I’m going to analyze the dataset of the Groningen dogs owners to see who are they - people who owns the dog in Groningen?

Let’s take a look at the data I collect and prepared, so maybe there are any correlations and relationships between different factors.

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import seaborn as sns

data = pd.read_csv("population_clean_data.csv") 

print(data.columns)
data.head()

Index(['buurtcode', 'buurtnaam', 'wijkcode', 'inwoners_per_km2',
       'perc_0_15_jaar', 'perc_15_25_jaar', 'perc_25_45_jaar',
       'perc_45_65_jaar', 'perc_65_en_ouder', 'perc_ongehuwd', 'perc_gehuwd',
       'perc_gescheid', 'perc_verweduwd', 'aantal_huishoudens',
       'perc_eenpersoonshuishoudens', 'perc_huishoudens_z_kind',
       'perc_huishoudens_met_kind', 'gemiddelde_huishoudsgrootte',
       'perc_west_migratie', 'perc_niet_west_migratie',
       'percentage_uit_marokko', 'perc_uit_antillen_en_aruba',
       'percentage_uit_suriname', 'percentage_uit_turkije',
       'perc_overige_nietwestersemigratie', 'bedrijven_landbouw',
       'gemid_won_waarde', 'perc_eengezinswoning', 'perc_meergezinswoning',
       'perc_koopwoningen', 'perc_huurwoningen',
       'perc_huurwoning_van_corporaties', 'perc_huurwoning_van_verhuurder',
       'perc_werknemers', 'perc_zelfstandigen',
       'gemiddeld_inkomen_per_inwoner', 'perc_huishoud_laag_inkomen',
       'perc_huishoud_hoog_inkomen', 'personen_met_bijstandsuitkering',
       'gewelds_misdrijven', 'autos_totaal', 'autos_per_huishouden',
       'motortweewielers', 'bevolking_totaal', 'percentage_minimahuishoudens',
       'aantal_minimahuishoudens', 'gemiddeld_huishoudensinkomen',
       'diefstal_af_uit_woning', 'diefstal_auto', 'diefstal_fiets',
       'diefstal_af_uit_bedrijf', 'diefstal_af_uit_auto',
       'aangiften_van_vernieling', 'aangiften_van_mishandeling', 'dogs_amount',
       'politie_report_aantal'],
      dtype='object')

	buurtcode	buurtnaam	wijkcode	inwoners_per_km2	perc_0_15_jaar	perc_15_25_jaar	perc_25_45_jaar	perc_45_65_jaar	perc_65_en_ouder	perc_ongehuwd	...	gemiddeld_huishoudensinkomen	diefstal_af_uit_woning	diefstal_auto	diefstal_fiets	diefstal_af_uit_bedrijf	diefstal_af_uit_auto	aangiften_van_vernieling	aangiften_van_mishandeling	dogs_amount	politie_report_aantal
0	BU000000	Binnenstad-Noord	WK001400	11891.0	2.0	46.0	35.0	11.0	6.0	86.0	...	30.0	25.0	1.0	176.0	17.0	8.0	25.0	35.0	37	287.0
1	BU000001	Binnenstad-Zuid	WK001400	12011.0	2.0	47.0	33.0	11.0	6.0	87.0	...	30.7	42.0	3.0	306.0	28.0	30.0	96.0	196.0	51	701.0
2	BU000002	Binnenstad-Oost	WK001400	14979.0	3.0	40.0	36.0	14.0	7.0	84.0	...	27.8	17.0	1.0	56.0	2.0	3.0	16.0	8.0	48	103.0
3	BU000003	Binnenstad-West	WK001400	17507.0	2.0	38.0	40.0	11.0	9.0	84.0	...	29.7	9.0	0.0	13.0	2.0	4.0	8.0	5.0	21	41.0
4	BU000005	Hortusbuurt-Ebbingekwartier	WK001400	12509.0	5.0	41.0	30.0	15.0	9.0	82.0	...	32.1	20.0	2.0	68.0	16.0	9.0	24.0	13.0	108	152.0

5 rows × 56 columns

First I’m going to get numeric columns and see the correlation matrix (but )

numeric_data = data.select_dtypes(include=[np.number])

corr_matrix = numeric_data.corr()
f, ax = plt.subplots(figsize = (20,20))
plt.title('Correlation of numeric factors', y = 1, size = 16)
sns.heatmap(corr_matrix, vmax = .8, annot_kws={'size': 8}, annot = True)

<AxesSubplot: title={'center': 'Correlation of numeric factors'}>

Looking at the picture of the correlation matrix (above) I can say that there are definately some correlations there, but because there are a lot of factors and poor readability of the matrix, it’s difficult to say anything spetial, so let’s take a closer look. We are interested in the dog’s owners, so I will take a dogs_amount column and show it’s correlation with other factors:

dogs_amount = corr_matrix[['dogs_amount']]
fig, ax = plt.subplots(figsize=(16, 16))
ax = sns.heatmap(dogs_amount, vmax = .8, annot_kws={'size': 10}, annot = True, linewidths=.5)

Now let’s get only strong correlated columns:

corr_values = [row for row in dogs_amount.items()][0][1]
columns = [column for column in numeric_data.columns if corr_values[column] > 0.6]

sns.pairplot(data = data,
             x_vars = ['dogs_amount'],
             y_vars = columns)

<seaborn.axisgrid.PairGrid at 0x13fbb8f90>

for column in columns:
    print(column + ": " + str(corr_values[column]))

perc_uit_antillen_en_aruba: 0.7000535645951397
personen_met_bijstandsuitkering: 0.722981338084339
autos_totaal: 0.8369100868143818
motortweewielers: 0.796743835950187
bevolking_totaal: 0.7475149616129847
aantal_minimahuishoudens: 0.709500199839632
aangiften_van_vernieling: 0.6056959022366318
dogs_amount: 1.0

As an answer to the question: “Who owns a dog in Groningen?” we can see from the charts above, most of the dog’s owners in Groningen have a car or two wheelers motorcycle, receive some welfare benefits and probably live in the area where they would experience destruction from other people more than in other places. Also I can see that probably people, who moved to the Netherlands from Netherlands Antilles and Aruba love dogs and have it at home more often them other people.