Project: Explore US Birth rates on different time scales¶
Questions to address:¶
- Analyse statistics related to the number of births in the US per day.
- The first step is to clean up the data and merge data files (data is separated into 1994 to 2003 and 2000 to 2014)
- Then different statistics are computed: births per day, per month, in different days of the week, etc.
Tools:¶
- read text from file and convert to list of strings
- manipulation of lists of strings
- frequency counts using dictionaries
- line and bar plots
Dataset: US birth rate per day (two files)¶
read the two files (US births on eachday from 1994 until 2014) and save the mean for the days where there is overlap
In [1]:
from IPython.display import display
def read_csv(file_name):
f=open(file_name,'r')
data=f.read()
string_list=data.split('\n')
header = string_list[0]
string_list=string_list[1:]
final_list=[]
for each_str in string_list:
int_fields=[]
string_fields=each_str.split(',')
for each_field in string_fields:
int_fields.append(int(each_field))
final_list.append(int_fields)
return (header, final_list)
header_1, cdc_list=read_csv('./data/US_births_1994-2003_CDC_NCHS.csv')
header_2,ssa_list=read_csv('./data/US_births_2000-2014_SSA.csv')
US_births = cdc_list
for each_day in ssa_list:
if each_day[0]>2003:
US_births.append(each_day)
else:
ii=0
for each_day_new in US_births:
if ((each_day_new[0]==each_day[0]) &
(each_day_new[1]==each_day[1]) &
(each_day_new[2]==each_day[2]) &
(each_day_new[3]==each_day[3])):
US_births[ii]=[each_day[0], each_day[1],
each_day[2], each_day[3],
(each_day[4]+each_day_new[4])/2]
ii+=1
print(header_1)
display(US_births[:12])
Analysis:¶
plotting defaults:
In [83]:
%matplotlib inline
import matplotlib.pyplot as plt
from matplotlib.ticker import MultipleLocator
colors = [(0/255,107/255,164/255), (255/255, 128/255, 14/255), 'red', 'green']
Create a function to count the number of births on any column of the data: year, month, date_of_month or day_of_week
In [84]:
def calc_birth_counts(input_list, column):
births_per_col={}
for each_element in input_list:
#each_element[4] is the number of births
birth_rate = each_element[4]
if each_element[column] in births_per_col:
births_per_col[each_element[column]]+=int(birth_rate)
else:
births_per_col[each_element[column]]=int(birth_rate)
return (births_per_col)
births_per_year = calc_birth_counts(US_births, 0)
births_per_month = calc_birth_counts(US_births, 1)
births_per_dom = calc_birth_counts(US_births, 2)
births_per_dow = calc_birth_counts(US_births, 3)
print(births_per_month)
Evolution of US birth rate over the years:¶
In [85]:
fig = plt.figure(figsize=(5,5))
ax = plt.subplot()
#axis mods
for key, spine in ax.spines.items():
if(key=='top' or key=='right'):
spine.set_visible(False)
ax.tick_params(left=False, right=False, top=False, bottom=False)
ax.xaxis.set_major_locator(MultipleLocator(5))
ax.set_xlim(1995, 2015); ax.set_ylim(3500000,4500000)
ax.set_xlabel('Year'); ax.set_ylabel('Number of Births')
ax.set_title('US Birth rate per year')
ax.plot(births_per_year.keys(), births_per_year.values(), c=colors[0], linewidth=3)
Out[85]:
Birth rates in the US peaked ~2007, this is likely related to the financial crisis.
US birth rate in different months:¶
In [88]:
fig = plt.figure(figsize=(5,5))
ax = plt.subplot()
#axis mods
for key, spine in ax.spines.items():
if(key=='top' or key=='right'):
spine.set_visible(False)
ax.tick_params(left=False, right=False, top=False, bottom=False)
ax.set_xlim(0, 13); ax.set_ylim(6400000,7700000)
ax.set_xlabel('Month'); ax.set_ylabel('Number of Births')
ax.set_title('US Birth rate in different Months')
Months = ['January', 'February', 'March', 'April', 'May', 'June',
'July', 'August', 'September', 'October', 'November', 'December']
tick_positions = range(1,13)
ax.set_xticks(tick_positions)
ax.set_xticklabels(Months, rotation=90)
ax.bar(births_per_month.keys(), births_per_month.values(),width=0.5, color=colors[0])
Out[88]:
In the years for which we have data available most babies are born in the summer. The low numbers in February are caused by the number of days in that month.
Explore the evolution of the number of births in a given month over the years:¶
In [89]:
def accross_years(input_list, column, element):
births_accross_years={}
for each_element in input_list:
birth_rate = each_element[4]
if each_element[column]==element:
if each_element[0] in births_accross_years:
births_accross_years[each_element[0]]+=birth_rate
else:
births_accross_years[each_element[0]]=birth_rate
return(births_accross_years)
births_in_a_month = []
for ii in range(1,12+1):
births_in_a_month.append(accross_years(US_births,1,ii))
fig = plt.figure(figsize=(5,5))
ax = plt.subplot()
#axis mods
for key, spine in ax.spines.items():
if(key=='top' or key=='right'):
spine.set_visible(False)
ax.tick_params(left=False, right=False, top=False, bottom=False)
ax.xaxis.set_major_locator(MultipleLocator(5))
ax.set_xlim(1995, 2015); ax.set_ylim(280000,400000)
ax.set_xlabel('Year'); ax.set_ylabel('Number of Births')
ax.set_title('US Birth rate in different months')
ax.plot(births_in_a_month[1].keys(), births_in_a_month[1].values(), c=colors[0], linewidth=3, label=Months[1])
ax.plot(births_in_a_month[7].keys(), births_in_a_month[7].values(), c=colors[1], linewidth=3, label=Months[7])
ax.plot(births_in_a_month[9].keys(), births_in_a_month[9].values(), c=colors[2], linewidth=3, label=Months[9])
plt.legend(loc='upper left')
Out[89]:
Explore the evolution of the number of births in different seasons over the years:¶
In [90]:
def accross_years(input_list, column, element):
births_accross_years={}
for each_element in input_list:
birth_rate = each_element[4]
if each_element[column] in [3*element+1, 3*element+2, 3*element+3]:
if each_element[0] in births_accross_years:
births_accross_years[each_element[0]]+=birth_rate
else:
births_accross_years[each_element[0]]=birth_rate
return(births_accross_years)
births_in_a_season = []
for ii in range(0,4):
births_in_a_season.append(accross_years(US_births,1,ii))
fig = plt.figure(figsize=(5,5))
ax = plt.subplot()
#axis mods
for key, spine in ax.spines.items():
if(key=='top' or key=='right'):
spine.set_visible(False)
ax.tick_params(left=False, right=False, top=False, bottom=False)
ax.xaxis.set_major_locator(MultipleLocator(5))
ax.set_xlim(1995, 2015); ax.set_ylim(900000,1200000)
ax.set_xlabel('Year'); ax.set_ylabel('Number of Births')
ax.set_title('US Birth rate in different months')
ax.plot(births_in_a_season[0].keys(), births_in_a_season[0].values(), c=colors[0], linewidth=3, label='Winter')
ax.plot(births_in_a_season[1].keys(), births_in_a_season[1].values(), c=colors[3], linewidth=3, label='Spring')
ax.plot(births_in_a_season[2].keys(), births_in_a_season[2].values(), c=colors[2], linewidth=3, label='Summer')
ax.plot(births_in_a_season[3].keys(), births_in_a_season[3].values(), c=colors[1], linewidth=3, label='Fall')
plt.legend(loc='upper left')
Out[90]:
There is a hint of an increase of Births in the fall in recent years. This could be related to people wanting their kids to be 6 years and 9-12 months rather than 6 years and 1-3 months when entering school.