analysis_process

May 20, 2025

[36]: import pandas as pd

import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns
from scipy import stats
import warnings #used to supress user warnings
#---------------------------------------------------------------------------------------------
#supresses 'user warnings' related to boolean series
warnings.filterwarnings(action='ignore', category=UserWarning,␣
↪message=r"Boolean Series.*")

#---------------------------------------------------------------------------------------------
#imported cleaned database
df = pd.read_csv(r"cleaned_student_depression_dataset.csv",index_col= 0,header␣
↪= 0 )

[37]: df.info()

<class 'pandas.core.frame.DataFrame'>
Index: 27857 entries, 0 to 27901
Data columns (total 18 columns):
# Column Non-Null Count Dtype
--- ------ -------------- -----
0 id 27857 non-null int64
1 Gender 27857 non-null object
2 Age 27857 non-null int64
3 City 27857 non-null object
4 Profession 27857 non-null object
5 Academic Pressure 27857 non-null int64
6 Work Pressure 27857 non-null int64
7 CGPA 27857 non-null float64
8 Study Satisfaction 27857 non-null int64
9 Job Satisfaction 27857 non-null int64
10 Sleep Duration 27857 non-null object
11 Dietary Habits 27857 non-null object
12 Degree 27857 non-null object
13 Have you ever had suicidal thoughts ? 27857 non-null object
14 Work/Study Hours 27857 non-null int64
15 Financial Stress 27857 non-null int64

1
 16 Family History of Mental Illness 27857 non-null object
17 Depression 27857 non-null bool
dtypes: bool(1), float64(1), int64(8), object(8)
memory usage: 3.9+ MB

[38]: df.head()

[38]: id Gender Age City Profession Academic Pressure \

Serial Number
0 1 Male 19 Delhi Student 4
1 2 Male 33 Visakhapatnam Student 5
2 8 Female 24 Bangalore Student 2
3 26 Male 31 Srinagar Student 3
4 30 Female 28 Varanasi Student 3

Work Pressure CGPA Study Satisfaction Job Satisfaction \

Serial Number
0 0 6.00 3 0
1 0 8.97 2 0
2 0 5.90 5 0
3 0 7.03 5 0
4 0 5.59 2 0

Sleep Duration Dietary Habits Degree \

Serial Number
0 '6-7 hours' Moderate B.Com
1 '5-6 hours' Healthy B.Pharm
2 '5-6 hours' Moderate BSc
3 'Less than 5 hours' Healthy BA
4 '7-8 hours' Moderate BCA

Have you ever had suicidal thoughts ? Work/Study Hours \

Serial Number
0 Yes 8
1 Yes 3
2 No 3
3 No 9
4 Yes 4

Financial Stress Family History of Mental Illness Depression

Serial Number
0 4 No True
1 1 No True
2 2 Yes False
3 1 Yes False
4 5 Yes True

[39]: df.describe() #description of cleaned database

2
[39]: id Age Academic Pressure Work Pressure \
count 27857.000000 27857.000000 27857.000000 27857.000000
mean 70443.316725 25.820835 3.141580 0.000431
std 40648.631003 4.906158 1.381802 0.044027
min 1.000000 18.000000 0.000000 0.000000
25% 35039.000000 21.000000 2.000000 0.000000
50% 70694.000000 25.000000 3.000000 0.000000
75% 105827.000000 30.000000 4.000000 0.000000
max 140699.000000 59.000000 5.000000 5.000000

CGPA Study Satisfaction Job Satisfaction Work/Study Hours \

count 27857.000000 27857.000000 27857.000000 27857.000000
mean 7.655911 2.944395 0.000682 7.157196
std 1.470837 1.360876 0.044429 3.707066
min 0.000000 0.000000 0.000000 0.000000
25% 6.280000 2.000000 0.000000 4.000000
50% 7.770000 3.000000 0.000000 8.000000
75% 8.920000 4.000000 0.000000 10.000000
max 10.000000 5.000000 4.000000 12.000000

Financial Stress
count 27857.000000
mean 3.140467
std 1.437145
min 1.000000
25% 2.000000
50% 3.000000
75% 4.000000
max 5.000000

[40]: # Inference 1
#---------------------------------------------------------------------------------------------
fig,axes = plt.subplots(nrows = 1, ncols = 2, figsize=(10, 6))
colors = ['#e87d5d','#62d997']
#---------------------------------------------------------------------------------------------
labels1 = ['Depressed','Non-depressed']
axes[0].bar(labels1, df['Depression'].value_counts(), width=0.4, color = colors)
axes[0].set_xticks(labels1,labels1,
rotation=0, ha='center')
axes[0].tick_params(axis='x', labelsize=10)
axes[0].set_title('Distribution of depression\n in individuals', size = 15)
axes[0].set_ylabel('Individuals (count)', size = 12)
#---------------------------------------------------------------------------------------------
explode = (0.05,0.05)
axes[1].pie(df['Depression'].value_counts(), labels=labels1,
autopct='%1.0f%%', colors=colors, explode=explode,
shadow=True, startangle = 30)

3
 axes[1].set_title('Distribution of depression\n in individuals',size = 15)
plt.subplots_adjust(left=0.1, right=0.9, top=0.9, bottom=0.1, wspace=0.1,␣
↪hspace=0.4)

#---------------------------------------------------------------------------------------------
plt.show()

[41]: # Inference 2
#---------------------------------------------------------------------------------------------
fig,axes = plt.subplots(nrows = 1, ncols = 2, figsize=(10, 6))
colors = ['#62add9','#d57bdb']
#---------------------------------------------------------------------------------------------
labels1 = df[df['Depression'] == True]['Gender'].value_counts().index
axes[0].bar(labels1, df[df['Depression'] == True]['Gender'].value_counts(),␣
↪width=0.4, color = colors)

axes[0].set_xticks(labels1,labels1,
rotation=0, ha='center')
axes[0].tick_params(axis='x', labelsize=10)
axes[0].set_title('Distribution of gender\n in depressed individuals', size =␣
↪15)

axes[0].set_ylabel('Individuals (count)', size = 12)

axes[0].set_xlabel('Gender', size = 12)
#---------------------------------------------------------------------------------------------
explode = (0.05,0.05)

4
 axes[1].pie(df[df['Depression'] == True]['Gender'].value_counts(),␣
↪labels=labels1,

autopct='%1.0f%%', colors=colors, explode=explode,

shadow=True, startangle = 30)
axes[1].set_title('Distribution of gender\n in depressed individuals',size = 15)
plt.subplots_adjust(left=0.1, right=0.9, top=0.9, bottom=0.1, wspace=0.1,␣
↪hspace=0.4)

#---------------------------------------------------------------------------------------------
plt.show()

[42]: # Inference 3
#---------------------------------------------------------------------------------------------
fig,axes = plt.subplots(nrows = 1, ncols = 2, figsize=(10, 6))
colors = ['#0390fc','#f032b7']
#---------------------------------------------------------------------------------------------
labels1 = df[df['Family History of Mental Illness'] == 'Yes']['Depression'].
↪value_counts().index

axes[0].bar(labels1, df[df['Family History of Mental Illness'] ==␣

↪'Yes']['Depression'].value_counts(),

width=0.4, color = ['#e87d5d','#62d997'])

axes[0].set_xticks(labels1,['Depressed','Non-depressed'],
rotation=0, ha='center')

5
 axes[0].tick_params(axis='x', labelsize=10)
axes[0].set_title('Distribution of individuals whose\n family had history of␣
↪mental illness', size = 15)

axes[0].set_ylabel('Individuals (count)', size = 12)

#---------------------------------------------------------------------------------------------
explode = (0.05,0.05)
axes[1].pie(df[df['Family History of Mental Illness'] == 'Yes']['Depression'].
↪value_counts(),

labels=['Depressed','Non-depressed'], autopct='%1.0f%%', colors=␣

↪['#e87d5d','#62d997'],

explode=explode, shadow=True, startangle = 30)

axes[1].set_title('Distribution of individuals whose\n family had history of␣
↪mental illness', size = 15)

plt.subplots_adjust(left=0.1, right=0.9, top=0.9, bottom=0.1, wspace=0.1,␣

↪hspace=0.4)

#---------------------------------------------------------------------------------------------
plt.show()

[43]: # Inference 4
#---------------------------------------------------------------------------------------------
fig,axes = plt.subplots(nrows = 2, ncols = 2, figsize=(15, 10))
colors = (0.2,0.4,0.2,0.6)
#---------------------------------------------------------------------------------------------
labels1 = df['Sleep Duration'].value_counts().index

6
axes[0,0].bar(labels1, df['Sleep Duration'].value_counts(), width=0.4, color =␣
↪colors)

axes[0,0].set_xticks(labels1, labels1, rotation=25, ha='center')

axes[0,0].tick_params(axis='x', labelsize=10)
axes[0,0].set_title('Sleep duration of all the individuals', size = 15)
axes[0,0].set_xlabel('Sleeping hours', size = 12)
axes[0,0].set_ylabel('Individuals (count)', size = 12)
#---------------------------------------------------------------------------------------------
labels2 = df[df['Depression'] == True]['Sleep Duration'].value_counts().index
axes[0,1].bar(labels2, df[df['Depression'] == True]['Sleep Duration'].
↪value_counts(), width=0.4, color = colors)

axes[0,1].set_xticks(labels2, labels2, rotation=30, ha='center')

axes[0,1].tick_params(axis='x', labelsize=10)
axes[0,1].set_title('Sleep duration of depressed individuals', size = 15)
axes[0,1].set_xlabel('Sleeping hours', size = 12)
axes[0,1].set_ylabel('Individuals (count)', size = 12)
#---------------------------------------------------------------------------------------------
labels3 = df[df['Depression'] == False]['Sleep Duration'].value_counts().index
axes[1,0].bar(labels3, df[df['Depression'] == False]['Sleep Duration'].
↪value_counts(), width=0.4, color = colors)

axes[1,0].set_xticks(labels3, labels3, rotation=30, ha='center')

axes[1,0].tick_params(axis='x', labelsize=10)
axes[1,0].set_title('Sleep duration of non-depressed individuals', size = 15)
axes[1,0].set_xlabel('Sleeping hours', size = 12)
axes[1,0].set_ylabel('Individuals (count)', size = 12)
#---------------------------------------------------------------------------------------------
explode = (0.05,0.05,0.05,0.05,0.05)
axes[1,1].pie(df['Sleep Duration'].value_counts(), autopct='%1.3f%%',␣
↪shadow=True, startangle = 30, labels = labels1,

explode = explode )
axes[1,1].set_title('Sleep duration of all individuals',size = 15)
plt.subplots_adjust(left=0.1, right=0.9, top=0.9, bottom=0.1, wspace=0.17,␣
↪hspace=0.53)

#---------------------------------------------------------------------------------------------
plt.show()

7
[44]: # Inference 5
#---------------------------------------------------------------------------------------------
fig,axes = plt.subplots(nrows = 2, ncols = 2, figsize=(15, 10))
colors = (0.2,0.4,0.2,0.6)
#---------------------------------------------------------------------------------------------
labels1 = df['Dietary Habits'].value_counts().index
axes[0,0].bar(labels1, df['Dietary Habits'].value_counts(), width=0.4,color =␣
↪colors)

axes[0,0].set_xticks(labels1, labels1, ha='center')

axes[0,0].tick_params(axis='x', labelsize=10)
axes[0,0].set_title('Dietary habits of all individuals', size = 15)
axes[0,0].set_ylabel('Individuals (count)', size = 12)
axes[0,0].set_xlabel('Dietary Habits', size = 12)
#---------------------------------------------------------------------------------------------
labels2 = df[df['Depression'] == True]['Dietary Habits'].value_counts().index
axes[0,1].bar(labels2, df[df['Depression'] == True]['Dietary Habits'].
↪value_counts(), width=0.4, color = colors)

axes[0,1].set_xticks(labels2, labels2, ha='center')

axes[0,1].tick_params(axis='x', labelsize=10)
axes[0,1].set_title('Dietary habits of depressed individuals', size = 15)

8
axes[0,1].set_ylabel('Individuals (count)', size = 12)
axes[0,1].set_xlabel('Dietary Habits', size = 12)
#---------------------------------------------------------------------------------------------
labels3 = df[df['Depression'] == False]['Dietary Habits'].value_counts().index
axes[1,0].bar(labels3, df[df['Depression'] == False]['Dietary Habits'].
↪value_counts(), width=0.4, color = colors)

axes[1,0].set_xticks(labels3, labels3, ha='center')

axes[1,0].tick_params(axis='x', labelsize=10)
axes[1,0].set_title('Dietary habits of non-depressed individuals', size = 15)
axes[1,0].set_ylabel('Individuals (count)', size = 12)
axes[1,0].set_xlabel('Dietary Habits', size = 12)
#---------------------------------------------------------------------------------------------
explode = (0.05,0.05,0.05,0.05)
axes[1,1].pie(df['Dietary Habits'].value_counts(), autopct='%1.2f%%',␣
↪shadow=True, startangle = 30, labels = labels1,

explode = explode)
axes[1,1].set_title('Dietary habits of all individuals', size = 15)
#---------------------------------------------------------------------------------------------
plt.subplots_adjust(left=0.1, right=0.9, top=0.9, bottom=0.1, wspace=0.17,␣
↪hspace=0.4)

plt.show()

9
[45]: # Inference 6
#---------------------------------------------------------------------------------------------
labels = ['age of all \nindividuals', 'age of depressed \nindividuals',
'age of non-depressed\n individuals']
colors = ['#995757', '#b0ae54', '#4bad6f']

fig, axes = plt.subplots(nrows=1,ncols=1,figsize=(5, 7))

axes.set_title('Age of individuals in different categories', size = 15)
axes.set_ylabel('Age', size = 12)

bplot = axes.boxplot([df['Age'],df[df['Depression'] ==␣

↪True]['Age'],df[df['Depression'] == False]['Age']], widths=0.80,

patch_artist=True, # allows color

tick_labels=labels)
# fills with colors
for patch, color in zip(bplot['boxes'], colors):
patch.set_facecolor(color)

for median in bplot['medians']:

median.set_color('black')

axes.tick_params(axis='x', labelrotation=25)
axes.set_yticks(range(16,64,2))
axes.set_yticklabels(range(16,64,2))

plt.show()

10
[46]: # Inference 7
#---------------------------------------------------------------------------------------------
labels = ['Academic Pressure \nof all individuals', 'Academic Pressure of␣
↪\ndepressed individuals',

'Academic Pressure of \nnon-depressed individuals']

colors = ['#536cb8', '#20465c', '#905799']

11
fig, axes = plt.subplots(nrows=1,ncols=1,figsize=(5, 7))
axes.set_title('Academic pressure of individuals in different categories', size␣
↪= 15)

axes.set_ylabel('Academic Pressure (on a scale of 0 to 5)', size = 12)

bplot = axes.boxplot([df['Academic Pressure'],df[df['Depression'] ==␣

↪True]['Academic Pressure'],

df[df['Depression'] == False]['Academic Pressure']],␣

↪widths=0.80,

patch_artist=True,
tick_labels=labels)

for patch, color in zip(bplot['boxes'], colors):

patch.set_facecolor(color)

for median in bplot['medians']:

median.set_color('black')

axes.tick_params(axis='x', labelrotation=25)
axes.set_yticks(np.arange(0,5.5,0.5))
axes.set_yticklabels(np.arange(0,5.5,0.5))

plt.show()

12
[47]: # Inference 8
#---------------------------------------------------------------------------------------------
labels = ['CGPA \nof all individuals', 'CGPA of \ndepressed individuals',
'CGPA of \nnon-depressed individuals']
colors = ['peachpuff', '#32a8a2', '#32a852']

fig, axes = plt.subplots(nrows=1,ncols=1,figsize=(5, 7))

axes.set_title('CGPA of individuals in different categories', size = 15)

13
axes.set_ylabel('CGPA', size = 12)
bplot = axes.boxplot([df['CGPA'],df[df['Depression'] == True]['CGPA'],
df[df['Depression'] == False]['CGPA']], widths=0.80,
patch_artist=True,
tick_labels=labels)

for patch, color in zip(bplot['boxes'], colors):

patch.set_facecolor(color)

for median in bplot['medians']:

median.set_color('black')

axes.tick_params(axis='x', labelrotation=25)
axes.set_yticks(np.arange(0,10.5,0.5))
axes.set_yticklabels(np.arange(0,10.5,0.5))

plt.show()

14
[48]: # Inference 9
#---------------------------------------------------------------------------------------------
fig, axes = plt.subplots(nrows=1,ncols=1,figsize=(20, 10))
labels = df['Degree'].unique()
collection = []
for x in df['Degree'].unique() :
collection.append(df[df['Degree']==x]['Work/Study Hours'])

15
 bplot = axes.boxplot(collection, widths=0.80,
patch_artist=True,
tick_labels=labels)

to_highlight = [1,3,6,10,17,27]
for x in to_highlight:
bplot['boxes'][x].set_facecolor('#cf9d7c')

for median in bplot['medians']:

median.set_color('black')

axes.tick_params(axis='x', labelrotation=90)
axes.set_title('Study hours of different courses', size = 15)
axes.set_xlabel('Degrees', size = 12)
axes.set_ylabel('Study hours', size = 12)
axes.set_yticks(np.arange(0,14,1))
axes.set_yticklabels(np.arange(0,14,1))

plt.show()

[49]: # Inference 10
#---------------------------------------------------------------------------------------------
fig, axes = plt.subplots(nrows=1,ncols=1,figsize=(20, 10))
labels = df['City'].unique()
collection = []

16
for x in df['City'].unique() :
collection.append(df[df['City']==x]['Academic Pressure'])

bplot = axes.boxplot(collection, widths=0.80,

patch_artist=True,
tick_labels=labels)

to_highlight = [0,4,8,10,13,14,17,22,24,28]
for x in to_highlight:
bplot['boxes'][x].set_facecolor('#cf9d7c')

for median in bplot['medians']:

median.set_color('black')

axes.tick_params(axis='x', labelrotation=90)
axes.set_title('Academic pressure of individuals in different cities', size =␣
↪15)

axes.set_xlabel('Cities', size = 12)

axes.set_ylabel('Academic pressure (on a scale of 0 to 5)', size = 12)
axes.set_yticks(np.arange(0,6,0.5))
axes.set_yticklabels(np.arange(0,6,0.5))

plt.show()

17
[50]: # Inference 11
#---------------------------------------------------------------------------------------------
labels = ['Male','Female']
colors = ['#707ccc', '#cc708d']

fig, axes = plt.subplots(nrows=1,ncols=3,figsize=(17, 8.5))

#---------------------------------------------------------------------------------------------
axes[0].set_title('Academic pressure of all \nindividuals of both genders',␣
↪size = 15)

axes[0].set_xlabel('Gender', size = 12)

axes[0].set_ylabel('Academic pressure (on a scale of 0 to 5)', size = 12)
bplot0 = axes[0].boxplot([df[df['Gender'] == 'Male']['Academic Pressure'],
df[df['Gender'] == 'Female']['Academic Pressure']],␣
↪widths=0.5,

patch_artist=True,
tick_labels=labels)

for patch, color in zip(bplot0['boxes'], colors):

patch.set_facecolor(color)

for median in bplot0['medians']:

median.set_color('black')

axes[0].tick_params(axis='x', labelrotation=25)
axes[0].set_yticks(np.arange(0,5.5,0.5))
axes[0].set_yticklabels(np.arange(0,5.5,0.5))
axes[0].set_aspect(0.5)
#---------------------------------------------------------------------------------------------
axes[1].set_title('Academic pressure of depressed \nindividuals of both␣
↪genders', size = 15)

axes[1].set_xlabel('Gender', size = 12)

axes[1].set_ylabel('Academic pressure (on a scale of 0 to 5)', size = 12)
bplot1 = axes[1].boxplot([df[df['Depression']==True][df['Gender'] ==␣
↪'Male']['Academic Pressure'],

df[df['Depression']==True][df['Gender'] ==␣
↪'Female']['Academic Pressure']], widths=0.5,

patch_artist=True,
tick_labels=labels)

for patch, color in zip(bplot1['boxes'], colors):

patch.set_facecolor(color)

for median in bplot1['medians']:

median.set_color('black')

axes[1].tick_params(axis='x', labelrotation=25)
axes[1].set_yticks(np.arange(0,5.5,0.5))

18
axes[1].set_yticklabels(np.arange(0,5.5,0.5))
axes[1].set_aspect(0.5)
#---------------------------------------------------------------------------------------------
axes[2].set_title('Academic pressure of non-depressed\n individuals of both␣
↪genders', size = 15)

axes[2].set_xlabel('Gender', size = 12)

axes[2].set_ylabel('Academic pressure (on a scale of 0 to 5)', size = 12)
bplot2 = axes[2].boxplot([df[df['Depression']==False][df['Gender'] ==␣
↪'Male']['Academic Pressure'],

df[df['Depression']==False][df['Gender'] ==␣
↪'Female']['Academic Pressure']], widths=0.5,

patch_artist=True,
tick_labels=labels)

for patch, color in zip(bplot2['boxes'], colors):

patch.set_facecolor(color)

for median in bplot2['medians']:

median.set_color('black')

axes[2].tick_params(axis='x', labelrotation=25)
axes[2].set_yticks(np.arange(0,5.5,0.5))
axes[2].set_yticklabels(np.arange(0,5.5,0.5))
axes[2].set_aspect(0.5)
#---------------------------------------------------------------------------------------------
plt.show()

19
[51]: # Inference 12
#---------------------------------------------------------------------------------------------
labels = ['Male','Female']
colors = ['#707ccc', '#cc708d']

fig, axes = plt.subplots(nrows=1,ncols=3,figsize=(12, 6.5))

#---------------------------------------------------------------------------------------------
axes[0].set_title('CGPA of all \nindividuals of both genders', size = 15)
axes[0].set_xlabel('Gender', size = 12)
axes[0].set_ylabel('CGPA', size = 12)
bplot0 = axes[0].boxplot([df[df['Gender'] == 'Male']['CGPA'],
df[df['Gender'] == 'Female']['CGPA']], widths=0.5,
patch_artist=True,
tick_labels=labels)

for patch, color in zip(bplot0['boxes'], colors):

patch.set_facecolor(color)

for median in bplot0['medians']:

median.set_color('black')

axes[0].set_yticks(np.arange(0,10.5,0.5))
axes[0].set_yticklabels(np.arange(0,10.5,0.5))
axes[0].set_aspect(0.35)
#---------------------------------------------------------------------------------------------
axes[1].set_title('CGPA of depressed\n individuals of both genders', size = 15)
axes[1].set_xlabel('Gender', size = 12)
axes[1].set_ylabel('CGPA', size = 12)
bplot1 = axes[1].boxplot([df[df['Depression']==True][df['Gender'] ==␣
↪'Male']['CGPA'],

df[df['Depression']==True][df['Gender'] ==␣
↪'Female']['CGPA']], widths=0.5,

patch_artist=True,
tick_labels=labels)

for patch, color in zip(bplot1['boxes'], colors):

patch.set_facecolor(color)

for median in bplot1['medians']:

median.set_color('black')

axes[1].set_yticks(np.arange(0,10.5,0.5))
axes[1].set_yticklabels(np.arange(0,10.5,0.5))
axes[1].set_aspect(0.35)
#---------------------------------------------------------------------------------------------
axes[2].set_title('CGPA of non-depressed\n individuals of both genders', size =␣
↪15)

20
 axes[2].set_xlabel('Gender', size = 12)
axes[2].set_ylabel('CGPA', size = 12)
bplot2 = axes[2].boxplot([df[df['Depression']==False][df['Gender'] ==␣
↪'Male']['CGPA'],

df[df['Depression']==False][df['Gender'] ==␣
↪'Female']['CGPA']], widths=0.5,

patch_artist=True,
tick_labels=labels)

for patch, color in zip(bplot2['boxes'], colors):

patch.set_facecolor(color)

for median in bplot2['medians']:

median.set_color('black')

axes[2].set_yticks(np.arange(0,10.5,0.5))
axes[2].set_yticklabels(np.arange(0,10.5,0.5))
axes[2].set_aspect(0.35)
#---------------------------------------------------------------------------------------------
plt.show()

[52]: # Inference 13
#---------------------------------------------------------------------------------------------
fig, axes = plt.subplots(nrows=1,ncols=1,figsize=(6,7.5))
labels = df['Sleep Duration'].unique()

21
collection = []
colors = ['#48db5e','#0390fc','#d93261','#07f57e','#17e3d5']
for x in df['Sleep Duration'].unique() :
collection.append(df[df['Sleep Duration']==x]['Age'])

bplot = axes.boxplot(collection, widths=0.80,

patch_artist=True,
tick_labels=labels)

for patch, color in zip(bplot['boxes'], colors):

patch.set_facecolor(color)

for median in bplot['medians']:

median.set_color('black')

axes.tick_params(axis='x', labelrotation=45)
axes.set_title('How age affects sleeping hours', size = 15)
axes.set_xlabel('Sleeping hours', size = 12)
axes.set_ylabel('Age', size = 12)
axes.set_yticks(np.arange(14,60,2))
axes.set_yticklabels(np.arange(14,60,2))

plt.show()

22
[53]: # Inference 14
#---------------------------------------------------------------------------------------------
fig,axes = plt.subplots(nrows = 1, ncols = 2, figsize=(10, 6))

23
#---------------------------------------------------------------------------------------------
labels1 = df['Have you ever had suicidal thoughts ?'].value_counts().index
colors = ['#e87d5d','#62add9']

axes[0].bar(labels1, df['Have you ever had suicidal thoughts ?'].

↪value_counts(), width=0.4, color = colors)

axes[0].set_xticks(labels1,['Have had suicidal \nthoughts','Do not have␣

↪\nsuicidal thoughts'],

rotation=0, ha='center')
axes[0].tick_params(axis='x', labelsize=10)
axes[0].set_title('Individuals with/without\n suicidal throughts', size = 15)
axes[0].set_ylabel('Individuals (count)',size = 12)
#---------------------------------------------------------------------------------------------
explode = (0.05,0.05)
axes[1].pie(df['Have you ever had suicidal thoughts ?'].value_counts(),␣
↪autopct='%1.3f%%', shadow=True, startangle = 30,

labels = ['Have had \nsuicidal \nthoughts','Do not \nhave␣

↪suicidal\nthoughts'], explode = explode, colors=colors)

axes[1].set_title('Individuals with/without\n suicidal throughts', size = 15)

plt.subplots_adjust(left=0.1, right=0.9, top=0.9, bottom=0.1, wspace=0.1,␣
↪hspace=0.4)

#---------------------------------------------------------------------------------------------
plt.show()

24
[54]: # Inference 15
#---------------------------------------------------------------------------------------------
fig, axes = plt.subplots(nrows=1,ncols=1,figsize=(6,7.5))
labels = sorted(df['Financial Stress'].unique())
collection = []
colors = ['#48db5e','#0390fc','#b4db48','#07f57e','#17e3d5']

for x in sorted(df['Financial Stress'].unique()) :

collection.append(df[df['Financial Stress']==x]['CGPA'])

bplot = axes.boxplot(collection, widths=0.80,

patch_artist=True,
tick_labels=labels)

for patch, color in zip(bplot['boxes'], colors):

patch.set_facecolor(color)

for median in bplot['medians']:

median.set_color('black')

axes.set_title('How financial stress affects CGPA', size = 15)

axes.tick_params(axis='x', labelrotation=0)
axes.set_ylabel('CGPA', size = 12)
axes.set_xlabel('Financial stress (on a scale of 1 to 5)', size = 12)
axes.set_yticks(np.arange(0,10.5,0.5))
axes.set_yticklabels(np.arange(0,10.5,0.5))

plt.show()

25
[55]: # Inference 16
#---------------------------------------------------------------------------------------------
fig, axes = plt.subplots(nrows=1,ncols=1,figsize=(6,4))

random_rows = df.sample(n=50, axis='rows')

x = random_rows['Work/Study Hours']
y = random_rows['Age']
z = np.polyfit(x,y, 1)

26
 p = np.poly1d(z)

axes.scatter(x,y,s =170, c = '#42424270')

axes.plot(x,p(x),'r--')
axes.set_xticks(np.arange(0,13,1))
axes.set_yticks(np.arange(16,61,3))
axes.set_title('Relation between age and work/study hours', size = 15)
axes.set_ylabel('Age', size = 12)
axes.set_xlabel('Work/study hours', size = 12)

plt.show()

[56]: # Inference 17
#---------------------------------------------------------------------------------------------
fig, axes = plt.subplots(nrows=1,ncols=1,figsize=(6,4))

random_rows = df.sample(n=50, axis='rows')

x = random_rows['Work/Study Hours']
y = random_rows['CGPA']
z = np.polyfit(x,y, 1)
p = np.poly1d(z)

27
 axes.scatter(x,y,s =170, c = '#42424270')
axes.plot(x,p(x),'r--')
axes.set_xticks(np.arange(0,13,1))
axes.set_yticks(np.arange(0,10.5,1))
axes.set_title('Relation between work/study hours and CGPA', size = 15)
axes.set_ylabel('CGPA', size = 12)
axes.set_xlabel('Work/study hours', size = 12)

plt.show()

[57]: # Inference 18
#---------------------------------------------------------------------------------------------
g = sns.pairplot(df[['Age','Academic Pressure','CGPA','Study␣
↪Satisfaction','Work/Study Hours','Financial Stress']])

g.figure.suptitle("Pair plot to show pairwise bivariate distribution", y=1)

plt.show()

28
[58]: # Inference 19
#---------------------------------------------------------------------------------------------
sns.kdeplot(data=df, x ='Academic Pressure')
plt.title('Density distribution for Academic Pressure')
plt.xlabel('Academic Pressure (on a scale of 0 to 5)')
plt.show()

29
[59]: df[df['Academic Pressure'] < 3]['Depression'].value_counts()

[59]: Depression
False 6475
True 2497
Name: count, dtype: int64

[60]: df[df['Academic Pressure'] > 3]['Depression'].value_counts()

[60]: Depression
True 9335
False 2103
Name: count, dtype: int64

[61]: # Inference 20
#---------------------------------------------------------------------------------------------
sns.kdeplot(data=df, x = 'Study Satisfaction')
plt.title('Density distribution for Study Satisfaction')
plt.xlabel('Study Satisfaction (on a scale of 0 to 5)')
plt.show()

30
[62]: df[df['Study Satisfaction'] < 4]['Depression'].value_counts()

[62]: Depression
True 10969
False 6123
Name: count, dtype: int64

[63]: df[df['Study Satisfaction'] > 3]['Depression'].value_counts()

[63]: Depression
False 5421
True 5344
Name: count, dtype: int64

[64]: # Inference 21
#---------------------------------------------------------------------------------------------
sns.kdeplot(data=df, x = 'Financial Stress')
plt.title('Density distribution for Financial Stress')
plt.xlabel('Financial Stress (on a scale of 0 to 5)')
plt.show()

31
[65]: df[df['Financial Stress'] > 4]['Depression'].value_counts()

[65]: Depression
True 5448
False 1257
Name: count, dtype: int64

[66]: # Inference 22
#---------------------------------------------------------------------------------------------
matrix = df[['Age','Academic Pressure','CGPA','Study Satisfaction','Work/Study␣
↪Hours','Financial Stress']]

values = pd.DataFrame(columns=['mean','mode','median','standard␣
↪deviation','confidence interval at 95%','standard error'],

index=['Age','Academic Pressure','CGPA','Study␣
↪Satisfaction','Work/Study Hours','Financial Stress'])

for x in matrix.columns :
values.loc[x,'mean'] = matrix[x].mean()
values.loc[x,'mode'] = matrix[x].mode()[0]

32
 values.loc[x,'median'] = matrix[x].median()
values.loc[x,'standard deviation'] = matrix[x].std()
values.loc[x,'standard error'] = matrix[x].sem()
interval = values.loc[x,'standard error'] * stats.t.ppf((1 + 0.95) / 2,␣
↪len(matrix[x]) - 1)

values.loc[x,'confidence interval at 95%'] = (values.loc[x,'mean'] -␣

↪interval, values.loc[x,'mean'] + interval)

values.head()

[66]: mean mode median standard deviation \

Age 25.820835 24 25.0 4.906158
Academic Pressure 3.14158 3 3.0 1.381802
CGPA 7.655911 8.04 7.77 1.470837
Study Satisfaction 2.944395 4 3.0 1.360876
Work/Study Hours 7.157196 10 8.0 3.707066

confidence interval at 95% standard error

Age (25.76321921075781, 25.878450746524024) 0.029395
Academic Pressure (3.125352936553525, 3.1578074899102004) 0.008279
CGPA (7.638638485585483, 7.673184216069394) 0.008812
Study Satisfaction (2.9284130546380416, 2.96037611863977) 0.008154
Work/Study Hours (7.113661520434242, 7.200729835418866) 0.022211

[67]: # Inference 23
#---------------------------------------------------------------------------------------------
corr_matrix = df[['Age','Academic Pressure','CGPA','Study Satisfaction','Work/
↪Study Hours','Financial Stress']].corr()

plt.imshow(corr_matrix, cmap='autumn_r', interpolation='nearest')

plt.colorbar()
plt.grid(False)
plt.title('Heat map of correlation coefficients\n betweeen each pair of␣
↪features')

tick_marks = np.arange(len(corr_matrix.columns))
plt.xticks(tick_marks, corr_matrix.columns, rotation=80)
plt.yticks(tick_marks, corr_matrix.index)

for i in range(len(corr_matrix.index)):
for j in range(len(corr_matrix.columns)):
plt.text(j, i, round(corr_matrix.iloc[i, j],3), ha="center",␣
↪va="center", color="black")

plt.show()

33
[68]: #for summary -- refer thesis/documentation

34