elenatorro/waterhack2018

Importamos librerias

import pandas as pd 
import matplotlib.pyplot as plt
%matplotlib inline

Households

general = pd.read_csv('/Users/servandotorres/Documents/data/households/general.csv')
economy =  pd.read_csv('/Users/servandotorres/Documents/data/households/economy.csv')
house_info = pd.read_csv('/Users/servandotorres/Documents/data/households/households_info.csv')
social_services_access = pd.read_csv('/Users/servandotorres/Documents/data/households/social_services_access.csv')
economy_1 = pd.read_csv('/Users/servandotorres/Documents/data/households/economy.csv', index_col='village')

Empezamos el analisis

Dataset general

general.head()

<style> .dataframe thead tr:only-child th { text-align: right; }

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

</style>

	id	gender	hh_head	relation_hh_head	gender_hh_head	district	ward	village	subvillage	house_condition
0	V4-01	Male	Yes	NaN	NaN	Muheza	Misalai	Kazita	Kwemuyu	The house is not good
1	V4-02	Female	Yes	NaN	NaN	Muheza	Misalai	Kazita	Kwemuyu	The house is not good
2	V4-03	Female	Yes	NaN	NaN	Muheza	Misalai	Kazita	Kwemuyu	The house is normal
3	V4-04	Female	Yes	NaN	NaN	Muheza	Misalai	Kazita	Kwemuyu	The house is normal
4	V4-05	Male	Yes	NaN	NaN	Muheza	Misalai	Kazita	Kwemuyu	Good house

general.isnull().sum()
#esta un poco sucio

id                    0
gender                0
hh_head               0
relation_hh_head    226
gender_hh_head      226
district              0
ward                  0
village               0
subvillage            0
house_condition      46
dtype: int64

general.head()
general.shape

(339, 10)

# Estadistica descriptiva 
general.describe()

<style> .dataframe thead tr:only-child th { text-align: right; }

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

</style>

	id	gender	hh_head	relation_hh_head	gender_hh_head	district	ward	village	subvillage	house_condition
count	339	339	339	113	113	339	339	339	339	293
unique	339	2	2	5	2	1	2	8	39	75
top	V3-09	Male	Yes	husband / wife	Male	Muheza	Misalai	Misalai	Mlalo	Normal house
freq	1	179	226	74	100	339	224	70	20	77

#Descricion categorica 
general.describe(include=[np.object, pd.Categorical]).T

<style> .dataframe thead tr:only-child th { text-align: right; }

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

</style>

	count	unique	top	freq
id	339	339	V3-09	1
gender	339	2	Male	179
hh_head	339	2	Yes	226
relation_hh_head	113	5	husband / wife	74
gender_hh_head	113	2	Male	100
district	339	1	Muheza	339
ward	339	2	Misalai	224
village	339	8	Misalai	70
subvillage	339	39	Mlalo	20
house_condition	293	75	Normal house	77

Economia

economy.head()

<style> .dataframe thead tr:only-child th { text-align: right; }

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

</style>

	id	village	acres_own_land	acres_rent_land	acres_not_paying_land	certificate	acres_cultivate_land	land_use	decision_maker	savings
0	V4-01	Kazita	7.0	0.0	0.0	No	7.0	Crop cultivation	father	Yes
1	V4-02	Kazita	0.5	0.0	0.0	No	0.5	Crop cultivation	mother	No
2	V4-03	Kazita	4.0	0.0	0.0	No	1.0	Crop cultivation	mother	Yes
3	V4-04	Kazita	1.5	0.0	0.0	Yes	1.5	Crop cultivation	both	Yes
4	V4-05	Kazita	0.0	0.0	0.0	No	0.0	NaN	father	Yes

Vision global de el tomador de decisiones

sns.countplot(y='decision_maker', data=economy)

<matplotlib.axes._subplots.AxesSubplot at 0x1131985c0>

#Valores nulos 
economy.isnull().sum()

id                       0
village                  0
acres_own_land           0
acres_rent_land          2
acres_not_paying_land    2
certificate              1
acres_cultivate_land     2
land_use                 6
decision_maker           0
savings                  0
dtype: int64

Superficie en propiedad por pueblo

ax = sns.barplot(x='village', y='acres_own_land', data=economy)
ax.figure.set_size_inches(16, 4)

ax = sns.barplot(x='village', y='acres_own_land', hue='land_use', 
                 data=economy, palette='Greys')
ax.figure.set_size_inches(16,4)

Propiedad

ax = sns.barplot(x='village', y='acres_own_land', data=economy)
ax.figure.set_size_inches(16, 4)

No paga

ax = sns.barplot(x='village', y='acres_not_paying_land', data=economy)
ax.figure.set_size_inches(16, 4)

Aquila

ax = sns.barplot(x='village', y='acres_rent_land', data=economy)
ax.figure.set_size_inches(16, 4)

house_info.head()

<style> .dataframe thead tr:only-child th { text-align: right; }

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

</style>

	id	village	household_member	gender	age	edu_level
0	V1-03	Misalai	1st member	male	80	Primary
1	V4-01	Kazita	2nd member	female	42	No formal education
2	V4-01	Kazita	3rd member	female	40	No formal education
3	V4-01	Kazita	1st member	male	55	Secundary
4	V4-01	Kazita	5th member	female	11	Primary

social_services_access.head()

<style> .dataframe thead tr:only-child th { text-align: right; }

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

</style>

	id	village	primary_school	service_quality	secundary_school	service_quality.1	market_place	service_quality.2	dispensary	service_quality.3	...	agriculture_storage	service_quality.8	agricultural_extension_service	service_quality.9	veterinary	service_quality.10	forest_service	service_quality.11	water_service	service_quality.12
0	V4-01	Kazita	yes	Bad	no	No	no	No	no	No	...	no	No	no	No	no	No	no	No	no	No
1	V4-02	Kazita	yes	Good	yes	Good	no	No	no	No	...	no	No	no	No	yes	Bad	yes	Bad	no	No
2	V4-03	Kazita	yes	Good	yes	Good	yes	Good	yes	Good	...	no	No	no	No	no	No	yes	Good	yes	Good
3	V4-04	Kazita	yes	Good	yes	Good	yes	Good	no	No	...	no	No	no	No	yes	Good	yes	Good	yes	Good
4	V4-05	Kazita	yes	Good	yes	Good	yes	Good	yes	Good	...	no	No	no	No	no	No	no	No	yes	Good

5 rows × 28 columns

Water access

drinking_water_distance = pd.read_csv('/Users/servandotorres/Documents/data/water-access/drinking_water_distance.csv')
water_collection = pd.read_csv('/Users/servandotorres/Documents/data/water-access/water_collection.csv')
water_collection = pd.read_csv('/Users/servandotorres/Documents/data/water-access/water_collection.csv')
water_payment = pd.read_csv('/Users/servandotorres/Documents/data/water-access/water_payment.csv')
water_quality = pd.read_csv('/Users/servandotorres/Documents/data/water-access/water_quality.csv')
water_sanitation = pd.read_csv('/Users/servandotorres/Documents/data/water-access/water_sanitation.csv')

drinking_water_distance['village']

import numpy as np
np.unique(drinking_water_distance['village'])

array(['8 =Kwelumbizi', 'Kazita', 'Kizerui', 'Kwemsoso', 'Mgambo',
       'Misalai', 'Shambangeda', 'Zirai'], dtype=object)

drinking_water_distance.head()

<style> .dataframe thead tr:only-child th { text-align: right; }

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

</style>

	id	village	rainy_season	dry_season
0	V4-01	Kazita	1- 2 hours	1- 2 hours
1	V4-02	Kazita	5 - 10 min	5 - 10 min
2	V4-03	Kazita	5 - 10 min	5 - 10 min
3	V4-04	Kazita	1- 2 hours	1- 2 hours
4	V4-05	Kazita	< 5 min	< 5 min

water_collection.head()
#water_collection.describe(include=[np.object, pd.Categorical]).T

<style> .dataframe thead tr:only-child th { text-align: right; }

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

</style>

	id	village	collector	no_buckets
0	V4-01	Kazita	Adult women	6 - 7
1	V4-02	Kazita	Adult women	1 - 2
2	V4-03	Kazita	Adult women	2 - 3
3	V4-04	Kazita	Adult women	5 - 6
4	V4-05	Kazita	Adult men	1 - 2

water_payment.head()

<style> .dataframe thead tr:only-child th { text-align: right; }

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

</style>

	id	village	rainy_season	dry_season
0	V4-01	Kazita	No	No
1	V4-02	Kazita	No	No
2	V4-03	Kazita	No	No
3	V4-04	Kazita	No	No
4	V4-05	Kazita	No	No

sns.countplot(y='dry_season', data=water_payment)

<matplotlib.axes._subplots.AxesSubplot at 0x11346a400>

sns.countplot(y='rainy_season', data=water_payment)

<matplotlib.axes._subplots.AxesSubplot at 0x1135ee710>

water_quality.describe(include=[np.object, pd.Categorical]).T

<style> .dataframe thead tr:only-child th { text-align: right; }

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

</style>

	count	unique	top	freq
id	339	339	V3-09	1
village	339	8	Misalai	70
dry_season	339	5	Clean	152
rainy_season	339	5	Clean	127

Water Sanitation

water_sanitation.head()

<style> .dataframe thead tr:only-child th { text-align: right; }

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

</style>

	id	village	floor_material	wall_material	roof_material	toilet_type	toilet_satisfaction	water_in_toilet	rubbish_pit	utensils_rack
0	V4-01	Kazita	earth	mud bricks	galvanised metal sheets / iron sheets	flush toilet with flushing system	fairly satisfied	Water bucket in the toilet	Yes	Yes
1	V4-02	Kazita	earth	mud bricks	galvanised metal sheets / iron sheets	simple pit latrine unimproved	unsatisifed	Water bucket in the toilet	Yes	Yes
2	V4-03	Kazita	earth	mud and poles or stones	galvanised metal sheets / iron sheets	flush toilet	very satisfied	Water bucket in the toilet	Yes	Yes
3	V4-04	Kazita	earth	mud and poles or stones	galvanised metal sheets / iron sheets	flush toilet with flushing system	fairly satisfied	No water in the toilet	No	Yes
4	V4-05	Kazita	earth	baked or burnt bricks	galvanised metal sheets / iron sheets	flush toilet with flushing system	fairly satisfied	Water bucket in the toilet	Yes	Yes

water_sanitation.describe(include=[np.object, pd.Categorical]).T

<style> .dataframe thead tr:only-child th { text-align: right; }

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

</style>

	count	unique	top	freq
id	339	339	V3-09	1
village	339	8	Misalai	70
floor_material	339	4	earth	261
wall_material	339	6	mud and poles or stones	180
roof_material	339	5	galvanised metal sheets / iron sheets	291
toilet_type	339	6	simple pit latrine unimproved	189
toilet_satisfaction	339	4	fairly satisfied	157
water_in_toilet	337	4	Water bucket in the toilet	311
rubbish_pit	339	2	Yes	211
utensils_rack	339	2	Yes	202

np.unique(water_quality['dry_season'])
water_quality.head()

<style> .dataframe thead tr:only-child th { text-align: right; }

.dataframe thead th {
    text-align: left;
}

.dataframe tbody tr th {
    vertical-align: top;
}

</style>

	id	village	dry_season	rainy_season
0	V4-01	Kazita	Quite clean	Quite clean
1	V4-02	Kazita	Quite clean	Clean
2	V4-03	Kazita	Clean	Clean
3	V4-04	Kazita	Clean	Clean
4	V4-05	Kazita	Quite clean	Quite clean

ax = sns.countplot(y='dry_season', data=water_quality)
ax

<matplotlib.axes._subplots.AxesSubplot at 0x11125cf98>

Calidad del agua por estacion

Sequía

sns.countplot(y='dry_season', data=water_quality)

<matplotlib.axes._subplots.AxesSubplot at 0x1a1acab860>

Lluvia

sns.countplot(y='rainy_season', data=water_quality)

<matplotlib.axes._subplots.AxesSubplot at 0x113643d30>