Literature Review Learner Analytics and Visualization Techniques

CHAPTER TWO

Literature Review: Learner Analytics and Visualization Techniques

2.1 The Concept of learner analytics and Visualization Techniques

The concepts of learner analytics and visualization techniques have gained increased importance within the academic sphere. This segment of the paper reviews the two main concepts and underscores their importance in learner analytics. In recent year, study has begun to grow on the need for better measurement, tracking, and visualizations of information about learners. Research on learning analytics in many areas has developed so that it has been able to describe the number of activities to assist in the understanding and optimization of learning and the environments where learning occurs. It is necessary to organize information categorically so as to facilitate the human in directing information (Calude & Maurer, 2011). Visualizations that use different organizational views entailing the same information can be a remarkably effective aid.

When it comes to co-located learning in a learning environment like the classroom, the learners find it way easier to cooperate with their teachers effectively. This is not the case in a digital environment or in an online classroom whereby both learners and teachers are separated by a substantial space but joined through technology. Thus, it is harder for teachers to assess the progress of learners and also be aware of what the learners require. Teachers would also find it harder to develop effective communication with their learners since such a plan does not support physical feedback from the learners. The learners also have difficulties in ensuring that what they are assigned to do meet the requirements of the teachers. However, there has been progress when it comes to learner analytics. This has evolved and has been able to describe various elements for instance tracking, visualization and analysis of data that has been obtained from learners. This data has been effective in analysis the behaviour of students.

With the increase in availability and also access to information, developments of a wide variety of visualization and data retrieval tools have been developed. This growth has been beneficial in learner analytics, especially to learners. Such tools are operated by users to identify data and knowledge that is relevant to the task at hand. This has helped users in the process of assessing, analyzing and visualizing information presented to them. Furthermore, it enables the users to internally consider the possible relationships among various diverse groups of data mentally (Leung 2011, p. 51).Sense-making activities in many cases usually result to accumulation of large amount of data, knowledge and information. This information is represented in form of external visual representations such as diagrams, graphs, documents and sketches. (Lytras, 2010).

The concepts of visual analytics and data visualizations have not been clearly distinguished previously. This is as a result of varied information that details how visualization tests have a relationship with visual analytics. Main objectives of information visualization involved creating effective interaction techniques for a given class of data and method of producing view. Visual analytics is more than visualization. It is a fundamental way to integrate visualization, human factor and data analysis (Green, 2009). The problem arises when coming up with a solution is hard to achieve if the goal is to combine best –fit automated analysis algorithms and visual and interaction techniques. Visual analytics aims to give higher priority to data analytics from the start and through all interactions.

The learning from users’ behaviour and effective use of visualizations should play a pivotal role in the analytical process. Learning analytics in many cases seeks to advance brilliant combination of analytical approaches and advanced visualization techniques, which may play a crucial role when it comes to semantic analysis. In cases that the information is semantically rich, there is an increased chance that the information could be visualized in a variety of ways or levels. This is usually the choice of the visualization developer who decided upon how they would need the information to be presented (Olej, Obrsalova, & Krupka, 2011).

2.2Learner Analytics Tools and Resources

At this point, it is certain that learner analytics tools play instrumental roles in relative procedures. This section reviews the various learner analytics tools and resources that are currently in the market. Studies conducted have been able to show that visualizations enable understanding and realization of patterns. There are several tools relevant to learner analytics. One such tool is SAM (Student Activity Meter) (Lytras, 2010). This tool is used to visualize the time spent on learning activities and material utilized in learning environments that are online. Under SAM, several types of visualizations exist which promote collaboration and understanding among learners. An example of such visualizations includes systems that are utilized in order to increase the awareness of resources that are used. These can be through the use of a time-line that provides a chronological account on the use of resource (Lytras, 2010). There is also second level that functions to classify data into indicators of a higher level. An example words counts are used to detail participation rates and are also related to a model.

SAM is tasked with the role of visualizing and analysis of activities that go on and resource are utilized. SAM utilizes the first as it shows the activities that learners attempt in a given period of time. It applies the second characteristics by use of basic statistics resulting from time utilization. There also other systems which visualize learner analytics that are used by SAM. CAMera is another such example used in visualizing the activities of a user and shows clear metric events. It utilizes the CAMera schema that entails capturing the interactions of a user through the use of tools and resources. SAM also uses computer based data even though in this case it focuses on higher-level indicators (Zimmermann & Cunningham, 2001). SAM as a tool is used not only by the teachers but by the learners in their individual learning environments. Teacher objectives are usually contained under SAM. Under SAM, the following objectives for teachers are supported.

Knowledge and understanding of the learning progress is one of the teacher’s objectives. This is a situation whereby the teachers are aware of what and how learners are doing so as to assess their progress. SAM provides visual overviews for the time that learner spend and the resources that these learners use .Comprehensive analysis of virtual learning cannot be achieved due to lack of face-to-face communication (Shrum & Glisan, 2009).Thus, relative visualizations provide reliable indicators for awareness. The visualizations can also be applied by teachers to find patterns and be able to identify potential problems. Information on learner time tracking is also used which allows teachers to assess their initial stage estimates. This allows them to determine how time is spent by students who take part in various activities. Such information is also used in statistical researches regarding the process of learning. From this, popular learning materials are determined which lead to resource discovery.

Learner goals are also targeted under SAM objectives. This involves self-monitoring whereby self-reflection and understanding is realized. Furthermore, time tracking can help the learner understand his/her time allocation in comparison to their peers and support occasionally to reveal how much time is spent to the teacher. One of the objectives of SAM is to visualize when, on which resources and for what period of time students have been working when compared to their peers. Thus, when it comes to such goals, particular emphasis is placed on resource proposal that provides resourceful learning material used by peers. This is very advantageous in learning that is self-regulated (Pozzi & Persico, 2010).

Back office software is also used in learner analytics. The creation of this tool shows that different interfaces and clients could be created in the process. It is used in collaboration with Moodle. Clients can interact with Moodle, increasing markedly in capacities, from being a monolithic platform to an interoperable application. Moodle is a web services layer that consists of a set of contracts that make use of certain functions defined in Moodle external libraries (Ewall, 2007). It is able to provide data on how learners are progressing which is quite helpful to teachers. The data is presented in a table for clarity in addition to use of bars for comparisons. One main characteristics of Moodle is that it is more advanced when compared to other tools.

Since the wide usage of the web and other technologies, the manner of teaching and learning has been changing. This development is no longer just due to technological changes that support new models of learning but also to new motivations, trends and learning models. Thus, when talking about learning analytics tools and resources, constant development must be considered. The idea for the basis of learner analytic tools use is that they can interact with various systems in the same instance, performing the same actions in several places (Shrun & Glisan, 2009).

There are various changes that take place in learning process. As a result, technology is tasked with the responsibility of providing solutions to these changes, for instance coming up with new tools used for learner analytics. Thus, application of external tools is easier to implement which interact with the LMS. It becomes easier to improve the functions of tools such as back office which lead to flexibility. Allowing opportunity of learning to the use of technology is encourage since it is advantageous in developing new ways of learning (O’Neil, 2008).

Single management systems can include course administration. This part provides the main features associated with courses such as creating, modifying, deleting and viewing important course details. The aim is to facilitate the users’ administration without accessing the platform. (Khosrowpou, 2006). User administration is also a control system within the back office tool. The client permits total control of Moodle users through an easy and intuitive interface. Client management is the part of the tool. It allows the choosing of protocols that will be used to connect with the platform within the tool. There is also the log administration that controls the activity in the Moodle, so that they view the logs happening in a course, a date or made by a user. This way, both teachers and students can be able to track their progress as the course continues. In addition to this, roles can be created, modified or deleted during the process (MacArthur, Graham, & Fitzgerald, 2008).

Jigsaw has also been identified as a learner analytical tool. It utilizes analytical systems that are visual to as to support investigative research. Jigsaw main objective is to maximize pixel use to take advantage of both the user’s high acuity central primary point and extensive peripheral field. Its two main goals are to move quickly through large document collections, permit investigators to operate efficiently and support hypothesis formation together with collection of information. This enhances credible decision making especially based on the defined hypotheses. In many case, it works with large collections of text documents or other reports and with the entities, which have been obtained from them. Jigsaw utilizes several windows effectively with representations carefully designed for investigative problems that are complicated.

The user is thought to be in an ‘information cockpit’ with multiple monitors located in front of and above the user. However, although Jigsaw has some linking and brushing to integrate the windows, it does not have the stable interaction Wire Vis employs (Richards & Lassonde, 2011). Human Capital Management rules put expectation Jigsaw users in a way that they would be less in motion but rather in need of cognitive effort. This is not the case in WireVis whereby cognitive effort would not be required for instance in window management connection. This is certainly an issue worthy of additional review and evaluation. Jigsaw is operational and straightforward in nature. The model provides a point of view for investigating these goals in that light. The interface of this tool permits direction integration with involves representations of reports and entities, changing details and focus. As with Wire Vis and other tools that have been described, simplicity and intuitiveness also seek to attain cognitive goals.

Finally, when it comes to making comparisons, Jigsaw uses a unique approach, employing an increasing, question-based planning to present in various aspects of information that are used for investigation and possible relationships, as compared with Wire Vis top-down visualization of the entire data set and its context. Undoubtedly, both approaches are reasonable and could be present in a general tool for complex problem solving, and will be subject to future study. Nonetheless, the Jigsaw tool would be an effective tool for learner analytics in diverse situations (Ewall, 2007).

2.3 Effectiveness of Learner Analytics

Learner analytics has been instrumental in assessing the learning process in different ways. Findings have provided a basement upon which objective decisions regarding viable improvements have been made. Successful assessment can be attributed to relative tools and resources. This section of the paper details the effectiveness of learner analytics as well as the tools that are used in this process. SAM was applied in an extensive online free course learning analytics so as to get feedback on the effectiveness of the method. A multi-method strategy is usually followed using a structured questionnaire, review of documents, case studies and semi-structured focus group discussions (Rogers, 2002).

In organizational and academic research, SAM should be introduced in a manner appropriate to the skill level and professional knowledge base. There should be a concern with learner satisfaction, learning outcomes and experiences by the learner for instance, the nature of interactions. These strategies would allow determining the effectiveness of the design, growth and establishment of learner analytics. Attention should also be placed on the importance of collection of information that may be utilized by the teacher or instructor (O’Neil, 2008). Any obstacles or barriers are also realized during this stage. From this, adjustments and correction can be made to ensure success in the end. Formative assessment is also done as to gather information during the early stages so as to determine if the efforts produce the intended outcome.

The learning and knowledge analytics (LAK) online course was organized in order to evaluate setups and demographics involved. Moodle was used extensively for communication and collaboration in academic and organizational learning. Evaluating this tool in this varied viewing elicited dynamic discussions. These discussions arose as a result of the outcomes that were expected and the methods to be used. The client activities of the Moodle system were visualized in SAM. Registered participants were up to 270 for the course and were primary researchers engaged in the learning analytic field and teachers who are concerned in learning analytics.

An online survey was also used and it was composed of two paths. The main objective of the case study was to get more details on the SAM in a comprehensive course and the perceived usefulness of SAM by learning analytics experts. Two dozen individuals, between the ages of 27 and 62 years old took part in the survey. A dozen of them are teaching courses, and the rest have been involved in teaching courses for more than 10 years. This was a complete test that would be able to inform the research being conducted and provide conclusive results.

Learning and Knowledge Analytics (LAK) teachers recognize the teaching subject as slightly different. Provision of providing feedback to students becomes the main concern. The LAK teachers are majorly interested in finding students who are doing as expected of them. The idea of locating the best student becomes less importance when weighted against Agricultural researchers. Usage of the documents is a perquisite by LAK and other Agricultural researchers. Knowing how and when online tools have been utilized and knowing if sources that are independent are used is rated a bit lower. Therefore, they are also concerned with document application within Moode (Smith & Sadler-Smith, 2006). Student application is also rated high. Collaboration and communication is also more important for LAK teachers. The time tracking issue is almost rated equal in both circumstances. Comparing with the goals set by the teachers the recognition and the support is also the most vital within this tool.

As far as the LAK teachers are concerned it is the duty of SAM to deal with provision of teacher feedback during learning to students. This is related to the capability of visual analytics. The time spending issue in many instances is less prioritized. The open questions, which detailed how to implement each of the visualization, provided practical insights on the use of SAM. For instance, one teacher would use the line chart in order to identify the likelihood and intensity of participation. Furthermore, another teacher can learn about sequential time that a course takes place. The classroom activity status in this case is verified by the line chart. Thus, the teacher would expect it to be amended if few students take part and few share little, while a large number is at the centre. Determining the effectiveness of SAM would be able to underscore the variables that impact on the learning program (Zimmermann & Cunningham, 2001).

When it comes to most of the other issues pertaining to the process of learning, the teachers cannot come up with a decision that they agree on as a whole. The second part of the survey was based upon both the teachers and the learners. When queried to assess the contributions of each visualization, inconclusive answers were produced. There is no statistical evidence that shows that both learners and the teachers would rate the visualizations differently (Cress, Dimirova, & Specht, 2009).

The learner analytics is used by the student for comparisons with peers. Three learners use the corresponding coordinates for comparison with the rest of the class for self-reflection so as to measure progress and growth motivation. In this scenario, a single person did not understand how the corresponding coordinate worked. This individual preferred the bar graph that showed the group he fitted in most. The bar graph indicated the rate of growth motivation and progress. It was based on the individual liking, which was not the intention of the study. The bar chart was perceived as redundant by one teacher, which arises as a result of the addition of the histograms within the parallel coordinates.

The recommendations to improve learner analytics in both academic and organizational learning have certainly proven to be useful. Eight of those participating wanted to continue using SAM in the field while four were not sure of their final decision. An open debate about how each individual liked SAM was asked. From the results, three mentioned the simplicity and the quantity they can see using the tool. Two of them enjoyed the rapid application at which the tool could be used. Two participants also liked the precise, accurate information based on circumstances that the tool provided. Furthermore, they also enjoyed the insightful outcomes of the tool.

Technology, as opposed to individual user simulation should be utilized when it comes to automation of the measurement process. This would lead to the reduction of the time to be used during the learning analytics process. The best practices in such cases is to establish a common set of relevant key performance indicators that are monitored and also measured on a regular basis for the learning organization. Current technology and also establish methods of data collection instruments should be utilized in order to produce results that are more detailed. For instance, it may involve collection of data from learners and teachers two to three months post-training. In the end, the data collected will be more effective

References

Claude, C., Maurer, II. (2011). Rainbow of computer science: dedicated to Hermann Maurer on

the occasion of his birthday. Berlin: Springer.

Ewall, K. (2007). A visual analytics environment for collaborative sense-making, Information

Visualization. Houston.

Green. T. M. (2009). Building and applying a human cognition model for visual analytics,

Information Visualization. Chicago: Oxford.

Khosrowpour, M. (2006). Emerging trends and challenges in information technology

management. IIershey: Idea Group.

Leung, H. (2011). Advances in web-based learning – ICWL 2011. Berlin: Springer.

Lytras, M.D (2010). Technology enhanced learning : quality of teaching and educational reform: First International Conference, TECH-EDUCATION 2010, Athens, Greece, May 19-21. Berlin: Springer.

MacArthur, C. A., Graham, S., & Fitzgerald, J. (2008). Handbook of writing research. New-York : Guilford.

Olej, V., Obrsalova, I., & Krupka, J. (2011). Environmental modeling for sustainable regional

development : systems approaches and advanced methods. Hershey: Information

Science Reference.

O’Neil, H. F. (2008). What works in distance learning : sample lesson based on guidelines.

Charlotte : IAP.

Pozzi, F., & Persico, D. (2010). Techniques for fostering collaboration in online learning

communities : theoretical and practical perspectives.IIershey : Information Science

Reference.

Richards, J. C, & Lassonde, C. A. (2011). Writing Strategies for All Primary Students :

Scaffolding Independent Writing with Differentiated Mini-Lessons, Grades K-3. San

Francisco: Jossey-Bass.

Rogers, P. L. (2002). Designing instruction for technology-enhanced learning. IIershey: IRM

Press .

Shrum, J. L., & Glisan, L. W. (2009). Teacher’s handbook : contextualized language instruction.

Boston: Heinle Cengage.

Smith, P. J., & Sadler-Smith, E. (2006). Learning in organizations : complexities and diversities.

London: Routledge.

Zimmennann, W., & Cunningham, S. (2001). Visualization in teaching and learning

mathematics : a project. Washington DC: Mathematical Association of America.