Generalizing Detection of Gaming the System Across a Tutoring Curriculum

Baker, Ryan S.; Corbett, Albert T.; Koedinger, Kenneth R.; Roll, Ido

doi:10.1007/11774303_40

Cited by 28 publications

(12 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Some of the EDM gaming-related work has focused on building machine detectors for gaming identification (Baker et al 2004a(Baker et al , 2006bWalonoski and Heffernan 2006a;Baker et al 2008a). For instance, Baker et al (2004a) used a density estimator with a set of Latent Response Models to predict how often a student would game the system.…”

Section: Edm Work On Detecting and Understanding The Causes Of Gamingmentioning

confidence: 99%

“…A name given to shallow reasoning in the context of an Intelligent Tutoring System (ITS) is gaming: "attempting to succeed by exploiting properties of the system rather than by learning the material" (Baker et al 2009). Not surprisingly, gaming is associated with reduced learning (Baker et al 2004b), and so there have been efforts to detect gaming (Baker et al 2006b(Baker et al , 2008a, understand gaming (Baker et al 2008b;Rodrigo et al 2008;Baker et al 2009) and reduce gaming (Murray and VanLehn 2005;Walonoski and Heffernan 2006b). While progress has been made on each front, challenges still remain.…”

Section: Introductionmentioning

confidence: 97%

See 1 more Smart Citation

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

Müldner

Burleson

Sande

et al. 2011

User Model User-Adap Inter

View full text Add to dashboard Cite

Students who exploit properties of an instructional system to make progress while avoiding learning are said to be "gaming" the system. In order to investigate what causes gaming and how it impacts students, we analyzed log data from two Intelligent Tutoring Systems (ITS). The primary analyses focused on six college physics classes using the Andes ITS for homework and test preparation, starting with the research question: What is a better predictor of gaming, problem or student? To address this question, we developed a computational gaming detector for automatically labeling the Andes data, and applied several data mining techniques, including machine learning of Bayesian network parameters. Contrary to some prior findings, the analyses indicated that student was a better predictor of gaming than problem. This result was surprising, so we tested and confirmed it with log data from a second ITS (the Algebra Cognitive Tutor) and population (high school students). Given that student was more predictive of gaming than problem, subsequent analyses focused on how students gamed and in turn benefited (or not) from instructional features of the environment, as well as how gaming in general influenced problem solving and learning outcomes.

show abstract

Section: Edm Work On Detecting and Understanding The Causes Of Gamingmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 97%

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

Müldner

Burleson

Sande

et al. 2011

User Model User-Adap Inter

View full text Add to dashboard Cite

show abstract

“…Kümeleme analizinde verinin kaç farklı kümeye ayrılacağının belirlenmesi konusunda ise iki farklı yaklaşım bulunmaktadır. Bu sayı ön tanımlı olarak belirtilmektedir ya da çeşitli algoritmalar kullanılarak veriye en uygun küme sayısı bulunmaktadır (Baker, Corbett, Koedinger ve Roll, 2006;Bienkowski, Feng ve Means, 2012).…”

Section: Veri Analiziunclassified

Çevri̇mi̇çi̇ Öğrenme Ortamindaki̇ Benzer Öğrenci̇ Gruplarinin Kümeleme Yöntemi̇ İle Beli̇rlenmesi̇

Akçapınar¹,

Altun²,

Aşkar³

2016

Eğitim Teknolojisi Kuram Ve Uygulama

View full text Add to dashboard Cite

ÖzBu çalışmanın amacı, çevrimiçi öğrenme ortamında benzer davranış örüntüsü sergileyen farklı öğrenci gruplarının kümeleme yöntemi ile belirlenmesidir. Çalışma, Türkiye'de bir devlet üniversitesinde Bilgisayar ve Öğretim Teknolojileri Eğitimi Bölümü'nde Bilgisayar Donanımı dersine kayıtlı 76 üniversite ikinci sınıf öğrencisi ile yürütülmüştür. Öğrenciler 14 hafta süresince yüz yüze derslere ek olarak çevrimiçi öğrenme ortamında, yansıma yazma, tartışma, soru-cevap, kaynak takibi vb. aktiviteler gerçekleştirmişlerdir. Analizlerde kullanılan veriler bu ortamın veri tabanından elde edilmiştir. Veriler iki farklı kümeleme algoritması ile analiz edilmiş ve sonuçları karşılaştırılmıştır. Aynı zamanda elde edilen farklı öğrenci gruplarının akademik performansları incelenerek etkileşim düzeyi ile akademik performans arasındaki ilişki analiz edilmiştir. Araştırma sonuçları öğrencilerin çevrimiçi öğrenme ortamındaki davranış örüntülerine göre üç farklı kümeye ayrılabileceğini göstermiştir. Bu kümelerin isimlendirilmesi konusunda ise iki farklı yaklaşım izlenmiştir. Kümeler, her bir kümede yer alan öğrencilerin aktivite düzeylerine göre (Çok aktif, Aktif, Aktif olmayan) ve öğrencilerin akademik performanslarına göre (Yüksek öğrenenler, Orta öğrenenler, Düşük öğrenenler) tanımlanmıştır.Anahtar Kelimeler: Öğrenci profili belirleme; çevrimiçi öğrenme; eğitsel veri madenciliği; kümeleme. IDENTIFYING DIFFERENT STUDENT PROFILES IN AN ONLINE LEARNING ENVIRONMENT WITH CLUSTER ANALYSIS AbstractThe purpose of this study is to explore how to group students who exhibit similar behavior patterns in an online learning environment via clustering. Relationships between these clusters and students' academic performance in each cluster were examined, as well. The * Bu çalışma birinci yazarın doktora tezinden üretilmiştir. Gökhan Akçapınar, Arif Altun, Petek Aşkar EĞİTİM TEKNOLOJİSİ Kuram ve Uygulama 47 study group consisted of 76 university students studying at the Computer Education and Instructional Technology Department at a state university in Turkey. In addition to the face to face classes during the course of 14 weeks, the students conducted several activities (writing reflections, participating in discussions, following course resources, etc.) in the online learning environment. The results of the cluster analysis indicated that students were ideally divided into three different groups according to the activities carried out in the online learning environment. Regarding the labeling of these clusters, two different methods were used. Firstly, the finalized clusters were named according to the students' activity levels (Non-active, Active, Very active). Then, these clusters were named with regards to students' academic performance (Low learners, Moderate learners, High learners).

show abstract

“…Education-oriented research will focus on the potential benefits of applying a certain technique rather than on the technique itself. Research oriented towards education will heavily use the educational jargon and will discuss methodology mainly in the Methodology chapter (e.g., Baker, Corbett, Koedinger, & Roll, 2006;Cohen & Nachmias, 2006;Sassoon & Nachmias, 1999;Superby, Vandamme, & Meskens, 2006).…”

Section: Educational Web Miningmentioning

confidence: 99%

Learning about Online Learning Processes and Students' Motivation through Web Usage Mining

Hershkovitz¹,

Nachmias

2009

IJELL

View full text Add to dashboard Cite

This study illustrates the potential of applying Web usage mining -the analysis of Web log filesin educational research. It consists of two sub-studies and focuses on two types of analysis, both related to the whole learning process: investigating one learner's activity in order to learn about her or his learning process, and examining the activity of a large group of learners, in order to develop a log-based motivation measure. Subjects were 674 adults who used an online learning unit as part of their preparations for the Psychometric Academic Entrance Exam and whose log files were drawn. The first sub-study aimed to illustrate the knowledge about the online learner that can be extracted from log files, and this resulted in a list of computable, non computable, and higher-level learning variables. In the second sub-study, a log-based motivation measuring tool was developed on the basis of a theoretical framework, a mechanism for computing relevant learning variables, and a clustering of these variables into three groups (associated with the theoretical framework). A discussion of the results, in the context of educational Web mining, is provided.

show abstract

Generalizing Detection of Gaming the System Across a Tutoring Curriculum

Cited by 28 publications

References 9 publications

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

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

Çevri̇mi̇çi̇ Öğrenme Ortamindaki̇ Benzer Öğrenci̇ Gruplarinin Kümeleme Yöntemi̇ İle Beli̇rlenmesi̇

Learning about Online Learning Processes and Students' Motivation through Web Usage Mining

Contact Info

Product

Resources

About