i-bashok-blog

[i-bashok-blog]

Data management and visualization, week 4

1. Program

import pandas import numpy

import seaborn import matplotlib.pyplot as plt

data  = pandas.read_csv ('addhealth_pds.csv', low_memory=False)

#set pandas to show all columns in dataframe

pandas.set_option('display.max_columns', None) #set pandas to show all rows in dataframe

pandas.set_option('display.max_rows', None)

#bug fix for display formats to avoid run time errors

pandas.set_option('display.float_format', lambda x:'%f'%x)

#record missing values to python missing (NaN) data["H1RE3"] = data ["H1RE3"].replace(6, numpy.nan) data["H1RE3"] = data ["H1RE3"].replace(7, numpy.nan) data["H1RE3"] = data ["H1RE3"].replace(8, numpy.nan) data["H1RE6"] = data ["H1RE6"].replace(6, numpy.nan) data["H1RE6"] = data ["H1RE6"].replace(8, numpy.nan) data["H1RE6"] = data ["H1RE6"].replace(7, numpy.nan)

data["H1ID1O"] = data ["H1ID1O"].replace(6, numpy.nan) data["H1ID1O"] = data ["H1ID1O"].replace(9, numpy.nan) data["H1ID1Q"] = data ["H1ID1Q"].replace(6, numpy.nan) data["H1ID1Q"] = data ["H1ID1Q"].replace(9, numpy.nan)

data["H1ID1Q"] = data ["H1ID1Q"].replace(8, numpy.nan)

recode1 = {1: 4, 2: 2, 3: 1, 4:0} data["USATTD"]= data["H1RE3"].map(recode1)

recode2 = {1: 48, 2: 24, 3: 12, 4:0} data["USATTDY"]= data["H1RE3"].map(recode2) print ('counts for USATTDY') c5 = data["USATTDY"].value_counts(sort=False, dropna=False) print (c5)

print ('percentages for USATTDY') p5 = data["USATTDY"].value_counts(sort=False, normalize=True) print (p5)

recode3 = {1: 30, 2: 4, 3: 1, 4:0} data["USPRAY"]= data["H1RE6"].map(recode3)

recode4 = {1: 365, 2: 48, 3: 12, 4:0} data["USPRAYY"]= data["H1RE6"].map(recode4) print ('counts for USPRAYY') c6 = data["USPRAYY"].value_counts(sort=False, dropna=False)

#univariable bar graph for categorical variables #first hange format from numeric to categoracal

sub2= data.copy()

sub2 ["USPRAYY"] = sub2 ["USPRAYY"].astype('category')

seaborn.countplot(x="USPRAYY", data = sub2) plt.xlabel ('Types of frequency prays') plt.title ('Types of frequency prays')

print  ('describe number of prays')

desc1= sub2 ["USPRAYY"].describe() print (desc1)

c1= sub2.groupby("USPRAYY").size()

print(c1)

print ('mode') mode1 = sub2["USPRAYY"].mode() print(mode1)

p1 = sub2.groupby("USPRAYY").size() * 100/ len(data) print (p1)

sub2 ["H1ID1Q"] = sub2 ["H1ID1Q"].astype('category')

seaborn.countplot(x="H1ID1Q", data = sub2) plt.xlabel ('Possibility of marriage') plt.title ('Possibility of marriage')

print  ('describe possibility of marriage')

desc2= sub2 ["H1ID1Q"].describe() print (desc2)

c2= sub2.groupby("H1ID1Q").size()

print(c1)

print ('mode') mode2 = sub2["H1ID1Q"].mode() print(mode2)

p2 = sub2.groupby("H1ID1Q").size() * 100/ len(data) print (p2) print (desc2)

sub2 ["H1ID1O"] = sub2 ["H1ID1O"].astype('category')

seaborn.countplot(x="H1ID1O", data = sub2) plt.xlabel ('Possibility of sex') plt.title ('Possibility of sex')

print  ('describe possibility of sex')

desc3= sub2 ["H1ID1O"].describe() print (desc3)

c3= sub2.groupby("H1ID1O").size()

print(c3)

print ('mode') mode3 = sub2["H1ID1O"].mode() print(mode3)

p3 = sub2.groupby("H1ID1O").size() * 100/ len(data) print (p3) print (desc3)

2. Outcome

counts for USATTDY

48.000000    2567

nan           894

12.000000    1105

0.000000      664

24.000000    1274

Name: USATTDY, dtype: int64

percentages for USATTDY

48.000000   0.457576

12.000000   0.196970

0.000000    0.118360

24.000000   0.227094

Name: USATTDY, dtype: float64

counts for USPRAYY

describe number of prays

count    5174.000000

unique      4.000000

top       365.000000

freq     2722.000000

Name: USPRAYY, dtype: float64

USPRAYY

0.000000       496

12.000000      586

48.000000     1370

365.000000    2722

dtype: int64

mode

0   365.000000

Name: USPRAYY, dtype: category

Categories (4, float64): [0.000000, 12.000000, 48.000000, 365.000000]

USPRAYY

0.000000      7.626076

12.000000     9.009840

48.000000    21.063961

365.000000   41.851169

dtype: float64

 mode

0   2.000000

Name: H1ID1Q, dtype: category

Categories (2, float64): [1.000000, 2.000000]

H1ID1Q

1.000000   22.893604

2.000000   75.199877

dtype: float64

count    6380.000000

unique      2.000000

top         2.000000

freq     4891.000000

Name: H1ID1Q, dtype: float64

count    5174.000000

unique      4.000000

top       365.000000

freq     2722.000000

Name: USPRAYY, dtype: float64

  Name: H1ID1Q, dtype: float64

describe possibility of sex

count    6436.000000

unique      3.000000

top         2.000000

freq     3533.000000

Name: H1ID1O, dtype: float64

H1ID1O

1.000000    2847

2.000000    3533

8.000000      56

dtype: int64

mode

0   2.000000

Name: H1ID1O, dtype: category

Categories (3, float64): [1.000000, 2.000000, 8.000000]

H1ID1O

1.000000   43.773063

2.000000   54.320418

8.000000    0.861009

dtype: float64

count    6436.000000

unique      3.000000

top         2.000000

freq     3533.000000

Name: H1ID1O, dtype: float64

[i-bashok-blog]

DATA MANAGEMENT, WEEK3

1. PROGRAM

import pandas

import numpy

data = pandas.read_csv ('addhealth_pds.csv', low_memory=False)

print ('counts for original H1RE3') c1 = data["H1RE3"].value_counts(sort=False)

print ('counts for original H1RE6')

c2 = data["H1RE6"].value_counts(sort=False)

print ('counts for original H1ID1O')

c3 = data["H1ID1O"].value_counts(sort=False)

print ('counts for original H1ID1Q')

c4 = data["H1ID1Q"].value_counts(sort=False) print (c1) print(c2) print(c3) print(c4)

# set missing data to nan data["H1RE3"] = data ["H1RE3"].replace(6, numpy.nan) data["H1RE3"] = data ["H1RE3"].replace(7, numpy.nan) data["H1RE3"] = data ["H1RE3"].replace(8, numpy.nan) data["H1RE6"] = data ["H1RE6"].replace(6, numpy.nan) data["H1RE6"] = data ["H1RE6"].replace(8, numpy.nan) data["H1RE6"] = data ["H1RE6"].replace(7, numpy.nan)

data["H1ID1O"] = data ["H1ID1O"].replace(6, numpy.nan) data["H1ID1O"] = data ["H1ID1O"].replace(9, numpy.nan)

data["H1ID1O"] = data ["H1ID1O"].replace(8, numpy.nan)

data["H1ID1Q"] = data ["H1ID1Q"].replace(6, numpy.nan) data["H1ID1Q"] = data ["H1ID1Q"].replace(9, numpy.nan) data["H1ID1Q"] = data ["H1ID1Q"].replace(8, numpy.nan)

print ('counts for  H1RE3 with 6,7 and 8 set to nan') p1 = data["H1RE3"].value_counts(sort=False, dropna=False)

print ('counts for  H1RE6 with 6,7 and 8 set to nan')

p2 = data["H1RE6"].value_counts(sort=False, dropna=False)

print ('counts for  H1ID1O with 6, 8 and 9 set to nan')

p3 = data["H1ID1O"].value_counts(sort=False, dropna=False)

print ('counts for  H1ID1Q with 6, 8 and 9 set to nan')

p4 = data["H1ID1Q"].value_counts(sort=False, dropna=False)

print(p1) print (p2) print (p3) print (p4)

recode1 = {1: 4, 2: 2, 3: 1, 4:0} data["USATTD"]= data["H1RE3"].map(recode1)

recode2 = {1: 48, 2: 24, 3: 12, 4:0} data["USATTDY"]= data["H1RE3"].map(recode2) print ('counts for USATTDY') c5 = data["USATTDY"].value_counts(sort=False, dropna=False) print (c5)

print ('percentages for USATTDY') p5 = data["USATTDY"].value_counts(sort=False, normalize=True) print (p5)

recode3 = {1: 30, 2: 4, 3: 1, 4:0} data["USPRAY"]= data["H1RE6"].map(recode3)

recode4 = {1: 365, 2: 48, 3: 12, 4:0} data["USPRAYY"]= data["H1RE6"].map(recode4) print ('counts for USPRAYY') c6 = data["USPRAYY"].value_counts(sort=False, dropna=False) print (c6)

print ('percentages for USPRAYY') p6 = data["USPRAYY"].value_counts(sort=False, normalize=True) print (p6)

sub3 = data[["USATTDY", "USPRAYY"]]

sub3.head(25) print(sub3)

2. OUTPUT

counts for  H1RE3 with 6 and 8 set to nan

nan   894

1.0  2567

2.0  1274

3.0  1105

4.0  664

Name: H1RE3, dtype: int64

counts for  H1RE6 with 6,7 and 8 set to nan

nan  890

1.0   2722

2.0   1370

3.0     586

4.0     496

5.0    440

Name: H1RE6, dtype: int64

counts for  H1ID1O with 6, 8 and 9 set to nan

nan  124

1.0   2847

2.0   3533

Name: H1ID1O, dtype: int64

counts for  H1ID1Q with 6, 8 and 9 set to nan

nan   124

1.0   1489

2.0    4891

Name: H1ID1Q, dtype: int64 counts for USATTDY

nan           894

0.0   664

12.0  1105

24.0  1274

48.0  2567

Name: USATTDY, dtype: int64

percentages for USATTDY

0.0  0.118360

12.0   0.196970

24.0   0.227094

48.0  0.457576

Name: USATTDY, dtype: float64 counts for USPRAYY

nan           1330

0.0     496

12.0    586

48.0    1370

365.0    2722

Name: USPRAYY, dtype: int64 percentages for USPRAYY

0.0     0.095864

12.0   0.113259

48.0   0.264785

365.0  0.526092

Name: USPRAYY, dtype: float64

3. Managing results

I have managed the missing data in all the variables I have chosen, have coded and recoded data. 

I have created 2 secondary variables but didn’t managed to group values within individual variables and make columns as it made no sense within my research  question.

4. The frequency distributions for variables

52,6% percent of the respondents pray every day (365 days per year), 46% attend religious services 48 times per year. 

Managing my data set is impossible to indicate the correlation between religious services attendance and romantic relationship so far.

[i-bashok-blog]

Data Management  Week2

Program

import pandas import numpy

data = pandas.read_csv('addhealth_pds.csv', low_memory=False)

data.columns = map(str.upper, data.columns)

print(len(data)) print(len(data.columns)) import pandas

import pandas

print('counts for H1RE3 frequency religious servises attendance for the past 12 months')

c1 = data["H1RE3"].value_counts(sort=False)

print (c1) print('counts for H1RE3 frequency religious servises attendance for the past 12 months, percentage')

p1 = data["H1RE3"].value_counts(sort=False, normalize=True) print (p1)

print('counts for H1ID1O would have sex in ideal romantic relarionship')

c2 = data["H1ID1O"].value_counts(sort=False) print (c2)

print('counts for H1ID1O would have sex in ideal romantic relarionship, percentage')

p2 = data["H1ID1O"].value_counts(sort=False, normalize=True) print (p2)

print('counts for H1ID1Q would get married in ideal romantic relarionship')

c3 = data["H1ID1Q"].value_counts(sort=False) print (c3)

print('counts for H1ID1Q would get married in ideal romantic relarionship, percentage')

p3 = data["H1ID1Q"].value_counts(sort=False, normalize=True) print (p3)

print('counts for H1RE6 the frequency of prays, 1 once a day, 2 once a week')

c4 = data["H1RE6"].value_counts(sort=False) print (c4)     print('counts for H1RE6 the frequency of prays, 1 once a day, 2 once a week,percentage')

p4 = data["H1RE6"].value_counts(sort=False, normalize=True) print (p4)

ct1 = data.groupby('H1RE3').size() print(ct1) pt1 = data.groupby('H1RE3').size() * 100 / len(data) print(pt1)

ct2 = data.groupby('H1ID1O').size() print(ct2) pt2 = data.groupby('H1ID1O').size() * 100 / len(data) print(pt2)

ct3 = data.groupby('H1ID1Q').size() print(ct3)

pt3 = data.groupby('H1ID1Q').size() * 100 / len(data) print(pt3)

ct4 = data.groupby('H1RE6').size()

print(ct4) pt4 = data.groupby('H1RE6').size() * 100 / len(data) print(pt4)

Frequency tables

H1RE3 1    2567 2    1274 3    1105 4     664 6       8 7     879 8       7 dtype: int64 H1RE3 1    39.468020 2    19.587946 3    16.989545 4    10.209102 6     0.123001 7    13.514760 8     0.107626 dtype: float64 H1ID1O 1    2847 2    3533 6      63 8      56 9       5 dtype: int64 H1ID1O 1    43.773063 2    54.320418 6     0.968635 8     0.861009 9     0.076876 dtype: float64 H1ID1Q 1    1489 2    4891 6      63 8      56 9       5 dtype: int64 H1ID1Q 1    22.893604 2    75.199877 6     0.968635 8     0.861009 9     0.076876 dtype: float64 H1RE6 1    2722 2    1370 3     586 4     496 5     440 6       6 7     879 8       5 dtype: int64 H1RE6 1    41.851169 2    21.063961 3     9.009840 4     7.626076 5     6.765068 6     0.092251 7    13.514760 8     0.076876 dtype: float64

Among 6504 responds almost 60% attends religion services and pray at least once per week.

43% will have sex in their ideal romantic relationship and only 22% imagine to get married.

We have almost no missing data among these questions.

[i-bashok-blog]

Coursera. Data Management and Visualization.

Assignment. Week1.

1. I have chosen the following data set: 

The National Longitudinal Study of Adolescent Health

(AddHealth) is a representative school-based survey of adolescents in grades 7-12 in the United States. 

2. My research question is

Does religion influence the pregnancy norms, ideal romantic relationship and marital timing of adolescents.

3. Hypotheses:

 Does the frequency of religious service’s attendance influence the ideal romantic relationship

Do adolescents who identify with a religious group tend to have stronger proscriptions regarding aspects of sexuality and family planning

Does the frequency of praying and teenager’s special activities in churches influence the pregnancy norms.

Does religious teenagers tend to have sex with their partners only if they treat them as future partner 

The more religious is the teenager the earlier is the ideal marriage age 

4. Variables and codebook

RELIGION

H1RE1 What is your religion

H1RE2 Do you agree or  disagree that the sacred scriptures of your religion are the word of God and  are completely without any mistakes? 

H1RE3 In the past 12 months,  how often did you attend religious services?

 H1RE4 How  important is religion to you?

H1RE6 How often do you pray? 

 H1RE7 Many churches,  synagogues, and other places of worship have special activities for  teenagers—such as youth groups, Bible classes, or choir. In the past 12  months, how often did you attend such youth activities?  

  PREGNENCY

H1RP1  Getting (If R is male,  add: someone) pregnant at this time in your life is one of the worst things that could happen to you.

H1RP2  It wouldn’t be all that bad if you got (IF R is male,  add: someone) pregnant at  this time in your life. 

 H1RP3  If you got the AIDS virus, you would suffer a great  deal. 

H1RP4  It would be a big hassle to do the things necessary  to completely protect yourself from getting a sexually transmitted disease. 

H1RP5  Imagine that sometime soon you were to have sexual  intercourse with someone just once, but were unable to use any method of  birth control for some reason. What is the chance that you would get (If R is male, add:  your partner) pregnant? 

H1RP6  Suppose that sometime soon you had sexual intercourse  for a whole month, as often as you wanted to, without using any protection.  What is the chance that you would get the AIDS virus? 

 IDEAL RELATIONSHIP

H1ID1A  We would go out together in a group. 

 H1ID1B  I would meet my partner’s parents.  H1ID1C I would tell other people that we were a couple. 

H1ID1D  I would see less of my other friends so I could spend  more time with my partner. 

 H1ID1E  We would go out together alone.  

H1ID1F  We would hold  hands. 

 H1ID1G  I would give my partner a present.  

H1ID1H  My partner would give me a present. 

 H1ID1I  I would tell my partner that I loved him or her. 

 H1ID1J  My partner would tell me that he or she loved me. 

 H1ID1K  We would think of ourselves as a couple. 

 H1ID1L We would talk about contraception or sexually  transmitted diseases. 

H1ID1M  We  would  kiss.

H1ID1N  We would touch each other under our clothing or with  no clothes on. 

H1ID1O  We would have  sex.  

H1ID1P  My partner or I would get pregnant. 

 H1ID1Q  We would get  married.   

5. I found the following research on my topic:

Racial, Socioeconomic, and Religious Influences on School-Level Teen Pregnancy Norms and Behaviors

http://www.academia.edu/2740650/Racial_Socioeconomic_and_Religious_Influences_on_School-Level_Teen_Pregnancy_Norms_and_Behaviors

The Influence of Religion in Adolescence on Adolescents’ Attitude toward Marital Timing

https://repositories.lib.utexas.edu/bitstream/handle/2152/ETD-UT-2012-08-6051/REDFORD-THESIS.pdf?sequence=1

6. Conclusions after the research.

Individuals who identify with a religious group tend to have stronger proscriptionsregarding aspects of sexuality including pornography ), family planning, and specific gender roles that may influence group norms regarding teen pregnancy.

The higher maternal religious attendance and salience levels, the less likely their children were to accept nonmarital sex or cohabitation.

 Maternal religiosity and affiliation frequently influenced adolescents’ attitude toward marital timing similarly to adolescent religiosity and affiliation.