Application Of Predictive Analytics For Student Performance In Higher Education

During Covid-19 Pandemic

A Project Report Submitted to Goa University

In partial fulfillment of the Requirement

For the degree of

Bachelor of Computer Applications

BY

SIDDESH SAWANT (9339)

SANDNYA SHET NARVEKAR (9340)

YOGITA TIRODKAR (9341)

SNEHAL VARAK (9342)

DATTARAJ VELGUENKAR (9343)

SAYUKTA VELIP (9344)

AMUL BORKAR (8316)

Guided by

(Ms. NISHA SAWANT)

I
 SPES's SHRI GOPAL GAONKAR MEMORIAL GOA MULTI-FACULTY COLLEGE

Affiliated to Goa University

CERTIFICATE

This is to certify that the project on

“Application Of Predictive Analytics For Student Performance In Higher Education

During COVID-19 Pandemic”

has been successfully completed and submitted by

SIDDESH SAWANT (9339)

SANDNYA SHET NARVEKAR (9340)

YOGITA TIRODKAR (9341)

SNEHAL VARAK (9342)
DATTARAJ VELGUENKAR (9343)
SAYUKTA VELIP (9344)
AMUL BORKAR (8316)

_______________ _________________

Ms. Nisha Sawant Dr. Shaikh M. Parvez Al Usmani

(Internal Guide) (Principal)

_______________ _________________

Ms. Nisha Sawant

(Project Coordinator) (External Examiner)

II
SPES's SHRI GOPAL GAONKAR MEMORIAL GOA MULTI-FACULTY COLLEGE

Affiliated to Goa University

DECLARATION BY CANDIDATES

We declare that this project report has been prepared by us and to the best of our knowledge,

it has not previously formed the basis for the award of any diploma or degree by this or any

other University.

Roll No Seat No. Name Signature

9339 SIDDESH DASHARATH SAWANT

9340 SANDNYA AVINASH SHET NARVEKAR

9341 YOGITA SHEKHAR TIRODKAR

9342 SNEHAL BABU VARAK

9343 DATTARAJ DIGAMBARVELGUENKAR

9344 SAYUKTA NARSINHA VELIP

8316 AMUL DHIRAJ MAHLE BORKAR

III
 SPES’s SHRI GOPAL GAONKAR MEMORIAL GOA MULTI-FACULTY COLLEGE

Affiliated to Goa University

CERTIFICATE BY THE SUPERVISOR

This is to certify that the project report is the record of the whole work done by the
candidates themselves under my guidance during the period of study and that to the best of
my knowledge, it has not previously formed the basis for the award of any diploma or degree
by this on any other university.

Name of college: SPES's Goa Multi-Faculty College

Course: Bachelor of Computer Applications (B.C.A.)

Academic Year: 2021-2022

(Ms. Nisha Sawant)

Project Guide

IV
 ACKNOWLEDGEMENT

We take this opportunity to thank all those who have helped us to complete the project in this
stipulated period of time.
We would like to express our special thanks of gratitude to our guide Mrs. Nisha Sawant and
our principal, Dr. Shaikh Mohammad Parvez Al-Usmani, who gave us the golden opportunity
to do this project on the topic “Application of Predictive Analytics For Student Performance
in Higher Education During Covid19 Pandemic”. It helped us in doing a lot of Research and
we came to know about a lot of things related to this topic. Special thanks to Dr. Hemant
Palivela Sir and Mr. Sarvesh Mani Tiwari Sir for their valuable inputs for our project. We
also like to thank our classmates, Mr. Sanjay Mourya and Mr.Alison Dias for helping us in
completion of our project.
Lastly, we express our sincere gratitude towards the teaching and non-teaching staff of the
college for timely guidance and valuable inputs.

V
 TABLE OF CONTENT
1. Introduction………………………………………………………………………(1)
1.1 Covid-19 and Pandemic…………………………..(1)
1.2 Education and Digitalization Of Education……………..(2)
1.3 Teaching and Learning in the Classroom……………..(2)
1.4 Online Teaching…………………..(2)
1.5 What Are The Advantages and Disadvantages Of Online Learning?...........(5,6)
1.6 Benefits Of Online Exam ..................………………………………………..(8)
1.7 Impact on students' performance…………………………………..(9)
1.8 Impact of Covid-19 on school education in India……………………..(10)
1.9 What Is Predictive Analytics?……………..(12)
1.10 Benefits of Predictive Analytics……………..(13)
1.11 Abstract……………..(14)
1.12 Objectives……………..(14)
1.13 Hypothesis……………..(15)
2. Review Of Literature And Research Methodology……………………………….(15)
2.1 Review Literature…………………………………………………….(15)
2.2 Research Methodology………………………………………….(28)
2.3Data Preprocessing and cleaning …………………………………………...(34)
2.4Data analysis…………………………………………………………………..(34)
2.5Data analytics………………………………………………….……………(35)
3. Data Analysis and Analyticss……………….……………(36)
3.1 Data Analysis……………….……………(36)
3.2 Data Analytics……………….……………(44)
4. Conclusion..………………………………………………………(49)
4.1 Hypothesis testing…………………………………………………………...(49)
4.2 Limitations of study………………………………………….……………...(50)
4.3 Future scope………………………………………………(50)
5. REFERENCES……………………………………………………………………..(5l)
6.GANTT CHART………………………………………..(53)

7. ANNEXURE………………………………………..(54)

VI
 INTRODUCTION
Covid-19 and Pandemic
Corona viruses are a large group of viruses that cause illness in humans and animals. Rarely,
animal corona viruses can evolve and infect people and then spread between people such as
has been seen with MERS and SARS. The outbreak of Novel coronavirus disease (COVID-
19) was initially noticed in a seafood market in Wuhan city in Hubei Province of China in
mid-December, 2019, has now spread to 215 countries/territories/areas worldwide. WHO
(under International Health Regulations) has declared this outbreak as a “Public Health
Emergency of International Concern” (PHEIC) on 30th January 2020. WHO subsequently
declared COVID-19 a pandemic on 11thMarch, 2020.
On 11 January, WHO received further information from the Chinese National Health
Commission that the outbreak is associated with exposures in one seafood market in Wuhan,
and that the Chinese authorities had identified a new type of coronavirus, which was isolated
on 7 January.
During the early stages of the outbreak, the number of cases doubled approximately every
seven and a half days.
On 30 January, with 7,818 confirmed cases across 19 countries, the WHO declared the
COVID-19 outbreak a Public Health Emergency of International Concern (PHEIC) and then
a pandemic on 11 March 2020 as Italy, Iran, South Korea, and Japan reported increasing
numbers of cases.
In March 2020, a nationwide lockdown was imposed, forcing all schools and colleges to close
and instructing students to abandon the usual classroom teaching style. This resulted in an
immediate increase in innovation and technology, which the teachers used to finish the
curriculum. However, due to a lack of infrastructure, productivity suffered. Remote learning
was not available in every family; according to the Global internet network research, just
24% of households had a consistent internet connection, while other remote areas remained
unreachable. As a result, the government was forced to ensure that sufficient benefits and
monetary funds were provided.
The shift to online teaching raised important challenges for educational institutions for
organizing examinations and ensuring the possibility for students to progress in their studies.
Electronic examinations (E-exams) were introduced as an effective mode of assessment with

Page 1 of 63
particular importance to provide immediate exams feedback, which is considered challenging
for academic staff members due to the increase in students’ number. Currently, E-exams are
considered a key element in distance education.

Education and Digitalization Of Education

Digitalization in education refers to the use of desktop computers, mobile devices, the
Internet, software applications, and other types of digital technology to teach students of all
ages. Digitalization in education refers to the use of digital technology to teach students.
Teaching and Learning in the Classroom
Classroom teaching and learning is a complex, multilayered, and social experience. Many
things happen at the same time, and the students and teacher must interpret and process these
events with considerable rapidity. Students can easily solve their difficulties with the help
from teachers and friends as face-to-face communication is more effective.
Online Teaching
Online teaching is the process of educating others on virtual platforms. This type of teaching
involves live classes, video conferencing, webinars, and other online tools. The online
applications are developed and designed to facilitate easy learning and better understanding.
The advancement in technology has improved the quality of distance education and increased
the trend of e-learning. Online tutors and learners are becoming comfortable with computers
and the internet. It is helping them to create a better virtual learning environment.
Online teaching is a student-centric methodology that increases students’ interest and
participation levels in virtual classrooms. The teaching skills play a great role in successful
interaction with the students.
The online teachers use digital resources to share with their students and are accessible to so
many people. It provides a lot of freedom to individuals to learn, teach, and develop skills at
their own pace. This type of teaching supports one-on-one basis interaction with the students.
It makes them comfortable with tough subjects.
Teaching online requires strong network connectivity and computing device systems.
Nowadays, more mobile applications are developed as they are easily accessible by a large
population. The mobile versions of the live teaching app have helped in increasing the
attendance and registration of students.

Page 2 of 63
The Classroom vs Online Classes
There are more advantages of taking classes in the classroom compared to taking them
online; such as being able to physically meet with the teacher, asking for feedback and seeing
their reaction.
The traditional way of taking class, which is called the classroom environment, has become a
lot easier to manage time because you are going to class everyday if not a little less. Students
have more time to meet with their peers and teacher and more time for homework to be done
in the classroom with the help of peers and your teacher. Many students learn best though the
face-to-face interaction or the hands-on approach that is provided by teachers and peers while
attending classes in the classroom environment. Online classes might be effective for some
students however many students learn by doing, seeing, and being able to interact with the
teacher and their peers.
The social construction visit approach is based on the assumption that individuals learn to
construct their knowledge and meanings through interaction with others. It holds that
knowledge is not presented to the individuals, but emerges from active dialogue where people
create their own learning paths and knowledge.
With online courses colleges and universities haves made tremendous impact on the
instruction and student learning. Distance education opportunities have brought the classroom
from the university or college settings to the home, allowing students the privilege of
pursuing college degrees without the inconvenience of actually traveling to campus to take
the course.
Thousands of students are earning their degree with little or no traditional participation and
no interaction with the teacher or their peers. Students who learn by hands on or best thought
the face-to-face interaction lose all this when they take classes online since you can’t
physically meet with you teacher or peers. Some students also lack the technological skills
needed to take online classes which causes some students to struggle. Taking classes online is
a lot cheaper than taking them in the classroom because of the cost of gas to travel to classes
and other financial expenses like baby sister, food and drinks and book bag. Some students
have a hard time getting internet access and a computer since the cost of getting a computer is
so expense.

Page 3 of 63
Group assignments in the classroom environment have a more set schedule with more time
for the project to do in classroom. This is a lot easier, because you can only physically meet
with each other and be able to talk face to face with each other compared to trying to do the
group assignment online, which could be challenging because of scheduling conflicts.
The weekly schedule you have in the classroom environment all depends on the college you
attend and the teacher.
The traditional way of taking classroom classes lets you have a total time of at least 60 hours
of face to face as to where the 8 weeks allows for only 32 hours of class. There are set
schedules in the classroom environment, but if the teacher feels their students do not
understand the concept, they can change the schedule or adapt the teaching.
In the classroom environment the teacher can see who is putting the effort in the group
assignment and give the correct credit to the students who do the work and effort. This allows
for the students that don’t put in effort get the grade they deserve. Meeting your teacher and
peers face to face allows for better understanding of feedback, which you receive from your
teacher and peers on assignments and allows you to better improve your assignment before
the final copy has to be turn in to the teacher.
Group Assignments in online classes are a lot hard to plan and takes a lot of more time to
figure out when everyone can work on the project together or just to figure out on whom does
what part. When you have to email each other or using another program to communicate to
each other and figure out how to plan between everyone schedules.
Individual Assignments
In the traditional classroom environment, the individual assignments can sometimes be hard
and you need help from the teacher
Assessment of student’s learning and student’s activities has been said that it takes up half of
the teacher’s professional time in the classroom, which is one of the many job responsibilities
of a teacher.
Tests
Taking tests in the classroom environment sometimes become a little overwhelming but it’s
also the best way to take the test since you have the ability to ask the teacher questions about
the test and be able to physical meet with other students and the teacher.

Page 4 of 63
Many teachers let the students in the classroom review the day before the test in class and
also before taking the class to allow the students the best grade they can get from studying.
Online classes some teachers don’t require a test in their classes since it’s online and the
teacher knows the students will use their books. Instead, they have them do more papers or
homework assignments which shows the teacher they are doing their work. Online classes let
the students take responsibility for their own leaning more than traditional instructional
approaches but some students have a hard time learning if they are not being pushed to study
for the tests and do the homework.
There are more advantages of taking classes in the classroom compared to taking them
online; such as being able to physically meet with the teacher, asking for feedback and seeing
their reaction.
What Are The Advantages Of Online Learning?
1. Efficiency
Online learning offers teachers an efficient way to deliver lessons to students. Online learning
has a number of tools such as videos, PDFs, podcasts, and teachers can use all these tools as
part of their lesson plans. By extending the lesson plan beyond traditional textbooks to
include online resources, teachers are able to become more efficient educators.
2. Accessibility Of Time And Place
Another advantage of online education is that it allows students to attend classes from any
location of their choice. It also allows schools to reach out to a more extensive network of
students, instead of being restricted by geographical boundaries. Additionally, online lectures
can be recorded, archived, and shared for future reference. This allows students to access the
learning material at a time of their comfort.
Thus, online learning offers students the accessibility of time and place in education.
3. Affordability
Another advantage of online learning is reduced financial costs. Online education is far more
affordable as compared to physical learning. This is because online learning eliminates the
cost points of student transportation, student meals, and most importantly, real estate.
Additionally, all the course or study materials are available online, thus creating a paperless
learning environment which is more affordable, while also being beneficial to the
environment.

Page 5 of 63
4. Improved Student Attendance
Since online classes can be taken from home or location of choice, there are fewer chances of
students missing out on lessons.
5. Suits A Variety Of Learning Styles
Every student has a different learning journey and a different learning style. Some students
are visual learners, while some students prefer to learn through audio. Similarly, some
students thrive in the classroom, and other students are solo learners who get distracted by
large groups.
The online learning system, with its range of options and resources, can be personalized in
many ways. It is the best way to create a perfect learning environment suited to the needs of
each student.
What Are The Disadvantages Of Online Learning?
1. Inability To Focus On Screens
For many students, one of the biggest challenges of online learning is the struggle with
focusing on the screen for long periods of time. With online learning, there is also a greater
chance for students to be easily distracted by social media or other sites. Therefore, it is
imperative for the teachers to keep their online classes crisp, engaging, and interactive to help
students stay focused on the lesson.
2. Technology Issues
Another key challenge of online classes is internet connectivity. While internet penetration
has grown in leaps and bounds over the past few years, in smaller cities and towns, a
consistent connection with decent speed is a problem. Without a consistent internet
connection for students or teachers, there can be a lack of continuity in learning for the child.
This is detrimental to the education process.
3. Sense Of Isolation
Students can learn a lot from being in the company of their peers. However, in an online
class, there are minimal physical interactions between students and teachers. This often
results in a sense of isolation for the students. In this situation, it is imperative that the school
allow for other forms of communication between the students, peers, and teachers. This can
include online messages, emails and video conferencing that will allow for face-to-face
interaction and reduce the sense of isolation.

Page 6 of 63
4. Teacher Training
Online learning requires teachers to have a basic understanding of using digital forms of
learning. However, this is not the case always. Very often, teachers have a very basic
understanding of technology. Sometimes, they don’t even have the necessary resources and
tools to conducts online classes.
To combat this, it is important for schools to invest in training teachers with the latest
technology updates so that they can conduct their online classes seamlessly.
5. Manage Screen Time
Many parents are concerned about the health hazards of having their children spend so many
hours staring at a screen. This increase in screen time is one of the biggest concerns and
disadvantages of online learning. Sometimes students also develop bad posture and other
physical problems due to staying hunched in front of a screen.
A good solution to this would be to give the students plenty of breaks from the screen to
refresh their mind and their body.
Benefits Of Online Exam
Exam Candidates are used to Digital.
Exam candidates can work on a computer in a way they are familiar with, rather than using
pen and paper. Candidates are at a disadvantage if they do not have access to computer
editing tools when writing e.g. cut and pasting text, or using a spreadsheet for accounts. It is
hard to switch to pen and paper when you are used to working digitally, and it can be very
tiring over long periods.
2. Reduced Administrative Burden.
Organisations greatly reduce the administrative burden of organising and running exams
when it moves online. Printing and circulating exams on paper and organising shipments of
completed scripts to markers is a time-consuming and costly process.
3. Quicker to Mark and Issue Results.
It is much quicker to mark online and candidates welcome receiving results quickly. With
auto-scorable questions, results are available immediately if wished.
4. Collaborative Question Authoring.

Page 7 of 63
Working online makes it easier for the examining body to manage all the tasks involved in
creating exams. For example, authoring questions can be done collaboratively, with clear
workflows for review and approval of questions before they are added to the question bank.
5. Automated Test Assembly Tools.
Once there is an approved bank of questions, exam papers can be created easily by selecting
questions or by using automated tools e.g. LOFT, or linear-on-the-fly testing, to create
automatic and randomized papers. You can read more about linear-on-the-fly testing here.
6. On-Screen Marking Tools.
Marking and moderating exam results is streamlined, secure and standardized when it is
managed online. Read more about on-screen marking here.
7. It’s More Environmentally Friendly.
It’s more environmentally friendly – with less paper, printing and transport used overall.
8. Scalable with Worldwide Reach
With online assessment, examining bodies can offer exams to candidates located over greater
geographical areas, as they are not restricted to offering exams in physical test centers.
9. Increased Security.
Another benefit of online assessment is greater security, as all exam papers, candidate details,
marks and results are digitally stored. In addition, only the people who should be able access
exam information can be granted the appropriate system access. It is much harder to control
access, storage and results management with paper-based exams.
10. Flexibility to Take Exams Anywhere
Remote invigilation, or online proctoring, gives candidates the ability to sit a secure and
invigilated exam from the comfort of their home, which is less stressful, and saves time and
money going to a test center. Read more about the benefits of online proctoring here.
11. Assessment Reporting.
Online assessment facilitates quick and clear reports on candidate results and progress. This
makes it easier to give useful feedback to candidates on how they are doing, areas where they
are strong and what areas of learning require attention.
12. Cost Effectiveness.

Page 8 of 63
Online assessment is more cost effective, due the vast reduction in administrative time
managing the whole exam creation, delivery and marking processes. You can read more
about other cost-efficiencies below.
13. Assessments that are Accessible to All
When you run assessments online, it is easier to adapt those assessments for candidates who
have disabilities, such as changing font size, background color etc. It is also easier to add
extra time for candidates who need special considerations. To learn more, visit this article on
accessibility in assessment.
Impact on students' performance
The COVID-19 pandemic has introduced uncertainty into major aspects of national and
global society, including for schools. For example, there is uncertainty about how school
closures last spring impacted student achievement, as well as how the rapid conversion of
most instruction to an online platform this academic year will continue to affect achievement.
Without data on how the virus impacts student learning, making informed decisions about
whether and when to return to in-person instruction remains difficult. Even now, education
leaders must grapple with seemingly impossible choices that balance health risks associated
with in-person learning against the educational needs of children, which may be better served
when kids are in their physical schools.
Amidst all this uncertainty, there is growing consensus that school closures in spring 2020
likely had negative effects on student learning. Based on historical learning trends and prior
research on how out-of-school-time affects learning, it was estimated that students would
potentially begin fall 2020 with roughly 70% of the learning gains in reading relative to a
typical school year. In mathematics, students were predicted to show even smaller learning
gains from the previous year, returning with less than 50% of typical gains. While these and
other similar forecasts presented a grim portrait of the challenges facing students and
educators this fall, they were nonetheless projections. The question remained: What would
learning trends in actual data from the 2020-21 school year really look like?
Impact of Covid-19 on school education in India
No one would have guessed that a virus-like Covid-19 would come and without
differentiating, it will alter the lifestyle of people. Due to Covid-19, many changes came to

Page 9 of 63
our world and it took some time for everyone to adopt the new normal. The Covid-19 impact
was everywhere, which resulted in the closure of Schools and other educational institutions.
Initially, most governments have decided to temporarily close the schools to reduce the
impact of Covid-19. Later it was reopened for a few grades, which increased the number of
infection rates and then closed again.
Though schools are closed, students are attending their classes through various education
initiatives like online classrooms, radio programs. Though it is a good thing happening on the
other side, there are lots of students who didn’t own the resources to attend the online classes
suffer a lot. Many students are struggling to obtain the gadgets required for online classes.
Teachers who are all experts in Blackboard, Chalk, books, and classroom teaching are really
new to this digital teaching, but they are adopting the new methods and handling it like a pro
to aid the students in the current position.
But on the negative side, many teachers are looking for an alternative job to support their
families.
Educated parents are supporting their children throughout the pandemic, but we require to
understand that there are some illiterate parents and their feeling of helplessness to help their
children in their education.
There are students in India who came to school just because they can get food. The great
midday meal scheme has helped many children who couldn’t bring their food from the home
to get their nutrition.
Because of the closure of the schools, many students were suffering from not having enough
food for their survival.
There is always a delay or cancellation of exams, which leads to confusion for many students
and there is no room for curriculum.
Most of the school-going children are involved in child labor to support their families.
There is a lot of chance that the education of female children and transgender children will
affect, as their parents may see, the financial and opportunity costs of doing so.
This pandemic has not only affected the students but also the Low-budget institutions and
schools, resulting in close-down the same.
There are both positive and negative matters happening around us amid the Covid-19.

Page 10 of 63
Technology paves the way for education, thus helping the students and teachers to connect
virtually through online classrooms, webinars, digital exams, and so on.
But the sad truth is that it is not available to many students all over the nation.
Everything is happening for the well-being of the students so that they can stay safe at home
without getting affected by the life-threatening virus.
We are not ready for this, but it came, so we have to go through this together, but we have to
update the infrastructure and should think of ways to undertake the situation and providing
education to every child amid the pandemic if we face something precisely like this in the
future to aid the forthcoming generation. Stay home. Stay safe.

What Is Predictive Analytics?

The term predictive analytics refers to the use of statistics and modeling techniques to make
predictions about future outcomes and performance. Predictive analytics looks at current and
historical data patterns to determine if those patterns are likely to emerge again. This allows
businesses and investors to adjust where they use their resources to take advantage of possible
future events. Predictive analysis can also be used to improve operational efficiencies and
reduce risk.

KEY TAKEAWAYS
Predictive analytics uses statistics and modeling techniques to determine future performance.
Industries and disciplines, such as insurance and marketing, use predictive techniques to
make important decisions.
Predictive models help make weather forecasts, develop video games, translate voice-to-text
messages, customer service decisions, and develop investment portfolios.
People often confuse predictive analytics with machine learning even though the two are
different disciplines.
Types of predictive models include decision trees, regression, and neural networks.

Uses of Predictive Analytics

Predictive analytics is a decision-making tool in a variety of industries.
Forecasting

Page 11 of 63
Forecasting is essential in manufacturing because it ensures the optimal utilization of
resources in a supply chain. Critical spokes of the supply chain wheel, whether it is inventory
management or the shop floor, require accurate forecasts for functioning.
Predictive modeling is often used to clean and optimize the quality of data used for such
forecasts. Modeling ensures that more data can be ingested by the system, including from
customer-facing operations, to ensure a more accurate forecast.
Credit
Credit scoring makes extensive use of predictive analytics. When a consumer or business
applies for credit, data on the applicant's credit history and the credit record of borrowers with
similar characteristics are used to predict the risk that the applicant might fail to perform on
any credit extended.
Underwriting
Data and predictive analytics play an important role in underwriting. Insurance companies
examine policy applicants to determine the likelihood of having to pay out for a future claim
based on the current risk pool of similar policyholders, as well as past events that have
resulted in payouts. Predictive models that consider characteristics in comparison to data
about past policyholders and claims are routinely used by actuaries.
Marketing
Individuals who work in this field look at how consumers have reacted to the overall
economy when planning on a new campaign. They can use these shifts in demographics to
determine if the current mix of products will entice consumers to make a purchase.
Active traders, meanwhile, look at a variety of metrics based on past events when deciding
whether to buy or sell a security. Moving averages, bands, and breakpoints are based on
historical data and are used to forecast future price movements.

Benefits of Predictive Analytics

There are numerous benefits to using predictive analysis. As mentioned above, using this type
of analysis can help entities when you need to make predictions about outcomes when there
are no other (and obvious) answers available.
Investors, financial professionals, and business leaders are able to use models to help reduce
risk. For instance, an investor and their advisor can use certain models to help craft an

Page 12 of 63
investment portfolio with minimal risk to the investor by taking certain factors into
consideration, such as age, capital, and goals.
There is a significant impact to cost reduction when models are used. Businesses can
determine the likelihood of success or failure of a product before it launches. Or they can set
aside capital for production improvements by using predictive techniques before the
manufacturing process begins.

Abstract
In an effort to mitigate the spread the of the Covid-19 pandemic, most educational institutions
around the world have been closed since March 2020. This has impacted more than 90% of
the world's student population. In this study we are going to predict whether the performance
of students during Covid-19 pandemic has improved or not.
We are going to split our data into training and testing dataset where 80% data is used in
training and 20% data is used for testing.
We are going to use six different algorithms to compute the accuracy of our predictions such
as RandomForestClassifier Algorithm, GaussianNB Algorithm, KNeighboursClassifier
Algorithm, LogisticRegression Algorithm, LinearDiscriminantAnalysis Algorithm,
DecisionTreeClassifier Algorithm.
We will also show the comparison of асcuraсу of each of these algorithms to find the most
accurate and the least accurate algorithm.

PROBLEM STATEMENT
To check whether students' performance in higher education has improved during covid
pandemic.

OBJECTIVES
1. The objective of this study is to analyze the changes in student’s performance after
COVID-19 outbreak.
2. To know the sources and reasons of stress during COVID-19 pandemic.
3. To reveal the various forms of online teaching learning modes.

Page 13 of 63
 4. To study the perceptions of teachers and students on online teaching learning during
covid-19 pandemic.
5. To examine the challenges faced by the teachers and students in adapting to the online
teaching –learning process during covid-19 pandemic.

HYPOTHESIS
1. During covid-19 pandemic performance of students in higher education has
improved.
2. Perception of teachers on online teaching learning is positive.
3. Perception of students on online teaching learning is positive.
4. Perception of teachers on online teaching learning is negative.
5. Perception of students on online teaching learning is negative.
6. Unstable network is the biggest hindrance in adapting to online teaching
learning process.

Page 14 of 63
REVIEW OF LITERATURE AND RESEARCH METHODOLOGY

LITERATURE REVIEW

1. Abdelsalam M.M., Ebitisam K.E., Shadi A., Hasan R. & Hadeel A. (2021). The Covid-
19 Pandemic And E-Learning: Challenges AND OPPORTUNITIES from The
Perspective Of Students And Instructors. Journal of Computing In Higher Education.
Doi.org/10.1007/s12528-021-09274-2
SUMMARY:-The spread of COVID-19 poses a threat to humanity, as this pandemic has
forced many global activities to close,including educational activities.To reduce the spread of
the virus, education institutions have been forced to switch to e-learning using available
educational platforms, despite the challenges facing this sudden transformation.The study
targets the society that includes students and teaching staff in the Information Technology
(IT) faculty at the University of Benghazi. The descriptive-analytical approach was applied
and the results were analyzed by statistical methods. Two types of questionnaires were
designed and distributed.,that is the student questionnaire and the instructor questionnaire.
Four dimensions have been highlighted to reach the expected results, i.e., the extent of using
e-learning during the COVID-19 pandemic, advantages, disadvantages and obstacles of
implementing E-learning in the IT faculty. The unexpected closure of educational institutions
as a result of the emergence of COVID-19 prompted the authorities to suggest adopting
alternatives to traditional learning methods in emergencies to ensure that students are not left
without studying and to prevent the epidemic from spreading. The formal learning system
with the help of electronic resources is known as e-learning. Whereas teaching can be inside
(or outside) the classrooms, the use of computer technology and the Internet is the main
component of e-learning In an educational environment, there are lots of learning-related
processes involved, and great amounts of potential rich data are generated in educational
institutions continuously in order to extract knowledge from those data for a better
understanding of learning-related E-learning is playing a vital role in the existing educational
setting, as it changes the entire education system and becomes one of the greatest preferred
topics for academics. Most students today want to study online and graduate from universities
and colleges around the world, but they cannot go anywhere because they reside in isolated

Page 15 of 63
places without good communication services. It also helped reduce the effort and travel
expenses and other expenses that accompany traditional learning.E-learning reduced
significantly the administrative effort, preparation and lectures recording, attendance, and
leaving classes. The student becomes a self-directed learner and learns simultaneously and
asynchronously at any time.

2.Dr. Wahab Ali (2020). Online and remote learning in higher education institutes: A
necessity in light of Covid-19 Pandemic. Higher Education Studies. Vol.10, No.3.
Summary: In the context of the COVID19 outbreak, governments and higher education
institutions around the world are launching various policy initiatives to continue teaching
activities to control the virus. However, has ambiguities and differences in what to teach, how
to teach, the workload of teachers and students, the teaching environment, and the impact on
educational equity. Large-scale national efforts to use technology to support distance
learning, distance education, and Online learning during the COVID19 pandemic are
emerging and developing rapidly. In the context of constantly emerging and changing
COVID19 , many countries are affected by COVID19. Similarly, many universities are
closed or about to close, and are moving towards online and distance learning. For example,
New York University in Shanghai and Duke Kunshan University provide examples of
successful adaptation and rapid deployment of educational technology products, such as the
Zoom and Moodle video conferencing platforms. During the protests and disagreements,
universities began to connect online in different ways, depending on their specific
institutional background. This means not only designing appropriate strategies for one's
specific environment, but also realizing that technical decisions will be formed in a way that
reflects existing differences, alliances, discourses, and opinions in one's specific institutions.
You may think it is the last thing, but when explaining the need to replace with a virtual
alternative to face-to-face teaching, it’s important to understand what information is being
sent. When protesting, the general term used has blended learning. The World Bank has noted
the fact that there are very few education systems, even the best performing education system
may not be, which can provide online learning for all students on such a large scale.
Technological progress often exceeds the ability of decision makers to keep up with costs and
infrastructure support. It should be determined to provide effective blended and online

Page 16 of 63
learning. In a previous study by the author, it revealed that approximately staff of higher
education institutions seem to be integrating ICT into their curricula, especially during
lectures and tutoring, but most of them still believes that trust is a factor that can be further
improved. This lack of trust may be due to the management of different learning tools and
platforms. There is no doubt that the enormous technological advancement in the world
requires a paradigm shift in the way we achieve educational goals and ambitions. Many
universities and educational institutions have adopted ICT tools, such as laptops, projectors,
tablets, smartphones, iPad, and interactive whiteboards. Moreover, The COVID19 pandemic
and social distancing requirements posed an undue challenge for the stakeholders connecting
to the internet because they must work with limited time and resources. You must determine
that adopting an online learning environment is not just a technical matter. This is a teaching
and teaching challenge. Therefore, proper preparation of textbooks and courses, as well as
Assessment knowledge is essential for online education. Technology is a delivery medium
that requires close cross-cooperation between teaching, content, and technical teams. Closing
the university and removing students and teachers from the classroom is an educational
change that requires the rapid mobilization of all university staff and resources.

3.Edy Budiman. (2020). Mobile Data Usage On Online Learning During Covid-19
Pandemic In Higher Education. iJIM. Vol. 14. No. 19.
SUMMARY:-The coronavirus caused more people to work or learn from home. The
unpreparedness of educational institutions in implementing online learning was the main
factor in this chaos. This study measures the amount of internet data usage in the Zoom
application as a communication video for online learning during the Covid-19 pandemic.It
aims to determine how much a student uses data and its correlation toward meeting duration
and feature usage.
The research involved 80 students.The results explained that there was a correlation between
the meeting duration with the feature usage toward the amount of mobile data usage. This
shows that greater the length of meeting duration and the feature usage, the greater the data
usage.
These results were used to strengthen the previous researcher's studies and become a
reference for decision-makers for information on the amount of internet data use in online

Page 17 of 63
learning, and other needs specifically for support to distribution of internet assistance
programs to educational actors (students, teachers, employees, etc.) during learning or work
from home due to Covid-19 pandemic.

4. F. Zheng, N. Abbas Khan, S. Hussain. (2020). The Covid-19 Pandemic And Digital
Higher Education: The Impact Of Students’ Proactive Personality On Social Capital
Through Internet Self-Efficacy And Online Interaction Quality. Children And Youth
Services Review. Doi:https://doi.org/10.1016/j.childyouth.2020.1055694
SUMMARY:The novel coronavirus (COVID-19) was declared a pandemic by the World
Health Organization on March 11, 2020.As a result, many countries in the world instruct their
citizens to stay at home, avoid very close physical contact, and take measures of social or
physical distancing.Similar instructions were issued to all educational institutions and these
institutions were instructed to arrange online classes. Therefore, virtual learning is the sole
option for students and teachers to communicate with each other.In a recent survey, it was
reported that during the COVID-19 crisis, 8.1% of the general public in China were under
moderate to severe stress. Personality is one of the factors that affect stress assessment and
response.on the one hand, the pandemic is life-threatening and the closure of educational
institutions is disrupting their studies. In developed countries, virtual learning is considered a
very effective alternative to traditional learning. In developing countries, digital learning
remains a challenge because of the limited scope of Internet services and related technologies.
Many scholars have conducted investigations and pointed out the poor Internet infrastructure,
acceptability of smartphones in education and the unfriendly digital class.

5.Ghada Refaat El Said. (2021). How Did Covid-19 Pandemic Affect Higher Education
Learning Experience? An Empirical Investigation of Learners’ academic Performance
at a University in a Developing Country. Advances in Human-Computer Interaction.
Vol. 2021. ID6649524.
SUMMARY:-COVID-19 has dramatically reshaped the way global education is
delivered.Millions of learners were affected by educational institution closures due to the
pandemic, which resulted in the largest online movement in the history of education.With this
sudden shift away from classrooms in many parts of the globe, universities had to rapidly

Page 18 of 63
shift to virtual and digital strategies.Many believe that the adoption of online distance
learning will persist after pandemic.Chi-square test was used to compare grade distribution
for both groups.This results suggested that there was no statistically significant difference in
students’grades.This study also included a survey of 435 students and interviews with a
sample of professors about their learning and teaching experience during the lockdown.Since
Egypt decided to make the shift to online distant learning in all future higher education plans,
the results of this research would be especially vital for universities in Egypt and other
developing countries.In 2019 and even before COVID-19, there was already high growth and
adoption in education technology, In 2015, the annual growth rate of online enrolment was
increasing with an extremely rapid rate of over 30% every year, and in 2019, the number of
students taking atleast one online course has been grown to 34.7 percent of the total learner’s
population worldwide This leads to a distinctive rise of distance learning,whereby teaching is
undertaken remotely and on digital platforms.According to Hodges wellplanned online
learning is totally different from shifting online in response to a crisis, as the speed with
which this shift is done could be shocking to faculty members and learners.Until now, there
have been few investigations into how universities all over the world have dealt with the
COVID-19.However, such stocktaking is necessary, in order to support higher education
institutions to overcome the lockdown, which is expected to continue in 2020 and
2021.Information on how other universities have behaved and a look at other countries and
the universities there can be very helpful.A large number of analyses and forecasts have
already been posted on the possible COVID-19 impact on higher education.Very few
empirical academic studies have been published about this phenomenon;the highest majority
of prepandemic literature which compared face-to-face with online distance courses was
conducted in higher educational institutions in developed countries, where information
technology infrastructure,resources, and support are available and reliable.

6. Haozhe J., Atiquil A.Y.M., Xiaoqing G. & Jonathan M.S.(2021). Online Learning
Satisfaction in Higher Education during the Covid-19 Pandemic: A Regional
Comparison between Eastern and Western Chinese Universities. Education and
Information Technologies. https://doi.org/10.1007/s1063-021-10519-x.

Page 19 of 63
Summary:The survey has been conducted because, the online learning platforms have played
an irreplaceable role in the massive practice of online learning in China during the COVID-19
Pandemic.
A total of 928 students from five universities in four Chinese provinces or municipalities
were surveyed through a sampling technique and analyzed through structural equation
modeling and Rasch model. Findings show that Chinese university students’ satisfaction with
online learning platforms is directly and indirectly impacted by their computer self-efficacy
and the perceived ease of use and usefulness of the platforms. Statistical analyses have
verified that Eastern and Western Chinese university students’ computer self-efficacy directly
impacts their perceived ease of use and usefulness of online learning platforms. This implies
that the university students’ perceived ease of use and usefulness of online learning platforms
depend on their beliefs in their individual capabilities to use it for study. With the
enhancement of computer self-efficacy, university students are likely to gradually accept the
benefits and advantages of online learning platforms. Eventually, they will likely find the use
of online learning platforms to be effortless In fact, perceived ease of use and usefulness are
frequently considered to be two crucial motivation variables in numerous TAM-based
models. Considering the impact of computer self-efficacy on these variables, we suggest it
may be better to take such individual psychological factors into consideration when assessing
new technology acceptance, adoption and satisfaction. The current study can contribute to
theoretical, methodical and practical understandings of university students’ satisfaction with
using online learning platforms, which have been recognized as irreplaceable emergency
educational tools

7. Joana P., Ariadna L., Frances S., Marc A. & Daniel A. (2021). A Methodology to
Study the University’s Online Teaching Activity from Virtual Platform Indicators: The
Effect Of The Covid-19 Pandemic at Universitat Politecnia De Catalunya. Sustainability
2021, 13, 5177. https://doi.org/10.3390/su13095177
Summary:The study has been conducted to define the difference between the online teaching
and virtual teaching platform.
The pandemic and quarantine increased public dependence on digital technology. In terms of
education, this dependence was even more acute, and the shortcomings of an educational

Page 20 of 63
system that had failed to achieve digital transformation were exposed [1]. Paper [1] points to
some short-, medium-, and long-term challenges
regarding universities and the pandemic:
•Short-term: finishing the course with short-term fixes. Universities adapt to the pandemic
and offer a teaching-learning process that is mediated by digital technologies.
•Medium-term: learning to live with the virus in a hybrid environment with virtuality as a
new scenario. Universities begin redesigning teaching to be face-to-face with tones of
virtuality.
•Long-term: the digital transformation of university teaching. Digitisation was appending task
and the pandemic forces new scenarios and the achievement of the digital transformation
demanded by society. The Directorate of Educational Innovation and the Department of
Management of Methodologies and Teaching Processes. Two of the objectives of the ICE at
the UPC are to: Provide effective support and advice for quality assurance on issues related to
the design of degrees, training development, and the teaching work of lecturers.
Improve teaching by increasingly exploiting the potential that is offered by technological
resources.The ICE is responsible for ensuring that the digital educational tools used
throughout the UPC are understood and used appropriately. Therefore, ICE played an
important role in the adaptation to the pandemic and in ensuring that teaching–learning tools
were effectively used in an environment of uncertainty and fear. Given that the teaching
model at the UPC had always been eminently face-to-face, the transition to online tools
always faced a risk of failure.
Finally, other useful guidelines to study key indicators include:
•Absolute values: total numbers.•
The relative values between terms: related to an analogous term, to measure relative changes.
Comparison between mean and median values of a set. The median value is the value
separating the top half from the lower half of a data set. The comparison between mean and
median value is quite fruitful, similar values imply a homogeneous behaviour because the
median is unaffected by extreme values. The median values give better information about the
central tendency of the sample, while the mean measures intensity. For example, to evaluate
the impact of the use of quizzes for all subjects:

Page 21 of 63
The mean measures the intensity of its use, a high mean value implies a greater use of
quizzes. The median value gives an idea of the use of quizzes for 50% of subjects.
•Temporal behaviour: measure the consecutive changes and changes over time. For instance,
some weekly indicators can be plotted for a term.
•A comparison of key indicators between units in bar diagrams or boxplots with the overall
mean/median level plotted to reveal the difference between them.
•Comparison of mean/median values of activity (or specific activity) between different units
in bar diagrams.
•Bivariate diagrams between the deliveries of students and lecturers per unit and the plot for
all units. Linear regression line and correlation coefficients measure the engagement of
students with the proposed deliveries by lecturers.
This is key to measuring feedback between lecturer and student.
Specifically, to discuss the effect of Covid-19 at a university, it would be useful to show the
results following these steps:
General indicators for the Covid-19 period: during the affected term.
Vision of the term affected before and during the quarantine.
Student response to the actions of lecturers by centers during affected terms.
Comparison between units during affected terms.
Comparison of terms affected with the homologous term the previous year. Under normal
conditions, some 17% of subjects show a high level of activity with more than 80 actions per
student and, during quarantine, almost twice as many subjects showed a high level of activity
on average. There is a surprising contrast between school centers or departments, where some
reveal that about half of their courses had high levels of activity. In general, all of the
indicators in the virtual classroom indicate that there was a good subject monitoring: both for
the lecturers (who needed to be trained) and for students.

8. Lokanath M., Tushar G. & Abha S.(2020). Online teaching-learning in higher

education during lockdown period of Covid-19 pandemic. International Journal of
Educational Research Open. 2020. 100012
Summary:During the lockdown period for novel corona virus disease (COVID19) in 2019,
the entire education system, from primary school to higher education, collapsed, not just in

Page 22 of 63
India, but globally. This research is a representation of the online teaching-learning model
adopted by Mizoram University in the teaching process and subsequent semester exams. He
hopes to provide an intellectually rich opportunity for future academic decision-making in
any adversity. The government of India began to seriously consider this issue, emphasizing
ICT and online education as part of the compulsory teaching process of higher education. The
central and state governments unanimously agreed that they will implement online education
nationwide, taking into account the need for hours. The formulation of the action plan took
into account our preparations for the online teaching model, the changing momentum of the
pandemic, and the availability of resources to implement the online teaching model. In order
to cooperate with the action plan, teachers independently prepared and trained to adapt to the
technology required by the online teaching mode. All teachers and students of Mizoram
University were the populations of the study. The research result advocates for universities to
provide adequate counseling services to maintain the mental health of high school students
in this pandemic. The results of the qualitative survey show that should strive to provide
students with free access to online education resources so that they can use their time in the
best way effectively during the lockdown period. Precisely now many classroom teachers are
trying to understand the ins and outs of distance learning for the first time and find free online
resources for schools. Due to the limited mobility of academic activities and limited exchange
programs between countries during the COVID19 blockade, the liberalization, privatization
and globalization of education have significantly deteriorated. Countries are facing the
paralysis of the policy to cope with the sudden changes in education planning, management
and organization during this pandemic, their technological infrastructure is broken, academic
incompetence and lack of resources; especially among them Low-income countries are most
affected by the setbacks because their funds have been exhausted. But it is obvious that
everyone must learn to survive the current crisis, because this is just the beginning; in the
long run, no one can ignore the digital transformation of universities. Developing a
multimodal approach to achieve course content goals for better learning outcomes may be a
better idea for dealing with the complexity of online education.

Page 23 of 63
9. Maria J.S., Sandro S. (2020). The Covid-19 Pandemic as an Opportunity to Foster the
Sustainable Development of Teaching In Higher Education. Sustainability 2020. 12.
8525; doi:10.3390/su12208525
SUMMARY:-This perspective paper aims to add to the knowledge already produced on this
topic, by arguing that these challenging conditions can be a pivotal moment of opportunity
for reshaping higher education, with the implementation, development and diffusion, among
academics and students of digital technologies. The methodology used to carry out this study
is qualitative, and the technique employed to analyze the data collected was content analysis.
Research studies, in diverse formats, already published on the COVID-19 topic and its
impacts were the elected data sources. While this pandemic is having profound implications
at the global level, its repercussions and future consequences are yet to be foreseen with some
degree of anticipation and assurance. The teaching and learning activities are now much more
developed via the Internet, and HEIs are using their specific internet platforms (e.g., Moodle)
or other online tools, such as Zoom or Google classroom, for example, and e-learning has
become, in much of them, the only possible mode of formal learning. Using a sample of
30,383 students from 62 countries, concluded that these institutional actors are overall
satisfied both with the transition from face-to-face to online teaching and the support they
received from their teachers throughout this process. Furthermore, the study concluded that
the students most affected by the COVID-19-relatededucational changes were male, part-
time, undergraduate and applied sciences students, as well as students with a lower living
standard.

10. Marion H., Melanie S., Michaela G., Barbel K., Svenja B., & Albert Z.(2020). Digital
readiness and its effect on higher education students’ socio-emotional perceptions in the
context of the Covid-19 pandemic. Journal of Research on Technology in education,
DOI:10.1080/15391523.2020.1846147
Summary:Efforts to control the spread of the COVID19 virus have affected all sectors of
society around the world, including higher education systems that shift to digital higher
education. Although has established a digital learning platform and a generally good
technical team of students, it is misleading to assume that the so-called "digital natives" have
the general ability to use technology in an academic environment. In the online learning

Page 24 of 63
environment, students need to feel that they are interacting with people, and has the
opportunity to develop personal relationships. A sense of belonging to a meaningful learning
community is emphasized as an important factor in the learning experience of students in
online learning, especially because it is difficult for to make their social presence
perceivable in an online environment. Research shows that online learning communities can
help create a sense of connection with other students, which can be seen as a resource for
building and growing knowledge. This research is in the special case of emergency remote
teaching. In other words, the students are not actively making up their minds or preparing for
the digital semester. Therefore, the study investigated the readiness of higher education
students for digital distance learning only in the context of the COVID19 pandemic. As an
indicator of the readiness of students for digital learning, the research focuses on the
availability of technical equipment, such as the availability of Note books, previous
technology-based learning experiences, and self-assessment skills for distance digital
learning. The research was conducted in the form of an online survey, and measurements
were taken. Just before the start of the 2020 summer semester, all students enrolled at
German universities were invited to participate in online surveys via email. They were told
that the survey is about students’ digital readiness and their emotional and social perceptions
for the next quarter. Regarding research question Q1, the preparation of students for digital
learning appears satisfactory. Less than 1% of students do not have personal computers, and
self-report skills, use of digital tools, and information-sharing behavior are at a high level
.Regarding research question Q2, students reported average scores for stress, overload,
minutes, and emotional loneliness. Values above the average of the proportions are reported
as happy. In other words, before the next semester, Finally, student readiness for digital
learning is related to the student's self-reported social sentiment value of (Research Question
Q3). Digital Learning Ready Students said they felt less nervous, overloaded, worried,
socially and emotionally lonely, but more joyful students appear to be satisfied with the
situation .In the future advancement of this research, social interaction should be considered.
Technology can promote social sharing, but in the final analysis, Social interactions affect
students' emotions, not technology. The COVID19 crisis has affected all areas of society,
including education: "Although this is a strong stress test for the education system, it is also

Page 25 of 63
an opportunity to develop alternative educational opportunities." Students receiving higher
education represent relatively good prerequisites.

11. Monika S., Ashish K., & Gursharan K. (2020). Research Perception, Motivation and
Attitude among Undergraduate Students: A Factor Analysis Approach. Procedia
Computer Science. Vol. 167. 185-192
Summary :The study highlights a survey on research perception , motivation and attitude
among UG students.In the process of human development, research and innovation have
played a key role since the beginning of civilization while perception is defined as a
procedure to identify sensual motivational environments and implications and interpret.
Recognition is considered and may result in a sublimation of recognition of a cognitive
person. The researcher conducted a survey and the responses collected were of 226 students
which states that there's no difference amongst male and female students in context to
creative thinking and attitude.
The relationship between research attitudes and creative thinking has been tested, and
statistical tests have been conducted on the differences between male and female students .
Their results showed that there were no significant differences between male and female
students in attitudes and creative thinking. Self discussed questionnaire was also shared
among the 385 management courses and engineering students via Google forms First section
of questionnaire comprised of demographic questions while second consisted of 10 items that
seeks to asses the students towards research. The results were calculated through Factor
analysis, Normality testing and Hypothesis testing .A descriptive and reasoned analysis of the
data is carried out to analyze the influence of factors such as the perception, motivation and
attitude of the students.

12. N. Kapasia, P. Paul, A. Roy, J. Saha, A. Zaveri, R. Mallick, B. Barman, P. Das, P.

Chouhan. (2020). Impact Of Lockdown On Learning Status Of Undergraduate And
Postgraduate Students During Covid-19 Pandemic In West Bengal, India. Children and
Youth Services Review. doi: https://doi.org/10.1016/j.childyouth.2020.105194
Summary:This article highlights the impact of lockdown during the Covid -19 pandemic , in
West Bengal ,India , the impact of the lockdown on the learning status of undergraduate and

Page 26 of 63
postgraduate students. With the outbreak of Pandemic being very destructive it led to
imposition of lockdown , closure of educational institutions and cessation of physical
activities. This situation made digital platform as a mandatory tool to carry educational
activities .The researcher conducted a survey amongst 232 PG and UG students studying in
various colleges of West Bengal which comprised topics like Problems regarding studying
amidst lockdown, Mode of learning , opinion of educational decisions .Most students use
Android mobile devices to participate in online learning. Students face various problems
related to depression, anxiety, poor Internet connection and unfavorable home learning
environment. Students in remote areas and marginalized areas primarily face enormous
learning challenges during this pandemic. Family income is another disadvantage of
education aid in online mode .Although a substantial proportion of students are using digital
platforms for learning.
Many of them face huge challenges in online study. During this hard times E- Learning can
be used as a solution to this Stumbling block .

Page 27 of 63
RESEARCH METHODOLOGY
To collect data on students’ learning curve during COVID-19, a survey was conducted via
online Google forms that were sent to students in Goa, belonging to different classes. The
survey consisted of questions for which options were provided to answer. The students had to
choose the best option that is true to them. This way, by circulating the survey from 30th
September 2021 to 10th October 2021, we received responses from different parts of the
State. After examining the data, we selected around 461 records with an aim to get a wide
variety of responses.
This project aims to depict how effective is the online education system in providing the
confidence about learning to students of different backgrounds and financial status, and how
well the influence, institute type, the gender of the students, and other social and
environmental factors were considered to have effective research on them.
The project aims to depict the students' learning curve in the pandemic. The study for this was
done on different parameters that are affecting the learning progress of the students. All the
data pertaining to social and environmental factors were considered and the analysis was
done in comparison to the students learning growth to obtain the curve. The study for all the
aims mentioned in the hypothesis was carried out and the effective graph was procured. The
acquired graph was then analyzed to know the students state and the problem faced by them
in this pandemic. For example, a student institute type versus learning progress graph was
drawn to understand how their institute type is affecting their learning from home during
COVID -19. The effect of online education on the students and the support level received by
them from their teachers for learning sitting at home could be determined by the analysis
done. Understanding these factors and their effect on the students' education and growth
becomes a dominant feature to be considered as they directly relate to the country’s economic
status.
1.Why did you choose this as your topic?
The spread of COVID-19 poses a threat to humanity. As this pandemic has forced many
global activities to close including educational activities. To reduce the spread of virus,
educational institutions have been forced to switch to e-mail learning using available
educational platforms. Despite the challenges facing this sudden transformation. Since

Page 28 of 63
educational institutions adopted e- learning, to know it's effects on students of higher
education we chose this as our topic.
2.How we collected the data?
We collected the data by circulating questionnaire through google forms among our
classmates, friends, relatives and other well wishers within Goa.
3. What the data represents?
There are in total 36 columns including timestamp which is default ,namely email address,
name, educational level, name of the institution, age, gender, taluka and there are few
columns which has subquestions like :-
A)-Adaptability to online classes
i-[I have got adapted to online classes]
ii-[Online classes are better than classroom teaching]
iii-[Teachers are putting lots of efforts]
iv-[E-learning tools are easy to use]
v-[Getting the notes and submitting the assignment is much easier in online mode]
vi-[I can do my tasks faster in online mode]
OPTIONS:- agree/strongly agree/neutral/disagree/strongly disagree.
B) -Issues in online classes
i-[Network is a measure issue even after one year]
ii-[Tools are not freely available or have many restrictions]
iii-[Have to be online all the time for activities and announcement related to classes]
iv-[Slow learners don’t get extra attentions]
v-[Resolving of doubt is not satisfactory]
C) Mental issues due to online classes
i-[Online co-curricular and extra curricular activities are not satisfactory]
ii-[Human interaction is minimize]
iii-[Resulting in technology addiction]
iv-[Increased screen time causes disturb sleep pattern]
vii-[Anxiety and depression when phone is not available to use]
viii-[Lots of distraction from other application during online classes]
D)-Quality of the online teaching learning mechanism

Page 29 of 63
i-[Knowledge based skills set is improving]
ii-[It has positive impact on my learning]
iii-[Evaluation mechanism needs improvement]
iv-[Attendance should not be compulsory]
v-[Resources are available 24x7]
E)Overall impact on personality
i-[Improve student teacher relation]
ii-[It has positive impact on my self confidence]
iii-[It has build new relationship with peers]
iv-[Reduce interaction with old friends]
v-[Mobile addiction]
vi-[Increase usage of social media apps]
All of the above questions have 5 options, among which only one has to be selected by the
person .
OPTIONS are :- Agree/Disagree/Neutral/Strongly agree/Strongly disagree.
4.Explaination of entire data set
1-Email Address
To check whether one person has filled the form only once.
2-Name
To identify the person.
3-Education level
To check the qualifications of the person who has filled the form.
4-Name of the institute
To check which institution the student is belongs to
5-Age
To know the level and age group of people who filled the form
6-Gender
To check how many number of male, female and others has filled the form.
7-Taluka
To check location of the person who have filled the form.
8-Adaptability to online classes

Page 30 of 63
This column has various sub-questions such as:-
i-I have got adapted to online classes
This question tells us whether the student has adapted to online classes or not
ii-online classes are better than classroom teaching]
To know the opinion of the student if online teaching is better than classroom teaching
iii-teachers are putting lots of efforts
Whether students believe that teachers are putting lots of efforts
iv-E-learning tools are easy to use
Whether E-learning tools like google classroom, zoom, google meet etc. are easy to use
v-Getting the notes and submitting the assignment is much easier in online mode
To know whether getting the notes and submitting the assignment is much easier in online
mode
vi-I can do my tasks faster in online mode
To know the opinion of the student whether the students can do their tasks faster in online
mode
For above question there are five options among which only one option is to be selected by
the respondent
OPTIONS:- Agree/Strongly agree/Neutral/Disagree/Strongly disagree.
9-Issues in online classes
This column has various sub-questions such as:-
i-Network is a major issue even after one year
This is to ask respondents whether they have network issues in attending online classes
ii-Tools are not freely available or have many restrictions
To know the functioning of mode of online classes
iii-Have to be online all the time for activities and announcement related to classes
Whether students need to be online all the time
iv-Slow learners don’t get extra attentions
Whether students who are weak in studies get extra attention
v-Resolving of doubt is not satisfactory
Whether students doubts are cleared properly

Page 31 of 63
For above question there are five options among which only one option is to be selected by
the person
If the person is fully agree with the statement then he /she should select strongly agree
If the person is disagree with the statement then he /she should select strongly disagree
If the person is not agree nor disagree with the statement then he /she should select neutral
10-Mental issues due to online classes
For above question there are five options among which only one option is to be selected by
the person
If the person is fully agree with the statement then he /she should select strongly agree
If the person is disagree with the statement then he /she should select strongly disagree
If the person is not agree nor disagree with the statement then he /she should select neutral
i-Online co-curricular and extra curricular activities are not satisfactory
Whether students are satisfied with co-curricular and extra curricular activities in online mode
ii-Human interaction is minimize
Whether students interaction with peers and staff and minimized
iii-Resulting in technology addiction
Whether students get addicted to mobiles
iv-Increased screen time causes disturb sleep pattern
Whether students sleeping pattern is getting disturb due to more use of mobile, pc
vii-Anxiety and depression when phone is not available to use
Whether students are filling anxiety and depression when phone is not available for
viii-Lots of distraction from other application during online classes
Whether students get distracted due to other applications on phone during online classes
11-Quality of the online teaching learning mechanism
Whether the quality of online teaching is good or bad in the opinion of the respondent
For above question there are five options among which only one option is to be selected by
the person
If the person is fully agree with the statement then he /she should select strongly agree
If the person is disagree with the statement then he /she should select strongly disagree
If the person is not agree nor disagree with the statement then he /she should select neutral
i-Knowledge based skills set is improving

Page 32 of 63
To know from students if they feel their knowledge base skills are improving or not
ii-It has positive impact on my learning
Whether the online classes has positive impact on learning in their opinion
iii-Evaluation mechanism needs improvement
Whether the marks evaluation mechanism in online education needs improvement in the
opinion of the respondents
iv-Attendance should not be compulsory
Opinion of students on whether attendance should be compulsory in online classes
v-Resources are available 24x7
Whether the resources to attend classes are available 24*7
12-Overall impact on personality
For above question there are five options among which only one option is to be selected by
the person
If the person is fully agree with the statement then he /she should select strongly agree
If the person is disagree with the statement then he /she should select strongly disagree
If the person is not agree nor disagree with the statement then he /she should select neutral

i-Improve student teacher relation

Whether online classes are improving relation between students and teacher
ii-It has positive impact on my self confidence
Whether there is positive impact on self confidence of students
iii-It has build new relationship with peers
Whether the students are able to build new relation with classmate and colleagues
iv-Reduce interaction with old friends
Whether online classes minimize interaction with old friend
v-Mobile addiction
Whether online classes lead students to mobile addiction
vi-Increase usage of social media apps
Whether more usage of internet leads to increase usage of social media apps

Page 33 of 63
Data Preprocessing and cleaning

We used excel sheet for our data pre-processing and jupyter notebook for analysis and
analytics. It is an open source web application that you can use to create and share documents
that contain live code, equations, visualizations, and text.

We did pre-processing using various steps. First, we quantified the string values to numeric
values, wherever applicable. Next, we filtered the data and removed outliers from the
required rows. This reduced the number of rows from 561 to 423 rows and then from filtered
data we plotted 5 graphs which were needed for further analysis.

Data Analysis

After data preprocessing and cleaning we performed feature selection. We reduced 22

columns from 35 columns to get the best accuracy. We then added 2 columns namely Sum
and Final Result. We calculated the total count of each row and stored it in the Sum Column
and the Final Result column is the mode of all answers for respective individual. (0) indicates
performance has not improved and (1) indicates performance has improved all this was done
in excel only.

Then in jupyter notebook we used the following code snippets to analyze our data.

import pandas as pd

The import pandas portion of the code tells Python to bring the pandas data analysis library
into your current environment. The as pd portion of the code then tells Python to give pandas
the alias of pd. This allows you to use pandas functions by simply typing pd.

import numpy as np

Can be used to perform a wide variety of mathematical operations on arrays. It adds powerful
data structures to Python that guarantee efficient calculations with arrays and matrices. The as
np portion of the code then tells Python to give NumPy the alias of np. This allows you to use
NumPy functions by simply typing np .

import seaborn as sns

Page 34 of 63
Seaborn lets us plot multiple scatter plots. It's a good option when you want to get a quick
overview of your data. It pairs all the continuous data and plots their correlation. It also plots
the distribution of the data.

Then we loaded the Final.csv file in the data frame, printed top 10 row, displayed the
datatypes, binning was done and then on target variable we plotted the bar graph from which
we came to know that the performance of 69 students has not improved and 354 students has
improved. Next analytics was done to know the accuracy of our model and which algorithm
gives how much accuracy.

Data Analytics

We performed feature selection on our dataset and dropped 22 columns from 35 columns to
get the best accuracy. Next we split the data into training and testing. After that we build
baseline models using Random Forest Classifier algorithm, and 5 other algorithms were used
to check the accuracy.

Page 35 of 63
 DATA ANALYSIS AND ANALYTICS

3.1. Data analysis

The students were given a multiple choice grid where various statements were provided with
five different choices (only one of these could be selected for each statement). The choices
for all the statements were Agree, Disagree, Neutral, Strongly Agree and Strongly Disagree.
The statements were divided into 5 broad categories namely Adaptability to online classes,
Issues in online classes, Mental issues due to online classes, Quality of online teaching-
learning mechanism and Overall impact on personality.

1.ADAPTABILITY TO ONLINE CLASSES

Under this category, six statements related to the adaptability of the students to the online
classes were given to the students for analyzing how many students have adopted to e-mail
learning. The objective here was to understand whether the students can adapt themselves to
e-learning. From the below graph we can see that 50% of the students have adopted to e-
learning. Most of the students strongly believe that online classes are not better than normal
classroom classes.80%students agrees that teachers are putting lots of efforts. 70% of the
responses have neutral opinion on the statement e-learning tools are easy to use .

Few students are disagree with getting the notes and submitting the assignmentsis much
easier in online mode. 50 % of the students says they can do their tasks faster in online mode.

Page 36 of 63
2.ISSUES IN ONLINE CLASSES

Under this criteria, five statements related to the issues in online classes were given to the
students for analyzing what are the various issues faced by them. From the below graph we
can see that network is the biggest problem even after one year for online classes. Very few
students are saying that e-learning tools are not freely available. We can see that more than
40% students have same opinion that they have to be online all the time for activities and
announcements related to online classes. Very few disagree with the statement that slow
learners don't get extra attention. Most of the students says that resolving of problem is not
satisfactory. The overall analysis shows that the students are having various issues in online
classes.

3.MENTAL ISSUES DUE TO ONLINE CLASSES

Under this category, six statements related to the mental issues of the students due to online
classes were given to the students for analyzing the mental issues due to online classes. From
the below graph we can see that more than 60% students are agreed saying online
cocurricular and extracurricular activities are not satisfactory. 80% students strongly believe
that human interaction is minimized due to online classes. Less than 10% students believe
that online classes are not resulting in technology addiction. Majority of the students agree
and admit that increased screen time results in disturbed sleep patterns are there are lots of
distraction from other applications during the online classes.

Page 37 of 63
4.QUALITY OF THE ONLINE TEACHING-LEARNING MECHANISM

Under this criteria, five statements related to the " OF THE ONLINE TEACHING-
LEARNING MECHANISM" were given to the students for analyzing what they feel about
the quality of e- learning mechanism. From the below graph we can see that very few
students believe that e-learning has improved their knowledge and other skills. Around 10%
students strongly believes that online classes has improved their learning. Majority of the
students says that improvement is needed for evaluation mechanism. Above 30 % students
disagrees that attendance should not be compulsory for online classes. Most of the students
don't have any opinion on availability of resources 24×7.

Page 38 of 63
5.OVERALL IMPACT ON PERSONALITY

Under this criteria, six statements related to the overall impact on personality of the students
were given to the students for analyzing how e-learning has impacted the personality of the
students. Very students strongly believe that online classes has good impact on their self
confidence. Equal number of students are disagreed that it has helped in building relationship
with peers and no opinion on this statement. Around 60% students agrees that online classes
has reduced interactions with old friends. Majority of the students strongly believes that
online classes resulted in mobile addiction and it also increased the usage of social media
apps. The overall analysis shows that the online classes has bad impact on student's
personality.

Page 39 of 63
After data preprocessing and cleaning we performed feature selection. We reduced 22
columns from 35 columns to get the best accuracy. We then added 2 columns namely Sum
and Final Result. We calculated the total count of each row and stored it in the Sum Column
and in the Final result column, we stored the result to check whether the performance has
improved (1) or not (0). The range to check the performance was between 26-50. We
calculated this range by adding the minimum and maximum acceptable values for each
column. We checked the result by using a condition that is if sum is greater than equal to 26
and less than equal to 50 then the performance has improved (1) else not improved (0).

Then in jupyter notebook we used following code snippets for analysis.

In above code we are importing various libraries which are required in our project. Such as

pandas – We can use pandas to perform various tasks like filtering our data according to
certain conditions or segmenting and segregating the data according to preference etc.

numpy - Can be used to perform a wide variety of mathematical operations on arrays. It adds
powerful data structures to Python that guarantee efficient calculations with arrays and
matrices. The as np portion of the code then tells Python to give NumPy the alias of np. This
allows you to use NumPy functions by simply typing np.

Page 40 of 63
Seaborn – is a data visualization library built on top of matplotlib and closely integrated with
pandas data structure in Python. Basically used for data visualization and exploratory data
analysis. It is also used for predictions.

In above code df is the data frame and we are storing the entire “Final.csv” datasets in that
data frame in order to read csv files and do operations on it. (csv– comma separated files).

Top 10 rows of data frame/set are returned(displayed). In this we can observe that numbering
starts from 0. So 0 to 9 (total 10 Rows) and all columns (total 14 Columns) and their names
Such as time V1 V2,V5,V6,V10,V11,V16,V17,V18,V19,V20,V24,Sum,Final_Result.

Page 41 of 63
Next we performed binning, in the above code we used pd.cut ()function to separate the array
elements into different bins and df.head () have printed the first 5 rows and all columns

df.head (13) have printed the first 13 rows and all columns

Page 42 of 63
.dtype – It will display the data type of all the variables as we can see in above output most of
them are numeric data only

By using this code we are going to understand target variable. “Target Variable” is the
variable whose values are to be predicted by other variables. So from this variable there will
be one binary (0 or 1) wherever the value is 1 performance improved and wherever the value
is 0 performance has not been improved. In we are also checking for outliers if present.

In above graph we ca say that there is a Final_Result feature on x-axis and on y-axis we have
count feature. In this 0 is Not_Improved performance of students and 1 is Improved
performance of students. We can also see that 1 (Improved performance) is above 350 (count)
and 0 (Not_Improved performance) is less i.e. 59(count).

Page 43 of 63
Data Analytics

It access the potential of analyzed data to make future predictions using advanced statistics
methods along with artificial intelligence. So in this we are going to make predictions based
on the given dataset.

Feature Selection

We performed feature selection on our dataset in which we dropped 22 columns from 35

columns to get the best accuracy.

In order to predict we need to first split the data.

Split the data in training and testing

In order to split the data we need to import more libraries which are required for splitting
data.

Once we have imported libraries next we will split the data.

In this code y value that is output or prediction we need to find or make. We are dropping
Final_Result.

After this we are splitting the data into training and testing with test_size 0.2 that means 80%
training and 20% testing.

After splitting next we are going to test the accuracy of our model using six different
algorithms.

Page 44 of 63
Building Models

1. So first we are going to see the RandomForestClassifier() algorithm.

In this code we are going to use RandomForestClassifier() algorithm. In model object we are
storing RandomForestClassifire() then we are fitting training data in model by using “.fit”
(x_train, y_train). Then in model predictions we are going to predict test (x_test) after this we
will print the accuracy score of our model where we will check the accuracy of y value
(y_test) then after this we will print classification report where we have (y_test) and
(model_predictions).
From output we can see the accuracy of our model that is 100% .
“Support” is whenever we say for 0 (Not_Improved) 12 times our answer was correct and
when we say for 1 (Improved) 73 times our answer was correct this is the meaning of support.

2. Next we are going to predict and see the accuracy of LogisticRegression() algorithm.

Page 45 of 63
By looking at the output of LogisticRegression() algorithm we can say that the accuracy
score remains same (accuracy 100%).
3. Next we are going to predict and see the accuracy of LinearDiscriminantAnalysis()
algorithm.

By looking at the output of LinearDiscriminantAnalysis() algorithm (accuracy 94.11%) we

can say that the accuracy score have been decreased as compare to RandomForestClassifier()
algorithm (accuracy 100%) and LogisticRegression() algorithm (accuracy 100%).

4. Next we are going to predict and see the accuracy of KNeighborsClassifier() algorithm.

Page 46 of 63
By looking at the output of KNeighborsClassifier() algorithm we can say that the accuracy
score remains same (accuracy 100%).

5. Next we are going to predict and see the accuracy of DecisionTreeClassifier() algorithm.

By looking at the output of DecisionTreeClassifier() algorithm we can say that the accuracy
score remains same (accuracy 100%).

6. Next we are going to predict and see the accuracy of GaussianNB() algorithm.

Page 47 of 63
In this output also we can see that GaussianNB() algorithms accuracy has been decreased that
is 95.29% as compare to RandomForestClassifier() algorithm, LogisticRegression()
algorithm, KNeighborsClassifier() algorithm and DecisionTreeClassifier() algorithm which
has accuracy score of 100% and accuracy of GaussianNB() algorithms is increased as
compare to LinearDiscriminantAnalysis() algorithm.

Page 48 of 63
 CONCLUSION
During covid-19 pandemic performance of students in higher education has improved. We
used Six different algorithms to test the accuracy of our model.
Hypothesis testing
1. During covid-19 pandemic performance of students in higher education has improved --
Accepted. Because the results shows 100% accuracy that the performance has improved
Since 73 times out of 73 our answer was correct .
2. Perception of teachers on online teaching learning is positive -- Accepted. Based on
personal interviews with teachers we can say that perception of teachers on online
teaching learning is positive.
3. Perception of students on online teaching learning is positive -- Accepted. Since
performance of 354 students out of 423 students has been improved.
4. Perception of teachers on online teaching learning is negative -- Rejected. Based on
personal interviews with teachers we can say that perception of teachers on online
teaching learning is negative.
5. Perception of students on online teaching learning is negative -- Rejected. Because
performance of 354 students out of 423 students has been improved so we reject this
hypothesis.
6. Unstable network is the biggest hindrance in adapting to online teaching learning process
-- Accepted. Because 209 out of 423 students has strongly agreed that unstable network
was the biggest issue during pandemic so we accept this hypothesis.
SR.NO. ALGORITHMS ACCURACY
1 RandomForestClassifier() Algorithm 100%
2 LogisticRegression() Algorithm 100%
3 LinearDiscriminantAnalysis() Algorithm 94.11%
4 KNeighboursClassifier() Algorithm 100%
5 DecisionTreeClassifier() Algorithm 100%
6 GaussianNB() Algorithm 95.29%

Page 49 of 63
From the above table we can conclude that except LinearDiscriminantAnalysis() algorithm
and GaussianNB() algorithm all four algorithms give 100% accuracy. Performance of 354
students has improved and performance of 69 students has not improved.

Limitations of study
1.The marks of students in the final examinations before and after pandemic are not taken
into consideration.
2. The study is limited to responses within the State of Goa.
3. The sample size is very small.
4. Responses of other stakeholders are not taken into consideration for predicting the students'
performance.
5. Most of the respondents are from colleges across Goa. Hence the number of responses
from school students, post graduate level students and research scholars is significantly less.

Future Scope
1. Same analysis and analytics can be performed on more number of students for more
accuracy.
2. We can collect data from more locations across the world.
3. More algorithms can be included such as Naive Bayes, Linear Regression, K-Means to
understand which algorithm gives more accuracy.
4. We can add exam marks for more precise predictions.

Page 50 of 63
 REFERENCES

1. Abdelsalam M.M., Ebitisam K.E., Shadi A., Hasan R. & Hadeel A. (2021). The Covid-19
Pandemic And E-Learning: Challenges AND OPPORTUNITIES from The Perspective Of
Students And Instructors. Journal of Computing In Higher Education.
Doi.org/10.1007/s12528-021-09274-2

2.Dr. Wahab Ali (2020). Online and remote learning in higher education institutes: A
necessity in light of Covid-19 Pandemic. Higher Education Studies. Vol.10, No.3.

3.Edy Budiman. (2020). Mobile Data Usage On Online Learning During Covid-19 Pandemic
In Higher Education. iJIM. Vol. 14. No. 19.

4. F. Zheng, N. Abbas Khan, S. Hussain. (2020). The Covid-19 Pandemic And Digital Higher
Education: The Impact Of Students’ Proactive Personality On Social Capital Through
Internet Self-Efficacy And Online Interaction Quality. Children And Youth Services Review.
Doi:https://doi.org/10.1016/j.childyouth.2020.1055694

5.Ghada Refaat El Said. (2021). How Did Covid-19 Pandemic Affect Higher Education
Learning Experience? An Empirical Investigation of Learners’ academic Performance at a
University in a Developing Country. Advances in Human-Computer Interaction. Vol. 2021.
ID6649524.

6. Haozhe J., Atiquil A.Y.M., Xiaoqing G. & Jonathan M.S.(2021). Online Learning
Satisfaction in Higher Education during the Covid-19 Pandemic: A Regional Comparison
between Eastern and Western Chinese Universities. Education and Information Technologies.
https://doi.org/10.1007/s1063-021-10519-x.

7. Joana P., Ariadna L., Frances S., Marc A. & Daniel A. (2021). A Methodology to Study
the University’s Online Teaching Activity from Virtual Platform Indicators: The Effect Of
The Covid-19 Pandemic at Universitat Politecnia De Catalunya. Sustainability 2021, 13,
5177. https://doi.org/10.3390/su13095177

Page 51 of 63
8. Lokanath M., Tushar G. & Abha S.(2020). Online teaching-learning in higher education
during lockdown period of Covid-19 pandemic. International Journal of Educational
Research Open. 2020. 100012

9. Maria J.S., Sandro S. (2020). The Covid-19 Pandemic as an Opportunity to Foster the
Sustainable Development of Teaching In Higher Education. Sustainability 2020. 12. 8525;
doi:10.3390/su12208525

10. Marion H., Melanie S., Michaela G., Barbel K., Svenja B., & Albert Z.(2020). Digital
readiness and its effect on higher education students’ socio-emotional perceptions in the
context of the Covid-19 pandemic. Journal of Research on Technology in education,
DOI:10.1080/15391523.2020.1846147

11. Monika S., Ashish K., & Gursharan K. (2020). Research Perception, Motivation and
Attitude among Undergraduate Students: A Factor Analysis Approach. Procedia Computer
Science. Vol. 167. 185-192

12. N. Kapasia, P. Paul, A. Roy, J. Saha, A. Zaveri, R. Mallick, B. Barman, P. Das, P.

Chouhan. (2020). Impact Of Lockdown On Learning Status Of Undergraduate And
Postgraduate Students During Covid-19 Pandemic In West Bengal, India. Children and
Youth Services Review. doi: https://doi.org/10.1016/j.childyouth.2020.105194.

Page 52 of 63
 GANTT CHART

Task Name
Q4 2021 Q1 2022

Sept Nov Dec Jan Mar Apr

Planning
Research
Collection of
data-set and
design
Implementation
Follow Up
Final Result

Page 53 of 63
ANNEXURE

Page 54 of 63
Page 55 of 63
Page 56 of 63
Page 57 of 63