What is Learning Analytics? Challenges and Types

In today's education, technology alone does not always contribute to immediate results for students or institutions. We need ways to evaluate education's efficacy in order to determine the efficacy of educational technologies. 

 

We will work to develop best practices for students, educators, and institutions to better understand how learning occurs. Learning analytics can help with both of these objectives.

 

Learning analytics will improve teaching by altering how we assist learners in their learning processes. Learning analytics applications are gradually being integrated into learning management systems in higher education. 

 

As a result, we must think about what Learning Analytics is and how we can contribute to the advancement of current learning technology work. While the field of learning analytics is very promising, it is still an unexplored term.

 

Also Read | Best Edtech Startups in the world

 

 

What is Learning Analytics?

 

Learning analytics is the measurement, collection, analysis, and reporting of data about learners and their contexts to understand and optimize learning and the environments in which it occurs. This general definition still holds true even as the field has grown. Learning analytics is both an academic field and a commercial market that has grown rapidly in the last decade. 

 

Learning Analytics, as a research and teaching field, combines learning (e.g. educational research, learning and assessment sciences, educational technology), analytics (e.g. statistics, visualization, computer/data sciences, AI), and Human-Centered Design (e.g. usability, participatory design, sociotechnical systems thinking).

 

How to use Learning Analytics

 

The EDUCAUSE Learning Initiative's Horizon Report: 2019 Higher Education Edition identifies learning analytics as one of the digital strategies and technologies expected to enter mainstream use in the near future. As a result, skilled data analysts are required to support these data-driven initiatives and ensure institutional success.

 

"The analytics programs that were once reserved for big businesses are now being widely used in higher education and K-12 institutions to measure student growth, inform curriculum decisions, and identify students at risk for failing a course or program.”

 

In addition to these practical applications, learning analytics is frequently used to:

 

• Measure key indicators of student performance

 

• Encourage student growth

 

• Recognize and improve the efficacy of teaching practices.

 

• Inform institutional strategy and decisions.
   

In order to successfully leverage the insights gleaned by the effort, a learning analytics initiative should be accompanied by strong communication between the analyst and educators.

 

Also Read | Types of Data in Statistics

 

Levels of Learning Analytics

 

There are four levels of Learning Analytics - Measurement, evaluation, advanced evaluation, and predictive and prescriptive analytics. Although each of these levels is correctly referred to as "analytics," they each mean something very different in terms of complexity, difficulty, and power.

 

1. Measurement

 

Analytics begins with measurement, or simply tracking things and recording values to tell us what happened. Measurement does not require complicated math or statistics, but it must start with data collection. Otherwise, no analytics can be performed.

 

2. Data Analysis

 

Once the data has been collected, it is time to begin evaluating it and determine whether the data indicates something positive or negative. At this level, we're using high-school math to aggregate the data and establish benchmarks (averages, means, modes, and basic statistics).

 

Most analytics today fall into the basic data evaluation category, which is fine. There's a lot of value here, as well as opportunities for big wins.

 

3. Advanced Assessment

 

As we progress through the advanced evaluation and apply college-level math, exciting things begin to happen. We're looking at correlations and regression analysis in this section.

 

We're using statistical techniques to figure out not only what happened, but also why it happened. Advanced evaluation generates causation theories, allowing us to focus on what works best and discard ineffective learning.

 

4. Analytics that Predicts and Prescribes

 

Predictive and prescriptive analytics are the most sophisticated levels of analytics, requiring graduate-level math and frequently relying on AI or machine learning-powered by large data sets. 

 

According to predictive analytics, "here's what's most likely to happen next based on what's happened in the past." Prescriptive analytics goes a step further and says, "Here's the action we should take to optimize the outcome based on what's most likely to happen next."

 

Finally, we rely on highly intelligent recommendation engines to deliver just the right learning, at just the right time, and in just the right way to significantly improve performance. We're not there yet as an industry, but we can get there if we start measuring and working our way up.

 

Also Read | Applications of Statistical Techniques

 

 

Challenges of Learning Analytics

 

Learning analytics, like any new technology, introduces new challenges that must be overcome in order to maximize its potential.

 

1. Recognizing the Issue

 

First and foremost, you must comprehend the problem that you are attempting to solve with learning analytics. Even the most advanced learning analytic program will not be able to assist you unless you know what you want to do with it.

 

Of course, there are limitations to what the data can help you with—not all learning takes place in a digital environment. You will be better able to solve the problem if you know which data will help you understand it.

 

2. Recognizing what needs to be built

 

Learning analytics, as a relatively new field, will necessitate organizations' being active builders of their learning analytics programs. Each company will have a unique digital environment and set of needs, necessitating a unique solution. 

 

Many trials and evaluations may be required in this area to adjust the program as it evolves. Content created by third-party vendors should adhere to the same learning data guidelines.

 

3. Recognize who needs Analytics

 

Many questions must be addressed as an organization develops its program:

 

Who are they creating this program for? Will this be used solely for new employee training? Is it for lifelong learning within the organization? Who will receive this information and act on it? Will there be dedicated roles within the company, or will this be a task for department heads and managers?

 

All of these answers will influence how the learning analytics program is developed and managed.

 

4. Setting aside enough time to create an effective program

 

It will not only take a long time to develop the first version of the learning analytics program, but an organization should also plan for multiple iterations of the program to be developed and implemented. As the program goes live, faults and flaws will be discovered that must be addressed. It is not a one-time task.

 

5. Dealing with massive amounts of Data

 

Data is available in a wide range of formats, types, and locations. Many organizations have found it difficult to develop a system capable of handling the demands of analyzing such a massive amount of disparate data. Performance issues may be prevalent, especially when a large number of learners are being tracked.

 

6. Developing a program that corresponds to Technical Capabilities

 

The field of learning analytics has opened many exciting doors. However, because it is a new field, there is much more speculation about what it can do than what it actually does.

 

Within an organization, there may be a belief that learning analytics can completely transform a training program or the way customer behavior is understood. While both of these scenarios are possible, it is also possible that there is a limit to what learning analytics can truly do, which can cause problems.

 

7. Security

 

Finally, security is a major issue in this field. Handling this volume of data will necessitate an equal level of security in both information storage and access.

 

An organization should take great care in creating an environment that ensures the safety and privacy of all those who use it. To comply with the EU GDPR and similar privacy laws, the organization should build a level of security that limits and separates rights based on roles and permissions.

Also Read | Data Democratization

 

 

Types of Learning Analytics

 

Below are types of learning analytics :

---

Types of Learning Analytics

---

 

1. Descriptive Analytics

 

Descriptive analytics will provide you with answers to what happened. A retailer, for example, will learn about the average monthly sales, while a healthcare provider will learn about the number of patients admitted in a week. 

 

Similarly, you can find the number of course enrollments, pass percentages, assessment scores, and so on in eLearning. Descriptive analytics gathers data from various sources to provide insights into past performance. This information can be used to make informed decisions about future training programs.

 

For example, if the data shows that dropout rates are increasing, you might improve the training content or switch to a more engaging learning strategy. These discoveries enable you to improve training programs and even eliminate courses that are a waste of the organization's resources.

 

2. Analytical Diagnostics

 

Diagnostic analytics can be used to drill down and ask why something occurred. To gain insights into a specific problem or opportunity, you can figure out the dependent elements as well as identify patterns.

 

For example, data from diagnostic analytics may show that a customer service eLearning course had low completion rates among senior executives while being effective for new hires.

 

Further investigation revealed that the course content was too basic for the senior executives, implying that the organization should roll out an advanced level customer service course for them.

 

In some ways, the deeper analysis highlighted the need to cater to learners' specific needs and provide a more personalized learning experience. This would help to ensure that the training program is not redundant while also positively impacting the performance of all learners.

 

3. Predictive Analytics

 

Predictive analytics, as the name implies, predicts what is likely to occur. It forecasts the future based on existing data findings. However, keep in mind that predictions are only estimates, and their accuracy is highly dependent on the quality of the data and the stability of the associated situations. As a result, careful data analysis is required.

 

Predictive analytics can assist in identifying potential difficulties that learners may encounter during their learning experience. This enables L&D managers to develop opportunities for early intervention and targeted support. Furthermore, predictive analytics can be used to improve training quality and increase engagement.

 

For example, suppose data from a post-course survey revealed that some learners preferred not to access the eLearning program from a desktop. 

 

Because most of them are short on time and frequently on the go, they prefer to access the training on their mobile devices at any time and from any location. Learner profiles and predictive analytics can help you zero in on and offer microlearning solutions that meet individual needs in this case.

 

4. Prescriptive Analytics

 

The goal of prescriptive analytics is to find answers to what questions should be asked. Simply put, it should aid in understanding why something will happen in addition to finding solutions to what will happen.

 

Furthermore, prescriptive analytics can assist you in strategically planning training interventions. In this scenario, simulations can be delivered in stages to assist learners in applying their learning in a simulated environment. This, in turn, would increase the training program's impact and value. Again, data drives today's world.

 

Learning analytics provide decision-makers with more information about how corporate training programs align with organizational goals and individual learning needs. There is a huge opportunity for L&D leaders and their stakeholders to make data-driven decisions and, more importantly, to use that data.

 

Lastly, learning analytics enables data-driven decisions about what and how an organization's employees learn to be made. Knowledge and application gaps are identified and closed, assisting with long-term content retention and business change. 

 

The application of this information, then, is critical. Learning analytics has many applications, but it allows for targeted improvement across different employee populations.