Testing a Potential Moderator(Python)

Introduction:
From the "gapminder" dataset, the variables "urbanrate" and "internetuserate" are going to be tested for their correlation with the potential moderator "politiyscore".

"polityscore" has been divided into 4 levels : No_Democracy, Unsatisfied_Democracy, Satisfied_Democracy and Good_Democracy. The purpose of this research is to investigate the effect of this variable on the correlation between "urbanrate" and "internetuserate".

Python Output1:

association between urbanrate and internetuserate for NO_Democracy countries
(0.60873731612685533, 0.0020528103271054998)
     
association between urbanrate and internetuserate for Unsatisfied_Democracy countries
(0.70570435251349417, 8.1192786139000039e-05)
     
association between urbanrate and internetuserate for Satisfied_Democracy countries
(0.45972994415542473, 0.047663537308730138)
association between urbanrate and internetuserate for Good_Democracy countries

(0.68152965983424918, 2.7020062268884203e-13)

The values of the 4 correlation coefficient

---------

Conclusion:
The change of value of the Pearson Correlation Coefficient due to the testing moderator-"polityscore" does not have obvious pattern or trend.  In other words, "polityscore" is not a significant moderator for the relationship between "urbanrate" and "internetuserate".

Python Code
import pandas
import numpy
import scipy.stats
import seaborn
import matplotlib.pyplot as plt

data = pandas.read_csv('gapminder.csv', low_memory=False)

data['urbanrate'] = data['urbanrate'].convert_objects(convert_numeric=True)
data['polityscore'] = data['polityscore'].convert_objects(convert_numeric=True)
data['internetuserate'] = data['internetuserate'].convert_objects(convert_numeric=True)
data['polityscore']=data['polityscore'].replace(' ', numpy.nan)

data_clean=data.dropna()

print (scipy.stats.pearsonr(data_clean['urbanrate'], data_clean['internetuserate']))

def politysco (row):
   if row['polityscore'] <= -5:
      return 1
   elif row['polityscore'] <= 0 :
      return 2
   elif row['polityscore'] <= 5:
      return 3
   elif row['polityscore'] > 5:
      return 4
data_clean['politysco'] = data_clean.apply (lambda row: politysco (row),axis=1)

chk1 = data_clean['politysco'].value_counts(sort=False, dropna=False)
print(chk1)

sub1=data_clean[(data_clean['politysco']== 1)]
sub2=data_clean[(data_clean['politysco']== 2)]
sub3=data_clean[(data_clean['politysco']== 3)]
sub4=data_clean [(data_clean['politysco']== 4)]

print ('association between urbanrate and internetuserate for NO_Democracy countries')
print (scipy.stats.pearsonr(sub1['urbanrate'], sub1['internetuserate']))
print ('       ')
print ('association between urbanrate and internetuserate for Unsatisfied_Democracy countries')
print (scipy.stats.pearsonr(sub2['urbanrate'], sub2['internetuserate']))
print ('       ')
print ('association between urbanrate and internetuserate for Satisfied_Democracy countries')
print (scipy.stats.pearsonr(sub3['urbanrate'], sub3['internetuserate']))
print ('association between urbanrate and internetuserate for Good_Democracy countries')
print (scipy.stats.pearsonr(sub4['urbanrate'], sub4['internetuserate']))




scat1 = seaborn.regplot(x="urbanrate", y="internetuserate", data=sub1)
plt.xlabel('Urban Rate')
plt.ylabel('Internet Use Rate')
plt.title('Scatterplot for the Association Between Urban Rate and Internet Use Rate forNO_Democracy countries countries')
print (scat1)
plt.show()

scat2 = seaborn.regplot(x="urbanrate", y="internetuserate", data=sub2)
plt.xlabel('Urban Rate')
plt.ylabel('Internet Use Rate')
plt.title('Scatterplot for the Association Between Urban Rate and Internet Use Rate for Unsatisfied_Democracy countries')
print (scat2)
plt.show()

scat3 = seaborn.regplot(x="urbanrate", y="internetuserate", data=sub3)
plt.xlabel('Urban Rate')
plt.ylabel('Internet Use Rate')
plt.title('Scatterplot for the Association Between Urban Rate and Internet Use Rate for Satisfied_Democracy countries countries')
print (scat3)
plt.show()

scat4 = seaborn.regplot(x="urbanrate", y="internetuserate", data=sub4)
plt.xlabel('Urban Rate')
plt.ylabel('Internet Use Rate')
plt.title('Scatterplot for the Association Between Urban Rate and Internet Use Rate for  Good_Democracy countries')
print (scat4)
plt.show()

a=scipy.stats.pearsonr(sub1['urbanrate'], sub1['internetuserate'])[0]
b=scipy.stats.pearsonr(sub2['urbanrate'], sub2['internetuserate'])[0]
c=scipy.stats.pearsonr(sub3['urbanrate'], sub3['internetuserate'])[0]
d=scipy.stats.pearsonr(sub4['urbanrate'], sub4['internetuserate'])[0]

x=pandas.DataFrame([a,b,c,d])

print('The values of the 4 correlation coefficient')

x.plot(kind='bar', legend=True)