In [1]:

    

import pandas as pd
import seaborn as sns
import matplotlib.pyplot as plt
sns.set(style="white")
sns.set_context("talk")


    

In [2]:

    

df = pd.read_csv('raw/2016-17-ClassCentral-Survey-data-noUserText.csv', decimal=',', encoding = "ISO-8859-1")


    

In [3]:

    

selectors = ['Timestamp', '# MOOCs Finished', 'How important is the ability to earn a certificate when you complete a MOOC?']


    

In [4]:

    

df_select = df.loc[:, selectors]
df_select.head()


    

    
Out[4]:







  

    

      

      
Timestamp
      
# MOOCs Finished
      
How important is the ability to earn a certificate when you complete a MOOC?
    
  
  

    

      
0
      
2016/11/28 11:59:22 AM PST
      
4.5
      
1.0
    
    

      
1
      
2016/12/01 6:45:40 AM PST
      
0
      
4.0
    
    

      
2
      
2016/12/01 6:54:23 AM PST
      
8
      
2.0
    
    

      
3
      
2016/12/01 7:17:09 AM PST
      
4.5
      
1.0
    
    

      
4
      
2016/12/01 7:24:51 AM PST
      
2.5
      
2.0
    
  

In [5]:

    

len(df_select)


    

    
Out[5]:





2491

In [6]:

    

table = pd.pivot_table(df_select, values = 'Timestamp', index = ['How important is the ability to earn a certificate when you complete a MOOC?'], 
                      columns =['# MOOCs Finished'], aggfunc = lambda x: len(x.unique()))


    

In [7]:

    

table


    

    
Out[7]:







  

    

      
# MOOCs Finished
      
0
      
1
      
15
      
2.5
      
25
      
4.5
      
8
    
    

      
How important is the ability to earn a certificate when you complete a MOOC?
      

      

      

      

      

      

      

    
  
  

    

      
1.0
      
108
      
44
      
40
      
75
      
36
      
59
      
53
    
    

      
2.0
      
83
      
48
      
15
      
72
      
13
      
30
      
25
    
    

      
3.0
      
190
      
63
      
22
      
105
      
30
      
62
      
42
    
    

      
4.0
      
176
      
62
      
25
      
85
      
26
      
68
      
43
    
    

      
5.0
      
351
      
83
      
27
      
97
      
34
      
56
      
36
    
  

In [8]:

    

table = table[['0', '1', '2.5', '4.5', '8', '15', '25']]
new_table = table.loc[:,:].div(table.sum(axis=0), axis=1).mul(100)


    

In [9]:

    

new_table.index


    

    
Out[9]:





Float64Index([1.0, 2.0, 3.0, 4.0, 5.0], dtype='float64', name='How important is the ability to earn a certificate when you complete a MOOC?')

In [10]:

    

#sns.heatmap(new_table.iloc[::-1], annot=True, fmt='2.0f', cbar = False, cmap="Greens")
sns.heatmap(new_table.iloc[[4, 0], :], annot=True, fmt='2.0f', cbar = False, cmap="Greens")


    

    
Out[10]:





<matplotlib.axes._subplots.AxesSubplot at 0x11321ceb8>

In [11]:

    

plt.show()