Is it a good move? Mining effective tutoring strategies from human–human tutorial dialogues

Lin, Jionghao; Singh, Shaveen; Sha, Lele; Tan, Wei; Lang, David; Gašević, Dragan; Chen, Guanliang

doi:10.1016/j.future.2021.09.001

Cited by 20 publications

(33 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Compared to the students who had made some prior progress on the problem before the start of a tutoring session, tutors were inclined to show more polite expressions in open questions but less polite in corrective feedback (i.e., negative feedback and positive feedback) to the students without prior progress. Lastly, by incorporating the politeness levels of instructional strategies with other documented factors (e.g., sentiment, time on task, and task complexity) [21], the GTB model achieved better performance of predicting student problem-solving performance compared to our previous work [21].…”

Section: Introductionmentioning

confidence: 82%

“…Additionally, we observed that more than 47% of the students made some progress on the problem before joining a tutoring session Table 2. The description of the instruction strategies derived from Brummernhenrich and Jucks [8] and Lin et al [21] Instructional Strategy in [8] Instructional Strategy in Our Work [21] 1 showed that tutors and students invested more effort in the Gap-bridged sessions than the other two. An explanation for this might be the more efforts tutors and students invested in tutorial sessions, the better problem-solving performance achieved by student.…”

Section: Descriptive Statistics Of the Datasetmentioning

confidence: 89%

“…We characterised instructional strategies by capturing the actions (i.e., dialogue acts) hidden behind tutors' utterances; this was informed by prior research [27,35,37]. To automatically identify the dialogue acts, our previous work [21] applied the BERT language model [1] to train a dialogue act classifier, which was built on 3,629 utterances (training, validation, and testing datasets in the ratio of 80%:10%:10%) from tutors and students. It should be noted that the dialogue acts classifier was trained on the same dataset as the current study.…”

Section: Inferring the Instructional Strategies From Instructional Co...mentioning

confidence: 99%

“…To infer the politeness levels of instructional strategies, we first used the dialogue act classifier developed in our previous work [21] to identify the hidden instructional strategies from tutors' utterances. Then, we used the politeness level scoring tool by [29] to identify the politeness levels of the same tutors' utterances.…”

Section: Inferring the Politeness From Instructional Communicationmentioning

confidence: 99%

“…In contrast, the predictors in GTB were constructed sequentially so that the prediction errors made by previous predictors can be corrected by the subsequent predictors. We selected GTB model to answer research RQ2 due to its accuracy in our previous study [21] in predicting student performance on the same dataset.…”

Section: Predictionmentioning

confidence: 99%

See 4 more Smart Citations

Exploring the Politeness of Instructional Strategies from Human-Human Online Tutoring Dialogues

Lin¹,

Raković

Lang³

et al. 2022

Preprint

Self Cite

View full text Add to dashboard Cite

Existing research indicates that students prefer to work with tutors who express politely in online human-human tutoring, but excessive polite expressions might lower tutoring efficacy. However, there is a shortage of understanding about the use of politeness in online tutoring and the extent to which the politeness of instructional strategies can contribute to students' achievement. To address these gaps, we conducted a study on a large-scale dataset (5,165 students and 116 qualified tutors in 18,203 online tutoring sessions) of both effective and ineffective human-human online tutorial dialogues. The study made use of a well-known dialogue act coding scheme to identify instructional strategies, relied on the linguistic politeness theory to analyse the politeness levels of the tutors' instructional strategies, and utilised Gradient Tree Boosting to evaluate the predictive power of these politeness levels in revealing students' problem-solving performance. The results demonstrated that human tutors used both polite and non-polite expressions in the instructional strategies. Tutors were inclined to express politely in the strategy of providing positive feedback but less politely while providing negative feedback and asking questions to evaluate students' understanding. Compared to the students with prior progress, tutors provided more polite open questions to the students without prior progress but less polite corrective feedback. Importantly, we showed that, compared to previous research, the accuracy of predicting student problem-solving performance can be improved by incorporating politeness levels of instructional strategies with other documented predictors (e.g., the sentiment of the utterances).

show abstract

Section: Introductionmentioning

confidence: 82%