IA Assignment

Usage of Internet and its impact on mental health

1
Contents

Introduction.........................................................................................................................................3

Aim:......................................................................................................................................................4

Rationale: Why this topic was chosen................................................................................................4

Data observation:.................................................................................................................................5

Data Analysis.......................................................................................................................................7

Outliner points.....................................................................................................................................7

Pearson’s Correlation Coefficent:......................................................................................................9

Interpretation.....................................................................................................................................11

Trendline analysis of stress level vs usage time...............................................................................11

Hypothesis Testing: T-Test...............................................................................................................13

Conclusion..........................................................................................................................................15

Biblography........................................................................................................................................16

2
Introduction

As we are heading towards future, much changes have been observed in our daily

lifestyle living such as automation and easy to do things have replaced our old habits of doing

things. One such advancement towards the access to the world at a faster rate is the

“Internet”. Just one click from our mobile phones or laptops and we can interact across the

world and can access millions of information for free. Our rapid increase in the daily internet

usage and social networking platforms such as Facebook, Instagram, Twitter etc. is showing

signs of some behavioral changes on us as we interact with different people of caste, religion

and culture. Several research is ongoing on prolonged usage of internet and social media and

its impact on mental health condition. Several studies indicate that various disorders and

depression condition could be developed over prolonged usage as it brings anxiety and

change in self-esteem among oneself.

Mental health is being defined as a state of wellbeing individually to be able to

perform everyday activities in a proper manner. Thus any external matter can influence the

psychological aspects of a human behavior and can lead to mental health disorders. Social

networking sites where people interact with eachother, post reels and show off a pseudo

lifestyle can bring significant psychological stress to someone which can lead to comparing

themselves among celebrities, influencers, friends etc. causing depression and overthinking

like issues.

Since all these factors arises a question about to research on the impact of

social media and internet usage over our mental health status and its consequences on a

human life. Thus it becomes an important aspect to study the impact of internet addiction on

our mental health and how the adverse effect can be avoided by controlling all the parameters

into it.

3
Aim:

Since these days a lot of things are going viral over the internet thus I was keen to

examine and research a bit on the issue of mental health over the usage of internet and this

research can give an expression about how time spend over internet can destroy the inner

peace of a person and how we can optimize it to limit the use of it. In this assignment I’ve

surveyed some questions related to internet usage, social interactions, stress, online time

spend and much more such statements which can help to understand how usage of internet

can affect our mental health status and how it can be correlated together.

Rationale: Why this topic was chosen

Over the past few years specially during and after covid-19 period I observed that my

time period over smartphones and social media network sites have increased drastically and I

started cutting myself off from the outside world. A mini world of mine is present right now

over my Instagram and Facebook account.

Figure 1 Psychological impact on usage of social media

4
 Even during night time, I spend time watching reels over the internet and there is a

change on sleeping pattern. So I decided to research on this topic of mental health impact so

as get more aware about its pros and cons fact and can improve myself over the usage of it

before it changes me completely to a different person.

Data observation:

To research about the idea behind this assignment, I begin surveying among my

known ones and chosen around 30 persons and asked them few questions along with options

to rate their usage of internet daily, number of hours spend online, social interactions and

their mental health condition after passing time over the internet.

I begin my survey by asking them the question “number of hours they spend daily on

the internet”. And to the results surprisingly people responded with an average usage of 5

hours which was shocking that an average person spends one-fourth of his or her day over

streaming the online content.

Along with this I was interested in knowing their state of mental health for that I

decided to ask few questions related to loneliness, depression, sadness etc. after using internet

over a period of some years and what changes they feel before and after a long usage of it.

Such parameters can be a potential method to judge the correlation of usage of internet and

their response to different stress levels.

I have enquired about the mental health condition of participants and their anxiety

levels or stress level and compared it with the number of hours they spend online. I have

prepared a following table by asking about the specific symptoms and how they feel and their

mood swings after surfing internets occasionally.

So, I decided to ask the fellow person the number of hours they spend over the internet on an

average and what is the stress level /anxiety they feel each day after surfing time on the
5
internet. For that I used a scale of 1 to 10 (1 for low stress upto 10 for highest stress levels).

From this we can draw a pattern of usage vs mental health challenges that could be faced.

Person Usage time daily on internet (Hours) Rating for stress/anxiety

1 3 3

2 4 4

3 6 3

4 5 5

5 7 5

6 6 7

7 8 6

8 6 3

9 8 7

10 10 6

11 2 3

12 7 5

13 5 6

14 5 4

15 9 6

16 7 8

17 4 5

18 9 3

19 11 7

20 9 5

Table 1 Survey of people over usage time period

6
Data Analysis

Scatt er diagram
9

7
Average usage of internet

0
0 2 4 6 8 10 12
Stess level rating

Figure 2 Avg. usage vs Stress level

Scatter diagram showing the distribution among the usage hours on internet over the stress

levels ratings that people have chosen over.

Outliner points

The outliner points on the data table indicates whether in the given data any points is outside

the lower and upper boundary limits and fall outside the data point zone such data points are

to be removed from the calculation as it will deviate the result from the actual path and cause

increase in error.

For both number of hours (x) and stress rating (y) variable the outliner set points are to be

calculated separately. To do that the Table for X and Y variable are set in ascending order

and medians for individual are calculated.

Lower limit ( ¿ )=Q1−1.5 × IQR

7
 Upper Limit ( UL )=Q3 +1.5× IQR

Q1 : Represent thelower quertile of the whisker box , median of the first half data set

Q3 : Represent theupper quertile of the whisker box median of the last half data set

Variable
Q 1= ( n+12 ) th Q 3= ( n+12 ) th Lower Boundary Upper Boundary

(LB) (UB)
For first half data set For next half data

set

Number of hours
Q 1= ( 10+2 1 ) th=5 th position
Q =(
3
10+1
2 ) Q −1.5 × IQR
1
th=5th position
Q3 +1.5 × IQR

(x) 5−1.5× 3.5=−0.25 ¿ 8.5+1.5 × 3.5=13.75

Q1=5+5 ¿ ¿ =5 Q 3=
8+ 9
=8.5
2 2

Stress rating (y)

Q 1= (
10+ 1
2 ) Q 3=
th=5 th position (
10+1
2 ) Q1−1.5 × IQR
th=5th position
Q3 +1.5 × IQR

3.5−1.5× 2.5=−0.25 ¿ 6+1.5 × 2.5=9.75

(3+ 4) (6+ 6)
¿ =3.5 ¿ =6
2 2

The following table shows that in the present data all the sample size of 20 points fall within

the Lower boundary and upper boundary range thus no outliner points was found which is to

be deleted from the table for our calculation of the Pearson’s coefficient in the following

steps.

Pearson’s Correlation Coefficent:

The following table shows a clear observation that as the time spend on the internet surfing

and chatting at social media results in a greater stress value or higher anxiety levels which

8
results in difficulty to focus on real time. Thus it can be fitted in a statistical observation

using Pearson’s coefficient ( r), which can be a useful tool to correlate both the variable.It

measures the intensity of linearity among two variables and how they are dependent to each

other.

Now I will correlate statistically about how variation the stress level is varying with

the number of hours spend over internet. The number of hours of internet usage is taken as X

and Stress /anxiety level rating is taken as variable Y.

No. X Y X-X (X −X ̅ )
2
Y-Y (Y −Y )
2
∑ (X −X )(Y −Y )
1 3 3 -3.55 12.6025 -2.05 4.2025 7.2775

2 4 4 -2.55 6.5025 -1.05 1.1025 2.6775

3 6 3 -0.55 0.3025 -2.05 4.2025 1.1275

4 5 5 -1.55 2.4025 -0.05 0.0025 0.0775

5 7 5 0.45 0.2025 -0.05 0.0025 -0.0225

6 6 7 -0.55 0.3025 1.95 3.8025 -1.0725

7 8 6 1.45 2.1025 0.95 0.9025 1.3775

8 6 3 -0.55 0.3025 -2.05 4.2025 1.1275

9 8 7 1.45 2.1025 1.95 3.8025 2.8275

10 10 6 3.45 11.9025 0.95 0.9025 3.2775

11 2 3 -4.55 20.7025 -2.05 4.2025 9.3275

12 7 5 0.45 0.2025 -0.05 0.0025 -0.0225

13 5 6 -1.55 2.4025 0.95 0.9025 -1.4725

14 5 4 -1.55 2.4025 -1.05 1.1025 1.6275

15 9 6 2.45 6.0025 0.95 09025 2.3275

16 7 8 0.45 0.2025 2.95 1.1025 1.3275

9
 17 4 5 -2.55 6.5025 -0.95 0.0025 2.4225

18 9 3 2.45 6.0025 -2.05 4.2025 -5.0225

19 11 7 4.45 20.7025 1.95 3.8025 8.6775

20 9 5 2.45 6.0025 -0.05 0.0025 -0.1225

6.55 5.05 109.85 39.35 37.745

Table 2 Statistical analysis of stress level and usage time

Calculating the mean values:

ΣX
X= =6.55
N

ΣY
Y= =5.05
N

Now from the above table we can calculate the Pearson’s correlation coefficient value to

check whether any correlation could be drawn or not.

The Pearson’s correlation coefficient (PCC) is given as:

Pearson ’ s coefficient (r )=
∑ (X− X )(Y −Y ) ¿
√ ∑ X− X ¿2 ∑ (Y −Y )2

37.745 37.745
r= = =0.5741
√ 109.85 ×39.35 65.746

10
Interpretation

Pearson’s coefficient value ranges in the range of -1 to +1, where +1 shows a extreme

positive correlation meaning that the trend is upward and both the variable are strongly

dependent on eachother. A value close to -1 shows that both are not correlated to eachother.

In the assignment the value of correlation comes out to be +0.5741 falls in the range

of 0 to +1 which means a positive correlation. Hence it can be observed that as we increase

the number of usage hours over the internet, slightly the anxiety or stress level also increases

simultaneously. As the Pearson’s coefficient value is somewhere close to 0.5 therefore it

cannot assume to be a strongest correlation but definitely a positive correlation which can

affect the daily routine of a human being and can affect the mental health of a person.

Trendline analysis of stress level vs usage time

As the Pearson coefficient comes out to be 0.5741 a positive trend line can be drawn

to fit the curve of stress level with number of usage hours, I am using a straight line method,

though a proper fitted curve will not be obtained as in the scatter diagram the points are

scattered in not a straight line upward.

Assuming a straight line equation line:

y=mx+c

where m=slope of theline

The gradient line slope of the equation is calculated by,

m=
∑ ( X− X)(Y −Y )
∑ ( X−X )
2

37.745
m= =0.003127
12067.022

11
 Thus equation becomes Y =0.003127 x +c , wherec is a constant

As the slope is positive it shows a positive gradient trend line and showing as we increases our time

period spend on internet the stress level increases simultaneously.

Taking a point when stress level is 5 and usage hour of internet is 7 hours to find the value of

constant.

5=0.003127 ( 7 ) +c

c=4.97811

Y =0.003127 x+ 4.97811

Figure 3 A straight line showing stress level with hours spend

usage hours vs stress level

9
8
7
6
stress level

5
4
3
2
1
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20
Number of hours on internet

In the above figure it can be observed that as the number of usage hours increases the

increase in stress level can be seen with the observation of the trendline.

Though the trendline doesn’t match exactly a straight line but it shows a trend where it can be

seen that an upward increment with a slope of 0.003127 is present indicating a positive

relation with hours spend over internet with the mental health state of an individual.

12
Hypothesis Testing: T-Test

After a brief analysis and studying the correlation of impact on mental health as we proceed

into increasing the number of hours. It’s time to test the correlation with the population

correlation coefficient (ρ) among the variable (x,y).

Null hypothesis H O : ρ=0

Alternatehypothesis H 1 : ρ≠ 0

The null hypothesis says that there is no correlation exist between x and y variable, where

alternative hypothesis shows that there is a linear relationship between x and y.

r √ n−2
t c=
√ 1−r 2
'
r : pearson s correlation coefficient

n : sample ¿ ¿

Here I am taking degree of freedom n-2. By calculating the the critical hypothesis value one

can check whether the population taken and the correlation found is correct or not.

In the assignment, sample size is taken as 30.

0.5741 √ 30−2
t stat = =3.7101
√ 1−0.57412

The value 3.7101 shows that ρ≠0 which means that it is an alternate hypothesis and not a null

hypothesis. Now considering a Significant value, the critical value of hypothesis can be

calculated

13
Figure 4 Degree of freedom vs significance level

From the above table taking significance level of 1% the critical value for degree of freedom

value 28 is 2.763 and the statistical value comes out to be 3.7101.

t statistical =3.7101>2.763

Thus with the 1% significance level of the hypothesis testing it can be concluding that the

hypothesis is valid and the impact on mental health or stress level is directly linked to the

number of usage hours over the internet.

14
Conclusion

The IA consist of studying the mental health issues related to internet usage and the time we

spend over the internet by measuring the intensity of stress or anxiety as we increase our time

consumption over the internet as we increase our screen time usage.

Online social media addiction is becoming a major stress level issue these days and

causing a real mess for many people. The addiction to show off, copy the styles of celebrities

and showing costly living is creating a major rift unconsciously inside the mind of people.

To study this, I took a survey over 30 people and asked them their time spend daily over the

internet and their stress level rating from 1 to 10 so as to correlate the stress level condition

with the number of hours they spend over the internet.

The finding of the study was that a slight positive correlation equals to pearson’s

coefficient 0.57 was emerged between the number of hours spend over internet with the stress

induced hence showing that stress could be increased as we increase the time period of our

internet surfing which can lead to various mental health related problems further.

A distinct observation was found that the result gives a hint about relationship of stress with

internet but other factors are also included such as family and friends’ behavior with

individuals, confidence and surroundings which can introduce stress in an individual thus

internet is not only a reason to induce stress but definitely it can cause headaches and a

comparative nature to judge people and ourself.

15
Biblography

1. Igor Pantic, (2014), Online social networking and mental health,

https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4183915

2. Jules J. Berman (2016), Understanding Your Data,

https://www.sciencedirect.com/topics/computer-science/pearson-correlation

3. Paraskevi Theofilou, (2024), The Relationship of Increased Internet Use to Mental

Health in Adults. A Quantitative Study in Greece,

https://www.researchgate.net/publication/377179781_The_Relationship_of_Increased

_Internet_Use_to_Mental_Health_in_Adults_A_Quantitative_Study_in_Greece

16