrated how such Dispositional Learning Analytics applications not only have great potential regarding predictive power, e.g. with the aim to promptly signal students at risk, but also provide both students and teacher with actionable feedback. The ability to link predictions, such as a risk for drop-out, with characterizations of learning dispositions, such as profiles of learning strategies, implies that the provision of learning feedback is not the end point, but can be extended to the design of learning interventions that address suboptimal learning dispositions. Building upon the case studies we developed in our previous research, we replicated the Dispositional Learning Analytics analyses in the most recent 17/18 cohort of students based on the learning processes of 1017 first-year students in a blended introductory quantitative course. We conclude that the outcomes of these analyses, such as boredom being an important learning emotion, planning and task management being crucial skills in the efficient use of digital learning tools, help both predict learning performance and design effective interventions.
Original languageEnglish
Title of host publicationMachine Learning Paradigms
Subtitle of host publicationAdvances in Learning Analytics
EditorsMaria Virvou, Efthimios Alepis, George A. Tsihrintzis, Lakhmi C. Jain
Place of PublicationCham, Switzerland
PublisherSpringer
Pages69-89
Number of pages21
Volume158
ISBN (Electronic)978-3-030-13743-4
ISBN (Print)978-3-030-13742-7
Publication statusPublished - 2020

Publication series

SeriesIntelligent Systems Reference Library book series
Volume158

Keywords

  • Blended learning
  • Dispositional learning analytics
  • E-tutorials
  • Learning feedback
  • Learning dispositions
  • Learning strategies

Cite this

Tempelaar, D., Nguyen, Q., & Rienties, B. (2020). Learning Feedback Based on Dispositional Learning Analytics. In M. Virvou, E. Alepis, G. A. Tsihrintzis, & L. C. Jain (Eds.), Machine Learning Paradigms: Advances in Learning Analytics (Vol. 158, pp. 69-89). Cham, Switzerland: Springer. Intelligent Systems Reference Library book series, Vol.. 158
Tempelaar, Dirk ; Nguyen, Quan ; Rienties, Bart. / Learning Feedback Based on Dispositional Learning Analytics. Machine Learning Paradigms: Advances in Learning Analytics. editor / Maria Virvou ; Efthimios Alepis ; George A. Tsihrintzis ; Lakhmi C. Jain. Vol. 158 Cham, Switzerland : Springer, 2020. pp. 69-89 (Intelligent Systems Reference Library book series, Vol. 158).
@inbook{e6e7b30645914f2c80719414b7bf2188,
title = "Learning Feedback Based on Dispositional Learning Analytics",
abstract = "The combination of trace data captured from technology-enhanced learning support systems, formative assessment data and learning disposition data based on self-report surveys, offers a very rich context for learning analytics applications. In previous research, we have demonstrated how such Dispositional Learning Analytics applications not only have great potential regarding predictive power, e.g. with the aim to promptly signal students at risk, but also provide both students and teacher with actionable feedback. The ability to link predictions, such as a risk for drop-out, with characterizations of learning dispositions, such as profiles of learning strategies, implies that the provision of learning feedback is not the end point, but can be extended to the design of learning interventions that address suboptimal learning dispositions. Building upon the case studies we developed in our previous research, we replicated the Dispositional Learning Analytics analyses in the most recent 17/18 cohort of students based on the learning processes of 1017 first-year students in a blended introductory quantitative course. We conclude that the outcomes of these analyses, such as boredom being an important learning emotion, planning and task management being crucial skills in the efficient use of digital learning tools, help both predict learning performance and design effective interventions.",
keywords = "Blended learning, Dispositional learning analytics, E-tutorials, Learning feedback, Learning dispositions, Learning strategies",
author = "Dirk Tempelaar and Quan Nguyen and Bart Rienties",
note = "Self collected survey and trace data",
year = "2020",
language = "English",
isbn = "978-3-030-13742-7",
volume = "158",
series = "Intelligent Systems Reference Library book series",
pages = "69--89",
editor = "Maria Virvou and Efthimios Alepis and Tsihrintzis, {George A.} and Jain, {Lakhmi C.}",
booktitle = "Machine Learning Paradigms",
publisher = "Springer",
address = "United States",

}

Tempelaar, D, Nguyen, Q & Rienties, B 2020, Learning Feedback Based on Dispositional Learning Analytics. in M Virvou, E Alepis, GA Tsihrintzis & LC Jain (eds), Machine Learning Paradigms: Advances in Learning Analytics. vol. 158, Springer, Cham, Switzerland, Intelligent Systems Reference Library book series, vol. 158, pp. 69-89.

Learning Feedback Based on Dispositional Learning Analytics. / Tempelaar, Dirk; Nguyen, Quan; Rienties, Bart.

Machine Learning Paradigms: Advances in Learning Analytics. ed. / Maria Virvou; Efthimios Alepis; George A. Tsihrintzis; Lakhmi C. Jain. Vol. 158 Cham, Switzerland : Springer, 2020. p. 69-89 (Intelligent Systems Reference Library book series, Vol. 158).

Research output: Chapter in Book/Report/Conference proceedingChapterAcademic

TY - CHAP

T1 - Learning Feedback Based on Dispositional Learning Analytics

AU - Tempelaar, Dirk

AU - Nguyen, Quan

AU - Rienties, Bart

N1 - Self collected survey and trace data

PY - 2020

Y1 - 2020

N2 - The combination of trace data captured from technology-enhanced learning support systems, formative assessment data and learning disposition data based on self-report surveys, offers a very rich context for learning analytics applications. In previous research, we have demonstrated how such Dispositional Learning Analytics applications not only have great potential regarding predictive power, e.g. with the aim to promptly signal students at risk, but also provide both students and teacher with actionable feedback. The ability to link predictions, such as a risk for drop-out, with characterizations of learning dispositions, such as profiles of learning strategies, implies that the provision of learning feedback is not the end point, but can be extended to the design of learning interventions that address suboptimal learning dispositions. Building upon the case studies we developed in our previous research, we replicated the Dispositional Learning Analytics analyses in the most recent 17/18 cohort of students based on the learning processes of 1017 first-year students in a blended introductory quantitative course. We conclude that the outcomes of these analyses, such as boredom being an important learning emotion, planning and task management being crucial skills in the efficient use of digital learning tools, help both predict learning performance and design effective interventions.

AB - The combination of trace data captured from technology-enhanced learning support systems, formative assessment data and learning disposition data based on self-report surveys, offers a very rich context for learning analytics applications. In previous research, we have demonstrated how such Dispositional Learning Analytics applications not only have great potential regarding predictive power, e.g. with the aim to promptly signal students at risk, but also provide both students and teacher with actionable feedback. The ability to link predictions, such as a risk for drop-out, with characterizations of learning dispositions, such as profiles of learning strategies, implies that the provision of learning feedback is not the end point, but can be extended to the design of learning interventions that address suboptimal learning dispositions. Building upon the case studies we developed in our previous research, we replicated the Dispositional Learning Analytics analyses in the most recent 17/18 cohort of students based on the learning processes of 1017 first-year students in a blended introductory quantitative course. We conclude that the outcomes of these analyses, such as boredom being an important learning emotion, planning and task management being crucial skills in the efficient use of digital learning tools, help both predict learning performance and design effective interventions.

KW - Blended learning

KW - Dispositional learning analytics

KW - E-tutorials

KW - Learning feedback

KW - Learning dispositions

KW - Learning strategies

M3 - Chapter

SN - 978-3-030-13742-7

VL - 158

T3 - Intelligent Systems Reference Library book series

SP - 69

EP - 89

BT - Machine Learning Paradigms

A2 - Virvou, Maria

A2 - Alepis, Efthimios

A2 - Tsihrintzis, George A.

A2 - Jain, Lakhmi C.

PB - Springer

CY - Cham, Switzerland

ER -

Tempelaar D, Nguyen Q, Rienties B. Learning Feedback Based on Dispositional Learning Analytics. In Virvou M, Alepis E, Tsihrintzis GA, Jain LC, editors, Machine Learning Paradigms: Advances in Learning Analytics. Vol. 158. Cham, Switzerland: Springer. 2020. p. 69-89. (Intelligent Systems Reference Library book series, Vol. 158).