Homework #5.html - In [1]: # File 'student-mat.csv'...

In [1]:
# File 'student-mat.csv' contains a list of students and
#their alcohol consumption habits along with some demographic data.
#Answer the following questions -
In [2]:
import os
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
import numpy as np
from scipy import stats
from statsmodels.stats import weightstats as stests
from scipy.stats import ttest_ind
from scipy.stats import ttest_1samp
from scipy.stats import norm
In [3]:
#Load the file “Movie-Ratings.csv”
student_mat = pd.read_csv("student-mat.csv")
display(student_mat)
sex age Mjob Fjob guardian
Workday
consumption
Weekend
consumption
absences
from
school
0 F 18
at_hom
e
teacher mother 1 1 6
1 F 17
at_hom
e
other father 1 1 4
2 F 15
at_hom
e
other mother 2 3 10
3 F 15 health services mother 1 1 2
4 F 16 other other father 1 2 4
... ... ... ... ... ... ... ... ...
390 M 20 services services other 4 5 11
391 M 17 services services mother 3 4 3
392 M 21 other other other 3 3 3
393 M 18 services other mother 3 4 0
394 M 19 other
at_hom
e
father 3 3 5
395 rows × 8 columns
In [4]:
#1 Is there a difference between mean workday alcohol consumption for students who

reported their
# mother as being at home compared to those whose mothers are working ? Look at
column Mjob. (10 points)
# Home
Home = student_mat.loc[(student_mat.Mjob == 'at_home')].reset_index(drop=True)
Work = student_mat.loc[(student_mat.Mjob != 'at_home')].reset_index(drop=True)
Work.loc[:,'Mjob'] = 'at_work'
compare_data = pd.concat([Home,Work]).reset_index(drop=True)
display(compare_data, len(Home),len(Work),len(compare_data))
sex age Mjob Fjob guardian
Workday
consumption
Weekend
consumption
absences
from
school
0 F 18
at_ho
me
teacher mother 1 1 6
1 F 17
at_ho
me
other father 1 1 4
2 F 15
at_ho
me
other mother 2 3 10
3 F 15
at_ho
me
other mother 1 1 8
4 F 16
at_ho
me
other mother 1 2 12
... ... ... ... ... ... ... ... ...
390 M 20
at_wo
rk
services other 4 5 11
391 M 17
at_wo
rk
services mother 3 4 3
392 M 21
at_wo
rk
other other 3 3 3
393 M 18
at_wo
rk
other mother 3 4 0
394 M 19
at_wo
rk
at_hom
e
father 3 3 5
395 rows × 8 columns
59
336
395
In [5]:
# Remove outliers outside of 6-sigma and re plot the data
#Home data

stdH = Home['Workday consumption'].std(ddof=1)
meanH = Home['Workday consumption'].mean()
#screening anything outside +/- 6 std dev
fence_low_stdH = meanH - 6*stdH
fence_high_stdH = meanH + 6*stdH
Home = Home.loc[(Home['Workday consumption'] > fence_low_stdH) & (Home['Workday
consumption'] < fence_high_stdH)]
#Work data
stdW = Work['Workday consumption'].std(ddof=1)
meanW = Work['Workday consumption'].mean()
#screening anything outside +/- 6 std dev
fence_low_stdW = meanW - 6*stdW
fence_high_stdW = meanW + 6*stdW
Work = Work.loc[(Work['Workday consumption'] > fence_low_stdW) & (Work['Workday
consumption'] < fence_high_stdW)]
compare_data = pd.concat([Home,Work]).reset_index(drop=True)
display(compare_data, len(Home),len(Work),len(compare_data))
sex age Mjob Fjob guardian
Workday
consumption
Weekend
consumption
absences
from
school
0 F 18
at_ho
me
teacher mother 1 1 6
1 F 17
at_ho
me
other father 1 1 4
2 F 15
at_ho
me
other mother 2 3 10
3 F 15
at_ho
me
other mother 1 1 8
4 F 16
at_ho
me
other mother 1 2 12
... ... ... ... ... ... ... ... ...