data-management-konny86

[data-management-konny86]

Summary description of the following solution for my 4rd assignment:

The first graph (Do you have children?) shows how many of the survey respondents have children or no children. NaN are not included.

The second graph (How many children do you have?) shows with a recoded result, that the biggest count of the survey respondents have 1-3 children, followed by 4-6 and last but not least 7-10.

The third graph (Where do your children live?) shows that the biggest count of the children live with both parents, followed by children living with at least one parent, and those who live without their parents.

The fourth graph (Which of the groups of children live predominantely with both parents) shows the following relationship:

Children in a family with totaly 9 kids, almost all live with both parents. Children in a family with totaly 2 or 6 kids, live with approximately 60 % with both parents.  None Children in a family with totaly 8 kids, do live with both parents. The reason for this seems to be the low number of respondents having 8 children. The rest of the children live in different frequency (like shown on the graph) with both parents.

[data-management-konny86]

Output of my 4th assignment:

[data-management-konny86]

Creating graphs for your data

Here is my solution for the 4th assignment.

The code:

#!/usr/bin/env python3

# -*- coding: utf-8 -*- """ Created on Wed Jun  9 14:17:38 2021 @author: konny My first Python program with the selected data set in relation to my research question """ import pandas import numpy import seaborn import matplotlib.pyplot as plt #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) data = pandas.read_csv('ool_pds.csv', low_memory=False) #upper-case all DataFrame column names data.columns = map(str.upper, data.columns) print() print('Content of the ool_pds.csv dataset:') print('rows:', len(data))            # Number of observations (rows)  print('columns:', len(data.columns))   # Number of variables (colums)   print('\n=============================================================================') # setting variables I will be working with to numeric (updated) data['W1_P17'] = pandas.to_numeric(data['W1_P17'])      #Do you have any biological or adopted children data['W1_P17A'] = pandas.to_numeric(data['W1_P17A'])    #How many children do you have data['W1_P19'] = pandas.to_numeric(data['W1_P19'])      #Where do your minor children (under 18) live # replace missing values to python missing (NaN) data['W1_P17']=data['W1_P17'].replace(-1, numpy.nan) data['W1_P17A']=data['W1_P17A'].replace(-1, numpy.nan) data['W1_P19']=data['W1_P19'].replace(-1, numpy.nan) #counts and percentages (i.e. frequency distributions) for each variable print('\ncounts for W1_P17 - \nDo you have any biological or adopted children, yes=1, no=2') c1 = data['W1_P17'].value_counts(sort=True, dropna=False) print (c1) print('\npercentages for W1_P17 - \nDo you have any biological or adopted children, yes=1, no=2') p1 = data['W1_P17'].value_counts(sort=True, normalize=True, dropna=False) print (p1,'\n') data['HAVECHILDS'] = data['W1_P17'] c1a=data.groupby('HAVECHILDS').size() print(c1a) data['HAVECHILDS'] = data['HAVECHILDS'].astype('category') print('\nDescription: People have children') desc1 = data['HAVECHILDS'].describe() print(desc1) print('\nHave childs category') c2a = data['HAVECHILDS'].value_counts(sort=False) print(c2a) plot1=seaborn.countplot(x='HAVECHILDS', data=data) plt.xlabel('yes=1, no=2') plt.title('Do you have children?') plt.show(plot1) print('\n=============================================================================') print('\ncounts for W1_P17A - Original data\nHow many biological or adopted children do you have?') c2 = data['W1_P17A'].value_counts(sort=True, dropna=False) print (c2) print('\npercentages for W1_P17A - Original data\nHow many biological or adopted children do you have?') p2 = data['W1_P17A'].value_counts(sort=True, normalize=True, dropna=False) print (p2) # recode the childrengroup recode2 = {1: '1-3', 2: '1-3', 3: '1-3', 4: '4-6', 5: '4-6', 6: '4-6', 7: '7-10', 8: '7-10', 9: '7-10', 10: '7-10'} data['CHILDRENGROUP']= data['W1_P17A'].map(recode2) print('\nHow many children do you have? - Recoding and grouping the data') c5 = data['CHILDRENGROUP'].value_counts(sort=True) print (c5) print('\nHow many children do you have? - Recoding and grouping the data - percentages') p5 = data['CHILDRENGROUP'].value_counts(sort=True, normalize=True) print (p5) # First change format from numeric to categorical plot2=seaborn.countplot(x="CHILDRENGROUP", data=data) plt.xlabel('Number of children') plt.title('How many children do you have?') plt.show(plot2) print('\n=============================================================================') print('\ncounts of W1_P19 - Original data\nWhere do your minor children live? \nLive with both parents=1\nLive with father or mother (different expressions)=2-4\nLive with no parent=5') c3 = data['W1_P19'].value_counts(sort=True, dropna=False) print (c3)15:50

print('\npercentages of W1_P19 - - Original data\nWhere do your minor children live? \nLive with both parents=1\nLive with father or mother (different expressions)=2-4\nLive with no parent=5') p3 = data['W1_P19'].value_counts(sort=True, dropna=False, normalize=True) print (p3) data['P3']=p3 # recode the parentgroup recode1 = {1: 'both parents', 2: 'father or mother', 3: 'father or mother', 4: 'father or mother', 5: 'relatives'} data['PARENTGROUP']= data['W1_P19'].map(recode1) #data['PARENTGROUP'] = data.apply(lambda row: PARENTGROUP(row),axis=1) # live with at least one of the parents (father/mother/both) # building a group print('\nWhere do your children live? - Recoding and grouping the data') c4 = data['PARENTGROUP'].value_counts(sort=True) print (c4) print('\nWhere do your children live? - Recoding and grouping the data - percentages') p4 = data['PARENTGROUP'].value_counts(sort=True, normalize=True) print (p4) plot3=seaborn.countplot(x="PARENTGROUP", data=data) plt.xlabel('The children live with...') plt.title('Where do your children live?') plt.show(plot3) print('\n=============================================================================') # Looking to see which group of children mostly live with both parents # recode the parentgroup # only value 1 (living with both parents) is needed recode3 = {1: 1, 2: 0, 3: 0, 4: 0, 5: 0} data['BOTHPARENTS']= data['W1_P19'].map(recode3) plot4=seaborn.catplot(x='W1_P17A', y='BOTHPARENTS', data=data, kind="bar", ci=None) plt.xlabel('Number of children') plt.ylabel('Children live with both parents') plt.title('Which children group live mostly with their both parents') plt.show(plot4)

[data-management-konny86]

Summary description of the following solution for my 3rd assignment:

The following frequency distributions show the original data which comes from the dataset and, in the next step, the recoded and grouped data. Furthermore the missing data is shown in the original data records as ‘NaN’

The data records of W1_P19 have many different characteristics for the question “Do children live with one parent, father or mother”. For example the mother or father lives alone or with a new partner together with their children etc. The important question for me is here: Do the children live with both parents, live with one parent (father or mother), or do not live with their parents at all (live with relatives). By recoding and grouping the original data it is possible to answer the question.

Analogously, this also applies to the data records of W1_W17A ‘How many children do you have’.

[data-management-konny86]

Making Data Management Decisions 

Here is my solution of the 3rd assignment:

My program:

#!/usr/bin/env python3 # -*- coding: utf-8 -*- """ Created on Wed Jun  9 14:17:38 2021 @author: konny My first Python program with the selected data set in relation to my research question """ import pandas import numpy data = pandas.read_csv('ool_pds.csv', low_memory=False) #upper-case all DataFrame column names data.columns = map(str.upper, data.columns) #bug fix for display formats to avoid run time errors pandas.set_option('display.float_format', lambda x:'%f'%x) print('Content of the ool_pds.csv dataset:') print('rows:', len(data))            # Number of observations (rows) Anzahl der Zeilen print('columns:', len(data.columns))   # Number of variables (colums)  Anzahl der Spalten print('\n=============================================================================') # setting variables I will be working with to numeric (updated) data['W1_P17'] = pandas.to_numeric(data['W1_P17'])      #Do you have any biological or adopted children data['W1_P17A'] = pandas.to_numeric(data['W1_P17A'])    #How many children do you have data['W1_P19'] = pandas.to_numeric(data['W1_P19'])      #Where do your minor children (under 18) live # replace missing values to python missing (NaN) # nan: specifies missing values in python data['W1_P17']=data['W1_P17'].replace(-1, numpy.nan) data['W1_P17A']=data['W1_P17A'].replace(-1, numpy.nan) data['W1_P19']=data['W1_P19'].replace(-1, numpy.nan) #counts and percentages (i.e. frequency distributions) for each variable print('\ncounts for W1_P17 - \nDo you have any biological or adopted children, yes=1, no=2') c1 = data['W1_P17'].value_counts(sort=True, dropna=False) print (c1) print('\npercentages for W1_P17 - \nDo you have any biological or adopted children, yes=1, no=2') p1 = data['W1_P17'].value_counts(sort=True, normalize=True, dropna=False) print (p1) print('\n=============================================================================') print('\ncounts for W1_P17A - Original data\nHow many biological or adopted children do you have?') c2 = data['W1_P17A'].value_counts(sort=True, dropna=False) print (c2) print('\npercentages for W1_P17A - Original data\nHow many biological or adopted children do you have?') p2 = data['W1_P17A'].value_counts(sort=True, normalize=True, dropna=False) print (p2) # recode the childrengroup recode2 = {1: 1, 2: 1, 3: 1, 4: 2, 5: 2, 6: 2, 7: 3, 8: 3, 9: 3, 10: 3} data['CHILDRENGROUP']= data['W1_P17A'].map(recode2) # building a group print('\nHow many Children do they have? - Recoding and grouping the data\n1-3 children = 1\n4-6 children = 2\n7-10 children = 3') c5 = data['CHILDRENGROUP'].value_counts(sort=True) print (c5) print('\nHow many Children do they have? - Recoding and grouping the data - percentages \n1-3 children = 1\n4-6 children = 2\n7-10 children = 3') p5 = data['CHILDRENGROUP'].value_counts(sort=True, normalize=True) print (p5) print('\n=============================================================================') print('\ncounts of W1_P19 - Original data\nWhere do your minor children live? \nLive with both parents=1\nLive with father or mother (different expressions)=2-4\nLive with no parent=5') c3 = data['W1_P19'].value_counts(sort=True, dropna=False) print (c3) print('\npercentages of W1_P19 - - Original data\nWhere do your minor children live? \nLive with both parents=1\nLive with father or mother (different expressions)=2-4\nLive with no parent=5') p3 = data['W1_P19'].value_counts(sort=True, dropna=False, normalize=True) print (p3) # recode the parentgroup recode1 = {1: 1, 2: 2, 3: 2, 4: 2, 5: 3} data['PARENTGROUP']= data['W1_P19'].map(recode1) # building a group print('\nWhere do the children live? - Recoding and grouping the data\n1=live with both parents\n2=live with father or mother\n3=live with ralatives') c4 = data['PARENTGROUP'].value_counts(sort=True) print (c4)

print('\nWhere do the children live? - Recoding and grouping the data - percentages\n1=live with both parents\n2=live with father or mother\n3=live with ralatives') p4 = data['PARENTGROUP'].value_counts(sort=True, normalize=True) print (p4)

========================================================

Results/Output: 

========================================================

Content of the ool_pds.csv dataset: rows: 2294 columns: 436 ============================================================================= counts for W1_P17 - Do you have any biological or adopted children, yes=1, no=2 1.000000    1304 2.000000     949 NaN           41 Name: W1_P17, dtype: int64 percentages for W1_P17 - Do you have any biological or adopted children, yes=1, no=2 1.000000   0.568439 2.000000   0.413688 NaN        0.017873 Name: W1_P17, dtype: float64 ============================================================================= counts for W1_P17A - Original data How many biological or adopted children do you have? NaN          999 2.000000     481 1.000000     306 3.000000     267 4.000000     138 5.000000      51 6.000000      25 7.000000      11 10.000000     10 8.000000       5 9.000000       1 Name: W1_P17A, dtype: int64 percentages for W1_P17A - Original data How many biological or adopted children do you have? NaN         0.435484 2.000000    0.209677 1.000000    0.133391 3.000000    0.116391 4.000000    0.060157 5.000000    0.022232 6.000000    0.010898 7.000000    0.004795 10.000000   0.004359 8.000000    0.002180 9.000000    0.000436 Name: W1_P17A, dtype: float64 How many Children do they have? - Recoding and grouping the data 1-3 children = 1 4-6 children = 2 7-10 children = 3 1.000000    1054 2.000000     214 3.000000      27 Name: CHILDRENGROUP, dtype: int64 How many Children do they have? - Recoding and grouping the data - percentages 1-3 children = 1 4-6 children = 2 7-10 children = 3 1.000000   0.813900 2.000000   0.165251 3.000000   0.020849 Name: CHILDRENGROUP, dtype: float64 ============================================================================= counts of W1_P19 - Original data Where do your minor children live? Live with both parents=1 Live with father or mother (different characteristics)=2-4 Live with no parent=5 NaN         1529 1.000000     414 4.000000     206 2.000000      74 5.000000      36 3.000000      35 Name: W1_P19, dtype: int64 percentages of W1_P19 -  Original data Where do your minor children live? Live with both parents=1 Live with father or mother (different characteristics)=2-4 Live with no parent=5 NaN        0.666521 1.000000   0.180471 4.000000   0.089799 2.000000   0.032258 5.000000   0.015693 3.000000   0.015257 Name: W1_P19, dtype: float64 Where do the children live? - Recoding and grouping the data 1=live with both parents 2=live with father or mother 3=live with ralatives 1.000000    414 2.000000    315 3.000000     36 Name: PARENTGROUP, dtype: int64 Where do the children live? - Recoding and grouping the data - percentages 1=live with both parents 2=live with father or mother 3=live with ralatives 1.000000   0.541176 2.000000   0.411765 3.000000   0.047059 Name: PARENTGROUP, dtype: float64

[data-management-konny86]

Here is mysolution of the 2nd assignment:

1) My Program:

#!/usr/bin/env python3 # -*- coding: utf-8 -*- """ Created on Wed Jun  9 14:17:38 2021

@author: Konny

My first Python program with the selected data set in relation to my research question """

import pandas import numpy

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

#upper-case all DataFrame column names data.columns = map(str.upper, data.columns)

#bug fix for display formats to avoid run time errors pandas.set_option('display.float_format', lambda x:'%f'%x)

print('Zeilen:', len(data))            # Number of observations (rows) Anzahl der Zeilen print('Spalten:', len(data.columns))   # Number of variables (colums)  Anzahl der Spalten

# Alternative Option für die Anzeige von Beobachtungen oder Zelen in einem Dataframe #print('Index:', len(data.index))

# checking the format of your variables print(data['W1_P17'].dtype)

# setting variables I will be working with to numeric (updated) data['W1_P17'] = pandas.to_numeric(data['W1_P17'])      #Do you have any biological or adopted children data['W1_P17A'] = pandas.to_numeric(data['W1_P17A'])    #How many children do you have data['W1_P19'] = pandas.to_numeric(data['W1_P19'])      #Where do your minor children (under 18) live

#counts and percentages (i.e. frequency distributions) for each variable print('counts for W1_P17 - Do you have any biological or adopted children, yes=1') c1 = data['W1_P17'].value_counts(sort=True) print (c1)

print('percentages for W1_P17 - Do you have any biological or adopted children, yes=1') p1 = data['W1_P17'].value_counts(sort=True, normalize=True) print (p1)

print('counts for W1_P17A - How many biological or adopted children do you have?') c2 = data['W1_P17A'].value_counts(sort=True) print (c2)

print('percentages for W1_P17A - How many biological or adopted children do you have?') p2 = data['W1_P17A'].value_counts(sort=True, normalize=True) print (p2)

print('counts for W1_P17A - Where do your minor children live? Live with both parents=1') c3 = data['W1_P19'].value_counts(sort=True, dropna=False) print (c3)

print('percentages for W1_P17A - Where do your minor children live? Live with both parents=1') p3 = data['W1_P19'].value_counts(sort=True, dropna=False, normalize=True) print (p3)

==========================================================

2) Output:

Zeilen: 2294 Spalten: 436 float64 counts for W1_P17 - Do you have any biological or adopted children, yes=1 1.000000     1304 2.000000      949 -1.000000      40 Name: W1_P17, dtype: int64 percentages for W1_P17 - Do you have any biological or adopted children, yes=1 1.000000    0.568687 2.000000    0.413868 -1.000000   0.017444 Name: W1_P17, dtype: float64 counts for W1_P17A - How many biological or adopted children do you have? 2.000000     481 1.000000     306 3.000000     267 4.000000     138 5.000000      51 6.000000      25 7.000000      11 10.000000     10 -1.000000      9 8.000000       5 9.000000       1 Name: W1_P17A, dtype: int64 percentages for W1_P17A - How many biological or adopted children do you have? 2.000000    0.368865 1.000000    0.234663 3.000000    0.204755 4.000000    0.105828 5.000000    0.039110 6.000000    0.019172 7.000000    0.008436 10.000000   0.007669 -1.000000   0.006902 8.000000    0.003834 9.000000    0.000767 Name: W1_P17A, dtype: float64 counts for W1_P17A - Where do your minor children live? Live with both parents=1 NaN          990 -1.000000    539 1.000000     414 4.000000     206 2.000000      74 5.000000      36 3.000000      35 Name: W1_P19, dtype: int64 percentages for W1_P17A - Where do your minor children live? Live with both parents=1 NaN         0.431561 -1.000000   0.234961 1.000000    0.180471 4.000000    0.089799 2.000000    0.032258 5.000000    0.015693 3.000000    0.015257 Name: W1_P19, dtype: float64

=========================================================

3) Description:

The most interesting information for me was that only 56 % of the people have an own or adopted child.  I expected a bigger quantity.

The next interesting piece of information was that families mostly have 2 children (own and adopted) and there are even families with up to 10 children.

The third statement tells us that only 18.0% of the children live together with both parents. However there is a big amount of missing data (’NaN’: 43.2%) and people who refused an answer (’-1′: 23.5%). 

[data-management-konny86]

I decided the following topics:

Step 3 - 5 (Codebook)    Created a new Codebook and keep it with my documents

Step 6 (Used searchterms and results):

Step 6 (Used References):

Summary of findings from the links above:

Thanks guys ;-)

[data-management-konny86]

Hello everybody this is my new blog ;-)