dataanalysisandinterpretation

[dataanalysisandinterpretation]

Code 3

""" Created on Thu Jun 17 21:01:53 2021

@author: MariaRomo """

import pandas import numpy import seaborn import matplotlib.pyplot as plt

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

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

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

data['TAB12MDX'] = pandas.to_numeric(data['TAB12MDX']) data['CHECK321'] = pandas.to_numeric(data['CHECK321']) data['S3AQ3B1'] = pandas.to_numeric(data['S3AQ3B1']) data['S3AQ3C1'] = pandas.to_numeric(data['S3AQ3C1']) data['AGE'] = pandas.to_numeric(data['AGE'])

sub1=data[(data['AGE']>=18) & (data['AGE']<=25) & (data['CHECK321']==1)]

sub2 = sub1.copy()

sub2['S3AQ3B1']=sub2['S3AQ3B1'].replace(9, numpy.nan) sub2['S3AQ3C1']=sub2['S3AQ3C1'].replace(99, numpy.nan)

recode1 = {1: 6, 2: 5, 3: 4, 4: 3, 5: 2, 6: 1} sub2['USFREQ']= sub2['S3AQ3B1'].map(recode1)

recode2 = {1: 30, 2: 22, 3: 14, 4: 5, 5: 2.5, 6: 1} sub2['USFREQMO']= sub2['S3AQ3B1'].map(recode2)

sub2['NUMCIGMO_EST']=sub2['USFREQMO'] * sub2['S3AQ3C1']

sub2["TAB12MDX"] = sub2["TAB12MDX"].astype('category')

seaborn.countplot(x="TAB12MDX", data=sub2) plt.xlabel('Nicotine Dependence past 12 months') plt.title('Nicotine Dependence in the Past 12 Months Among Young Adult Smokers in the NESARC Study')

seaborn.distplot(sub2["NUMCIGMO_EST"].dropna(), kde=False); plt.xlabel('Number of Cigarettes per Month') plt.title('Estimated Number of Cigarettes per Month among Young Adult Smokers in the NESARC Study')

print ('describe number of cigarettes smoked per month') desc1 = sub2['NUMCIGMO_EST'].describe() print (desc1)

c1= sub2.groupby('NUMCIGMO_EST').size() print (c1)

print ('describe nicotine dependence') desc2 = sub2['TAB12MDX'].describe() print (desc2)

c1= sub2.groupby('TAB12MDX').size() print (c1)

p1 = sub2.groupby('TAB12MDX').size() * 100 / len(data) print (p1)

c2 = sub2.groupby('NUMCIGMO_EST').size() print (c2)

p2 = sub2.groupby('NUMCIGMO_EST').size() * 100 / len(data) print (p2)

sub2['PACKSPERMONTH']=sub2['NUMCIGMO_EST'] / 20

c2= sub2.groupby('PACKSPERMONTH').size() print (c2)

sub2['PACKCATEGORY'] = pandas.cut(sub2.PACKSPERMONTH, [0, 5, 10, 20, 30, 147])

# change format from numeric to categorical sub2['PACKCATEGORY'] = sub2['PACKCATEGORY'].astype('category')

print ('pack category counts') c7 = sub2['PACKCATEGORY'].value_counts(sort=False, dropna=True) print(c7)

print ('describe PACKCATEGORY') desc3 = sub2['PACKCATEGORY'].describe() print (desc3)

sub2['TAB12MDX'] = pandas.to_numeric(sub2['TAB12MDX'])

# bivariate bar graph C->Q seaborn.catplot(x="PACKCATEGORY", y="TAB12MDX", data=sub2, kind="bar", ci=None) plt.xlabel('Packs per Month') plt.ylabel('Proportion Nicotine Dependent')

Univariate: Number of cigarettes per month

Bivariate graphs: Correlation between Packs per months and Proportion of Nicotine Dependent

[dataanalysisandinterpretation]

Code 2

#!/usr/bin/env python3 # -*- coding: utf-8 -*- """ Created on Wed Jun  9 18:27:23 2021

@author: MariaRomo """

import pandas import numpy

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

#pasar todo a mayusculas o minusculas data.columns=map(str.upper, data.columns)

#runtime error, poner esto pandas.set_option('display.float_format',lambda x:'%f'%x)

#setting variables you will be working with to numeric data['TAB12MDX'] = pandas.to_numeric(data['TAB12MDX']) data['CHECK321'] = pandas.to_numeric(data['CHECK321']) data['S3AQ3B1'] = pandas.to_numeric(data['S3AQ3B1']) data['S3AQ3C1'] = pandas.to_numeric(data['S3AQ3C1']) data['AGE'] = pandas.to_numeric(data['AGE'])

#Filtrar información sub1= data[(data['AGE']>=18)&(data['AGE']<=25)&(data['CHECK321']==1)]

#copy of new subset data sub2=sub1.copy()

print('counts for S3AQ3B1') c5 = sub2['S3AQ3B1'].value_counts(sort=False) print(c5)

sub2['S3AQ3B1']= sub2['S3AQ3B1'].replace(9, numpy.nan)

print('counts for S3AQ3B1') c6 = sub2['S3AQ3B1'].value_counts(sort=False, dropna=False) print(c6)

print('counts for S3AQ3C1') c7 = sub2['S3AQ3C1'].value_counts(sort=False) print(c7)

#Quitar los 99 del códgio sub2['S3AQ3C1']= sub2['S3AQ3C1'].replace(99, numpy.nan)

print('counts for S3AQ3C1') c8 = sub2['S3AQ3C1'].value_counts(sort=False, dropna=False) print(c8)

data.loc[(data['S2AQ3']!=9)&(data['S2AQ8A'].isnull()),'S2AQ8A']=11

recode1={1:6, 2:5, 3:4, 4:3, 5:2, 6:1} sub2['USFREQ']=sub2['S3AQ3B1'].map(recode1)

recode2={1:30, 2:22, 3:14, 4:5, 5:2.5, 6:1} sub2['USFREQMO']=sub2['S3AQ3B1'].map(recode2)

print('counts for USFREQMO') c9 = sub2['USFREQMO'].value_counts(sort=False, dropna=False) print(c9)

sub2['NUMCIGMO_EST'] = sub2['USFREQMO'] * sub2['S3AQ3C1']

sub3= sub2[['IDNUM', 'S3AQ3C1', 'USFREQMO', 'NUMCIGMO_EST']] a=sub3.head(n=25) print(a)

sub2['AGEGROUP3']=pandas.cut(sub2.AGE,[17,20,22,25])   print(pandas.crosstab(sub2['AGEGROUP3'],sub2['AGE']))   

Output:

runfile('/Users/MariaRomo/Documents/NESARC/EjercicioMR2.py', wdir='/Users/MariaRomo/Documents/NESARC') counts for S3AQ3B1 9.000000       3 6.000000      71 3.000000      91 2.000000      68 5.000000      65 1.000000    1320 4.000000      88 Name: S3AQ3B1, dtype: int64 counts for S3AQ3B1 NaN            3 6.000000      71 3.000000      91 2.000000      68 5.000000      65 1.000000    1320 4.000000      88 Name: S3AQ3B1, dtype: int64 counts for S3AQ3C1 15.000000     99 98.000000      1 9.000000       6 20.000000    365 30.000000     38 19.000000      1 17.000000      2 80.000000      1 24.000000      1 11.000000      3 28.000000      1 35.000000      1 16.000000      5 99.000000      9 60.000000      2 6.000000      60 10.000000    387 3.000000     114 40.000000     30 12.000000     25 8.000000      42 13.000000      7 25.000000     13 2.000000     111 7.000000      45 27.000000      1 18.000000      3 5.000000     163 1.000000      83 14.000000      3 4.000000      84 Name: S3AQ3C1, dtype: int64 counts for S3AQ3C1 NaN            9 15.000000     99 98.000000      1 9.000000       6 20.000000    365 30.000000     38 19.000000      1 17.000000      2 80.000000      1 24.000000      1 11.000000      3 28.000000      1 35.000000      1 16.000000      5 60.000000      2 6.000000      60 10.000000    387 3.000000     114 40.000000     30 12.000000     25 8.000000      42 13.000000      7 25.000000     13 2.000000     111 7.000000      45 27.000000      1 18.000000      3 5.000000     163 1.000000      83 14.000000      3 4.000000      84 Name: S3AQ3C1, dtype: int64 counts for USFREQMO NaN             3 2.500000       65 30.000000    1320 22.000000      68 5.000000       88 1.000000       71 14.000000      91 Name: USFREQMO, dtype: int64     IDNUM   S3AQ3C1  USFREQMO  NUMCIGMO_EST 20      21  3.000000 30.000000     90.000000 76      77  3.000000 22.000000     66.000000 102    103 10.000000 30.000000    300.000000 121    122 10.000000 30.000000    300.000000 135    136 20.000000 30.000000    600.000000 149    150  5.000000 30.000000    150.000000 154    155  8.000000 30.000000    240.000000 173    174  1.000000 30.000000     30.000000 177    178 10.000000 30.000000    300.000000 183    184 20.000000 30.000000    600.000000 187    188  2.000000  5.000000     10.000000 209    210  3.000000 30.000000     90.000000 219    220  5.000000 14.000000     70.000000 222    223  1.000000 30.000000     30.000000 278    279 98.000000 30.000000   2940.000000 336    337 20.000000 30.000000    600.000000 363    364 20.000000 30.000000    600.000000 398    399  2.000000 22.000000     44.000000 412    413  5.000000 30.000000    150.000000 417    418 20.000000 30.000000    600.000000 508    509 30.000000 30.000000    900.000000 511    512  1.000000  2.500000      2.500000 519    520 20.000000 30.000000    600.000000 522    523 10.000000 30.000000    300.000000 529    530  4.000000 30.000000    120.000000 AGEGROUP3                                         (17, 20]   161  200  221    0    0    0    0    0 (20, 22]     0    0    0  239  228    0    0    0 (22, 25]     0    0    0    0    0  231  241  185

Summary:

In this code I used frequency distribution in all the variables to see the output and if the variable was doing what I was expecting.

I used recode, coding out missing data, creating secondary variables, aggregate, grouping and  split variables.

[dataanalysisandinterpretation]

Code

Code:

import pandas import numpy

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

#Upper or lower case data.columns=map(str.upper, data.columns)

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

print(len(data)) #imprime la cantidad de rows u observaciones print(len(data.columns)) #imprime la cantidad de columnas o variables

#Checking the format of the variables, use print to see the format data['ETHRACE2A'].dtype

#setting variables you will be working with to numeric data['TAB12MDX'] = pandas.to_numeric(data['TAB12MDX']) data['CHECK321'] = pandas.to_numeric(data['CHECK321']) data['S3AQ3B1'] = pandas.to_numeric(data['S3AQ3B1']) data['S3AQ3C1'] = pandas.to_numeric(data['S3AQ3C1']) data['AGE'] = pandas.to_numeric(data['AGE'])

#Another option for displaying observations or rows in a data frame #Counts and percentages (i.e. frequency distributions) for each variable

print('counts for TAB12MDX - nicotine dependence in the past 12 months, yes = 1') c1 = data['TAB12MDX'].value_counts(sort=False) print (c1)

print('percentages for TAB12MDX nicotine dependence in the past 12 months, yes = 1') p1 = data['TAB12MDX'].value_counts(sort=False, normalize=True) print (p1)

print('counts for CHECK321 - smoked in the past year, yes = 1') c2 = data['CHECK321'].value_counts(sort=False) print(c2)

print('percentages for CHECK321 smoked in the past year, yes = 1') p2 = data['CHECK321'].value_counts(sort=False, normalize=True) print (p2)

print('counts for S3AQ3B1 - usual frequency when smoked cigarettes') c3 = data['S3AQ3B1'].value_counts(sort=False) print(c3)

print('percentages for S3AQ3B1 usual frequency when smoked cigarettes') p3 = data['S3AQ3B1'].value_counts(sort=False, normalize=True) print (p3)

#dropna- to display missing data

print('counts for S3AQ3C1 - usual quantity when smoke cigarettes') c4 = data['S3AQ3C1'].value_counts(sort=False, dropna=False) print(c4)

print('percentages for S3AQ3C1 usual quantity when smoke cigarettes') p4 = data['S3AQ3C1'].value_counts(sort=False, dropna=False, normalize=True) print (p4)

#By groups is another approach to frequency distributions both counts and percetages

print('counts for TAB12MDX - nicotine dependence in the past 12 months, yes = 1') ct1= data.groupby('TAB12MDX').size() print(ct1)

print('percentages for TAB12MDX nicotine dependence in the past 12 months, yes = 1') pt1= data.groupby('TAB12MDX').size() * 100/len(data) print(pt1)

#Subset data sub1= data[(data['AGE']>=18)&(data['AGE']<=25)&(data['CHECK321']==1)] sub2=sub1.copy()

print('counts for AGE ') c5 = sub2['AGE'].value_counts(sort=False) print(c5)

print('percentages for AGE') p5 = sub2['AGE'].value_counts(sort=False, normalize=True) print (p5)

print('counts for CHECK321 - smoked in the past year, yes = 1') c6 = sub2['CHECK321'].value_counts(sort=False) print(c6)

print('percentages for CHECK321 smoked in the past year, yes = 1') p6 = sub2['CHECK321'].value_counts(sort=False, normalize=True) print (p6)

Output:

runfile('/Users/MariaRomo/Documents/NESARC/EjercicioMR1.py', wdir='/Users/MariaRomo/Documents/NESARC') 43093 3010 counts for TAB12MDX - nicotine dependence in the past 12 months, yes = 1 0    38131 1     4962 Name: TAB12MDX, dtype: int64 percentages for TAB12MDX nicotine dependence in the past 12 months, yes = 1 0   0.884854 1   0.115146 Name: TAB12MDX, dtype: float64 counts for CHECK321 - smoked in the past year, yes = 1 2.000000    8078 9.000000      22 1.000000    9913 Name: CHECK321, dtype: int64 percentages for CHECK321 smoked in the past year, yes = 1 2.000000   0.448454 9.000000   0.001221 1.000000   0.550325 Name: CHECK321, dtype: float64 counts for S3AQ3B1 - usual frequency when smoked cigarettes 4.000000      747 2.000000      460 5.000000      409 9.000000      102 3.000000      687 6.000000      772 1.000000    14836 Name: S3AQ3B1, dtype: int64 percentages for S3AQ3B1 usual frequency when smoked cigarettes 4.000000   0.041470 2.000000   0.025537 5.000000   0.022706 9.000000   0.005663 3.000000   0.038139 6.000000   0.042858 1.000000   0.823627 Name: S3AQ3B1, dtype: float64 counts for S3AQ3C1 - usual quantity when smoke cigarettes NaN          25080 28.000000        3 4.000000       573 60.000000      241 8.000000       299 7.000000       269 24.000000        7 22.000000       10 33.000000        1 50.000000      106 2.000000       884 34.000000        1 15.000000      851 5.000000      1070 98.000000       15 99.000000      262 40.000000      993 9.000000        49 55.000000        2 75.000000        2 39.000000        1 20.000000     5366 19.000000        5 29.000000        3 25.000000      155 45.000000        8 21.000000        1 35.000000       30 23.000000        2 14.000000       25 30.000000      909 70.000000       12 27.000000        2 37.000000        2 3.000000       923 6.000000       463 18.000000       59 17.000000       22 11.000000       23 10.000000     3077 57.000000        1 1.000000       934 66.000000        1 12.000000      230 13.000000       34 80.000000       47 16.000000       40 Name: S3AQ3C1, dtype: int64 percentages for S3AQ3C1 usual quantity when smoke cigarettes NaN         0.581997 28.000000   0.000070 4.000000    0.013297 60.000000   0.005593 8.000000    0.006938 7.000000    0.006242 24.000000   0.000162 22.000000   0.000232 33.000000   0.000023 50.000000   0.002460 2.000000    0.020514 34.000000   0.000023 15.000000   0.019748 5.000000    0.024830 98.000000   0.000348 99.000000   0.006080 40.000000   0.023043 9.000000    0.001137 55.000000   0.000046 75.000000   0.000046 39.000000   0.000023 20.000000   0.124521 19.000000   0.000116 29.000000   0.000070 25.000000   0.003597 45.000000   0.000186 21.000000   0.000023 35.000000   0.000696 23.000000   0.000046 14.000000   0.000580 30.000000   0.021094 70.000000   0.000278 27.000000   0.000046 37.000000   0.000046 3.000000    0.021419 6.000000    0.010744 18.000000   0.001369 17.000000   0.000511 11.000000   0.000534 10.000000   0.071404 57.000000   0.000023 1.000000    0.021674 66.000000   0.000023 12.000000   0.005337 13.000000   0.000789 80.000000   0.001091 16.000000   0.000928 Name: S3AQ3C1, dtype: float64 counts for TAB12MDX - nicotine dependence in the past 12 months, yes = 1 TAB12MDX 0    38131 1     4962 dtype: int64 percentages for TAB12MDX nicotine dependence in the past 12 months, yes = 1 TAB12MDX 0   88.485369 1   11.514631 dtype: float64 counts for AGE 18    161 19    200 20    221 21    239 22    228 23    231 24    241 25    185 Name: AGE, dtype: int64 percentages for AGE 18   0.094373 19   0.117233 20   0.129543 21   0.140094 22   0.133646 23   0.135404 24   0.141266 25   0.108441 Name: AGE, dtype: float64 counts for CHECK321 - smoked in the past year, yes = 1 1.000000    1706 Name: CHECK321, dtype: int64 percentages for CHECK321 smoked in the past year, yes = 1 1.000000   1.000000 Name: CHECK321, dtype: float64

[dataanalysisandinterpretation]

Association between smoking behavior and nicotine dependence

I'm a student and I want to understand the methodology and analysis with the support of the university. This is the reason why I decided to follow the example given.

Data set selected: NESARC

Question and hypothesis: 

How much does a person need to smoke to become nicotine dependent?

The more individuals smoke, the more likely they are to have nicotine dependence.

Variables:

Topic: Nicotine dependence

 Nicotine dependence in the last 12 months

 Nicotine dependence - Lifetime

2nd Topic: Smoking behavior 

 Cigarette smoking status

 Usual frequency when smoke cigarettes

 Usual quantity when smoke cigarettes 

Literature

In the research carried out, I found a work that focuses on deciphering the consumption by individuals who are nicotine dependent and those who have a psychiatric disorder. It is interesting for me to see the relationship between a psychiatric disorder and cigarette consumption, since we would expect from them a different behavior in almost every action due to their illness, but when talking about nicotine addiction, both behave in the same way, so I suppose there must be a pattern in people's behavior that leads the brain to demand this stimulus, even when it is under a psychiatric disorder. One of the conclusions of the article is that 70% of the cigarettes consumed in the United States come from people who are nicotine dependent and psychiatrically ill individuals.

Another article I read talks more about chemistry and how nicotine works, but what I found interesting was the correlation it explains between the chemistry and the smoking behavior nicotine can detach in people. It establishes that a gene found in nicotine exerts differential effects during the different stages of smoking, this information can give more details about a person's behavior before they become nicotine dependent.

Reference

Grant BF, Hasin DS, Chou SP, Stinson FS, Dawson DA. Nicotine Dependence and Psychiatric Disorders in the United States: Results From the National Epidemiologic Survey on Alcohol and RelatedConditions. Arch Gen Psychiatry. 2004;61(11):1107–1115. https://jamanetwork.com/journals/jamapsychiatry/article-abstract/482090

Viba Malaiyandi BSc, Edward M. Sellers MD, PhD, Rachel F. Tyndale PhD. Implications of CYP2A6 Genetic Variation for Smoking Behaviors and Nicotine Dependence. Perspectives in Clinical Pharmacology (2005) 77, 145–158; doi: 10.1016/j.clpt.2004.10.011