Green Space and Performance

Explore the relationship between green space access and different health conditions

import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
import seaborn as sns
from sklearn.preprocessing import OneHotEncoder

# load data
health = pd.read_csv('../data/CitieSHealth_BCN_DATA_PanelStudy_20220414.csv')
health.head()

	ID_Zenodo	date_all	year	month	day	dayoftheweek	hour	mentalhealth_survey	occurrence_mental	bienestar	...	education	covid_work	covid_mood	covid_sleep	covid_espacios	covid_aire	covid_motor	covid_electric	covid_bikewalk	covid_public_trans
0	71	22190	2020	10	2	4	18	Yes	1.0	9.0	...	Universitario	Ha empeorado mucho	Ha empeorado mucho	Ha empeorado un poco	Le doy más importancia que antes	Le doy más importancia que antes	Lo utilizo igual que antes	Lo utilizo igual que antes	Lo utilizo más que antes	Lo utilizo igual que antes
1	112	22202	2020	10	14	2	22	Yes	2.0	8.0	...	Universitario	Ha empeorado mucho	Ha empeorado un poco	No ha cambiado	Le doy más importancia que antes	Le doy más importancia que antes	Lo utilizo igual que antes	NaN	Lo utilizo más que antes	Lo utilizo menos que antes
2	110	22217	2020	10	29	3	18	Yes	10.0	9.0	...	Universitario	Ha empeorado mucho	No ha cambiado	No ha cambiado	Le doy más importancia que antes	Le doy más importancia que antes	Lo utilizo igual que antes	Lo utilizo más que antes	Lo utilizo más que antes	Lo utilizo menos que antes
3	115	22222	2020	11	3	1	18	Yes	14.0	3.0	...	Universitario	Ha empeorado mucho	Ha empeorado un poco	Ha empeorado un poco	Le doy más importancia que antes	Le doy más importancia que antes	Lo utilizo igual que antes	Lo utilizo igual que antes	Lo utilizo igual que antes	Lo utilizo menos que antes
4	135	22231	2020	11	12	3	22	Yes	12.0	9.0	...	Universitario	Ha empeorado mucho	Ha empeorado un poco	Ha empeorado un poco	Le doy más importancia que antes	No ha cambiado	NaN	NaN	Lo utilizo menos que antes	Lo utilizo más que antes

5 rows × 95 columns

# The list of variables that we are intersted in
variables = ['access_greenbluespaces_300mbuff',
             'performance', 'estres', 'energia', 'sueno', 
             'horasfuera'] #'dieta', 'alcohol', 'drogas', 'enfermo', 'smoke', 'psycho',]

renamed_variables = {'access_greenbluespaces_300mbuff': 'access_greenblue_spaces',
                     'estres': 'stress_level', 'energia': 'energy_level',
                     'sueno':'sleep_quality', 'horasfuera': 'time _outdoors',
                     'dieta': 'diet', 'drogas':'drug_use', 'enfermo': 'illness'
                     }

health_clean = health.loc[:, variables].rename(columns = renamed_variables).dropna()

health_clean.head()

	access_greenblue_spaces	performance	stress_level	energy_level	sleep_quality	time _outdoors
1	Yes	51.240993	5.0	9.0	8.0	5.0
2	Yes	56.006067	8.0	5.0	9.0	8.0
3	Yes	58.177117	5.0	2.0	2.0	5.0
4	Yes	71.481757	6.0	9.0	2.0	22.0
6	No	46.047583	5.0	5.0	10.0	3.0

def ohe(data, column):
    enc = OneHotEncoder()
    enc.fit(data[column])
    encoded_data = pd.DataFrame(enc.transform(data[column]).toarray().astype(int))
    encoded_data.columns = enc.get_feature_names_out()
    encoded_data = encoded_data.set_index(data.index)
    return encoded_data

numerical = list((health_clean.dtypes[health_clean.dtypes == 'float64'].index) | (health_clean.dtypes[health_clean.dtypes == 'int64'].index))
categorical = list((health_clean.dtypes[health_clean.dtypes != 'float64'].index) & (health_clean.dtypes[health_clean.dtypes != 'int64'].index))

one_hot_health_clean = pd.concat([health_clean[numerical], ohe(health_clean[categorical], categorical)], axis=1)
one_hot_health_clean.head()

/tmp/ipykernel_9643/101148290.py:9: FutureWarning: Index.__or__ operating as a set operation is deprecated, in the future this will be a logical operation matching Series.__or__.  Use index.union(other) instead.
  numerical = list((health_clean.dtypes[health_clean.dtypes == 'float64'].index) | (health_clean.dtypes[health_clean.dtypes == 'int64'].index))
/tmp/ipykernel_9643/101148290.py:10: FutureWarning: Index.__and__ operating as a set operation is deprecated, in the future this will be a logical operation matching Series.__and__.  Use index.intersection(other) instead.
  categorical = list((health_clean.dtypes[health_clean.dtypes != 'float64'].index) & (health_clean.dtypes[health_clean.dtypes != 'int64'].index))

	performance	stress_level	energy_level	sleep_quality	time _outdoors	access_greenblue_spaces_No	access_greenblue_spaces_Yes
1	51.240993	5.0	9.0	8.0	5.0	0	1
2	56.006067	8.0	5.0	9.0	8.0	0	1
3	58.177117	5.0	2.0	2.0	5.0	0	1
4	71.481757	6.0	9.0	2.0	22.0	0	1
6	46.047583	5.0	5.0	10.0	3.0	1	0

sns.pairplot(one_hot_health_clean, hue = "access_greenblue_spaces_Yes")
plt.savefig('../outputs/relation_btw_health_and_green')

sns.violinplot(data = health_clean, y = 'performance', x = 'access_greenblue_spaces');
plt.savefig('../outputs/performance vs green_access')

sns.histplot(data = health_clean, 
             x = 'stress_level', 
             hue = 'access_greenblue_spaces', 
             kde = True,
             stat = "density", 
             common_norm = False);
plt.savefig('../outputs/hist_of_stress_level')

sns.histplot(data = health_clean, 
             x = 'energy_level', 
             hue = 'access_greenblue_spaces', 
             kde = True,
             stat = "density", 
             common_norm = False);	
plt.savefig('../outputs/hist_of_energy_level')

sns.histplot(data = health_clean, 
             x = 'sleep_quality', 
             hue = 'access_greenblue_spaces', 
             kde = True,
             stat = "density", 
             common_norm = False);
plt.savefig('../outputs/hist_of_sleep_quality')

sns.histplot(data = health_clean, 
             x = 'time _outdoors', 
             hue = 'access_greenblue_spaces', 
             kde = True,
             stat = "density", 
             common_norm = False);
plt.savefig('../outputs/hist_of_outdoor_time')