Students’ sense-making of personalised feedback based on learning analytics

Abstract

Although technological advances have brought about new opportunities for scaling feedback to students, there remain challenges in how such feedback is presented and interpreted. There is a need to better understand how students make sense of such feedback to adapt self-regulated learning processes. This study examined students’ sense-making of learning analytics–based personalised feedback across four courses. Results from a combination of thematic analysis and epistemic network analysis show an association between student perceptions of their personalised feedback and how these map to subsequent self-described self-regulated learning processes. Most notably, the results indicate that personalised feedback, elaborated by personal messages from course instructors, helps students refine or strengthen important forethought processes of goal-setting, as well as to reduce procrastination. The results highlight the need for instructors to increase the dialogic element in personalised feedback in order to reduce defensive reactions from students who hold to their own learning strategies. This approach may prompt reflection on the suitability of students’ current learning strategies and achievement of associated learning goals.
 Implications for practice or policy:
 
 Personalised feedback based on learning analytics should be informed by an understanding of students’ self-regulated learning.
 Instructors implementing personalised feedback should align this closely with the course curriculum.
 Instructors implementing personalised feedback in their courses should consider the relational element of feedback by using a positive tone.
 Personalised feedback can be further enhanced by increasing the dialogic element and by including more information about learning strategies.

A Systematic Literature Review of Empirical Research on Epistemic Network Analysis in Education

“…The remaining articles ranged from a single coder to nine coders. The manual coding was supported with software tools in 19 articles (25%), for example, ncoder automatic coding [26], [45], [47], [49], [73], [82], MAXQDA 2018 [84], [90]- [92] and NVivo [67], [104], automated coding algorithm [48], [77], Python [44], and linguistic tools (LIWC and Coh-Metrix) [59].…”

“…Transcripts of recorded interviews were used in 16 articles (21.1%) to explore the epistemic frames in the community of practice [13], [60]- [62], games and assignment interviews [14], [20], [21], [63], [64], students' sense-making of feedback [65]- [67] and reading strategies [68], [69].…”

“…Different levels of experience of participants were compared in four studies [48], [61], [77], [100]. Finally, only three studies (3.9%) compared different courses [66], [67], [104] and two studies (2.6%) compared students' perceptions of feedback [44], and LA dashboards [65]. The coding of the selected articles was deductive, inductive, or a combination of both (Figure 6).…”

“…Other theoretical frameworks were reported in the reviewed studies, namely, knowledge building [83], the control-value theory of achievement emotion [66], and Language processing theory [47]. Finally, the achievement goal theory framework was reported [67].…”

Developing capabilities: Lifelong learning in the age of AI

“…Learning analytics (LA), an inter‐disciplinary domain of applied research, has evolved in the past ten years to utilise these data for informing, learning and teaching practices (Joksimović et al., 2019). Applications of LA include searching for insights to improve learning design (Macfadyen et al., 2020), providing feedback (Pardo et al., 2017), including feedback about self‐regulation practices inferred from behaviour traces (Lim et al., 2020) and recommendations of learning resources (Bodily & Verbert, 2017), among others. These developments form the unique offering for LLL, broadly conceived, as they enable feedback processes for individual and institutional insights around how learning can be better supported and improved.…”