The effect of Gross National per capita on Expected years schooling children aged 6 of China in 2019

Introduction

Children's education is always a big concern for many families in the world, and so does in China, a competitive society. Then, what are the factors influencing expected schooling years of children? There is much research studying about factors influencing students' education achievement. For example, Gobena (2018) pointed out that "there was a weak positive relationship between Family Total Income per Month (FTIPM) and student academic achievement measured by CGPA". How about family's economic situation influencing expected children's schooling years when they are young in China? If there are gender differences in terms of this? In this project, I try to focus on the relationship of Gross National Income per capita and Expected years of schooling children (aged 6) in China. I specifically select year 2019 as a sample.

Data and descriptive analysis

The data used in the analysis is the subnational human development index data from the Global Data Lab. I choose to analyze the relationship of Gross National Income per capita in thousands of US Dollars and Expected years schooling children aged 6 of China in 2019.

#Import Data df = pd.read_csv("SHDI-SGDI-Total 5.0.csv") df

iso_codeobject

USA2.9%

IND2%

184 others95.1%

countryobject

AFG

Afghanistan

AFG

Afghanistan

AFG

Afghanistan

AFG

Afghanistan

AFG

Afghanistan

AFG

Afghanistan

AFG

Afghanistan

AFG

Afghanistan

AFG

Afghanistan

AFG

Afghanistan

#Prepare Data df1 = df[df.level != 'National'] dfChina=df1.query("country == 'China'") dfChina_2019=dfChina.query("year == 2019") dfChina_2019.head()

iso_codeobject

countryobject

9212

CHN

China

9213

CHN

China

9214

CHN

China

9215

CHN

China

9216

CHN

China

#Select Data needed df_n = dfChina_2019[['region', 'shdi','esch','eschf' ,'eschm','lgnic','lgnicf','lgnicm']].round(2) df_n

regionobject

Beijing3.2%

Tianjin3.2%

29 others93.5%

shdifloat64

0.61 - 0.9

9212

Beijing

0.9

9213

Tianjin

0.84

9214

Hebei

0.74

9215

Shanxi

0.75

9216

Inner Mongolia

0.77

9217

Liaoning

0.77

9218

Jilin

0.74

9219

Heilongjiang

0.74

9220

Shanghai

0.87

9221

Jiangsu

0.8

df_n.describe().round(2)

shdifloat64

eschfloat64

count

mean

0.75

13.94

std

0.06

1.58

min

0.61

8.77

25%

0.73

13.28

50%

0.75

14.02

75%

0.77

14.4

max

0.9

Exploratory data analysis

Strip plots

px.strip(df_n,x = 'esch', hover_name= 'region')

px.strip(dfChina_2019, x = 'esch', y = 'region', range_x= [5, 20], color = 'region', hover_name= 'region', hover_data = ['region'], labels=dict(esch='Expected years schooling children aged 6') )

px.strip(dfChina_2019, x = 'lgnic', y = 'region', range_x= [8, 12], color = 'region', hover_name= 'region', hover_data = ['region'], labels=dict(lgnic='Gross National Income per capita') )

The strip plots illustrate Gross National Income Per capital and Expected years schooling children aged 6 of China in 2019. We can find that some provinces or cities with more Gross National Income per capita, such as Beijing and Shanghai, also have longer Expected years of schooling children aged 6. While some provinces with less Gross National Income per capita, such as Guizhou and Gansu, also have shorter Expected years of schooling children aged 6.

Regression Analysis

The effect of Gross National Income per capita on Expected years schooling children aged 6

Since the coefficient for the variable esch (expected years schooling children aged 6) is 0.144, with a significant p-value, there is a positive effect of Gross National Income per capita on Expected years schooling children aged 6 of China in 2019: With Gross National per capita increase by one thousand dollars, the expected years of schooling children aged 6 will get would increase by 0.144 years.

smf.ols(formula='lgnic ~ esch', data=df_n).fit().summary()

px.scatter(df_n, x ='lgnic', y ='esch', log_x = False, log_y = False, hover_data= ['region', 'lgnic'], trendline = "ols", trendline_options=dict(log_x = False, log_y = False), marginal_x = "box", marginal_y = "box" )

The effect of Gross National Income per capita on Expected years schooling children aged 6 (of male)

With Gross National Per capita increase by one thousand dollars, the expected years of schooling boys aged 6 will get would increase by 0.172 years.

df_n['lgnic'] = pd.to_numeric(df_n['lgnic']) df_n['eschm'] = pd.to_numeric(df_n['eschm']) smf.ols(formula='lgnic ~ eschm', data=df_n).fit().summary()

The effect of Gross National Income per capita on Expected years schooling children aged 6 (of female)

With Gross National Per capita increase by one thousand dollars, the expected years of schooling girls aged 6 will get would increase by 0.118 years.

df_n['lgnic'] = pd.to_numeric(df_n['lgnic']) df_n['eschf'] = pd.to_numeric(df_n['eschf']) smf.ols(formula='lgnic ~ eschf', data=df_n).fit().summary()

Concluding remarks

Through the regression analysis, I found that:

There is a positive effect of Gross National Income per capita on Expected years schooling children aged 6 of China in 2019: With Gross National per capita increase by one thousand dollars, the expected years of schooling children aged 6 will get would increase by 0.144 years.

The effect of Gross National per capita on Expected years schooling children aged 6 is more pronounced for males than for females.