초록 ▼

Ⅰ. 연구의 개요
본 연구는 현재 중등 단계의 온라인 교육을 이용하고 있는 학습자의 학습데이터 수집현황을 살펴보고 학습데이터의 관리 및 활용 체계를 마련하여 학습자 맞춤형 교육지원을 위한 방안 및 과제를 도출하는데 목적이 있다. 이를 위해 Ⅰ장에서는 학습분석과 맞춤형교육 관련 선행 연구 검토 및 문헌 분석을 통해 이론적 기초를 탐색하였고 현재 서비스중인 중등 단계 온라인 교육 현황을 살펴보았다. 연구 방향 설정 및 학습분석 사례 조사를 위해 영역별 전문가 자문단 구성을 통해 수시 전문가 협의회를 운영하였고 학습분석 기반교육지원

Ⅰ. 연구의 개요
본 연구는 현재 중등 단계의 온라인 교육을 이용하고 있는 학습자의 학습데이터 수집현황을 살펴보고 학습데이터의 관리 및 활용 체계를 마련하여 학습자 맞춤형 교육지원을 위한 방안 및 과제를 도출하는데 목적이 있다. 이를 위해 Ⅰ장에서는 학습분석과 맞춤형교육 관련 선행 연구 검토 및 문헌 분석을 통해 이론적 기초를 탐색하였고 현재 서비스중인 중등 단계 온라인 교육 현황을 살펴보았다. 연구 방향 설정 및 학습분석 사례 조사를 위해 영역별 전문가 자문단 구성을 통해 수시 전문가 협의회를 운영하였고 학습분석 기반교육지원 서비스에 대한 학습자, 교사, 운영기관 관계자, 교육부 및 시도교육청의 요구를 파악하고자 초점집단면접(FGI)을 실시하였다. 더불어 중등 단계 온라인 교육의 한계 및 문제점 진단과 학습분석 기반 맞춤형 교육지원 방안 및 정책과제 도출을 위해 전문가 델파이 조사 및 정책토론회를 실시하였다. 이를 통해 학습분석 기반 맞춤형 교육지원 방안을 크게 학습 데이터 수집 활용을 위한 법 제도 개선 방안, 학습분석 기반 맞춤형 교육지원 서비스 방안, 학습분석을 위한 학습 데이터 수집 개선 방안, 학습분석 기반 맞춤형 교육지원을 위한 콘텐츠 개선 방안, 학습분석 기반 맞춤형 교육지원을 위한 시스템개선 방안의 5가지로 도출하고자 하였다.

(출처 : . 연구의 개요 6p)

Abstract ▼

The purpose of this study is to explore the current practice of data collection regarding students that use secondary online education and to develop a system of data management and utilization for the purpose of deriving plans and tasks for supporting personalized learning.

Chapter One explo

The purpose of this study is to explore the current practice of data collection regarding students that use secondary online education and to develop a system of data management and utilization for the purpose of deriving plans and tasks for supporting personalized learning.

Chapter One explores and analyzes extant literature on customized education to develop a theoretical framework as well as to assess current online secondary education practices. A number of expert conferences consisted of area-specific expert advisory groups were convened with a view to setting a holistic andappropriate research direction and gathering information on suitable cases oflearning analysis. In addition, a focus group interview was conducted to identify the divergent needs of learners, teachers, operators, and the Ministry of Education and the Provincial Office of Education in regard to learning analysis education support services. Further, expert delphi surveys and policy debates were conducted in order to identify issues in secondary-level online education and to devise customized education support plans as well as policy tasks based on learninganalytics. Five action items that were derived from the various methodologies are 1) improvement of the legal system to enable utilization of data on student learning, 2) using data to develop customized training support services, 3)improvement of data collection practice, 4) enhancement of contents to supportcustomized training as informed by learning analytics, and 5) using learninganalytics to better personalize education support system.

Chapter 2 analyzes literature on current uses of learning analytics and personalized education, laws and systems related to the collection and use of personal information big data, and examples of application of learning analytics.
Following implications are discussed based on this study’s findings. First, as learning analytics encompasses evaluation and instant feedback of the learning process, prediction and prescription of learning performance, discovery of optimal learning paths and improved efficiency, optimization of learning environments, and support level-specific decision making, it must inform and present specific plans, strategies and tasks to achieve its objectives. Second, learning analytics must take comprehensive and holistic approach across macro (national and regional),meso (institutions) and micro (student) levels. Third, the current practice of collecting data on online education based on learning data models, such as the Edu Graph model of IMS Global, should be assessed and enhanced. Fourth, latesttechnology should be utilized to enable efficient learning analysis. Further, the unit of analysis should be changed from ‘groups’ to ‘individuals’ in all phases ofteaching in order to utilize learning analytics for customized education support.
Fifth, efforts should be made to better align online secondary education and learning analytics platform with national government agenda.

As for legal codes and systems concerning learning analytics and related collection and utilization of personal information, first, the potential issues of legal and ethical problems that may arise in utilizing the learner’s personal data and the importance of training data scientists must be noted given that the learning analytics presupposes the collection of learning data. Second, the concept and scope of personal information should be redefined so that various log data accumulated in the system can be analyzed and utilized to enhance the process of individualized learning. At the same time, the data sovereignty should also be strengthened so that students as owners of their personal information can maketheir own decisions about whether and how their data are utilized for their benefit. Third, a comprehensive system must be devised to enable thecombination, integration, and streamlining of personal information and data.

We studied various cases of learning analytics application in overseas K-12 and higher-level education and drew implications for appropriate objectives andmethodologies. First, learning analytics should be used to improve academic performance of students with poor academic record due to the lack of academic opportunities and to prevent them from leaving school system. Second, learning analytics can help understand patterns in learners’ learning resource utilization, learning behavior and characteristics, and their relationship with academic outcomes. These information in turn can be utilized to enhance the effectiveness of the curriculum and teaching methods. Third, teachers, by utilizing learning analytics, can better plan and establish personalized study plans and even receivedsuitable support by accumulating and analyzing data on students’ learningexperiences, learning patterns, interests, and career planning. Fourth, learninganalytics must contribute to better outcomes of education policy and provideevidence for redesigning classroom learning.

Chapter 3 surveys the current secondary-level online education with a view to identifying methodologies and potential projects for personalized education support. To that end, the background and purpose, operating system and subjects,content and systems, collection and utilization of learning data of learning analytics are examined. We found that online secondary education for each policy program has various backgrounds and purposes to meet the educational needs ofregular, out-of-school, and adult students. Open Secondary Schools aim at providing educational opportunities for under-educated adult learners while online classes were offered for students that attend regular schools with limited choice of subjects. Online joint curriculum provides real-time video classes andjoint courses for small-scale and advanced subjects whereas remote class systems enabled learning for the physically handicapped students or those with chronic illnesses. There also are e-schools that compensate for student athletes’ loss of class and provide learning support for out-of-school adolescents and preschoolersfor primary and secondary education diplomas.

The overall online education apparatus revolves around the Korean Education Development Institute in charge of running the overall secondary online education while the Ministry of Education and the Provincial Office of Education providepolicy-making and budget support services to schools and educational institutions,teachers and students. These parties play varying roles depending on the natureof involved policy projects.

Secondary-level online education contents are mostly those of Open Secondary Schools and are widely in use. Various elective learning and career related contents are also widely utilized depending on the needs of certain policy programs. Different online learning management systems - Learning Management System (LMS), Leaning Content Management System (LCMS), item management systems, evaluation management system, video class platform – have beendeployed to assist phase-specific needs.

The current data collection and utilization practice were classified according to the typology the Edu Graph model of IMS Global adopts and consequently were examined in four areas: learning content, learning activities, operations, learnerprofiles and career data. Our findings indicate that learning data collected fromeach online education program are mostly basic log data and different programsuse different variable names and structure.

Implications derived from the analysis of the current secondary online education practices are as follows. First, online educational support needs to be personalized depending on the needs and goals of each user. Second, reducinginefficiency that result from running separate operating systems requires fullintegration of the online operating systems. Third, the scope and role of utilizingeducational contents such as open market must be taken into account in designingpersonalized online education. Fourth, a radical shift in the perception of typesof educational contents is called for so that various approaches, such as videolectures, digital textbooks, and live video classes, can become integral parts. Fifth,new indicators of learning analytics other than basic log information must be defined in order to ensure extraction and utilization of meaningful data. Finally,creative technological integration and application are needed to enable automated learning analytics.

Chapter 4 of this report summarizes the findings from focus group interviews(FGI) that were conducted to identify the expected benefits of and specificdemands for customized online education. The interviews were also intended toprovide insights into how the 6 online-based secondary education services beingpromoted by the Digital Education Research Center of Korea Education Development Institute are perceived. In addition, the results of the expert delphi survey are summarized for the purpose of devising practical plans and tasks for implementing online education services customization based on learning analytics.
The 20 subjects of the FGIs include officials from the Ministry of Education and the Provincial Office of Education, as well as staff, teachers, and students from the institutions that use online secondary education services. The expert delphi survey was conducted in two stages with a total of 20 experts from relevantinstitutions, academia and industry. Our findings of the FGIs and expert delphi survey are as follows.

First, a shift in overall perception toward online education services is necessary.
The emphasis must be placed on achieving practical educational effectiveness rather than the current role of supplementing regular school education or expanding educational opportunities. To this end, new education contents and methods must facilitate teacher-student as well as student-student interaction and encompass measures of evaluation as appropriate for each curriculum.

Second, from the legal and institutional standpoint, relevant legal codes and systems must be swiftly amended to keep pace with changes in the online education environment. An active and open vision to provide students with practical help must be combined with strict enforcement of privacy protection andcopyrights in development and utilization of online education content. Prevention of personal data breach requires higher level of ethics regarding personal information among policy stakeholders as well as more stringent legal andinstitutional regulations.

Third, the diagnosis of academic capability and learning patterns must be made at the level of individual students rather than groups (e.g., Open Secondary Schoolstudents, students with health issues, and student athletes, etc.). A motivatingenvironment that facilitates self-regulated learning will provide students withcareer exploration and education.

Fourth, diverse and effective utilization of online resources such as the development of modular content will necessitate utilizing of data on learners’ use of content. Modular content will need to reflect information about the hierarchyand relationships of content as well as students’ usage (e.g., indicating segmentsmost repeatedly played by the top 1% of students).

Fifth, exerts are needed for effective use of the learning analytics. Systematic collection, management and utilization of both structured and unstructured datawill be required in order to derive meaningful analytical results from policy programs. In addition, Learning Management System (LMS) and learning analytics platforms must be standardized for interoperability.

Chapter 5 presents plans for learning analytics-based personalized online education support in five areas. The plans for each area are summarized below.

The first concerns modifications in the legal codes and systems for collecting and utilizing learning data. First of all, a shift is needed in the personalinformation and data utilization paradigm such that an emphasis is placed on “safeutilization” instead of “protection.” In this context, personal information shouldbe redefined based on identifiability along with a principle as to how to treat learning data as unidentifiable ‘behavioral information,’ separate from personal information. The most important thing in the collection and utilization of the learning data is to ensure that users of online education as owners of their information are informed and thus sufficiently aware of exactly what educational benefits can be gained from data analysis as well as potential side effects including possible ethical problems. Additionally, students must be allowed to opt out of data use at any time. In other words, students must be notified of learner profiling in advance and allowed to refuse profiling. To be more specific, we may considerintroducing MyData platform as a system that enables learners to control and utilize their personal information and data for their own learning and using it asa basis for the establishment of a broader learning data integration and utilization system.

The second plan is utilizing learning analytics for a personalized education support service. In this study, we identify five areas of online education where personalized training support services can be particularly useful: 1) identifying and providing support for students of potential academic underachievement, 2) preventing school drop-out, 3) tracking the level of participation in onlinelearning, 4) providing customized learning plans, and 5) measuring and comparing student performance. In addition, we propose detailed customized education support services for each online education policy program.

The third plan concerns how to improve the collection of data to allow for learning analytics. We analyzed the current practice of learning data collection per each online education policy program and indicated areas with a room forimprovement. Across all policy programs, collection and utilization of students’ online learning experience is essential, and a detailed plan for using learning analytics to provide personalized education support is presented for each policy program.

The fourth plan deals with improving educational contents based on the learning analytics to support customized education. First, in order to foster more interactive learning, we propose a new instructional design in the aspects of students-instructional material, student-instructor, student-student interactionfrom learning analytics point of view. In addition, we emphasize the importance of making accurate ‘diagnose’ based on learning analytics in personalizededucation support and along with it, present the need for development of evaluative indicators. Further, we propose plans to modularize educationalcontent in order to expand the quality of the contents to support customized education as well as plans to secure and develop educational contents to support general education.

Last plan concerns a system improvement to support customized education based on learning analytics. The plans for system improvement are presented separately for 2) information system and 2) teaching and learning support system.
First, the improvement plan as an information system addresses capacity expansion and changes in database structure to enable big data analysis, data structure for rapid big data analysis and timely service, analysis algorithms,computing performance, automated learning analytics through AI and machine learning, and building a privacy system. The plan for system improvement as ateaching and learning support system deals with expandability to accommodate a future-oriented learning environment, creating synergy between andco-evolution of learning analysis, learning management system and content management system, as well as the need for innovative improvements in visualizeduser interfaces including the design and development of dashboard.

Chapter 6 identify core principles of online education policy necessary for successful implementation of personalized education support through leveraginglearning analytics and propose an implementation road map. Policy suggestionsare largely divided into addressing limitations and problems of online education,creating an environment for application of the learning analysis in online education, and promoting continued research on learning analytics.

In terms of improving the limitations and problems of online education, we propose a policy shift to emphasize the quality of online education, a shift in theperception of educational content, and the integration of online educational operating systems. In terms of education quality, diverse instructional methods canbe integrated, and appropriate student evaluation measures for online education need to be devised in ways that value students’ actual academic achievements.
A shift in the perception toward online educational calls for a break away from negative stereotypes about traditional e-learning formats primarily in the form of video contents. That is, online education contents can be provide in many different forms including video-based teaching materials. The plan to integrateprogram-specific online education operating systems aim at providing moreinclusive online education by creating a more flexible, unified operating system.
In terms of creating an environment that allow for application of learning analytics in online education, we propose ways to ease restrictions on the collection and utilization of personal information, to create an integrated system that enable data linkage from a learning analytics perspective, to produce learning analytics experts, and to establish a system for creating and sharing sustainable educational content. Finally, to address the needs for continued research on learning analytics,we propose to establish a learning analytics system as informed by teaching andlearning theory, take a thorough inventory of currently disaggregated systemresources, and to conduct research to lead the trend of technological development in the area.

(출처 : Abstract 268p)

목차 Contents

• 표지 ... 1
• 머리말 ... 4
• 연구요약 ... 6
• 목차 ... 16
• 표목차 ... 18
• 그림목차 ... 19
• Ⅰ. 연구의 개요 ... 22
• 1. 연구의 필요성 및 목적 ... 24
• 2. 연구 내용 ... 30
• 3. 연구 방법 ... 32
• Ⅱ. 선행 연구 및 문헌 분석 ... 36
• 1. 학습분석과 맞춤형 교육 ... 38
• 2. 학습분석 동향과 개인정보·빅데이터 수집·활용 법제도 현황 ... 53
• 3. 학습분석 적용 사례 ... 61
• Ⅲ. 중등 단계 온라인 교육 현황 ... 86
• 1. 온라인 교육의 배경 및 목적 ... 89
• 2. 온라인 교육의 운영 체제 ... 91
• 3. 온라인 교육 콘텐츠 및 시스템 현황 ... 93
• 4. 학습 데이터 수집 및 활용 현황 ... 100
• 5. 소결: 함의점 ... 107
• Ⅳ. 학습분석 기반 맞춤형 교육지원을 위한 조사 분석 ... 110
• 1. 초점집단면접(FGI) ... 112
• 2. 전문가델파이 ... 160
• Ⅴ. 학습분석 기반 맞춤형 교육지원 방안 ... 184
• 1. 학습 데이터 수집·활용을 위한 법·제도 개선 방안 ... 187
• 2. 학습분석 기반 맞춤형 교육지원 서비스 방안 ... 193
• 3. 학습분석을 위한 학습 데이터 수집 개선 방안 ... 225
• 4. 학습분석 기반 맞춤형 교육지원을 위한 콘텐츠 개선 방안 ... 232
• 5. 학습분석 기반 맞춤형 교육지원을 위한 시스템 개선 방안 ... 238
• Ⅵ. 정책 제언 및 추진 로드맵 ... 244
• 1. 정책 제언 ... 246
• 2. 추진 로드맵 ... 259
• 참고문헌 ... 262
• Abstract ... 268
• 부록 ... 278
• 부록1. FGI 면담지 ... 280
• 부록2. 전문가델파이 조사지(1~2차) ... 290
• 부록3. 전문가델파이 1차 조사 결과(2차 조사 비연계 항목) ... 299
• 부록4. 전문가델파이 2차 조사결과(추가문항) ... 302
• 끝페이지 ... 315