Learning Analytics and their Application in Technology-enhanced Professional Learning
Learning Analytics and their Application in Technology-enhanced Professional Learning
复制标题
学习分析及其在技术增强的专业学习中的应用
DOI：
--
复制
发表时间：
2013
期刊：
影响因子：
--
通讯作者：
A. Margaryan
中科院分区：
文献类型：
--
作者：
Bettina Berendt;Riina Vuorikari;A. Littlejohn;A. Margaryan
学习分析（LA）是“测量，收集，分析和报告有关学习者及其背景的数据，以了解和优化学习及其发生的环境”（Siemens & Gasevic，2012）。最初，“分析”指的是一种使用数据来支持决策和理解领域的方法。LA的基本组成部分是（1）数据，（2）目标或（研究）问题，可选地基于教育理论，（3）提供关于目标实现或（研究）结构的信息的测量，可选地（4）使用这些值作为变量的描述性或预测性模型，以及（5）计算这些测量值的计算模型和例程，从给定的数据建模结果。LA系统还包括（6）向选定的利益相关者报告这些结果的自动或半自动方式。可选地，（7）结果可以部署在某些应用程序功能中。1）和3）的例子是用于测量学习者行为和知识的“点击流”数据，或支撑领域模型的文本数据。目标（2）可以是描述学习者之间的协作。描述性或预测性模型（4）可以包括用于预测学习者是否“处于辍学风险”的学习者简档或模型。确定这些度量（5）的计算模型的范围从通过聚类技术的简单计数到分类器学习。这些模型可能是纯统计的（将测量的变量关联起来），也可能是理论的（例如，解释为什么具有某种行为的人有辍学的风险，以及这种行为和风险与学习有什么关系）。（6）的典型选择是仪表板，当结果报告给教师或信息作为反馈给学习者时。（7）的一个例子是使用学习者模型为用户提供个性化的学习资源，这些资源被认为对个人的学习有用。有关组件的相关模型，请参见（Greller & Drachsler，2012）。
Learning analytics (LA) is the "measurement, collection, analysis and reporting of data about learners and their contexts, for purposes of understanding and optimising learning and the environments in which it occurs" (Siemens & Gasevic, 2012). Originally, “analytic” refers to a way of using data to support decision-making and understanding a domain. Essential LA components are (1) data, (2) goals or (research) questions, optionally based on educational theory, (3) measures that give information about goal attainment or (research) construct, optionally (4) descriptive or predictive models that use these values as variables, and (5) computing models and routines that compute these measures’ values, modelling results from the given data. LA systems also comprise (6) automatic or semi-automatic ways of reporting these results to the chosen stakeholders. Optionally, (7) the results can be deployed within some application functionality. Examples of 1) and 3) are ‘clickstream’ data used to measure learner behaviour and knowledge, or text data underpinning domain models. The goal (2) could be to depict collaboration between learners. The descriptive or predictive models (4) may comprise learner profiles or models for predicting whether a learner is ‘at risk of dropping out’. The computing models that determine these measures (5) range from simple counts via clustering techniques to classifier learning. The models may be purely statistical (correlating measured variables) or refer to theory (which would, for example, explain why someone with certain behaviour is at risk of dropping out, and what the behaviour and the risk have to do with learning). A typical choice for (6) is dashboards, when the results are reported to teachers or information is given as feedback to learners. An example of (7) is the use of learner models to offer users personalised learning resources that are assumed useful for an individual’s learning. For a related model of components, see (Greller & Drachsler, 2012).