DEV Community

Ronald R
Ronald R

Posted on

DATA ANALYSIS - Student Mental Health - A Reading

For this Blog I decided to make something different, something special and simple.

I decided to analyze a Data set for Students Mental Health that I got from Kaggle, I didnt want to over complicate things so I just went on and started analyzing the file.

For this Analysis I decided to use Jupyter Notebook and a mixture of SQL instead of straight up analysing it with Notebook. I wanted to challenge myself.

After quite a few research I was able to analyze the data.

The Goal

I am to figure out which year of students is more likely to have exprience trouble with mental health and somehow correlate it with stress.

This was a very interesting Analysis just because I have always wondered why are students on their final years are more relax as to oppose to students in their first and with this analysis I was to able to see it in a bigger picture.

The Strategy

Total Number of Students resulted in atleast a 100

Image description

After Cleaning the Data making sure that there was no NaN or null numbers.
First off I queried the Male students getting their

  • Gender
  • Ager
  • Course
  • Year
  • CGPA

To access the size of the Data set

query = "SELECT Gender, Age, Course, Year, CGPA FROM student_mental_health_data_cleaned WHERE Gender = 'Male'"
Enter fullscreen mode Exit fullscreen mode

I realize that there where only about 20 something student that were 'Male'

After which, I made the decision to just use both genders at the same time. Since the sample size ratio between 'Male' to 'Female' is fairly out balanced.

Next, i queried the data of students that has Depression and Anxiety

query = "SELECT Gender, Age, Year, Depression, Anxiety, Panic FROM student_mental_health_data WHERE (Depression = 'Yes' OR Anxiety = 'Yes')"
Enter fullscreen mode Exit fullscreen mode

The result showed that out of a 100 students half experience some sorth of Mental Health issue.

"Please note that the number of responses can still vary depending on whether or not they are diagnosed or undiagnosed, for this analysis we will take the responses as face value to determine an outcome"

There are three major Factors that can affect the result

  • Uneven sample size
  • Professional Diagnosis
  • Emotional affection

After which we tallied the results of the dataset that we queried

Image description

The result showed that out of 50 students

21 of students are Year 1
15 of students are Year 2
13 of students are Year 3
2 of students are year 4

If we dissect the result it showed that Year 1 and Year 2 students are the most to exprience mental health.

A study done by Gretchen Geng and Richard Midford in 2015 titled "Investigating First and Second Year Education Studentsโ€™ Stress
Level" provides a more insight on why this is because it was somehow related to completion of assessments because of their busy schedule at schools.
This finding corelates with the research that first and second year studentsโ€™ high stress and eventualy lead to mental health issues at a stagerring high levels comes from a lack of time to study because of multiple factors as well as the ability to cope to a new and unfamilliar environment."

Year 1 can be factored in because they have the most number of students that participated in the survey. Though Year 2 in terms of ratio can be considered aas the most affected students and the chances of mental health exprience can occur.

Stress can be factored in the situation and the data set since accoring to Geng and Midford their coping mechanism is still developing as oppose to Year 3/4 students.

Summary

We can say that out of all the students Year 2 is more likely to expirence mental health issue. Though we still have to factor in the situation of each students, but by taking it in face value the result shows us which students in terms of ratio is the most affected.

Image description

And with that I move on to my next Data Analysis maybe this time instead of using SQL we'll just use python analysis.

Top comments (0)