How to use big data to improve student success after COVID-19 by Nazish Jamali

 Introduction

The COVID-19 pandemic starting in China had disturbed the educational system throughout the world. During this pandemic situation from kindergarten to 12th grade (K-12), students faced a lot of challenges. These challenges can be studied by using big data.

The term big data defines that the analyzing the collection of data that is enormous in size to uncover the secret patterns arithmetically is known as big data. By using big data educational institutes can run their operations and generate the quality for their programs. The data sets require innovative technological techniques for their efficient use. In this article, the author defines the techniques hat how to use big data to improve the success of K-12 students after the COVID-19 pandemic.

Big data is different from your regular data because it comes in large volumes, it comes from multiple places, and also it comes in a very speedy manner. Big data can shape the students' future. The root of big data in education comes from two different places.

The first one is the student information system (SIS), this system tells the academic background of students, and tells their status that where are they in the academic area, and also tells their performance along with the demographics. The second one is the learning management system (LMS), this system provides information about the behavior of students.

Big data has three different levels by using these levels with proper techniques we can improve the students from (K-12) success.

How to use big data to improve student success after COVID-19 by Nazish Jamali

1.     Micro-level big data

At this level, we can get data about the student’s interaction and their learning environment with the help of clickstream data. At this level, you will know about the student’s actions and how many times they click or preview a particular course or video, these actions are recorded. This is the students’ self-regulated learning from which we can understand how they are trying to learn so they can improve their success. This level is divided into six parts.

a.      Knowledge

At this level, we can identify how performance relates to complex cognitive skills. The automated detectors show how students interact and transfer knowledge from one domain to the next. With this data, instructors can analyze the success of students.

b.      Metacognition and self-regulation

This information is acquired from the learning management system (LMS) like Blackboard or moocs. In this part, we know about students and how they are managing their time, their tendency to procrastinate, and how they review and preview course materials.

c.       Identifying affective states

In this category, we get information that helps the instructors to understand non-cognitive constructs which are related to the student's engagement, motivation, and effects, for example during the COVID-19 how students engaged with online learning, how they feel, what was their interest and frustration. By analyzing these variables instructors can know the barriers and will try the improve the success of students after COVID-19.

d.      Knowledge inference

In this part, instructors identify and evaluate students’ knowledge or latent knowledge estimation. By using big data while the COVID-19 situation instructors easily examine how students try to analyze particular problems that were given online.

e.       Actionable knowledge

here administration closely monitors the students’ progress that how they became dis-engaged from online courses during COVID-19, and by using detectors instructors can find out how can improve students' engagement and motivation for their success.

f.       Cluster information

By using big data instructors can cluster information as they can get enormous data and try to personalize that data for the improvement of students’ success.

2.     Meso level big data or text data

At this level, we can get information about students through digital writing such as from discussion forums, online assignments, and social media interactions. Whatever the students have written and analyze that and make that into big data. By using linguistic tools instructors can analyze students understanding of topics through writing by online forums, LMS, or social media. By identifying students writing skills instructors will work hard for improving students’ success. This level is divided into four components:

a.      Cognitive

Assessing and evaluating students’ cognitive function, knowledge and skills from their writing level by using big data to work on their success.

b.      Social processes

We get information by critical assessment about social interaction between students and teachers from online discussion forums, social media, and videos.

c.       Behavioral

related to the student's course engagement and resource seeking behavior for example, students that view lecture videos sometimes pause sometimes they fast forward they rewind so we try to understand exactly what is this related to their behavior and how is this all tied to the way that they're understanding the course material

d.      Student’s self-concept or motivation

At this level, we know the students feel that how they are finding a course, and how they engage in learning activities, is the impact dropout rate in the course. This critical information gives us an idea about the learner's self-concept. By analyzing this data instructors can work for improving their success.

3.     Macro-level big data

This is known as institutional data, in this level we get information about students from the institutions such as admission data, class schedules, enrollments, and term grades. It is not often updated so, we can get information about students yearly or semester-wise. This gives administrators information that how to understand and deal with students.

In general, the three different ways of using big data to improve students’ success after COVID-19 are: accessing this data, analyzing this data, and using this data to make decisions. We can also face challenges while using these methods for studying the data. By following the above all steps the instructors can easily acquire information about the students' strengths and weaknesses, progress, grades, writing power, manners, behaviors, demographics, admission data, enrollments, dropout rate, past grades, social interactions, and cognitive power by using big data. Once the instructor collects the academic background information about students from the period of COVID-19. Then the instructor will analyze the collected information and he/she will know about the learner’s strengths and weaknesses so, he/she will work hard and implement new ways of teaching as students can get motivated in the teaching and learning process as it will improve student’s success after COVID-19 by using big data. Big data can utilize to construct a preemptive early warning system so this gives an idea if students are on the cusp of failing so, the instructor needs to intervene so they can understand exactly what the students are facing. This is necessary because it provides course guidance, we can look at past information to understand how students could improve for their success in the future.