Displaying Münster’s socioeconomic and educational situation geographically

We believe that the first step in correcting social inequality is identifying it. Thus, our goal was to investigate the possibility of a connection between social inequality and the availability of educational facilities in Münster, through geographically displayed data.

The resulting (heat)maps can be viewed at the following website:
https://techlabsgruppe132024-2025-ecuxxxpfegpswbxlwfq2kk.streamlit.app/

The code for the Project can be viewed at the following website:
https://github.com/Varadaux/Techlabs_Gruppe13_2024-2025

Introduction

Our group’s hypothesis was that the social inequality between the various city districts of Münster could be linked to a significant difference in the availability of nearby educational facilities. If a city district only/mainly has schools for lower educational levels in close-proximity, it could be that the inhabitants of these districts fare less well socioeconomically. Conversely, the inhabitants of city districts with only/mainly schools for higher educational levels in close-proximity could be faring much better socioeconomically in comparison. The reasoning behind this hypothesis is, that the lower educational degrees are less sought after and typically offer lower paying jobs in comparison to the higher educational degrees.

If this hypothesis were true, the perhaps the situation in the city districts which are faring less well socioeconomically could be improved by improving the educational opportunities in their proximity.

Methodology

To investigate our hypothesis, we first collected data on the schools in Münster, as well as socio-economic data for the city districts of Münster. The data was mainly sourced from the official data website of the city of Münster https://opendata.stadt-muenster.de. Missing data points, such as missing addresses or city districts, were supplemented via manual web searches. Regarding the socio-economical data, we focused on percentage metrics such as the unemployment in the various city districts. Conversely, we focused on absolute metrics for the educational data. This is because we only had data on the number of students in the various schools, but did not know in which districts the students lived, or the number of children living in Münster’s city districts. The data we collected was from the school year 2021/2022 because this was the most recent available data. To display the data as a heat map with the shapes of Münster’s city districts, we also gathered data on their shapes and geographical locations.

After collecting and supplementing our data, we organized it into Pandas DataFrames and GeoPandas GeoData frames according to the city districts. This allowed us to visualize the data in the form of bar charts and heatmaps. To create the heatmaps, we first created a basic map with Folium and then added the data and a legend as various layers.

Results

The main result of this project are two Heatmaps displaying the student numbers of different school-types and the social situations in the city districts of Münster respectively. Additionally, the project also produced a third map which displays the exact location of educational facilities, such as schools.

Analyzing the data reveals of the school year of 2021/2022 reveals the following insights: The city districts with the highest percentages of unemployment were the districts either possessing a school which offered the lowest educational degree, or which were located adjacent to districts with such a school. However, because the district of residency for the students of the schools which offered the lowest educational degree is unknown, it is unclear to what degree such school impacts the socioeconomic of a district, if at all. Further complicating the results, the districts with the highest unemployment rates also possessed schools which offered higher educational degrees or were adjacent to districts with such schools. Also, in these cases, the student number of the schools with the higher educational degrees were multitudes larger than that of the schools which offered the lowest educational degree. Lastly, some of the districts with the lowest unemployment rates only had elementary schools.

The main significant correlation which was found in the data was that between the percentage of children with a migration background in a city district and percentages of unemployment. For approximately two thirds of the 45 districts, dividing the percentage of children with a migration background by a factor of 10 yields approximate percentage of unemployment. However, it should be stated that the available data does not clearly suggest that the migration background of the parents is the reason for their unemployment.