rjhts13 - Tumblr blog

[rjhts13]

ANOVA TEST

Model Interpretation for ANOVA:

When examining the association between the employment rate(quantitative variable) and economic type of country(categorical variable),an Analysis of Variance (ANOVA) revealed that there is a significant  difference between the number of working population in poor countries and the rest of the world. On comparing the employment rate between three types of countries classified as rich,poor and moderate(based on the income per person value) we get a p-value of 3.64e-13.

On performing the POST hoc test, we can see that there is a significant difference in the working classes around the world.

CODE

RESULTS

[rjhts13]

Visualization Of My Data

I am adding the count plots to check for the data in my dataset.

The below countplot helps us to visualize how the countries are seperated on the basis of economy.

The below countplot helps us to visually categorize the suicide rates based on their frequency.

It can be seen that maximum number of suicides fall under high frequency and the least is under low frequency.

There is a difference in the number of working class women under different economic conditions.

Starting from A and ending at G, the number of female employees are separated on the basis of income. Women with medium income form the highest trend and women with high income form the lowest trend.

High Income Women Employees -  11.65%

Medium Income Female Employees - 68.5%

Low Income Female Employees - 19.855%

Now the female employment based on the country’s economy and suicide rates can be visualized with the help of the below mentioned plot.

It can be observed the number of suicides is maximum among the rich category and minimum among the middle-class category. Also there is a linear relationship between the number of female employees and suicide rates. It can be evidently seen that the female suicide rate and country’s economy have a relationship.

 From the below plot it can be seen that there is a negative relationship between male and female employment.

[rjhts13]

Interpretation on Suicides and Female Employment

I have split the countries in my data set into three categories depending on the income per person column,namely :-

a) Rich

b)Middle class

c)Poor

I have collapsed the responses for female employment rate and employment rate to create number of female and male employees. It can be seen that the number of female employees are much greater than number of male employees in rich and poor countries. The number of suicides are most observed in the economically rich countries than others.

Here is the output data set,

On further observation it an be seen that number of suicides are more in highly infected and polluted areas(i.e) in areas with greater co2 emissions and HIV rates. It can also be observed that the number of alcohol users are more prone to suicide than people with lesser consumption. It leads us to think whether the deaths are wrongly interpreted as suicides or suicides are caused due to the above mentioned external factors. It was also studied that urban rate does not influence the female employment rate.

I am attaching the data set here,

[rjhts13]

Data Analysis Project

PROGRAM

# -*- coding: utf-8 -*- import pandas import numpy data = pandas.read_csv(r'E:\Data Science\Assignment\GapMinder.csv',low_memory=False,encoding='utf-8') pandas.set_option('display.float_format',lambda x:'%f'%x) print('----------------------Counts-----------------------') data['femaleemployrate']=data['femaleemployrate'] a1 = data['femaleemployrate'].value_counts(sort=False,dropna=False) print(a1) print('--------------------------------Percentages--------------------------------') #a2 = data['femaleemployrate'].value_counts(sort=False,normalize=True) #print(a2) a3=data.groupby('femaleemployrate').size()*100/len(data) #for percentage print(a3) print('---------------------Counts-------------------------') b1=data['incomeperperson'].value_counts(sort=False) print(b1) print('-------------------------Percentages------------------') b2=data['incomeperperson'].value_counts(sort=False,normalize=True) print(b2) print('---------------------Counts-------------------------') c1 = data['suicideper100th'].value_counts(sort=False,dropna=False) print(c1) print('-------------------------Percentages------------------') c2=data['suicideper100th'].value_counts(sort=False,normalize=True) print(c2)

OUTPUT

The number of working females were compared to the number of suicides that occurred in that place. The suicide rates and female employment rates are compared and there is an indirect relationship between female employment rate and suicide rate depending on the country they live in. Economically wealthy countries suffer lower suicide rates than economically backward countries. Both these types of countries seems to have a good employment rate.The data set has many unique values and the data has to be refined further for knowing the exact values. In countries with female employment rate between (40-75)%  there is a suicide rate of  (5-13)% whereas female employment rate below 40% has a suicide rate of below 5%.

[rjhts13]

After looking through the codebook of GapMinder that includes one year of numerous country-level indicators of health, wealth and development, I have decided that I am particularly interested in the employment rate of different countries. While employment rate was the initial variable, on further study it is observed that female employment rate has huge differences for different countries. I decide that I am most interested in exploring the association between female employment rate and countries. I add to my codebook variables reflecting female employment rate (e.g. female employment rate and country). On second review of the codebook it is observed that suicide rate and female employment rate also has a relationship. Is female employment rate being affected by the country and suicide rate?

Literature review

Female labor force participation and suicide by panelJeffrey A.Burr Patricia L.McCall and EvePowell-Griner states the dependence of female employment rate and suicide rate

Gender Gaps in Unemployment Rates in OECD Countries shows the dependence of female employment rate vs male employment rate in various countries.

A note on the changing relationship between fertility and female employment rates in developed countries  by Namkee Ahn & Pedro Mira  shows the relationship between female employment rate,fertility rate and countries.

Hypothesis

From the above journals and publications we can see that female unemployment rate is more in middle-east and Mediterranean countries. From the researches held it is observed that female employment rate is greatly affected by conservative societal ideas and suicide rates. The suicide rates of working class females greatly differs for different countries, which suggests that female employment rate is majorly affected by the countries they live in.