An analysis of students’ gaming behaviors in an intelligent tutoring system: predictors and impacts

Müldner, Kasia; Burleson, Winslow; Sande, Brett van de; VanLehn, Kurt

doi:10.1007/s11257-010-9086-0

Cited by 67 publications

(47 citation statements)

References 39 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Interventions, such as the one described in Stott 9 , which include a component of software engagement, can be undermined by learners' attempts to make progress through the software without engaging in the intended cognitive activity -so-called 'gaming the system' 27 . Howland et al 28 provide three distinguishing criteria for software usage to promote meaningful learning: it should be used actively, intentionally and constructively.…”

Section: Conceptual and Theoretical Frameworkmentioning

confidence: 99%

influence of science reading comprehension on South African township learners’ learning of science

Stott¹,

Beelders²

2019

S. Afr. J. Sci.

View full text Add to dashboard Cite

The majority of South African township learners have poor reading comprehension skills, which is known to impact negatively on their understanding of content subjects such as science, although the extent of the impact is not fully understood. We explored this impact, as well as the extent to which reading comprehension accounted for the differential effectiveness observed for out-of-class, text-dependent science intervention programmes. Eye movement and mouse-click data were collected from 65 Grade-8 and Grade-9 township learners as they read texts and answered electronic quizzes about electric circuits and lightning on a computer fitted with eye-tracking hardware and software. These data were used to describe the learners' reading and question-answering patterns and derive a composite English for science and technology (EST) reading comprehension index for each learner. Correlations were sought between this index and the learners' Natural Sciences marks and the benefit gained from two previous out-of-school science intervention programmes. Most learners were able to engage meaningfully with a less text-rich, moderately familiar quiz, but there was a prevalence of reading avoidance, guessing and reliance on superficial text features to answer questions for a more text-rich, unfamiliar quiz. Moderate to strong correlations were found between the EST index and both Natural Sciences and intervention marks. The findings suggest that while a significant number of higher achieving township learners possess sufficient levels of EST reading comprehension skills to benefit from textbased interventions, the majority require help in developing EST reading comprehension skills to enhance the likelihood of the intervention's success. Significance:•From this study, we infer that a small group of South African township learners, identifiable by their relatively high Natural Sciences marks, are able to read English science texts with sufficient comprehension to be able to benefit from text-dependent interventions, including engagement with self-study interactive software. The majority, however, read such texts at the frustration level, making it unlikely for interventions to be effective if they rely on the learner being able to engage in independent reading.

show abstract

Section: Conceptual and Theoretical Frameworkmentioning

confidence: 99%

influence of science reading comprehension on South African township learners’ learning of science

Stott¹,

Beelders²

2019

S. Afr. J. Sci.

View full text Add to dashboard Cite

show abstract

“…The lesson here is that even if a model can easily be ported from one ITS to another, it might not work, so even after porting it, the model should be tested to measure how the new context and new data impacts the ported model. An example of how to measure a model generality can be found in Paquette et al (2015) where a model for gaming was tested across different ITS for the same domain or Muldner et al (2011) where a similar comparison was made to determine which was a better predictor for gaming behavior: the problem or the student.…”

Section: Context Sensitivity: a Case In Which Ported Models Mights Nomentioning

confidence: 99%

An Analysis of Student Model Portability

Aguirre

Uresti

Boulay

2016

Int J Artif Intell Educ

View full text Add to dashboard Cite

Sharing user information between systems is an area of interest for every field involving personalization. Recommender Systems are more advanced in this aspect than Intelligent Tutoring Systems (ITSs) and Intelligent Learning Environments (ILEs). A reason for this is that the user models of Intelligent Tutoring Systems and Intelligent Learning Environments, i.e. their student models, tend to be more heterogeneous and complex than traditional models used in Recommender Systems. To share and reuse student models we must first understand the restrictions for porting or reusing student models in new ITSs or ILEs. This paper proposes a classification of student models in terms of their portability. Portability is measured via each model's accessibility, complexity, architecture, popularity, and description. We use this classification to analyse and then grade student models that have been published in the AIED, EDM and ITS research communities in 2013 and 2014. The classification is intended to be used by researchers both as a methodology to measure the portability of a student model and as a guide to find existing reusable models.

show abstract

“…Unfortunately, students tended to ask for hints rapidly, ignoring them until Mr Davis indicated how to fix the model. This form of help abuse, sometimes called gaming the system, is quite common in other tutoring systems as well (Baker, Corbett, Koedinger, & Wagner, 2004;Muldner, Burleson, van de Sande, & VanLehn, 2011). This led Biswas et al to modify Mr Davis so that he gave hints on the process of model construction rather than the product, i.e.…”

Section: Tutoringmentioning

confidence: 99%

Model construction as a learning activity: a design space and review

VanLehn

2013

Interactive Learning Environments

Self Cite

View full text Add to dashboard Cite

Modeling is becoming increasingly important both as a way to learn science and mathematics, and as a useful cognitive skill. Although many learning activities qualify as "modeling", this article focuses on activities where (1) students construct a model rather than explore a given model, (2) the model is expressed in a formal language rather than drawings, physical objects or natural language texts and (3) the model's predictions are generated by executing it on a computer. Most research on such learning activities has focused on getting students to successfully construct models, which they find very difficult to do. In the hope that new research can find ways to remove this bottleneck, this article attempts to list all the major ideas that have appeared in the literature and might be useful to those developing new learning activities involving model construction. The ideas are organized into a design space with five dimensions: (1) modeling language types, (2) ways for describing the systems that students should model, (3) instructional objectives and their corresponding assessments, (4) common student difficulties and (5) types of scaffolding.

show abstract

An analysis of students’ gaming behaviors in an intelligent tutoring system: predictors and impacts

Cited by 67 publications

References 39 publications

influence of science reading comprehension on South African township learners’ learning of science

influence of science reading comprehension on South African township learners’ learning of science

An Analysis of Student Model Portability

Model construction as a learning activity: a design space and review

Contact Info

Product

Resources

About