Students’ Social Media Addiction

Introduction

This presentation will illustrate the analysis of students’ social media addiction data. This dataset contains records of students’ social media behavior and related life outcomes.

Presentation Contents:

  • Data Dictionary

  • Exploratory Data Analysis

  • Linear Regression

  • Logistic Regression

Data Dictionary

The Social_Media_Addiction dataset includes:

Variable Description
Age Student’s Age
Gender The student’s self-reported gender (Male or Female)
Academic_Level The student’s current academic affiliation
Country The country of residence where the student completed the survey
Daily_Hours The average number of hours per day the student spends on social media platforms
Platform The social media platform on which the student spends the most time
Affects_Academic_Performance Self‐reported impact on academics (Yes/No)
Sleep_Hours The respondent’s average nightly sleep duration in hours
Mental_Health_Score Self‐rated mental health (1 = poor to 10 = excellent)
Relationship_Status Single / In Relationship / Complicated
Conflicts Number of relationship conflicts due to social media
Addicted_Score Social Media Addiction Score (1 = low to 10 = high)

Load Packages & Data

First, load the necessary libraries.

library(ggplot2)
library(GGally)
library(tidyverse)
library(lattice)

EDA - Summary of Data

library(readr)
Social_Media_Addiction <- read_csv("Social media addiction - Social Media Addiction (1).csv")
Social_Media_Addiction <-mutate(Social_Media_Addiction,
                                across(c(Gender,
                                         Academic_Level,
                                         Country,
                                         Platform,
                                         Affects_Academic_Performance,
                                         Relationship_Status),factor))
summary(Social_Media_Addiction)
   Student_ID       Age           Gender          Academic_Level    Country   
 Min.   :  1   Min.   :18.00   Female:353   Graduate     :325    India  : 53  
 1st Qu.:177   1st Qu.:19.00   Male  :352   High School  : 27    USA    : 40  
 Median :353   Median :21.00                Undergraduate:353    Canada : 34  
 Mean   :353   Mean   :20.66                                     Denmark: 27  
 3rd Qu.:529   3rd Qu.:22.00                                     France : 27  
 Max.   :705   Max.   :24.00                                     Ireland: 27  
                                                                 (Other):497  
  Daily_Hours         Platform   Affects_Academic_Performance  Sleep_Hours   
 Min.   :1.500   Instagram:249   No :252                      Min.   :3.800  
 1st Qu.:4.100   TikTok   :154   Yes:453                      1st Qu.:6.000  
 Median :4.800   Facebook :123                                Median :6.900  
 Mean   :4.919   WhatsApp : 54                                Mean   :6.869  
 3rd Qu.:5.800   Twitter  : 30                                3rd Qu.:7.700  
 Max.   :8.500   LinkedIn : 21                                Max.   :9.600  
                 (Other)  : 74                                               
 Mental_Health_Score      Relationship_Status   Conflicts    Addicted_Score 
 Min.   :4.000       Complicated    : 32      Min.   :0.00   Min.   :2.000  
 1st Qu.:5.000       In Relationship:289      1st Qu.:2.00   1st Qu.:5.000  
 Median :6.000       Single         :384      Median :3.00   Median :7.000  
 Mean   :6.227                                Mean   :2.85   Mean   :6.437  
 3rd Qu.:7.000                                3rd Qu.:4.00   3rd Qu.:8.000  
 Max.   :9.000                                Max.   :5.00   Max.   :9.000

EDA - Distribution of Age

ggplot(Social_Media_Addiction)+
  aes(Age)+
  geom_histogram(binwidth = 0.5)

  • From 18 years old to 24 years old.

  • 18 years old is the fewest respondents.

ggplot(Social_Media_Addiction)+
  aes(Age)+
  geom_histogram(binwidth = 0.5)+
  coord_cartesian(xlim = c(18,23), ylim = c(130,170))

  • Most frequent response was 22 years old.

EDA - Distribution of the platforms used

ggplot(Social_Media_Addiction)+
  aes(Platform, fill = Gender)+
  geom_bar()+
  theme(axis.text.x = element_text(angle = 45, hjust = 1))

  • Instagram is the most popular app, followed by TikTok.

  • More female use Instagram and more male use Facebook.

  • Instagram and TikTok are highly addictive apps.

EDA - Daily Usage Hours vs Sleep Hours

ggplot(Social_Media_Addiction, aes(x = Daily_Hours, y = Sleep_Hours)) +
  geom_point()

  • Gender does not play a role.

  • As the amount of usage time increases the amount of time of sleeping is decreasing.

Linear Regression - Sleep Hours vs Mental Health Score

ggplot(Social_Media_Addiction)+
  aes(Sleep_Hours, Mental_Health_Score)+
  geom_jitter()+
  geom_smooth(method = "lm") +
  labs(title = "Sleep Hours vs Mental Health Score")

  • The more sleep you get, the better your mental health will be.

  • The more one uses social media, the worse one’s mental health becomes.

Conclusion

  • The more time spent using social media, the less time spent sleeping.

  • Less sleep was associated with worse mental health.

  • Most of the students surveyed were between the ages of 19 and 22.

  • Most used social media was Instagram.

    Therefore, Attention should be paid to students’ social media use and its impact on their health.