In-depth analysis on Suicide in India (2001-2014)
- Sharuk
- Sep 9, 2022
- 11 min read
Table of content:
Tools used in this project:
Python for Data cleaning.
Excel for Data analysis.
Tableau for Visualization and Dashboards.
In-depth analysis on Suicide in India (2001-2014), Yes, I understand it’s a morbid topic but I want to know, is there any kind of underlying factors influences these unfortunate demises and possible solution for prevention.
About the Data:
All 15 primary datasets and few secondary datasets are collected from data.gov.in
Data cleaning process were carried out using Python in Jupiter lab. Issues encountered and solutions applied on cleaning data can be seen in My GitHub repository. Lots of Data wrangling and specific section of data cleaning done using Power query in excel as well.
Data analysis carried out in excel. Worksheets can be found here.
Our multi-faceted data includes,
Circumstance of suicide
Education status of the victims
Professional profile of the victims
Marital status of the victims
Means adopted by the victim to commit suicide.
Before we go into specificity of the cases, Let's explore the data at hand for better understanding. For which I have prepared a dataset which represent the common denominator of all 5 datasets.
General
Which region has the highest suicide rate?
Region by No. of victims
Maharashtra accounts for 12.4% of all the suicide cases in India (2001-2014). Which makes it the state with most suicide cases.
Followed by West Bengal and Tamil Nadu with 11.81% and 11.41% respectively.
But this isn’t adequate enough to know which State/UT has the highest suicide rate. For that we need suicide rate per capita.
Region by Victims for every 100k
For calculations, click here.
Puducherry (UT) has the highest suicide rate with 49 victims for every 100k (lakh) people. About 3.7 times that of the National average.
Followed by Andaman and Nicobar (UT) with 37 victims for every 100k people.
Bihar has lowest suicide per capita, with only 0.8 victim for every 100k people.
Region with rising/decline of Suicide cases over the years?
Total victim by Year
From 2001 to 2014 India saw 21.3% increase in suicide cases.
With R-squared value of 0.91
Suicide cases in Gujarat by Year
Gujarat saw 50.8% increase in suicide cases.
With R-squared value of 0.92
Suicide cases in Kerala by Year
Kerala is the only state which has any form of downward trajectory over the years with 11.7% decrease in suicide cases.
With R-squared value of 0.84.
Which zone of India has the highest suicide rate?
Suicide victim for every 100k by Zones
Southern zone of India has the highest suicide per capita by far.
Central zone has the lowest suicide rate.
What is the gender distribution in suicide cases?
Male to Female victims (%)
No. of victims by gender over the years
No. of female victims remains almost the same over 14 years, whereas no. of male victims has a steady upward trajectory.
Cases of male victims increased by about 34.4% with R-squared value of 0.97, whereas, cases of female victims increased by mere 0.78%.
Male to Female victim for every 10k
Male suicide rate is 26.3% higher than that of female suicide rate.
Findings:
Suicide cases of men, steadily increase over the years, whereas for suicide cases of women stayed the same.
Southern zone of India has the highest suicide rate and central zone has the lowest suicide rate.
Now let's go into specifics.
Suicide Circumstances
What are the most common circumstances which provoked these acts?
Total victims by Suicide Circumstances
About 23.9% of the victims had “Family Problem” as the cause for suicide.
Followed by “Prolonged illness” with 13.2% and “Mental illness” with 6.38%.
Let’s broadly categorize all of the circumstances and analyze.
Total victims by Categorized Circumstances
Almost ¼ of the cause for suicide is “Family Problems”.
With “Medical issue” accounts for above 1/5th of the causes of suicide.
“Dowry dispute” makes up about 50% of the suicide victims in “Marital issues” category.
“Prolonged illness” makes up about 62.2% of the victims in “Medical issue” category.
Which circumstance has highest gender disproportionality.
No. of victims by Gender and Circumstances.
The above chart shows total number suicide victim by circumstances on gender basis. But, this isn’t helpful to finding out whether gender has any disproportional effects on suicide circumstances.
Male to Female victim ratio
The overall percentage for male and female victim is 64.48% and 35.52% respectively.
We’ll use this ratio as a reference. From here on we use this ratio as rule of thumb, i.e., any data point that deviate significantly from this ratio means there is Gender disproportionality.
So the total suicide cases number converted into percentage with calculated fields in Tableau.
Circumstances by victims on gender basis in %
The reference line represent the ratio.
Men makes up 96.3% of “Drug abuse/addiction” suicide victims.
Similarly, men accounted for 88.2% of “Bankruptcy”, 87.4 % of “Professional Problem” and 86.4% of “Unemployment” suicide victims.
Women makes up about 97.6% of “Dowry Dispute” suicide victims.
Likewise, Women accounted for 96.2% of “Illegitimate Pregnancy”, 69.9% of “Impotency/infertility” and 67.5% of “Physical abuse” suicide victims.
i.e., Women, who only accounts for 35.5% of overall suicide victims makes up about 97.6% of “Dowry Dispute” victims.
i.e., Men, who makes up about 64.5% of overall suicide victims accounted for 96.3% of “Drug abuse/addiction” victims.
% of victim by categorized circumstances on gender basis
The categorical grouping of circumstances shows that
In career related issues, male suicide rate is higher than their female counterparts.
In marital issues, female suicide rate is higher than their female counterparts.
Which circumstance of suicide increased/decreased over the years.
“Drug abuse” victims over the years
Suicide of “drug abuse” victim is increased by 157% over the years
R-squared value of 0.91
“Illegitimate pregnancy” victims over the years
84.8% decrease in “illegitimate pregnancy” over the years
R-squared value of 0.69
“Family problems” victims over the years
“Family problem” category is increased by 46% over the years
R-square value of 0.86
Findings:
Family problem is the major cause for known suicide cases.
In career related issues, male suicide rate is higher than their female counterparts.
With “Bankruptcy”, “Professional problem” and “Unemployment” as the major career related cause for suicide.
In marital issues, female suicide rate is higher than their male counterparts.
With “Dowry dispute”, “Illegitimate pregnancy” and “Impotency” as the major marital related cause for suicide.
Suicide case related to “Drug abuse” soared over the years and almost all of victims are men.
Educational status
Does educational level have any effect over suicide?
Suicide victims by Education level
Almost 48% of the suicide victims completed “Primary” and “Secondary” schooling.
Victims who completed “Diploma”, “Graduate” and “Post Graduate and above” combined, only accounts for 3.82% of the total suicide case.
From the above chart, it seems like that more educated individual, accounts for lower number of suicide cases. But there is bias in this data, we need to take into consideration of population distribution among these educational levels to get more accurate picture.
We are going to need census data on education level of Indian population in this time period.
Calculation:
No. of suicide victims for every 10k people of the demography
Middle school qualification has the highest suicide rate. For every 10k people with Middle school level of education 36 of them have committed suicide.
Similarly, 34 out of every 10k people with Secondary level of education were the victims.
It seems like people who are not educated are less likely to commit suicide just as highly educated. It is the people who are in between the spectrum who are most likely to commit suicide.
Is there any gender disproportionality here?
Total victim by education level on gender basis
Now let’s calculate victim per capita on educational level.
Calculation:
Victim for every 10k by gender on educational basis
Men with “No education” have twice the suicide rate than Women with “No education”.
As the educational level increases the difference between male and female suicide per capita keeps on decreasing. Not that suicide per capita decreases in fact it increases rather the gender disproportionality decreases
Which educational qualification saw increase/decrease in suicide cases over the years?
Victim by year on Education level over the year
Both Secondary and Hr. Secondary demography have consistent upward trajectory over years.
Victim with “Secondary” level education increased by 64.8% and R-square value of 0.98.
Victim with “Hr. Secondary” level education increased by 113.8% has R-square value of 0.97.
Findings
People with “No education” have the lower suicide rate just as people with “Graduate and above” education qualification. It’s the people in-between the spectrum has the highest suicide rates.
As the educational level increases the difference between male and female suicide rate decreases. Not that suicide rate itself decreases (in fact it increases) rather the gender disproportionality decreases.
Suicide case of people who completed “Secondary” and “Hr. secondary” level education keeps on increasing over the years.
Professional profile
Which type of professional has the highest suicide rate?
Professional profile by Total victims
House wife demography makes up almost 1/5 of the know suicide cases.
Between 2001-2013 about 5,61,992 Women committed suicide, among them about 3,07,985 lived their lives as House wife, which means about 54.8% of women who committed suicide were house wife. It can be argued that about 78% of Indian women tend to be house wife.
Followed by “Farming/Agriculture” with 13.3% of all suicide cases.
“Service (Government)” and “Public Sector Under taking” have low suicide rate.
Is there any gender disproportionality in here?
Professional profile by total victims on gender basis
Around 90% percentage of suicide committed in the demography of “Self-employed” and “Professional activity” are men.
Women accounts for cent percent of the Housewife who committed suicide.
By definition housewife category excludes men. Its worth mentioning, between 2001-2013 about 5,61,992 Women committed suicide, among them about 3,07,985 lived their lives as House wife, which means about 54.8% of women who committed suicide were house wife. It can be argued that about 78% of Indian women tend to be house wife.
Which profession saw significant increase/decrease in suicide cases over the years?
Professional profile by years
No. of victims in “Service (Private)” and “Student” are increasing over the years with R-squared value of 0.83 and 0.73 respectively.
Whereas for “Farming/Agriculture” victims counts are decreasing over the years with R-squared value of 0.63
Findings
“House wife” profession has the highest suicide cases.
“Farming/Agriculture” has the second most suicide cases and “Service (Government)” and “Public sector undertaking” the lowest suicide cases.
Almost all of the Suicide victim who’s “Self-Employed”, “Professional activity” are men.
As the “Farming/Agriculture” profession victims cases decreases, “Service (Private)” and “Student” victims cases increases over the years.
Marital status
Which type of marital life has the highest suicide cases?
Total victims by Marital status
The chart shows marital people committed more suicide by far. But we know this data isn’t sufficient enough to support that statement.
Calculation:
Victim for every 10k people by marital status
Divorced people have highest suicide rate by far. Which is 6.3 times more than married people which has the second highest suicide rate.
Is there any gender disproportionality in here?
Calculation:
Population of widowed women is 252.4% higher than the population of widowed men.
Similarly, divorced women population is 116 % higher than the population of divorced men.
Victim for every 10k by marital status on gender basis
Divorced men have highest suicide rate by far, which is 3 times more than Divorced women.
Similarly widowed men have 3 times higher suicide rate than that of widowed women.
Findings
Divorced people have highest suicide rate by far, followed by married people.
Population of widowed women is thrice the population of widowed men. Similarly, divorced women population is twice the population of divorced men.
Divorced men have three times more suicide per capita than divorced women. Similarly widowed men have three times more suicide per capita than widowed women.
Mean adopted
What is the most commonly used method to commit suicide?
Means adopted by no. of victims
About 1/3 of the victims used hanging as the mean to commit suicide
1/3 of the victims used consuming poison as the mean.
Does gender factors into this?
Mean adopted by total victim % on gender basis
Among victims who committed suicide “By over alcoholism” over 93% of were men. Similarly, about 86% who used “By machine” as a mean were men.
Among victims who committed suicide “By fire/self-immolation” over 65 % were women.
Which “Means adopted” saw significant increase/decrease in suicide cases over the years?
Victim by means adopted over the year
“Consuming sleeping pills” as the mean is decreased by about 50% over the years
Victim by means adopted over the year
“By hanging” as the mean have consistent upward trajectory about 80% over the years.
Findings
Most common means adopted for suicide is by hanging and by consuming poison.
Almost all of the victims who committed suicide “By over alcoholism” are men.
Most of the victims who committed suicide “By fire/self-immolation” were women.
Conclusion
Summing up Findings
General:
Suicide cases of men, steadily increase over the years, whereas for suicide cases of women stayed the same.
Southern zone of India has the highest suicide rate and central zone has the lowest suicide rate.
Circumstance of the suicide:
Family problem is the major cause for known suicide cases.
In career related issues, male suicide rate is higher than their female counterparts.
With “Bankruptcy”, “Professional problem” and “Unemployment” as the major career related cause for suicide.
In marital issues, female suicide rate is higher than their male counterparts.
With “Dowry dispute”, “Illegitimate pregnancy” and “Impotency” as the major marital related cause for suicide.
Suicide case related to “Drug abuse” soared, and almost all of victims are men.
Educational status:
People with “No education” have the lower suicide rate just as people with “Graduate and above” education qualification. It’s the people in-between the spectrum has the highest suicide rates.
As the educational level increases the difference between male and female suicide rate decreases. Not that suicide rate itself decreases (in fact it increases) rather the gender disproportionality decreases.
Suicide case of people who completed “Secondary” and “Hr. secondary” level education keeps on increasing over the years.
Professional profile:
“House wife” profession has the highest suicide cases.
“Farming/Agriculture” has the second most suicide cases and “Service (Government)” and “Public sector undertaking” the lowest suicide cases.
Almost all of the Suicide victim who’s “Self-Employed”, “Professional activity” are men.
As the “Farming/Agriculture” profession victims cases decreases, “Service (Private)” and “Student” victims cases increases over the years.
Marital status:
Divorced people have highest suicide rate by far, followed by married people.
Population of widowed women is thrice the population of widowed men. Similarly, divorced women population is twice the population of divorced men.
Divorced men have three times more suicide per capita than divorced women. Similarly widowed men have three times more suicide per capita than widowed women.
Means adopted by the victim:
Most common means adopted for suicide is by hanging and by consuming poison.
Almost all of the victims who committed suicide “By over alcoholism” are men.
Most of the victims who committed suicide “By fire/self-immolation” were women.
Interpretation of Findings and Possible solutions
Why does Southern zone of India have highest suicide rate?
From the “Educational level” analysis, we know that people with No education and people with highest level education has the lowest suicide rates and people who in-between the spectrum has the higher suicide rates.
Hypothesis:
People with "No education" have lower suicide rate, thus, region where a significant portion of the population are "uneducated" will have lower suicide rate.
With that in mind, let’s calculate people with no education in states/UT (in percentage) and take average based on zones. Plotting theses values with “Victim for every 100k” values, we get,
Southern zone has the lowest percentage of people with no education and has highest suicide rate, whereas central zone has the lowest suicide rate and higher percentage of people with no education.
There is definitely correlation here, as the percentage of people with no education decreases the suicide rate increases. In other words, as the percentage of educated people increases the suicide rate also increases.
Addressing issues in men's career:
From the “Professional profile” analysis, we know that the suicide rate for victim with government related jobs are low. After all working for government does have its exclusive perks. It provides a sense of social security. Monthly salary on time, pension schemes, medical insurance, housing facilities etc. Government jobs definitely provides stable life. This claim is further supported by the fact that the professions which has highest suicide rate like “Farming”, “Service (Private)” and “Self-employed” are the kind of jobs that lack the social security factor that Government job has.
By having jobs (including private jobs) to integrate social safety net factors like medical insurance, pension schemes, etc., will significantly reduce the suicide rate of men.
Addressing issues in women's marital life:
From the “Circumstances” analysis we know that women are disproportionately affected in marital issues. “Dowry dispute” has the highest gender disproportionality by far. These kinds of issues require social reforms.
Links
Data collected from data.gov.in
Project's GitHub repository
Project's Excel worksheets
Tableau Dashboard
Comments