Classification of learners in an adaptive learning management system

JOURNAL OF SCIENCE OF HNUE Interdisciplinary Science, 2013, Vol. 58, No. 5, pp. 159-165 This paper is available online at CLASSIFICATION OF LEARNERS IN AN ADAPTIVE LEARNING MANAGEMENT SYSTEM Nguyen Thi Huong Giang School of Engineering Pedagogy, Hanoi University of Science & Technology Abstract. In this paper, we show the significant benefits adopting the Adaptive Learning Management System (ALMS) to improve the teaching quality for online courses. We set up a problem to realize and orient students in the ALMS. To solve this problem, we first show the psychological fundamentals for classifying students. We then point out why the learning orientation model should be selected as a tool for identifying students who are taking online courses. Finally, we propose a means to applying this model to improve individualized learning in adaptive online courses. Keywords: E-learning, ALMS, classification of students, Learning Orientation Model (LOM). 1. Introduction Over the past ten years, the Adaptive Learning Management System (ALMS) has undergone significant changes. There are three different approaches to the design and construction of the ALMS: 1) develop a normal adaptive learning environment, 2) support solutions for delivering adaptive courses and 3) develop an adaptive learning environment based on standards. In this paper, we introduce how to deliver material to be learned according to student’s ability in ALMS. Our solution will improve the effectiveness of applying ICT in education and training in Vietnam since it will support some advantages of traditional teaching when providing online courses such as pedagogic communications and interactions. In traditional LMSs, there are some major activities: processing the user’s login/logout, inserting a new subject, updating a learning module package in a subject and creating and inserting a new learning module package. In order to developing ALMS, we propose to build a new function which takes into account the student’s ability: collecting Received February 02, 2013 Accepted May 29, 2013. Contact Nguyen Thi Huong Giang, e-mail address: giang.nguyenthihuong@mail.hust.edu.vn 159 Nguyen Thi Huong Giang information about a student’s ability, processing the collected information and making decisions on the learning content and pedagogic methods that are suitable for each student. In the scope of this paper, we propose using the learning orientation model as a tool for classifying online students. In the next section, we will introduce the psychological basic for classifying students. 2. Content 2.1. Models of classifying learners Classification of students in teaching and learning is very important for delivering subject material and learning strategies according to a student’s ability. Therefore, a high number of classification models which are now proposed. These improve the capability to improve the quality of education. In the scope of this paper, we analyze and assess some models in the following categories: the classifying model based on a student’s internal characteristics, the classifying model based on methods and strategies of learning, and the complex classifying model that will discover the most suitable model for identifying students in ALMS. In the first group, there are three models: Myers-Briggs, Gregorc and Herrmann. The Myers-Briggs model relates to personality types. The Gregorc mind style model is based on the two dimensions of dealing with preferences for perception and ordering. These can be combined into four basic mediation channels which lead to four types of students. The third is the Herrmann “whole brain” model. This model distinguishes between four modes which are based on the differences between the left and right cerebral hemispheres of the human brain. The second group of models relates to learning strategies. “Learning strategies can be seen as short term methods that students apply in a particular situation. These strategies can change with the time, teacher, subject, and situation. When learning strategies are frequently used by students, learning styles can be derived from these strategies” [4]. Therefore, learning strategies are important factors to identify individual students. These models include four major types: the Pask’s Serialist /Holist /Versatilist Model, the model of Entwistle’s Deep, the Surface and Strategic Learning Approach, the Kolb model which models the learning process and incorporates the important role of experience in this process, and the Honey and Mumford model. The third group of classification models are the complex models. These models consider the research results of previous classification models. This is very important for classifying students because both students and the learning process are influenced by many internal and external factors from the learning environment. In this group, we assess the model of Felder–Silverman and that of Margaret Martinez. Because there are many different models that can be used to classify students, it is very necessary to assess the contribution of the model for individualized learning 160 Classification of learners in an adaptive learning management system in online courses. This helps find the most suitable model for online training. Basic to this assessment are the properties of classifying students who are taking online courses. These properties are drawn from analyzing the necessary skills of online learners, the fundamentals of classification, the requirements of delivering learning objects, and the capability of supporting the automatic classification. The result of the assessment of the above classification models is presented in Table 1. (1) The necessary skills of online students According to Clark and Mayer online students need to have metacognitive skills. These are the ability to set learning goals, to determine how to reach their goals and to make adjustments where necessary. Students with poor metacognitive skills need more direction whereas students with good metacognitive skills tend to be more self-sufficient learners. This skill-set has been described elsewhere as qualities of a "self-directed learner" [5]. As a result, the classification model in online-courses should be related to the skills of self-directed learners. Table 1. Assessment of classification models for identifying online students No. Model of classifying learners Skills of “self-directed learners” Fundamentals of classification Delivering learning objects Supporting the automatic classification (1) (2) (3) (4) 1 Model of Myers-Briggs - Personality of learners - + 2 Model of Gregorc - Mind of learners - + 3 Model of Herrmann - Structure of human brain - + 4 Model of Pask - Learning methods + + 5 Model of Entwistle - Learning methods + + 6 Model of Kolb - Learning methods - + 7 Model of Honey - Mumford - Behaviour and thinking of learners - + 8 Model of Felder-Silverman - Combination of models of Kolb, Myers-Briggs, Pask,. . . + + 9 Model of Margaret Martinez + Combination of properties of learning orientation - + (2) The fundamentals of classification As mentioned above, the classification of students has been described in many different models. These models use different or similar fundamentals for classification. As 161 Nguyen Thi Huong Giang a result, plenty of classification models exist, each integrating some aspects of learning, and some overlapping each other. Therefore, the most suitable model to classify students is the model that combines and inherits from other models. (3) The requirements of delivering subject material Online students have a variety of learning levels so courses cannot be too strict about requiring a minimum learning level of online participants. Courses should be developed such that the desire of "anyone wishing to learn is met." However, if there is "no pupil illiteracy, there is no way to convey learning content to students.” And yet people at all levels can participate in online courses if they wish to and so online courses must have a strategy to provide content that is adapted to each student. Thus, the model of classifying learners must be associated with the effective delivery of subject material. (4) The capability of supporting automatic classification In online courses, the characteristics of students regarding their ability to use information technology, ability to self-study, physiological characteristics, and learning methods cannot be identified by the teacher. Online courses should have a reasonable way of obtaining this information as a basis for classifying students. Therefore, the classification tool in online learning courses must have automatic features that will substitute for teachers. As the result, the classification model needs to support the ability to make an automatic classification by the questionnaire or specific identifying characteristics. 2.2. Applying the “learning orientation model” to identify online students In recent research done on classifying students of online courses, many approaches on how to control the individual learning differences between groups of students are introduced. There are many classification models that are proposed as mentioned above. Based on the assessment of these models for identifying online students presented in Table 1, the most suitable model is the model of Margaret Martinez on learning orientation (LOM). According to Martinez [6], “this model focuses on the dominant factors that impact self-motivation, self-directedness, and learning autonomy. It is based on research into the neurobiology of learning and memory, and incorporates the dominant impact of emotions, intentions, and social factors, as well as cognitive issues. The model explores design of the online learning environment, online presentation of instruction, the role of the instructor, and expected outcomes. It also describes strategies to help learners improve online learning ability as they become more self-motivated, self-managed, independent learners”. Therefore, this model has more advantages for developing a tool of classifying students than other models (see Table 1) such as: supporting the discovery of the necessary skills of self-directed students, the ability of automatic classification and combining the fundamental factors to classifying students. However, the weak point of this model in individualized learning is that there is no description of the delivery of subject material 162 Classification of learners in an adaptive learning management system in online courses. Therefore, we propose that it is necessary to combine the theory of Don Clark on instruction presentation [7] into this model of Martinez. In the model of Martinez, students are divided into four groups based on their psychological properties in online courses: transforming students, performing students, conforming students and resistant students. The design for delivery of learning materials following Don Clark’s guide is introduced in four approaches: the first is applied by the delivery method of deductive-inquisition in which the contents are presented as general information and has students find and produce examples, the second group is applied by the delivery method of inductive-inquisition that presents examples and then has students abstract general information, the third group is applied by the delivery method of deductive-exposition that presents general information and then presents some examples and the last group is applied by the delivery method of inductive-exposition that presents examples and then presents general information. Applying the Learning Orientation Model in identifying students in an adaptive online-course, we propose a process of identification as seen in Figure 1. In this process, a classification of students happens in two modes: static and dynamic. The static classification is based on information obtained from a questionnaire completed by the student at the beginning of the course. This information helps ALMS identifies a suitable group of students and then delivers the appropriate subject materials to the student at the beginning of the course. However, during the learning process, the results of tests or the student’s activities can show that the student belongs to another learning group. Therefore, the dynamic classification is essential to deliver subject materials that suits the the student’s ability in the learning process. As a result, this process is an effective way to individualize learning in adaptive online courses. The delivery of subject material is divided into four paths, one going each learner group following the LOM classification. In each learning node (unit), a student is placed along a certain learning path. At the end of this learning path, the student must take the path determined in the output test. The result of the output test and the data that’s collected from all of the student’s activities in this node are stored in the system database and becomes dynamic information for the student’s next learning node. In summary, both the classification of students and the delivery of learning materials oriented to each group of students will support the fundamentals of building ALMS effectively and feasibly. 3. Conclusion In this paper, we have shown the application of a Learning Orientation Model (LOM) to classify students taking online courses. This includes the fundamentals of classification, the reason to select the LOM to identify students and the process of identification using LOM in an adaptive online course. Our study is an important step to develop ALMS as well as to support pedagogical designs for learning content in 163 Nguyen Thi Huong Giang Figure 1. The process of identifying students in an adaptive online-course (Using the Learning Orientation Model) the ALMS. Our result is fundamental to implement the software for the adaptive tool and complete the problem of developing ALMS. The success of building and running ALMS will promote the advantages taking of online courses and, step-by-step, upgrade the quality of online courses in Vietnam. REFERENCES [1] Peter Brusilovsky, John Eklund, Elmar Schwarz, 1998. Web based education for All: A tool for development adaptive courseware. [2] Norazah Yosof, Paridah Samsuri, 2002. Domain-Expert Repository Management for Adaptive. Hypermedia Learning System. 164 Classification of learners in an adaptive learning management system [3] Guzmán, E., Conejo, R., Pérez-de-la-Cruz, J.L., 2007. Adaptive testing for hierarchical student models. User Modeling and User-Adapted Interaction, Volume 17, Numbers 1-2 / March. [4] Harrer, A., McLaren B.M., Walker, E., Bollen, L., Sewall, J., 2006. Creating cognitive tutors for collaborative learning: steps toward realization. User Modeling and User-Adapted Interaction, Volume 16, Numbers 3-4 / September. [5] De Bra, P., Smits, D., Stash, N., 2006. Creating and Delivering Adaptive Courses with AHA!, In W. W. Nejdl and K. Tochtermnann (eds.). Innovative Approaches for Learning and Knowledge Sharing. LNCS 4227, pp 8-33. [6] Jesus G. Boticario, Olga C. Santos, 2007. An open IMS-based user modelling approach for developing adaptive learning management systems. Journal of Interactive Media in Education [7] Santos, O.C., Boticario, J.G., Barrera, C., 2005. Alfanet: an adaptive and standard-based learning environment built upon dotLRN and other open source developments. Calvo, RA; R.A. Ellis and D. Peters “Internationalisation and eLearning Systems: .LRN Case Studies”. Foro Hispano dotLRN – Online Educa. May 9-11, 2005. Madrid (Spain). [8] Prof.Dr. Nguyen Viet Huong, Pham Minh Viet, Nguyen Thi Huong Giang, 2007. Setting up the activities in an adaptive learning management system (ALMS), Proceedings, HS-IC. International Conference on Scientific Research in Open Universities, 4-6 November, Cat Ba Island, Vietnam. [9] Nguyen Thi Huong Giang, Nguyen Thi Viet Huong, 2010. Proposal of building adaptive learning management system in Vietnamese online courses. Journal of Science and Technology for Technical Universities, Vol. 75. 165